UREAphos: Supramolecular bidentate ligands for asymmetric hydrogenation

Article abstract of DOI:10.1039/b614571j

Supramolecular bidentate phosphite ligands are presented as a new class of ligands for rhodium catalysed asymmetric hydrogenation. The Royal Society of Chemistry.

Full text of DOI:10.1039/b614571j

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
UREAphos: supramolecular bidentate ligands for asymmetric
hydrogenation{
Albertus J. Sandee, Alida M. van der Burg and Joost N. H. Reek*
Received (in Cambridge, UK) 6th October 2006, Accepted 6th November 2006
First published as an Advance Article on the web 24th November 2006
DOI: 10.1039/b614571j
Supramolecular bidentate phosphite ligands are presented as a
new class of ligands for rhodium catalysed asymmetric
hydrogenation.
Chiral bidentate ligands1 have dominated the field of asymmetric
transition metal catalysis for more than 30 years. However, since
the recent breakthrough by Feringa and de Vries,2 Reetz3 and
Pringle,4 showing that metal complexes based on monodentate
ligands can provide highly enantioselective catalysts, many
examples using monodentate ligands have appeared in the
literature.5 The main reason for the preference for monodentate
over bidentate ligands is their relative ease of preparation,
facilitating the preparation of large catalyst libraries. Recently,
we6 and others7 have introduced a new class of bidentate ligands
that form by a self-assembly process of two monodentate ligands.8
After the successful exploration of our SUPRAphos6c,d,e library
Chart 1
based on porphyrin appended ligands we decided to extend our
recently introduced nonchiral urea-based homo-bidentate ligands9
to chiral analogues for use in asymmetric conversions.10 This class
should lead to a novel class of supramolecular bidentate ligands
that is easily accessible. Here we report the synthesis of urea
appended chiral phosphite building blocks and their successful use
in rhodium catalysed hydrogenation of various substrates.
We, and the group of Love7f have independently introduced
novel phosphine ligands containing appended urea groups. These
self-complementary hydrogen bond motifs enable the formation of
supramolecular bidentate ligands.9 We showed that bidentate
ligands form in situ by just mixing two equivalents of ligand A in
the presence of a palladium precursor to provide [Pd(A)2MeCl]
(Scheme 1).
based on the bisnaphthol backbone, and a (chiral) spacer to con-
nect these functions (Chart 1, top). Small differences in spacer and
motifwereappliedtocreateasmallseriesofligandbuildingblocks.{
F is the only ligand containing an R-bisnaphthol backbone; all
the others contain the S-bisnaphthol backbone. The spacers of
building blocks D, E, F and G contain additional chirality (see
Chart 1 for details). F is the only ligand that contains a thiourea
motif, G is utilised with an indole-amide binding motif whereas
all other systems are functionalised with the urea binding motif.
It is important to note that all ligand building blocks are easily
accessible and amenable to simple variation to facilitate the
preparation of large, diverse libraries at a later stage. Indeed, all
compounds were obtained via a simple two-step synthetic
procedure, consisting of a coupling of an amino-alcohol with an
iso(thio)cyanate to obtain the urea-alcohol that is subsequently
reacted with 2,29-bisnaphthol phosphorochloridite to obtain the
urea containing phosphites. These ligands were characterised and
[Rh(D)2(NBD)](BF4) was investigated as an archetypal complex
by NMR and IR to study the structure of this type of complexes
and support the presence of intramolecular H-bonding between
the two ligands attached to the metal centre (see ESI). In analogy
to A (and complexes thereof) studied previously,9 it was observed
that at low concentration (5 mM) the Ar–NH moiety of ligand D
is predominantly in the non-associated state (N–H stretch at
3430 cm21). In contrast, in [Rh(D)2(NBD)](BF4) the majority of
the N–H is found to be in the hydrogen bonded form (N–H stretch
The current ligand building blocks (Chart 1, B–G){ consist of a
similar urea-type hydrogen bond motif, a chiral phosphite ligand
Scheme 1 Formation of a palladium complex based on a supramole-
cular bidentate ligand (A).
van’t Hoff Institute for Molecular Sciences, University of Amsterdam,
Nieuwe Achtergracht 166, 1018 WV Amsterdam, The Netherlands.
E-mail: reek@science.uva.nl; Fax: (+31)20-525-6422;
Tel: (+31)20-525-6437
{ Electronic supplementary information (ESI) available: Experimental
details. See DOI: 10.1039/b614571j
1
at 3360 cm21). As expected, the Ar–NH-signal in the H-NMR
spectrum varies with the ligand concentration (from 5 mM to
40 mM), indicating the formation of intermolecular hydrogen
bonds. In the complex [Rh(D)2(NBD)](BF4), this signal does not
864 | Chem. Commun., 2007, 864–866
This journal is ß The Royal Society of Chemistry 2007

View Article Online
shift in this concentration range, since it is involved in an
intramolecular hydrogen bond (see ESI).11 These experiments
clearly demonstrate that these phosphite ligands behave similarly
to A; they self-aggregate at high concentrations in solution or form
intramolecular hydrogen bonds when coordinated to a transition
metal. In contrast to what is observed for [Pd(A)2MeCl] and in
agreement with what is generally observed for rhodium–phosphite
complexes,12a the ligands are coordinated in a cis-configuration
prepared from the R-bisnaphthol. The difference in ee obtained
with the catalysts based on E and C (entries 5 and 3) (92.7,
16.6% ee, respectively) indicates that the introduction of chirality in
the spacer can result in huge improvements of the selectivity.
Comparison of the selectivity induced by [Rh(G)2] and [Rh(D)2]
(95.8% vs. 46%) (entries 7 and 4) shows that a change in binding
motif can also have major consequences. The reason for the large
effect is currently not clear, but it is likely that the increased acidity
of one of the N–H moieties in ligand G increases the affinity
between the building blocks, thereby making the complex more
rigid. At this stage we cannot rule out the possibility that steric
effects introduced by the indole, although it is positioned remotely,
are the origin of the effect observed. Surprisingly, the rhodium
complex formed from ligand B, which contains a simple C3 amino-
alcohol spacer and an n-butyl urea binding motif, did not give any
conversion in this catalytic process. Since this ligand has the most
flexible spacer (C3 and no substituents) between the ligand and the
urea, it is likely that intramolecular urea coordination to rhodium
metal reduces the catalytic activity of the complex.
about the metal centre in [Rh(D)2(NBD)](BF4) as evidenced by 31
P
NMR.12b
Next, the ligands were explored as supramolecular bidentate
ligands in rhodium catalysed asymmetric hydrogenation reactions
(Table 1).{ Two of the six ligands provided rhodium complexes
that hydrogenated dimethyl itaconate in more than 90% ee, and
most catalysts gave 100% conversion of this substrate. The most
selective catalyst was [Rh(G)2] (Table 1 entry 7). It provided an
enantioselectivity of 95.8% to the S-product at 100% conversion
after a reaction of 18 hours at room temperature.
The huge difference in catalyst performance between metal
complexes based on E and F (entries 5 and 6) must be caused by
the thio-urea functional group since this is the main difference
between the ligands. An explanation for this is that the sulfur of
the thio-urea can also coordinate to rhodium, potentially giving
rise to PS-coordination complexes.14 In addition, thio-ureas show
much weaker self-association behaviour than oxygen based
ureas.13 The current results suggest that the thio-urea is not a
suitable binding motif for the formation of supramolecular ligands
for rhodium catalysed hydrogenation.
In general, large differences are observed, making the current
approach and building blocks very interesting since extension of
the library is straightforward due to the availability of chiral
amino-alcohols and isocyanides.
After these interesting initial results we also studied the
hydrogenation of N-(3,4-dihydro-2-naphthalenyl)-acetamide (2),
which is a much more challenging substrate both in terms of
conversion and selectivity.15 Upon applying rhodium catalysts
based on A–G, conversions were obtained between 0.4 to 100%
and the selectivity ranged from 37.9% for the S-enantiomer to
76.5% for the R-enantiomer. Also in the hydrogenation of 2 the
enantioselective outcome of the catalysis is determined by the
chirality of the bisnaphthol backbone; only [Rh(F)2] resulted in
preference for the formation of the S-product (entry 13). The
difference in results for catalysts based on C, D and E in this
conversion (providing 76.5, 60.7% and 52.5 ee, respectively)
(entries 10, 11 and 12) shows the influence of the nature of the
amino-alcohol spacer on the enantioselectivity. The nature of the
binding motif also appeared important in the hydrogenation of 2.
The thio-urea motif again slows down the catalysis; [Rh(F)2] only
gave 0.4% conversion (entry 13). Remarkably, the amide-indole
functionalised ligand G provides a catalyst [Rh(G)2] that gives an
almost racemic product mixture (1.5% ee only) (entry 14), whereas
in the hydrogenation of 1 this ligand was found to provide the
most selective catalyst. Similar to the hydrogenation of 1, the
ability to vary the amino-alcohol spacer as well as the binding
motif provides new handles for catalyst fine-tuning. The most
selective catalyst, [Rh(C)2], provided 76.5% ee towards the
R-product (entry 10), which makes this the second best rhodium
catalyst in terms of selectivity (obtained at conversion of
12.3%).6e,15 The application of monodentate phosphoramidite
ligands results only in very low ee (up to 34%) in this reaction as
was previously reported.15b
The absolute chirality of the product is clearly determined by the
nature of the bisnaphthol since all catalysts yielded the S-product
except that based on F, which was the only building block
Table 1 Rhodium catalysed asymmetric hydrogenation of various
substrates using UREA-phos ligands
Conversion
[%]
Ee [%]
(config)
Entry
Liganda
Substrate
1
2
3
4
5
6
7
8
9
A
B
C
D
E
F(R)
G
A
B
C
D
E
F(R)
G
A
B
C
D
1
1
1
1
1
1
1
2
2
2
2
2
2
2
3
3
3
3
3
3
100
0
0
0
100
100
100
100
100
100
4
12.3
4.1
34.1
0.4
26.1
100
0
16.6 (S)
46 (S)
92.7 (S)
13.8 (R)
95.8 (S)
0
4 (R)
10
11
12
13
14
15
16
17
18
19
20
a
76.5 (R)
60.7 (R)
52.5 (R)
37.9 (S)
1.5 (R)
0
In the hydrogenation of methyl 2-acetamidoacrylate (3) we
found similar trends. Four catalysts provided full conversion while
[Rh(F)2] (entry 20) and [Rh(B)2] (entry 16) showed low and no
conversion respectively. The chirality of the bisnaphthol deter-
mines the chirality of the product while fine-tuning is established
by the choice of the amino-alcohol spacer. Both [Rh(C)2] and
[Rh(D)2] provided the product in high selectivity; 93.6% and 92.3%
0
100
100
100
36.9
93.6 (R)
92.3 (R)
82.1 (R)
44.6 (S)
E
F(R)
All ligands based on S-bisnaphthol phosphite backbone, unless
noted differently.
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 864–866 | 865

View Article Online
J. N. H. Reek, Angew. Chem., Int. Ed., 2003, 42, 5619; (c) V. F. Slagt,
M. Ro¨der, P. C. J. Kamer, P. W. N. M. van Leeuwen and J. N. H. Reek,
J. Am. Chem. Soc., 2004, 126, 4056; (d) J. N. H. Reek, M. Ro¨der,
P. E. Goudriaan, P. C. J. Kamer, P. W. N. M. van Leeuwen and
V. F. Slagt, J. Organomet. Chem., 2005, 605, 4505; (e) X.-B. Jiang,
L. Lefort, P. E. Goudriaan, A. H. M. de Vries, P. W. N. M. van
Leeuwen, J. G. de Vries and J. N. H. Reek, Angew. Chem., Int. Ed.,
2006, 45, 1223.
7 (a) B. Breit and W. Seiche, J. Am. Chem. Soc., 2003, 125, 6608; (b)
B. Breit and W. Seiche, Angew. Chem., Int. Ed., 2005, 44, 1640; (c)
B. Breit and W. Seiche, J. Am. Chem. Soc., 2003, 125, 6608; (d)
F. Chevallier and B. Breit, Angew. Chem., Int. Ed., 2006, 45, 1599; (e)
J. M. Takacs, D. S. Reddy, S. A. Moteki, D. Wu and H. Palencia,
J. Am. Chem. Soc., 2004, 126, 4494; (f) P. A. Duckmanton, A. J. Blake
and J. B. Love, Inorg. Chem., 2005, 44, 7708.
ee respectively (entries 17 and 18). In this process a large phenyl-
group on the a-position in the spacer (in E) appeared to have a
negative effect on the selectivity.
In conclusion, we have introduced a new class of supramole-
cular bidentate phosphite ligands that was successfully applied in
the rhodium catalysed asymmetric hydrogenation of various
substrates. The small series employed in this contribution have
already provided hydrogenation catalysts that are highly selective.
The easy accessibility of these ligands and the huge potential for
catalyst tuning make them highly suitable for combinatorial
approaches and high throughput screening experimentation. We
are currently developing a fully automated preparation protocol
for this new class of ligands and these results will be reported in
due course.
8 For recent reviews on this subject see: (a) M. J. Wilkinson,
P. W. N. M. van Leeuwen and J. N. H. Reek, Org. Biomol. Chem.,
2005, 3, 2371; (b) B. Breit, Angew. Chem., Int. Ed., 2005, 44, 6816; (c)
A. J. Sandee and J. N. H. Reek, Dalton Trans., 2006, 3385.
9 L. K. Knight, Z. Freixa, P. W. N. M. van Leeuwen and J. N. H. Reek,
Organometallics, 2006, 25, 954.
Engelhard De Meern B.V. and the Ministry of Economic
Affairs are gratefully acknowledged for financial support.
10 During the preparation of this manuscript an example of asymmetric
hydrogenation using self-assembled ligands by means of hydrogen
bonds was reported: M. Weis, C. Waloch, W. Seiche and B. Breit, J. Am.
Chem. Soc., 2006, 128, 4188.
11 Upon increasing the concentration further, higher order aggregates are
observed. These are, however, of no significance to catalysis since that is
typically performed at lower concentrations.
12 Observed 31P NMR signal for [Rh(D)2(NBD)](BF4): 31P NMR (CDCl3,
162.0 MHz) d 164.4 (d, JRhP = 220 Hz) which is typical for cis-
bisphosphite complexes, see: P. C. J. Kamer, J. N. H. Reek and
P. W. N. M. van Leeuwen, Rhodium Phosphite Catalysts, Catal. Met.
Complexes, 2000, 22, 35.
Notes and references
{ All reactions were performed in CH2Cl2, Rh : L = 1 : 2.4, Rh : substrate
= 1 : 100, 10 bar of H2, 18 h at room temperature.
1 For a recent review on bidentate ligands in asymmetric hydrogenation,
see: W. Tang and X. Zhang, Chem. Rev., 2003, 103, 3029 and references
cited therein.
2 M. van den Berg, A. J. Minnaard, E. P. Schudde, J. van Esch,
A. H. M. de Vries, J. G. de Vries and B. L. Feringa, J. Am. Chem. Soc.,
2000, 122, 11539.
13 P. Bu¨hlmann, S. Nishizawa, K. P. Xiao and Y. Umezawa, Tetrahedron,
1997, 53, 1647.
3 M. T. Reetz and G. Mehler, Angew. Chem., Int. Ed., 2000, 39, 3889.
4 C. Claver, E. Fernandez, A. Gillon, K. Heslop, D. J. Hyett,
A. Martorell, A. G. Orpen and P. G. Pringle, Chem. Commun., 2000,
961.
5 For a recent review on monodentate ligands, see: (a) T. Jerphagnon,
J.-L. Renaud and C. Bruneau, Tetrahedron: Asymmetry, 2004, 15, 2101
and references cited therein; (b) J. G. de Vries, in Handbook of Chiral
Chemicals, ed. D. Ager, S. Laneman, CRC Press, New York, 2nd edn,
2005.
14 Hydrogenation catalysis using PS-coordinating ligands including
thiourea functional groups has been previously reported by
Tiripicchio et al.: D. Cauzzi, M. Costa, N. Cucci, C. Graiff,
F. Grandi, G. Predieri, A. Tiripicchio and R. Zanoni, J. Organomet.
Chem., 2000, 593, 431.
15 (a) Z. Zhang, G. Zhu, Q. Jiang, D. Xiao and X. Zhang, J. Org. Chem.,
1999, 64, 1774; (b) H. Bernsmann, M. van den Berg, R. Hoen,
A. J. Minnaard, G. Mehler, M. T. Reetz, J. G. de Vries and B.
L. Feringa, J. Org. Chem., 2005, 70, 943.
6 (a) V. F. Slagt, P. W. N. M. van Leeuwen and J. N. H. Reek, Chem.
Commun., 2003, 2474; (b) V. F. Slagt, P. W. N. M. van Leeuwen and
866 | Chem. Commun., 2007, 864–866
This journal is ß The Royal Society of Chemistry 2007