Supraphos: A supramolecular strategy to prepare bidentate ligands

Article abstract of DOI:10.1016/j.jorganchem.2005.02.026

Herein, we report a new strategy for the preparation of chelating bidentate ligands, which involves the mixing of two mondentate ligands functionalized with complementary binding sites. The assembly process is based on selective metal-ligand interactions

Full text of DOI:10.1016/j.jorganchem.2005.02.026

Journal of Organometallic Chemistry 690 (2005) 4505–4516
www.elsevier.com/locate/jorganchem
q
Supraphos: A supramolecular strategy to prepare bidentate ligands
*
Joost N.H. Reek , Marc Roder, P. Elsbeth Goudriaan, Paul C.J. Kamer,
¨
Piet W.N.M. van Leeuwen, Vincent F. Slagt
Van ꢀt Hoff Institute for Molecular Sciences, University of Amsterdam, Nieuwe Achtergracht 166, 1018 WV Amsterdam, The Netherlands
Received 28 September 2004; received in revised form 26 January 2005; accepted 26 January 2005
Available online 18 April 2005
Abstract
Herein, we report a new strategy for the preparation of chelating bidentate ligands, which involves the mixing of two mondentate
ligands functionalized with complementary binding sites. The assembly process is based on selective metal–ligand interactions
employing phosphite zinc(II) porphyrins 1–6 and the nitrogen-containing phosphorus ligands b–i (Scheme 1). Only 14 monodentate
ligands were utilized to generate a library of 48 palladium catalysts based on supraphos-type bidentate ligands. The characterization
of rhodium complexes based on representative Supramolecular bidentate ligands and the comparison of their performance in the
hydroformylation of styrene will be presented. The current library of catalysts was tested in the asymmetric palladium-catalyzed
alkylation of rac-1,3-diphenyl-2-propenyl acetate, which resulted in a large variety in the observed enantioselectivity for the different
catalysts. Importantly, small variations in the supraphos building blocks, lead to large differences in the enantioselectivity imposed
by the catalyst, the most selective catalyst producing 97% ee.
ꢀ 2005 Elsevier B.V. All rights reserved.
Keywords: Catalysis; Supramolecular chemistry; Combinatorial chemistry; Palladium; Allylic substitution; Hydroformylation
1. Introduction
chelating ligands. Methodologies that enable the synthe-
sis of sufficiently large libraries of bidentate phosphorus
containing ligands are scarce [3], and new strategies are
required to prepare diverse catalyst libraries that can be
used for high throughput experimentation (HTE) [4].
Traditionally, bidentate ligands are prepared by attach-
ing donor atoms to a ligand backbone. Herein we report
a supramolecular strategy to generate bidentate ligands
that involves the simple mixing of monodentate ligands
[5], which is particularly suitable for combinatorial ap-
proaches and HTE [4,6]. The strategy, schematically dis-
played in Fig. 1, is based on the utilization of ligands
functionalized with complementary binding sites. By
simply mixing the ligands in solution, chelating biden-
tate ligands spontaneously form by assembly. In this re-
port, we employ the commonly used axial coordination
of nitrogen donor ligands to zinc(II) porphyrin building
blocks as the binding motif. Breit and Seiche [7] recently
reported a similar strategy to form chelating bisphos-
Bidentate chelating ligands comprise an important
class of compounds for the construction of transition
metal catalysts [1]. However, compared to monodentate
ligands [2], the synthesis is generally more complex and
time consuming. This is especially inconvenient for reac-
tions that need intensive ligand-optimization, and as a
result substantial effort has recently been devoted to
the re-investigation of the potential of monodentate li-
gands. Although some very high enantioselectivities
have been observed with monodentate ligands, there will
always remain classes of reactions that require bidentate
q
This paper was first reported on July 7, 2004 at the XIVth
International Conference on Homogeneous Catalysts in Munich/
Germany.
*
Corresponding author. Fax: +31 20 525 6422.
E-mail address: reek@science.uva.nl (J.N.H. Reek).
0022-328X/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.02.026

4506
J.N.H. Reek et al. / Journal of Organometallic Chemistry 690 (2005) 4505–4516
Fig. 1. A schematic representation of the supramolecular strategy to form bidentate ligands via assembly. Two ligands (L1 and L2) with
complementary binding motifs (black and grey) will associate to form a bidentate chelating ligand.
phine ligands by the homo-assembly of two identical
monodenate phosphine ligands using a self-complemen-
tary hydrogen bond motif. Interestingly, the selectivities
observed for the hydroformylation of alkenes by the
chelating supramolecular ligand are comparable with
the best traditional covalent bidentate ligands. Takcas
et al. [8] prepared a library of chiral diphosphite ligands
based on phosphite functionalized chiral bisoxazoline li-
gands via the assembly of heteroleptic zinc complexes.
The library of palladium catalysts was explored for
allylic alkylation reaction and interestingly enantiomeric
excesses between 20% and 99% were observed.
ligand 1 Æ b is formed in situ. UV–vis titrations [13,14]
in toluene and NMR-spectroscopy experiments show
that the pyridyl moiety of b selectively coordinates to
zinc(II) porphyrin 1 with a binding constant of
K(1 Æ b) = 1.6 · 103 Mꢀ1. The binding constant of meta-
phosphite zinc(II) porphyrin 2 with b is slightly higher,
K(2 Æ b) = 2.1 · 103 Mꢀ1. These results indicate that via
selective pyridine–zinc interactions, two monodentate
phosphorus ligands form a bidentate ligand assembly.
Previously, we reported that the binding of the nitrogen
donor atoms to the zinc(II) porphyrin is very selective as
the phosphorus donor atoms do not coordinate to the
zinc metal center and therefore they are available for
transition metal coordination [9]. In order to investigate
the coordination properties of this new class of ligands,
their rhodium complexes were prepared and studied
with high-pressure NMR spectroscopy (Table 1). The
NMR-spectroscopy experiments performed in toluene-
d8 under 20 bars of H2/CO, show that these ligand sys-
tems do indeed coordinate to transition metals in a
bidentate fashion.
ortho-Phosphite zinc(II) porphyrin 1 in the presence
of a rhodium precursor, [Rh(acac)(CO)2], forms a
monoligated rhodium phosphite complex 7, as is the
case for other bulky phosphite ligands [15]. The one-
to-one mixture of 3-pyridyldiphenylphosphine b and 1
resulted in the generation of the assembled bidentate
phosphine–phosphite ligand 1 Æ b, which results in the
formation of a new bidentate phosphine–phosphite rho-
dium complex 8 (Scheme 2).
2. Results and discussion
We have shown previously [9] that zinc(II) porphy-
rins and nitrogen donor ligands offer a complementary
motif that is suitable for the formation of bidentate li-
gands, since the binding is sufficiently strong and selec-
tive, even under conditions relevant to catalysis. For
the current work we prepared six phosphite-functional-
ized porphyrins (1–6) that in combination with eight
monodentate phosphorus ligands (b–i) generates a li-
brary of 48 chelating ligands formed by assembly
(Scheme 1). The synthesis of the phosphite–porphyrins
is straightforward; it involves a reaction between
mono-hydroxy-zinc(II) TPP [10] (TPP = tetraphenyl
porphyrin) and a phosphorochloridite [11]. The phos-
phite products were readily isolated after a purification
over a short alumina column. Monodentate ligands a–i
were commercially available or prepared via known or
standard procedures (see Section 4) [12].
The rhodium–phosphorus coupling constants indi-
cate that the assembled bidentate phosphine–phosphite
ligand 1 Æ b coordinates in an equatorial–equatorial (ee)
fashion to the rhodium metal center (Figs. 2 and 3).
As was evident from the NMR spectra ortho-phosphite
zinc(II) porphyrin 1 in the presence of a stoichiometric
amount of triphenylphosphine a, yielded a mixture of
two different rhodium complexes, respectively phos-
phine–phosphite complex 9 and a small amount of bis-
phosphine complex 10. The coordination geometry of
2.1. The assembly of phosphine–phosphite ligands and the
formation of metal complexes
By simply mixing two monodentate ligands, i.e.,
ortho-phosphite zinc(II) porphyrin
pyridyldiphenylphosphine b, an assembled bidentate
1
and meta-

J.N.H. Reek et al. / Journal of Organometallic Chemistry 690 (2005) 4505–4516
4507
Scheme 1. Building blocks used to make bidentate supraphos ligands via assembly.
Table 1
Selected 1H and 31P NMR data of high-pressure NMR-spectroscopy of various rhodium complexes at T = 25 ꢁC
Complex
d(31P)-phosphine (ppm)
d(31P)-phosphite (ppm)
JRh–P (Hz)
JRh–P03 (Hz)
JP–PO3 (Hz)
d(1H)–Rh–H (ppm)
c
c
c
c
[Rh(1)(acac)(CO)]a (7)
[HRh(1 Æ b)(CO)2] (8)
[HRh(1)(a)(CO)2] (9)
[HRh(a)2(CO)2] (10)
[HRh(2)2(CO)2] (11)
[HRh(2 Æ b)(CO)2] (12)
a
127.8
144.3
179.8
300
265
29.4
30.1
143
b
153
b
ꢀ10.6
ꢀ9.8
ꢀ9.1
ꢀ9.3
ꢀ10.8
b
c
c
c
c
b
37.9
c
140
c
150.4
148.6
243
b
b
31.4
Data of the [Rh(acac)(CO)(1)] complex; due to its low solubility no satisfactory NMR-data were obtained for the HRh(CO)3(1) complex.
b
c
Broad signals were observed.
Not present.
phosphine–phosphite complex 9 was found to be a mix-
complex (12). The interaction of the nitrogen donor
atom with the zinc(II) porphyrin is crucial for the for-
mation of this complex, since a one-to-one mixture of
2 and triphenylphosphine a in the presence of rhodium
does not yield a phosphine–phosphite complex, but only
the bisphosphine rhodium complex 10. These assembled
bidentate ligands chelate to a transition metal in a sim-
ilar fashion as their covalently linked analogues.
To obtain further proof that these assemblies indeed
exhibit chelation towards transition metals UV–vis titra-
tions (in dichloromethane) were performed to determine
binding constants of the pyridine–zinc(II) porphyrin
complex in the presence and absence of one equivalent
of rhodium phosphine complexes. In the presence of
HRh(PPh3)3CO, the first triphenylphosphine readily
exchanges with the phosphite-ligand (1) due to the
ture of equatorial–equatorial (ee) and equatorial–apical
(ea) coordination modes which are in fast exchange on
the NMR time-scale, leading to broader signals [16].
Interestingly, the addition of one equivalent of 3-pyr-
idyldiphenylphosphine b to this mixture results in pyri-
dine coordination to the zinc(II) porphyrin template of
1, and consequently the predominant rhodium species
in solution is that based on phosphine–phosphite biden-
tate 1 Æ b (Fig. 3).
High-pressure NMR spectra revealed that meta-
phosphite zinc(II) porphyrin 2 in the presence of a rho-
dium precursor, [Rh(acac)(CO)2], yields a bis-phosphite
rhodium complex (11). The addition of 3-pyridyldiphe-
nylphosphine b to this mixture also results in the forma-
tion of a bidentate phosphine–phosphite rhodium

4508
J.N.H. Reek et al. / Journal of Organometallic Chemistry 690 (2005) 4505–4516
Scheme 2. Assembly of rhodium complex 8 [HRh(1 Æ b)(CO)2] (PM3 optimized structure) consisting of phosphite zinc(II) porphyrin template 1 and
3-pyridyldiphenylphosphine b, which can also be prepared from phosphine–phosphite complex 9 ([HRh(1)(a)(CO)2]) (a = triprenylphosphine).
UV–vis titrations of ortho-phosphite zinc(II) porphyrin
1 and 3-pyridyldiphenylphosphine b in dichloromethane
indicate that the presence of a rhodium complex,
[HRh(a)3(CO)], increased the association constant by a
factor of seventeen (K = 3.8 · 103 Mꢀ1 to K =
64.5·103 Mꢀ1) (Table 2), which corresponds to a 7 kJ/
Fig. 2. Coordination modes of [HRh(P)2(CO)2]: equatorial–equatorial
(ee) and equatorial–apical (ea), P = phosphorus ligand.
mol difference in binding free energies (20.5 and
27.5 kJ/mol, respectively). For meta-phosphite zinc(II)
porphyrin 2 and b a similar increase was observed.
stronger binding properties of the phosphite. Phosphine
2.2. Catalysis
ligand b competes with triphenylphosphine to coordi-
nate to the complex, and in comparison to triphenyl-
phosphine, ligand b will bind more strongly to the
metal center on account of the chelate effect. The chelate
energy is therefore expressed as an increase in binding
energy, and if the chelate effect is negligible we expect
to find similar binding constants for both experiments.
2.2.1. Rhodium catalyzed hydroformylation
After determining that bidentate chelating ligands
can form by self-assembly, their behavior in catalysis
was investigated. For this purpose we studied a selection
of the assemblies for the rhodium-catalyzed hydro-
Fig. 3. High-pressure 31P NMR spectrum of the assembled bidentate complex 8 [HRh(1 Æ b)(CO)2] in toluene-d8.

J.N.H. Reek et al. / Journal of Organometallic Chemistry 690 (2005) 4505–4516
4509
Table 2
tate 1Æb shows a similar increase in selectivity for the
branched product, but the activity of the catalyst is high-
er then that of complex 9. In the mixture of the three
monodentate ligands, triphenylphosphine a, b and 1
(in equal amounts), the chelating assembled bidentate li-
gand should predominantly coordinate to the rhodium
center and indeed similar catalytic results were observed
in the presence or absence of a. This demonstrates that
the chelate species forms under reaction conditions.
The meta-phosphite zinc (II) porphyrin 2 used as
monodentate ligand results in the formation of bis-
phosphite rhodium complexes (11), which generates rea-
sonable activity and low selectivity in the
hydroformylation of styrene (TOF = 1060 and b/l =
3.6). Ligand 2 in the presence of triphenylphosphine a
yields predominately [HRh(CO)2(a)2] 10, as was evident
from the high-pressure NMR spectra. Consequently, a
similar activity and selectivity is observed for the hydro-
formylation of styrene as in the absence of 2. In con-
trast, upon employing the assembled bidentate ligand
2 Æ b, the activity and selectivity changes producing com-
parable results as observed for 1 Æ b. The addition of a to
the solution of bidentate assembled ligand 2 Æ b does not
alter the catalytic properties of the assembled rhodium
complex based on 2 Æ b, again showing that the chelate
effect is maintained under catalytic conditions.
Binding constants of phosphite zinc(II) porphyrin templates and meta-
pyridyldiphenylphosphine in dichloromethane as determined with
UV–vis spectroscopy
Complex
Binding constant (K-value, Mꢀ1
)
1 Æ b
3.8 · 103
64.5 · 103
6.9 · 103
43.0 · 103
1 Æ b + [HRh(a)3(CO)]
2 Æ b
2 Æ b + [HRh(a)3(CO)]
Scheme 3. The rhodium catalyzed hydroformylation of styrene.
formylation [17] of styrene (Scheme 3) and the complete
library for the asymmetric allylic substitution.
The rhodium catalyzed hydroformylation of styrene
was performed in toluene at 80 ꢁC under 20 bars of
H2/CO (1/1). The results displayed in Table 3 show that
key features like selectivity and activity vary consider-
ably by using different building blocks. ortho-Phosphite
zinc(II) porphyrin 1 yields a rhodium catalyst with high
activity and low regioselectivity for the branched alde-
hyde in the hydroformylation of styrene (TOF = 2900
and b/l = 2.6). According to the complexation studies,
this very bulky phosphite forms monoligated complexes,
explaining the high activity. Mixing triphenylphosphine
a and 1 yields a catalyst with a low activity and a higher
b:l ratio. The [HRh(CO)2(1)(a)] 9, observed in the high-
pressure NMR spectroscopy experiments, apparently
undergoes catalysis with a low turn over frequency for
hydroformylation. Interestingly, the assembled biden-
2.2.2. Palladium catalyzed allylic alkylation
The supramolecular ligand library based on mono-
dentate phosphorus ligands a–i, and 1–6 was tested in
the palladium catalyzed asymmetric allylic alkylation
[18] of rac-1,3-diphenyl-2-propenyl acetate using di-
methyl malonate as the nucleophile (Fig. 4). The matrix
of ligands gave rise to a catalyst library of 60 members,
of which 48 were based on supramolecular bidentate li-
gands, simply by mixing stock-solutions of the 16
monodentate ligands. In Fig. 4, the enantiomeric excess
(ee) of the catalysts based on the various ligand assem-
blies is displayed and this figure clearly demonstrates a
large variety in selectivity induced by the various mem-
bers of the ligand library. For the tested catalysts, the
enantiomeric excess ranged from 87% (S) to 86% (R)
(Table 4). Importantly, the ee strongly depends on the
ligand assembly, and both of the components are cru-
cially important.
Unexpectedly, upon using (S)-ortho 3 as the monod-
entate ligand, we observed a high ee (85%, S); for this
reaction only a few monodentate ligands are known to
generate products with high eeꢀs [2c,19]. In contrast,
the palladium catalyst based on (S)-meta 5 resulted in
a much lower selectivity (18% ee, S). To explain the
large difference in performance, we compared the struc-
ture of the complexes based on these ligands with those
based on the chiral MeO–MOP (2-diphenylphosphino-
20-methoxy-1,10-binaphthyl) monodentate ligands of
Hayashi, which also give high eeꢀs. The solid state struc-
Table 3
Rhodium catalyzed hydroformylation of styrene using various supra-
phos ligandsa
Ligandb
TOFc
b/ld
Branched (%)
1
1 + a
2900
135
2.6
11.0
10.4
10.5
3.6
72
92
91
91
78
83
90
90
83
83
1 Æ b
1 Æ b + a
2
2 + a
398
375
1060
1730
449
5.0
9.3
2 Æ b
2 Æ b + a
461
9.4
5.0
a
b
a
1750
1850
5.1
[Rh(acac)(CO)2] = 0.83 mmol/l, styrene/rhodium = 12000, pres-
sure = 20 bar (CO/H2 = 1/1), T = 80 ꢁC.
Phosphite/rhodium = 25, phosphite/phosphine = 1.
b
c
TOF = average turn over frequency = (mol aldehyde) (mol Rh)ꢀ1
h
ꢀ1, the reaction was stopped after 1 h.
d
b/l = branched/linear.

4510
J.N.H. Reek et al. / Journal of Organometallic Chemistry 690 (2005) 4505–4516
Fig. 4. Enantioselectivity results obtained for asymmetric palladium-catalyzed allylic alkylation of 1,3-diphenylallyl acetate and dimethyl malonate
from a catalyst library of self-assembled bidentate ligands based on 1–6 and a–i.
Table 4
Allylic alkylation of 1,3-diphenylallyl acetate using different palladium
rin ring of this ligand is also in close proximity to the
palladium center, and shielding one coordination site
catalyst assemblies at T = ꢀ20 ꢁC: ee and product configuration are
of the complex (Fig. 5). The calculated structure is very
givena
similar to the solid state structure of the palladium allyl
Ligand
1
2
3
4
5
6
chloride complex with the MeO–MOP ligand that was
determined by X-ray diffraction. In addition, from the
calculated structure of the complex it is evident that only
one bulky ligand can coordinate to the metal. It is inter-
esting to note that the second ligand coordinated this
palladium center, which in this complex is a chloride,
0
0
87 (S)
0
86 (R)
0
20 (S)
0
48 (R)
0
a
b
c
d
e
f
0
0
0
0
47 (R)
0
47 (S)
0
59 (S)
0
17 (R)
0
0
0
0
0
40 (S)
23 (S)
4 (S)
45 (S)
42 (R)
11 (S)
38 (R)
23 (R)
3 (R)
45 (R)
43 (S)
10 (R)
11 (R)
10 (S)
22 (R)
37 (S)
43 (S)
27 (S)
33 (S)
5 (S)
0
0
0
˚
0
0
is only 3.8 A from the zinc atom of the porphyrin ring.
g
h
i
40 (S)
40 (R)
28 (S)
36 (S)
34 (R)
28 (S)
40 (S)
39 (R)
28 (S)
This suggests a pre-organization that facilitates the for-
mation of the supramolecular chelating bidentate with
for example b. In the case of (S)-meta 5 phosphite li-
gand, the porphyrin ring is located substantially further
away from the palladium center, and this ligand is not
sufficiently bulky to prevent the second ligand from
coordinating to the metal center, providing a plausible
explanation for the difference in selectivity obtained with
ligands 3 and 5.
Interestingly, with bidentate ligand assembly 3 Æ b,
the opposite product in lower enantiomeric excess
(47% ee, R) was obtained in contrast to monodentate
3. In the bidentate phosphine–phosphite system, the at-
tack of the nucleophile occurs trans to the phosphine
a
[[Pd(allyl)Cl]2 ] = 0.1 mmol/l, [1–6] = 0.6 mmol/l, [a–i] = 0.6 mmol/l,
the reaction was stopped after 24 h, T = 25 ꢁC, complete conversion
was obtained in all cases.
ture of the palladium allyl chloride complex supported
by the MeO–MOP ligand shows that one of the naph-
thalene moieties is located close to the palladium center,
which is considered to be crucial for the creation of a
chiral environment [2c,20]. We have calculated the struc-
ture of the palladium allyl chloride complex with ligand
3 using PM3 calculations and we found that the porphy-

J.N.H. Reek et al. / Journal of Organometallic Chemistry 690 (2005) 4505–4516
4511
Fig. 5. PM3 optimized molecular structures of 3Pd(allyl)Cl (left) and Pd(allyl chloride) complex with the MeO–MOP ligand (right). The palladium
center, shown as CPK-model, is clearly located above the porphyrin ring in a similar fashion as previously observed for MOP ligands.
Table 5
Allylic alkylation of 1,3-diphenylallyl acetate using different palladium
ligand [21,22], whereas in 3Pd(allyl)Cl the attack will
be trans to the phosphite providing an explanation
for the reversal in product selectivity. Bidentate ligand
catalyst assemblies at T = ꢀ20 ꢁCa
Ligand
Conversion (%)
eeb (%)
assembly 5 Æ b produced a higher ee of the same prod-
uct as 5 (59% ee, S), indicating that the formation of
the chelate ligand with b has a completely different ef-
fect. Remarkably, bulky phosphite 6 generates prefer-
entially the opposite enantiomeric product to that
observed for the same reaction using ligand 3 (48%
ee R). However, in the presence of co-ligand (b, d, e,
f, g, h), which forms a chelating bidentate ligand, it
does result in the same product as the analogous
assemblies based on 3. As expected, nonchiral 1 and
2 produced products with eeꢀs (up to 40%) only in com-
bination with chiral building blocks g–i. It is interesting
to note that all reactions in the presence of triphenyl-
phosphine (a) resulted in racemic product, indicating
that the catalysis is dominated by palladium triphenyl-
phosphine species, illustrating the importance of the
presence of the nitrogen–zinc interaction to obtain che-
lating ligands. The introduction of two chiral centers in
the supraphos-ligand, i.e., the combinations 3g, 4g, 5g,
6g, 3h, 4h, 5h, 6h, 3i, 4i, 5i, 6i, does not result in signif-
icantly higher enantioselectivities as those obtained
with only one chiral center.
From the initial screening experiments some ‘‘hits’’
were identified, and these catalyst systems were subse-
quently studied at ꢀ20 ꢁC (Table 5). Under these condi-
tions the palladium catalyst supported by 3 produced
products with very high enantioselectivity (97% (S)).
The catalyst based on ligand assembly 3 Æ b gave under
these conditions an enantiomeric excess of 60% (R).
Interestingly, 3 Æ b proved to yield a more active catalyst
than that based on 3, since the yield after 24 h was 100%
compared to 56%. A change in the building blocks of the
supraphos ligand not only affects the selectivity but also
the activity of the catalyst. At ꢀ20 ꢁC the catalyst based
on the bidentate assembly 3 Æ d resulted in the formation
3
56
100
100
100
54
97 (S)
60 (R)
0
3 Æ b
3 Æ c
3 Æ d
4
44 (S)
96 (R)
60 (S)
42 (S)
70 (S)
4 Æ b
5
100
73
5 Æ b
a
40
[[Pd(allyl)Cl]2] = 0.100 mmol/l, [phosphite] = 0.6 mmol/l, [phos-
phine] = 0.6 mmol/l, the reaction was stopped after 43 h, T = ꢀ20 ꢁC.
b
ee = enantiomeric excess.
of the S-product with an enantiomeric excess of 44%. In
analogy to previous reports of covalently linked phos-
phine–phosphite ligands [22], a small difference in the
length of the bridge between the phosphine and phos-
phite resulted in a large difference in enantioselectivity.
In fact, the difference between supraphos ligands 3 Æ b
and 3 Æ d is only one methylene unit and for the palla-
dium catalyzed allylic substitution of 1,3-diphenylallyl
acetate this difference results in the formation of the
opposite enantiomers (3 Æ b 60% (R) and 3 Æ d 44% (S)).
At ꢀ20 ꢁC ligand 5 resulted in a catalyst that produces
42% ee (S), whereas the bidentate assembly 5 Æ b gener-
ates the poduct in 70% ee (S). In this case, the metal
complex of assembly 5 Æ b is a slightly slower catalyst
(40% yield compared to 73% for 5). Importantly, small
changes in the assembly of the phosphite zinc(II) por-
phyrin and the phosphorus ligands a–i have a large
influence on the enantioselectivity as well as the activity
of the catalyst system. The diversity of the relatively
small supramolecular catalyst library is already suffi-
cient to yield catalysts with selectivities ranging from
97% (S) to 60% (R).

4512
J.N.H. Reek et al. / Journal of Organometallic Chemistry 690 (2005) 4505–4516
identified and these promising catalyst systems gener-
ated products with high enantioselectivities up to
97% at ꢀ20 ꢁC. In addition, it was found that small
changes in the structure of the components of the
assembled catalyst affected the enantioselectivity enor-
mously. For example, the assembly of 3-pyridyldiphe-
nylphosphine b and phosphite zinc(II) porphyrin 3,
preferentially leads to the formation of the R-product
(60% ee), whereas the catalyst based on 3 gives the S-
product in 97% ee. Importantly, the supraphos ligand
b Æ 3 is tolerant towards excess of the nonchiral com-
ponent b and triphenylphosphine, clearly demonstrat-
ing the effect of the chelate. In the control
experiments with monodentate ligands the presence
of triphenylphosphine resulted in racemic products.
Some of the supraphos ligands were used in rho-
dium-catalyzed hydroformylation of styrene indicating
that the class of compounds can be used for various
transformations. So far we have only explored phos-
phite zinc(II) porphyrin 1–6 and the nitrogen donor li-
gands a–i, but many other (a)chiral building blocks
can be employed for the ligand assemblies. We are
currently preparing larger libraries and we are explor-
ing other noncovalent interactions such as hydrogen
bonds for their applicability to this new concept.
The advantage of the approach has been clearly dem-
onstrated as the size of the ligand library grows expo-
nentially with the number of building blocks.
Fig. 6. Palladium catalyzed allylic alkylation at various ratios of 3/b
(3/palladium = 1) at 25 ꢁC.
The stability of the chelating supraphos ligand was
investigated in the presence of an excess of nonchiral
component b, by monitoring the ee of the product for
palladium catalyzed allylic substitution of 1,3-diphe-
nyl-allyl acetate at various 3/b ratios. Its clear from
Fig. 6 that the enantioselectivity remains around 50%
for all b/3 ratios between 1 and 30, which in the latter
case means that 90 equivalents of b with respect to pal-
ladium is present in the solution. This shows that under
all these conditions the catalysis is dominated by the pal-
ladium complex of supraphos ligand 3 Æ b. For compar-
ison, in all the experiments performed with
triphenylphosphine (a, Table 4) we obtained racemic
products, indicating that in these experiments the catal-
ysis was dominated by triphenylphosphine based palla-
dium complexes. Control experiments with supraphos
3 Æ b as the ligand in the presence and absence of triphe-
nylphosphine produced identical results, also indicating
that the chelate effect of the supraphos 3 Æ b ligand is suf-
ficient to form exclusively 3 Æ b ligated palladium com-
plex under these conditions.
4. Experimental
4.1. General procedures
Unless stated otherwise, reactions were carried out
under an atmosphere of argon using standard Schlenk
techniques. THF, hexane and diethyl ether were distilled
from sodium benzophenone ketyl, CH2Cl2, isopropanol
and methanol were distilled from CaH2 and toluene was
distilled from sodium under nitrogen. NMR spectra (1H,
31P{1H} and 13C{1H}) were measured on a Bruker DRX
300 MHz and Varian Mercury 300 MHz; CDCl3 was
used as the solvent, if not further specified. Mass spectra
were recorded on a JEOL JMS SX/SX102A four sector
mass spectrometer; for FAB-MS, 3-nitrobenzyl alcohol
was used as the matrix. UV–vis spectroscopy experi-
ments were performed on a HP 8453 UV/Visible System.
Elemental analyses were obtained on an Elementar Var-
io EL apparatus. Gas chromatographic analyses were
run on an Interscience HR GC Mega 2 apparatus
(split/splitless injector, J&W Scientific, DB-l J&W 30 m
column, film thickness 3.0 lm, carrier gas 70 kPa He,
FID Detector) equipped with a Hewlett packard Data
system (Chrom-card). Chiral HPLC analyses were car-
3. Conclusion
We present here an alternative strategy for the
covalent synthesis of bidentate ligands, which consti-
tutes the assembly of two monodentate ligands with
complementary binding motifs, and the new class of
ligands we termed supraphos. In the current contribu-
tion the assembly processes are based on selective
metal–ligand interactions, i.e., the axial nitrogen coor-
dination to zinc(II) porphyrins, and we employed
phosphorus based mondentate ligands. For this pur-
pose monodentate phosphite zinc(II) porphyrin ligands
1–6 and phosphorus ligands a–i have been prepared.
Upon mixing these monodentate ligands self-assem-
bled bidentate ligands are formed instantaneously, as
proven by NMR and UV–vis spectroscopy. Based on
only 6 + 8 monodentate ligands, a catalyst library con-
sisting of 48 bidentate assembled ligands has been pre-
pared and was successfully employed in the palladium
catalyzed asymmetric allylic alkylation. Some hits were
ried out using
(O.46 · 25 cm).
a
Daicel Chiralcel-OD column

J.N.H. Reek et al. / Journal of Organometallic Chemistry 690 (2005) 4505–4516
4513
4.2. Materials
143.04, 140.42, 137.02, 136.09, 134.67, 133.07, 132.23,
130.68, 129.12, 128.45, 127.75, 126.80, 125.54, 125.07,
124.60, 122.56, 122.39, 122.13, 121.41, 119.73, 35.68,
34.85, 31.70, 31.47; HRMS (FAB+): m/z Calc. for
With exception of the compounds given below, all re-
agents were purchased from commercial suppliers and
used without further purification. Diisopropylethyl-
amine (dipea) and triethylamine were distilled from
CaH2 under argon. The following compounds were syn-
thesized according to published procedures: phosphoro-
chloridites [11], pyridylphosphines b and c [12a,b],
phosphines d–f [12], hydroxyl porphyrins [10] and zin-
c(II) porphyrins were prepared according to the method
of Adler [23].
C72H68N4O3PZn
([MH+]):
1131.4321;
Obsd.:
1131.4329. Anal. Calc. for C72H67N4O3PZn: C, 76.35;
H, 5.96; N, 4.95. Found: C, 76.83; H, 6.35; N, 4.68%.
4.2.3. Synthesis of (S)(1,10-binaphthyl-2,20-diyl)
(5-(phenyl-2-yl)-10,15,20-tris(phenyl) zinc(II)
porphyrin) phosphite 3
5-(2-Hydroxyphenyl)-10,15,20-tris(phenyl)
zinc(II)
porphyrin (1.59 g, 2.30 mmol), azeotropically dried
with toluene (3 · 5 ml), and diisopropylethylamine
(4.0 ml, 23.0 mmol) were dissolved in THF (80 ml)
and the solution was cooled to ꢀ40 ꢁC. Freshly pre-
pared (S)-2,20-binaphtol phosphorochloridite (0.73 g,
2.09 mmol) was dissolved in THF (20 ml) and added
dropwise, stirring was continued for 15 min. The cool-
ing bath was removed and the solution was allowed
to warm to room temperature, stirring was continued
for 30 min. The reaction mixture was filtered and the
solvent evaporated. The crude product was purified
by flash column chromatography under Argon (basic
alumina; CH2Cl2) to remove the excess of hydroxyl-
porphyrin, giving 3 (0.887 g, 0.88 mmol, 42%) as a pur-
ple-red solid: 1H NMR (300 MHz): d 8.94 (d, 6H,
J = 4.5 Hz), 8.89 (d, 1H, J = 4.5 Hz), 8.84 (d, 1H,
J = 4.5 Hz), 8.23 (m, 6H), 8.13 (m, 2H), 7.77 (m,
11H), 7.66 (m, 2H), 7.55 (m, 1H), 7.16 (m, 1H), 6.91
(d, 1H, J = 8.7 Hz), 6.84 (m, 1H), 6.60 (d, 1H,
J = 8.1 Hz), 6.45 (m, 2H), 6.38 (m, 1H), 5.72 (d, 1H,
J = 8.7 Hz), 5.60 (d, 1H, J = 8.4 Hz); 31P NMR
(121.5 MHz): d 145.15; 13C-ATP (75.465 MHz): d
150.62, 150.50, 150.22, 146.43, 145.32, 143.25, 135.79,
134.71, 132.42, 132.12, 131.50–130.22, 128.61, 127.53–
126.63, 125.50, 125.21, 122.56, 121.85, 121.07; HRMS
(FAB+): m/z Calc. for C64H40N4O3PZn ([MH+]):
1007.2130; Obsd.: 1007.2144. Anal. Calc. for
C64H39N4O3PZn: C, 76.23; H, 3.90; N, 5.56. Found:
C, 76.08; H, 4.16; N, 5.42%.
4.2.1. Synthesis of (3,30-5,50-tetra-tert-butyl-1,10-
biphenyl-2,20-diyl)(5-(phenyl-2-yl)-10,15,20-tris
(phenyl) zinc(II) porphyrin) phosphite 1
5-(2-Hydroxyphenyl)-10,15,20-tris(phenyl) zinc(II)
porphyrin (1.50 g, 2.16 mmol), azeotropically dried with
toluene (3 · 5 ml), and diisopropylethylamine (3.76 ml,
21.6 mmol) were dissolved in THF (80 ml) and the solu-
tion was cooled to 0 ꢁC. Freshly prepared 3,30-5,50-tetra-
tert-butyl-1,10-biphenyl-2,20-diyl
phosphorochloridite
(0.93 g, 1.96 mmol) was dissolved in THF (20 ml) and
added dropwise, stirring was continued for 15 min.
The cooling bath was removed and the solution was al-
lowed to warm to room temperature, stirring was con-
tinued for 30 min. The reaction mixture was filtered
and the solvent evaporated. The crude product was puri-
fied by flash column chromatography under Argon
(basic alumina; CH2C12) to remove the excess of hydro-
xyl-porphyrin, yielding 1 (1.02 g, 0.90 mmol, 46%) as a
1
purple-red solid: H NMR (300 MHz): d 8.90 (m, 8H),
8.22 (m, 6H), 8.09 (m, 1H), 7.76 (m, 9H), 7.64 (m, 1H)
7.59 (m, 1H), 7.47 (m, 1H), 7.19 (d, 2H, J = 4.2 Hz),
6.93 (d, 2H, J = 4.2 Hz), 1.21 (s, 18H), 0.93 (s, 18H);
31P NMR (121.5 MHz): d 133.4; 13C-ATP (75.465
MHz): d 150.67, 150.41, 150.32, 146.63, 145.53, 143.35,
135.89, 134.71, 132.46, 132.46, 132.05, 129.27, 127.61,
126.96, 126.72, 126.68, 122.56, 121.85, 121.07, 35.24,
31.62, 30.95; HRMS (FAB+): m/z Calc. for C72H68-
N4O3PZn ([MH+]): 1131.4320; Obsd.: 1131.4314. Anal.
Calc. for C72H67N4O3PZn: C, 76.35; H, 5.96; N, 4.95.
Found: C, 76.26; H, 6. 15; N, 5.06%.
4.2.4. Synthesis of (R)(1,10-binaphthyl-2,20-diyl)
(5-(phenyl-2-yl)-10,15,20-tris(phenyl)
zinc(II)porphyrin)phosphite 4
This compound was prepared as described for 3,
using freshly prepared (R)-2,20-binaphtol phosphoro-
1
chloridite. Yield (46%) as a purple-red solid: H NMR
(300 MHz): d 8.94 (d, 6H, J = 4.5 Hz), 8.89 (d, 1H,
J = 4.5 Hz), 8.84 (d, 1H, J = 4.5 Hz), 8.22 (m, 6H),
8.14 (m, 2H), 7.78 (m, 11H), 7.64 (m, 2H), 7.55
(m, 1H), 7.16 (m, 1H), 6.91 (d, 1H, J = 8.4 Hz), 6.84
(m, 1H), 6.60 (d, IH, J = 8.1 Hz), 6.45 (m, 2H), 6.38
(m, 1H), 5.71 (d, 1H, J = 8.7 Hz), 5.59 (d, 1H,
J = 8.4 Hz); 31P NMR (121.5 MHz): d 145.13; 13C-
ATP (75.465 MHz): d 150.62, 150.50, 150.22, 146.43,
4.2.2. Synthesis of (3,30-5,50-tetra-tert-butyl-1,10-
biphenyl-2,20-diyl)(5-(phenyl-3-yl)-10,15,20-tris
(phenyl) zinc(II) porphyrin) phosphite 2
This compound was prepared as described for 1,
using 5-(3-hydrophenyl)-10,15,20-tris(phenyl) zinc(II)
1
porphyrin. Yield (47%) as a purple-red solid: H NMR
(300 MHz): d 8.96–8.94 (m, 8H), 8.25–8.20 (m, 6H),
8.01 (s, 1H), 7.95 (d, 1H, J = 8.7 Hz), 7.82–7.77 (m,
9H), 7.67–7.62 (m, 1H), 7.49–7.45 (m, 1H), 7.42 (s,
2H), 7.23 (s, 2H), 1.50–1.47 (m, 18H), 1.36–1.32 (s,
18H); 31P NMR (121.5 MHz): d 139.02; 13C-ATP
(75.465 MHz): 150.77, 150.48, 150.22, 146.98, 144.66,

4514
J.N.H. Reek et al. / Journal of Organometallic Chemistry 690 (2005) 4505–4516
145.32, 143.25, 135.79, 134.71, 132.42, 132.12, 131.50–
130.22, 128.61, 127.53–126.63, 125.50, 125.21, 122.56,
121.85, 121.07; HRMS (FAB+): m/z Calc. for
(S)-2,20-binaphtol
phosphorochloridite
(5.3 g,
15.1 mmol) was dissolved in THF (20 ml) and added
dropwise. The cooling bath was removed and the solu-
tion was allowed to warm to room temperature, stirring
was continued for 1 h. The reaction mixture was filtered
and the solvent evaporated. A mixture of toluene/hex-
ane 1/3 (40 ml) was added to extract the product. After
filtration the solvent was removed in vacuo, giving g
(5.4 g, 13.2 mmol, 87%) as a white solid: 0H NMR
(300 MHz): d 8.54 (d, 1H, J = 2.4 Hz), 8.40 (d, 1H,
J = 4.2 Hz), 8.02 (d, 1H, J = 8.7 Hz), 7.95 (d, 1H,
J = 8.7 Hz), 7.92 (d, 2H, J = 8.7 Hz), 7.56-7.16 (m,
C64H40N4O3PZn
([MH+]):
1007.2130;
Obsd.:
1007.2136. Anal. Calc. for C64H39N4O3PZn: C, 76.23;
H, 3.90; N, 5.56. Found: C, 76.29; H, 4.14; N, 5.22%.
4.2.5. Synthesis of (S)(1,10-binaphthyl-2,20-diyl)
(5-(phenyl-3-yl)-10,15,20-tris(phenyl) zinc(II)
porphyrin) phosphite 5
This compound was prepared as described for 3,
using 5-(3-hydroxyphenyl)-10,15,20-tris(phenyl) zinc(II)
porphyrin and freshly prepared (S)-2,20-binaphtol
phosphorochloridite. Yield (39%) as a purple-red solid:
1H NMR (300 MHz): d 8.97–8.94 (m, 6H), 8.83–8.78
(m, 2H), 8.24-8.21 (m, 6H), 8.04–7.79 (m, 6H), 7.77 (s,
9H) 7.70–7.15 (m, 10H); 31P NMR (121.5 MHz): d
142.49; 13C-ATP (75.465 MHz): d 150.53 (C), 150.28
(C), 144.83 (C), 143.01 (C), 134.67 (CH), 132.39 (CH),
132.02 (CH), 131.49 (C), 131.05 (CH), 130.76 (CH),
130.23 (CH), 128.61 (CH), 127.83 (CH), 127.31 (CH),
127.26 (CH), 126.83 (CH), 126.83 (CH), 126.63 (CH),
125.50 (CH), 125.23 (CH), 121.99 (CH), 121.49 (C),
120.19 (CH); HRMS (FAB+): m/z Calc. for
C64H40O3PZn ([MH+]): 1007.2129; Obsd.: 1007.2109.
Anal. Calc. for C64H39N4O3Zn: C, 76.23; H, 3.90; N,
5.56. Found: C, 76.36; H, 4.08; N, 5.42%.
10H); 31P NMR (121.5 MHz):
d
143.05; 13C
(75.465 MHz): 148.88 (d, Jcp = 6.1 Hz), 147.52 (d,
Jcp = 4.8 Hz), 146.154 (d, Jcp = 1.4 Hz), 145.80 (s),
142.71 (d, Jcp = 7.3 Hz), 133.02 (s), 132.75 (s), 132.01
(s), 131.55 (s), 130.97 (s), 130.36 (s), 129.29 (s), 128.67
(d, Jcp = 4.8 Hz), 128.48 (s), 127.945 (s), 127.83 (s),
127.28 (s), 127.19 (s), 126.80 (s), 126.67 (s), 125.68 (s),
125.49 (s), 124.49 (s) 121.77 (s), 121.59 (s). HRMS
(FAB+): m/z Calc. for C25H17NO3P ([MH+]):
410.0946; Obsd.: 410.0952. Anal. Calc. for
C25H16NO3P: C, 73.35; H, 3.94; N, 3.42. Found: C,
73.20; H, 4.16; N, 3.25%.
4.2.8. Synthesis of (R)(1,10-binaphthyl-2,20-diyl)
(3-pyridyl) phosphite h
This compound was prepared as described for g,
using 3-hydroxypyridine and freshly prepared (R)-2,20-
binaphtol phosphorochloridite. Yield (78%) as a pur-
ple-red solid: 1H NMR (300 MHz): d 8.54 (d, 1H,
J = 2.4 Hz), 8.40 (d, 1H, J = 4.2 Hz), 8.02 (d, 1H,
J = 8.7 Hz), 7.95 (d, 1H, J = 8.7 Hz), 7.92 (d, 2H,
J = 8.7 Hz), 7.56–7.16 (m, 10H); 31P NMR
(121.5 MHz): d 143.05; 13C-ATP (75.465 MHz): 148.88
(d, Jcp = 6.1 Hz), 147.52 (d, Jcp = 4.8 Hz), 146.84 (d,
Jcp = 1,4 Hz), 145.80 (s), 142.71 (d, Jcp = 7.3 Hz),
133.02 (s), 132.75 (s), 132.01 (s), 131.55 (s), 130.97 (s),
130.36 (s), 129.29 (s), 128.67 (d, Jcp = 4.8 Hz), 128.48
(s), 127.945 (s), 127.83 (s), 127.28 (s), 127.19 (s),
126.80 (s), 126.67 (s), 125.68 (s), 125.49 (s), 124.49 (s)
121.77 (s), 121.59 (s). HRMS (FAB+): m/z Calc. for
C25H17NO3P ([MH+]): 410.0946; Obsd.: 410.0938. Anal.
Calc. for C25H16NO3P: C, 73.35; H, 3.94; N, 3.42.
Found: C, 73.64; H, 4.38; N, 3.05%.
4.2.6. Synthesis of (S)(3,30-bis(trimethylsilyl)-1,1-
binaphthyl-2,20-diyl)(5-(phenyl-2-yl)-10,15,20-tris
(phenyl) zinc(II) porphyrin) phosphite 6
This compound was prepared as described for 3,
using freshly prepared (S)-3,30-bis(trimethylsilyl)-2,20-
binaphtol phosphorochloridite. Yield (51%) as a pur-
ple-red solid: 1H NMR (300 MHz): d 9.06 (d, 2H),
8.96 (d, 2H, J = 5.4 Hz), 8.85 (d, 2H, J = 5.1 Hz), 8.83
(m, 2H), 8.66 (m, 1H), 8.49 (m, 3H), 8.23 (m, 2H),
8.12 (m, 2H), 8.02 (m, 1H), 7.94 (d, 2H, J = 5.4 Hz),
7.84–7.55 (m, 9H), 7.45 (m, 1H), 7.37 (m, 2H), 7.22–
7.11 (m, 4H), 6.63 (m, 1H), 8.94 (m, 1H), 5.06 (m,
1H), 4.50 (m, 1H), 0.51 (s, 9H), 0.03 (s, 9H); 31P
NMR (121.5 MHz): d 150.03; 13C-ATP (75.465 MHz):
d 150.7, 150.5, 150.4, 145.4, 135.7, 134.7, 132.4, 132.1,
131.5–130.2, 128.7, 127.6–126.5, 125.4, 125.1, 122.5,
121.9, 121.1, 0.35, 0.20; HRMS (FAB+): m/z Calc. for
C70H56N4O3PSi2Zn ([MH+]): 1151.2930; Obsd.:
1151.2920. Anal. Calc. for C70H55N4O3PSi2Zn: C,
72.92; H, 4.81; N, 4.86. Found: C, 72.96; H, 4.83; N,
4.71%.
4.2.9. Synthesis of (S)(3,30-bis(trimethylsilyl)-1,10-
binaphthyl-2,20-diyl)(3-pyridyl) phosphite i
This compound was prepared as described for g,
using freshly prepared (S)-3,30-bis(trimethylsilyl)-2,20-
binaphtol phosphorochloridite. Yield (66%) as a white
4.2.7. Synthesis of (S)(1,10-binaphthyl-2,20-diyl)
(3-pyridyl) phosphite g
1
solid: H NMR (300 MHz): d 8.30 (d, 1H, J = 2.7 Hz),
3-Hydroxypyridine (1.44 g, 15.1 mmol), azeotropi-
cally dried with toluene (3 · 5 ml), and triethylamine
(2.3 ml, 16 mmol) were dissolved in THF (40 ml) and
the solution was cooled to ꢀ40 ꢁC. Freshly prepared
8.27 (d, 1H, J = 5.1 Hz), 8.08 (d, 2H, J = 2.7 Hz), 7.92
(dd, 2H, J = 2.7, 8.1 Hz), 7.45–7.40 (m, 2H), 7.39–7.31
(m, 1H), 7.26–7.09 (m, 5H), 0.43 (s, 9H), 0.38 (s, 9H);
31P NMR (121.5 MHz): d 138.99; HRMS (FAB+):

J.N.H. Reek et al. / Journal of Organometallic Chemistry 690 (2005) 4505–4516
4515
m/z Calc. for C31H33NO33PSi2 ([MH+]): 554.1737;
Obsd.: 554.1725. Anal. Calc. for C31H32NO3PSi2: C,
67.24; H, 5.82; N, 2.53. Found: C, 67.18; H, 5.89;
N, 2.42%.
(b) J.M. Brown, in: E.N. Jacobsen, A. Pfaltz, H. Yamamoto
(Eds.), Comprehensive Asymmetric Catalysis, vol. 1, Springer,
Berlin, 1999 (Chapter 5.1);
(c) G.W. Parshall, S.D. Ittel, Homogeneous catalysis, Wiley, New
York, 1992;
(d) P.W.N.M van Leeuwen, P.C.J. Kamer, J.N.H. Reek, P.
Dierkes, Chem. Rev. 100 (2000) 2741.
4.3. Catalysis
[2] (a) I.V. Komarov, A. Borner, Angew. Chem., Int. Ed. 40 (2001)
1197;
The hydroformylation experiments were performed as
follows. A stainless steel 25 ml autoclave, equipped with
a teflon stirring bar, was charged with 0.42 lmol of
[Rh(acac)(CO)2], 10.4 lmol of phosphine and 0.017 ml
of dipea in 4.0 ml of toluene. The solution was incubated
for 1 under 20 bar CO/H2 (1:1). The pressure was re-
duced to 1 bar and a mixture of 0.34 ml styrene and
0.17 ml of decane in 0.67 ml of toluene was added. Sub-
sequently the CO/H2 pressure was repressurized to
20 bar. The mixture was stirred at 80 ꢁC, for 1 h. The
autoclave was cooled down to 0 ꢁC in ice and the pres-
sure was reduced to 1.0 bar. A sample was taken and
the conversion was checked by GC analysis of the crude
product after filtration over a plug of silica to remove
the catalyst.
The allylic alkylation experiments were performed as
follows. Under Schlenk conditions 0.50 lmol of [Pd(al-
lyl)Cl]2, 3.0 lmol phosphite and 3.0 lmol phosphine were
dissolved in 5.0 ml of CH2C12 and stirred for 30 min.
Respectively, 50 lmol 1,3-diphenylallylacetate, 150 lmol
dimethyl malonate, 150 lmol BSA and 50 lmol decane
and a pinch of KOAc were added. The mixture was stirred
at 25 ꢁC (ꢀ20 ꢁC) and after 24 h the reaction was stopped
by adding a saturated aqueous ammonium chloride solu-
tion. Subsequently, 5.0 ml of petroleum ether was added
and the solution was washed once more with a saturated
NH4Cl solution. The organic phase was dried over
Na2SO4, filtered and the conversion was determined by
GC analysis. The solution was chromatographed (SiO2;
petroleum ether/CH2Cl2 = 1/1) to give analytically pure
products [24]. Enantiomeric purities were determined by
chiral HPLC (OD column, eluent 0.5% iso-propanol in
hexane tR(R) = 33.2 min and ts(S) = 34.9 min).
(b) M. van den Berg, A.J. Minnaard, E.P. Schudde, J. van Esch,
A.H.M. de Vries, J.G. de Vries, B.L. Feringa, J. Am. Chem.
Soc. 122 (2000) 11539;
(c) T. Hayashi, Acc. Chem. Res. 33 (2000) 354.
[3] J.-B. Hu, G. Zhao, Z.-D. Ding, Angew. Chem., Int. Ed. 40 (2001)
1109.
[4] (a) J.P. Stambuli, S.R. Stauffer, K.H. Shaughnessy, J.F. Hartwig,
J. Am. Chem. Soc. 123 (2001) 2677;
(b) R.F. Harris, A.A.J. Natio, G.T. Copeland, S.J. Miller, J. Am.
Chem. Soc. 122 (2000) 11270;
(c) K.H. Shaughnessy, P. Kim, J.F. Hartwig, J. Am. Chem. Soc.
121 (1999) 2123;
(d) T.-Y. Yue, W.A. Nugent, J. Am. Chem. Soc. 124 (2002)
13692;
(e) M.T. Reetz, K.M. Kuhling, A. Deege, H. Hinrichs, D. Belder,
Angew. Chem., Int. Ed. 39 (2000) 3891;
(f) M.T. Reetz, Angew. Chem., Int. Ed. 41 (2002) 1335;
(g) J.G. de Vries, A.H.M. de Vries, Eur. J. Org. Chem. (2003) 799;
(h) S.R. Stauffer, J.F. Hartwig, J. Am. Chem. Soc. 125 (2003)
6977.
[5] A part of the work has been published as a short communication
V.F. Slagt, M. Ro¨der, P.C.J. Kamer, P.W.N.M. van Leeuwen,
J.N.H. Reek, J. Am. Chem. Soc. 126 (2004) 4056.
[6] (a) M.T. Reetz, T. Sell, A. Meiswinkel, G. Mehler, Angew.
Chem., Int. Ed. 42 (2003) 790;
(b) M.T. Reetz, Angew. Chem., Int. Ed. 40 (2001) 284;
(c) M.T. Reetz, Angew. Chem., Int. Ed. 41 (2002) 1335;
(d) S.J. Taylor, J.P. Morken, Science 280 (1998) 267;
(e) M.B. Francis, E.N. Jacobsen, Angew. Chem., Int. Ed. 38
(1999) 937;
(f) R.H. Crabtree, Chem. Commun. (1999) 1611;
(g) M.B. Francis, T.F. Jamison, E.N. Jacobsen, Curr. Opin.
Chem. Biol. 2 (1998) 422.
[7] B. Breit, W. Seiche, J. Am. Chem. Soc. 125 (2003) 6608.
[8] J.M. Takcas, D.S. Reddy, S.A. Moteki, D. Wu, H. Palencia, J.
Am. Chem. Soc. 126 (2004) 4494.
[9] (a) V.F. Slagt, J.N.H. Reek, P.C.J. Kamer, P.W.N.M. van
Leeuwen, Angew. Chem., Int. Ed. 40 (2001) 4271;
(b) V.F. Slagt, P.W.N.M. van Leeuwen, J.N.H. Reek, Angew.
Chem., Int. Ed. 42 (2003) 5619;
(c) V.F. Slagt, P.W.N.M. van Leeuwen, J.N.H. Reek, Chem.
Commun. (2003) 2474;
4.4. High-pressure NMR-experiments
(d) V.F. Slagt, P.C.J. Kamer, P.W.N.M. van Leeuwen, J.N.H.
Reek, J. Am. Chem. Soc. 126 (2004) 1526.
In a typical experiment, the high pressure NMR tube
was filled with 30 lmol of [Rh(acac)(CO)2], 75 lmol of
phosphite zinc(II) porphyrin, 75 lmol phosphine and
1.5 ml of toluene-d8. The tube was purged three times
with 15 bar of CO/H2 (1:1), pressurized to approxi-
mately 20 bar, heated to 80 ꢁC and incubated for 1 h.
Measurements were performed at 25 ꢁC.
[10] (a) A.N. Cammidge, K.M. Lifsey, Tetrahedron Lett. 41 (2000)
6655;
(b) H. Fujii, T. Yoshimura, H. Kamada, Chem. Lett. (1996) 581.
[11] G.J.H. Buisman, L.A. van der Veen, A. Klootwijk, W.G.J. de
Lange, P.C.J. Kamer, P.W.N.M. van Leeuwen, D. Vogt, Organo-
metallics 16 (1997) 2929.
[12] (a) R.J. Bowen, A.C. Garner, S.J. Berners-Price, I.D. Jenkins,
R.E. Sue, J. Organomet. Chem. 554 (1998) 181;
(b) A. Buhling, P.C.J. Kamer, P.W.N.M. van Leeuwen, J. Mol.
Catal. A 98 (1995) 69;
References
(c) G.R. Newkome, Chem. Rev. 93 (1993) 2067, and references
cited therein;
(d) K. Kellner, W. Hanke, J. Organomet. Chem. 326 (1987) C9–
C12;
[1] (a) P.A. Chaloner, M.A. Esteruelas, F. Jooꢀ, L.A. Oro, Homo-
geneous hydrogenation, Kluwer, Dordrecht, 1994;

4516
J.N.H. Reek et al. / Journal of Organometallic Chemistry 690 (2005) 4505–4516
(e) K. Kellner, W. Hanke, A. Tzschach, Zeit. fur Chem. 24 (1984)
193;
[18] (a) J. Tsuji, Tetrahedron 42 (1986) 4361;
(b) B.M. Trost, D.L. Van Vranken, Chem. Rev. 96 (1996) 395;
(c) G. Helmchen, A. Pfaltz, Acc. Chem. Res. 33 (2000) 336;
(d) A. Pfaltz, Acc. Chem. Res. 26 (1993) 339;
(f) L. Maier, Helv. Chim. Acta 48 (1965) 1034;
(g) G. Bissky, G.-V. Roschenthaler, E. Lork, J. Barten, M.
Medebielle, V. Staninets, A.A. Kolomeitsev, J. Fluorine Chem.
109 (2001) 173;
(e) B.M. Trost, M.L. Crawley, Chem. Rev. 103 (2003) 2921.
[19] (a) T. Hayashi, M. Kawatsura, Chem. Commun. (1997) 561;
(b) T. Hayashi, M. Kawatsura, Y. Uozumi, J. Am. Chem. Soc.
120 (1998) 1681;
(h) 3-Pyridylmethyldiphenylphosphine and imidazol-1-yl-meth-
yldiphenylphosphine were kindly donated by Shell Research and
Technology center, Amsterdam, The Netherlands.
(c) Y. Hamada, N. Seto, H. Ohmori, K. Hatano, Tetrahedron
Lett. 37 (1996) 7565;
[13] (a) U. Michelsen, C.A. Hunter, Angew. Chem., Int. Ed. 39 (2000)
764;
(d) C.W. Edwards, M.R. Shipton, N.W. Alcock, H. Clase, M.
Wills, Tetrahedron 59 (2003) 6473;
(b) G. Szintay, A. Horvath, Inorg. Chem. Acta. 310 (2000)
175.
(e) V.N. Tsarev, S.E. Lyubimov, A.A. Shiryaev, S.V. Zheglov,
O.G. Bondarev, V.A. Davankov, A.A. Kabro, S.K. Moiseev, V.N.
Kalinin, K.N. Gavrilov, Eur. J. Org. Chem. (2004) 2214.
[20] (a) For detailed analysis of these effects see I.J.S. Fairlamb, G.C.
[14] P.N. Taylor, H.L. Anderson, J. Am. Chem. Soc. 121 (1999)
11538.
[15] T. Jongsma, G. Challa, P.W.N.M. van Leeuwen, J. Organomet.
Chem. 421 (1991) 121.
ˇ
Lloyd-Jones, S. Vyskocil, P. Kocovsky, Chem. Eur. J. 8 (2002)
ˇ
ˇ
´
[16] Low temperature 31P NMR spectroscopy (T = ꢀ20 ꢁC) of phos-
4443;
(b) P. Dotta, P.G.A. Kumar, P.S. Pregosin, A. Albinati, S. Rizzato,
Organometallics 22 (2003) 5345.
phine–phosphite complex 9 shows a double doublet (JRh–P
=
131 Hz, JP–PO3 = 111 Hz) for the phosphine. This suggests that the
phosphine is coordinated in the equatorial plane of the rhodium
complex. At the given temperature the phosphite signal is a broad
multiplet and the phosphine–phosphite coupling constant (JP–
PO3 = 111 Hz) suggests that phosphite coordination to the rho-
dium is at the equatorial and apical position in fast exchange on
the NMR spectroscopy time-scale. The phosphorus-signal of the
phosphite is at rather low field (d = 179.8 ppm) which is attributed
to steric constraints, as is also found for several other phosphine–
phosphite rhodium complexes..
´
[21] R. Pretot, A. Pfaltz, Angew. Chem., Int. Ed. 37 (1998) 323.
[22] S. Deerenberg, H.S. Schrekker, G.P.F. van Strijdonck, P.C.J.
Kamer, P.W.N.M. van Leeuwen, J. Fraanje, K. Goubitz, J. Org.
Chem. 65 (2000) 4810.
[23] A.D. Adler, F.R. Longo, F. Kampas, J. Kim, J. Inorg. Nucl.
Chem. 32 (1970) 2443.
[24] (a) Product characterization data were found in agreement with
the reported literature, see for example: P. von Matt, O. Loiseleur,
G. Koch, A. Pfaltz, C. Lefeber, T. Feucht, G. Helmchen,
Tetrahedron Asymm. 5 (1994) 573;
[17] P.W.N.M. van Leeuwen, C. Claver (Eds.), Rhodium Catalyzed
Hydroformylation, Kluwer Academic Publishers, Dordrecht,
2000.
(b) J.F. Bower, R. Jumnah, A.C. Williams, J.M.J. Williams, J.
Chem. Soc., Perkin Trans. 1 (1997) 1411.