Self-assembly of a confined rhodium catalyst for asymmetric hydroformylation of unfunctionalized internal alkenes

Article abstract of DOI:10.1021/ja211455j

A chiral supramolecular ligand has been assembled and applied to the rhodium-catalyzed asymmetric hydroformylation of unfunctionalized internal alkenes. Spatial confinement of the metal center within a chiral pocket results in reversed regioselectivity and remarkable enantioselectivities.

Full text of DOI:10.1021/ja211455j

Communication
pubs.acs.org/JACS
Self-Assembly of a Confined Rhodium Catalyst for Asymmetric
Hydroformylation of Unfunctionalized Internal Alkenes
Tendai Gadzikwa, Rosalba Bellini, Henk L. Dekker, and Joost N. H. Reek*
van’t Hoff Institute for Molecular Sciences, University of Amsterdam, Science Park 904, 1098 XH, Amsterdam, The Netherlands
*
S Supporting Information
variant of this reaction as well. We explored this possibility to
modest success with a supramolecularly generated
3
1−35
ABSTRACT: A chiral supramolecular ligand has been
assembled and applied to the rhodium-catalyzed asym-
metric hydroformylation of unfunctionalized internal
alkenes. Spatial confinement of the metal center within a
chiral pocket results in reversed regioselectivity and
remarkable enantioselectivities.
monoligating capsule ligand, though the improved enantiose-
I
lectivities were due to altered Rh coordination rather than
36
active-site encapsulation. We then postulated that, to generate
a chiral system that would exhibit significant confinement
effects, it was necessary to assemble a structurally robust ligand
that would bind the metal center within a cavity in such a way
that would minimize conformational freedom, thus enforcing
the spatial preferences of the chiral pocket most effectively. To
this end, we self-assembled a multivalently bound bidentate
n transition metal catalysis, the enantioselective trans-
formation of unfunctionalized internal alkenes presents a
I
1
,2
I
daunting challenge. Substrates of this class lack moieties for
ancillary coordination to the metal center and so must rely on
nonbonding interactions with the catalyst for the induction of
chirality. To be selective, the catalyst should distinguish two
linear alkyl groups whose electronics are nearly identical, and so
must be capable of differentiating the alkyl groups' lengths. For
the hydroformylation reaction in particular, these substrates
pose the further complication that internal alkenes are much
less reactive than terminal olefins, and substrate isomerization is
a side reaction that needs to be suppressed as it compromises
ligand that anchors a Rh catalytic center within a restricted
chiral pocket. The remarkable enantioselectivities afforded by
this ligand are reported herein.
We constructed a supramolecular “box” using bis-
II
37,38
[
Zn (salphen)] (3, Figure 1) as a template
and a chiral
phosphorus-based ligand as the dipyridyl pillar. Boxes
assembled from 3 and various dipyridyl linkers were previously
shown to be extremely robust, forming spontaneously, with
binding constants of at least 1020 M . We conjectured that
employing 3-pyridyl-substituted monodentate phosphoramidite
ligands (2) as pillars of such a “box” would bring two
phosphorus atoms into close proximity, affording a self-
assembled chiral bidentate ligand (1, Figure 1). We foresaw
−3
3
,4
selectivity.
Elevated temperatures are often required to
achieve adequate activities, which, however, also leads to
lower selectivities. To circumvent these difficulties, efforts in
the hydroformylation of internal olefins are generally focused
on directing the migration of the π-bond to the chain terminus,
I
that, once a Rh atom was bis-ligated and the active catalyst
generated, the internal pocket of the “box” would be sterically
crowded, as with our previously reported nonchiral capsule,
decreasing the conformational freedom around the active site,
as desired for good selectivity. A computational model of the
active catalyst does indeed indicate that bis-ligation is feasible,
with the metal center embedded within a chiral pocket (1-
CORhH-octene, Figure 1).
5
−9
followed by hydroformylation.
However, this strategy
precludes the production of the more valuable chiral aldehydes,
yielding achiral linear aldehydes preferentially. Thus, a major
achievement in the hydroformylation of unfunctionalized
internal alkenes would be the generation of a catalyst that
hydroformylates the internal bond specifically with high
1
0−18
A pure crystalline powder of 1 was obtained via slow
evaporation of an equimolar solution of 3 and 2 in acetone. The
enantioselectivity.
Supported by the evidence of myriad
examples in the enzymatic world, we believed that such specific
discrimination could be best achieved with the catalytically
1
H NMR spectrum of dissolved 1 confirmed that 3 and 2 are
1
9,20
present in a 1:1 ratio (see Supporting Information (SI) Figure
S1). To ascertain whether 1 is a discrete architecture or a
coordination polymer, 2 was titrated against 3, monitored by
circular dichroism spectroscopy. The titration curve (SI, Figure
S3) had a single inflection point (1:1) with no further increase
beyond this point, as would be expected for a discrete
active metal located within a sterically restrictive pocket.
The encapsulation of transition metal catalysts in this manner
has been shown to give rise to substrate selectivity in a variety of
2
1
22,23
catalyzed reactions, including epoxidation
and C−H
2
4,25
activation.
led to enhanced product selectivities
In fewer examples, active-site confinement has
26−28
or has altered product
39
2
9
structure. A diffusion NMR experiment was conducted to
establish the number of molecules in the suprastructure (SI,
Figure S4). The obtained diffusivity indicated a discrete
assembly with a hydrodynamic radius of about 1 nm,
profiles altogether. We previously reported an encapsulated
catalyst for the selective hydroformylation of unfunctionalized
3
0
internal alkenes. This supramolecular catalyst exhibits
unprecedented regioselectivity, with the ability to distinguish
the C3 and C2 carbons of 2-octene with great precision,
resulting in 9:1 selectivity. Based on this system, we anticipated
that encapsulation would improve selectivity for the asymmetric
Received: September 15, 2011
Published: January 23, 2012
©
2012 American Chemical Society
2860
dx.doi.org/10.1021/ja211455j | J. Am. Chem. Soc. 2012, 134, 2860−2863

Journal of the American Chemical Society
Communication
catalysis at slightly higher temperature and for a longer time
(
40 °C for 5 d) and found that we could increase the
conversions to 36% and 79% respectively for the trans and cis
substrates while retaining good selectivities (Table 1, entries 3
and 8).
Table 1. Evaluation of Ligands for the Hydroformylation of
a
trans- and cis-2-Octene
Figure 1. Construction of supramolecular ligand 1 from the assembly
of 3-PyMonoPhos (2) with bis[Zn(salphen)] (3): metalation of 1 to
a
All reactions were performed at 25 °C in toluene with 20 bar 1:1
1-Rh. Bottom left: Spartan PM3 calculated structure of the octene
CO/H , 0.5 mol % Rh(acac)(CO) , 2.5 mol % L (4.5 mol % 2), and
8
are reported. At 40 °C for 5 d.
substrate coordinated to the active catalyst. Black, 2; gray, 3; teal,
rhodium; green, cis-2-octene; red, oxygen.
2
2
b
4 h reaction time. GC conversions (based on dodecane standard)
c
corresponding most closely to a 2+2 assembly, as determined
by the computational model.
The compound 1-Rh was prepared by incubating 1 and
The substrate scope of 1 was then evaluated using various
unfunctionalized alkenes. Performing hydroformylation reac-
tions on a series of cis- and trans-2-alkenes, using ligand 1, we
found that our selectivities were general for all 2-olefins: the
innermost aldehyde was produced preferentially, with the trans-
Rh(acac)(CO) (1:1) at 80 °C overnight. The major product,
2
3
1
1
-Rh, was characterized by P NMR (SI, Figure S5), the
sharper peaks of which (compared to those of 1: SI, Figure S2)
2
-olefins affording 80% of the major enantiomer, and the cis-2-
suggested a structure with less conformational freedom. The
olefins 90% (Table 2, entries 1−6). For the symmetric 3-
hexenes, both cis and trans substrates were converted with
almost equal selectivity, i.e., 80% of the major enantiomer
3
1
P NMR spectrum has a single doublet (JP−Rh = 290 Hz),
indicating the formation of the bis-ligated square planar RhI
complex. Electrospray ionization mass spectrometry corrobo-
rated the NMR evidence for bis-ligation, clearly showing a peak
(
2
Table 2, entries 7 and 8). For all substrates, 1 outshone ligand
, with the monodentate ligand performing exactly as it did in
2
+
corresponding to the [Rh(acac)(1)] species (SI, Figure S6).
Satisfied that we had achieved our goal in terms of
assembling a chiral encapsulating bidentate ligand, we explored
its utility in the Rh-catalyzed asymmetric hydroformylation of
both cis- and trans-2-octene. For comparison, the monodentate
ligand 2 was also investigated. Using ligand 2, both alkene
isomers were converted to aldehyde a preferentially, as
the case of the octenes (SI, Table S1). The performance of our
ligand in the hydroformylation of terminal alkenes was also
investigated. Ligand 1 was not particularly selective for 1-
a
Table 2. Evaluation of the Substrate Scope of Ligand 1
b
c
4
0
entry
substrate
% conv
% inner
er
expected. However, upon fixation of ligand 2 into box
structure 1, we observed a reversal of regioselectivity, resulting
in a modest preference for aldehyde b. The more remarkable
change, however, was in the increase in enantiomeric ratio (er):
using ligand 1 in the hydroformylation of trans-2-octene
resulted in an increase in the ratio of the R enantiomer of b
from 61 to 86%. For cis-2-octene, we observed a more marked
increase in enantioselectivity, from an er of 48:52 to a record-
breaking 93:7. The significance of these results is apparent
when compared to those of the commercially available (R,R)-
Chiraphite and (S,S)-DIOP, as ligand 1 clearly outperforms
both chiral bidentate ligands significantly. In the case of cis-2-
octene, 1 affords R-b as 65% of all possible products, while the
other ligands deliver about 20%. The activities observed with
ligand 1, however, are relatively low: 10% and 20% conversion
for trans- and cis-2-octene, respectively. We performed the
1
2
3
4
5
6
7
8
9
trans-2-nonene
cis-2-nonene
trans-2-octene
cis-2-octene
trans-2-heptene
cis-2-heptene
trans-3-hexene
cis-3-hexene
1-octene
13
26
10
20
9
48:52
39:61
40:60
30:70
45:55
33:67
na
81:19
90:10
86:14
93:7
83:17
91:9
24
19
11
77
80
19:81
22:78
nd
na
56:44
5:95
d
1
0
styrene
73:27
a
All reactions were performed at 25 °C in toluene with 20 bar 1:1
CO/H , 0.5 mol % Rh(acac)(CO) , 2.5 mol % 1, and 84 h reaction
time. GC conversions (based on dodecane standard) are reported.
The er values were determined by chiral GC only for the innermost
aldehyde. S:R.
2
2
b
c
d
2
861
dx.doi.org/10.1021/ja211455j | J. Am. Chem. Soc. 2012, 134, 2860−2863

Journal of the American Chemical Society
Communication
octene; however, moderate selectivity was obtained for styrene,
for which ligand 2 showed no enantioselectivity whatsoever.
Under hydroformylation conditions, an excess of ligand is
required to suppress the formation of ligand-free rhodium, an
active and nonselective catalyst that compromises the
(7) Bronger, R. P. J; Kamer, P. C. J.; van Leeuwen, P. W. N. M.
Organometallics 2003, 22, 5358−5369.
(8) Yu, S.; Chie, Y.-M.; Guan, Z.-h.; Zhang, X. Org. Lett. 2008, 10,
3
(
469−3472.
9) Seayad, A.; Ahmed, M.; Klein, H.; Jackstell, R.; Gross, T.; Beller,
M. Science 2002, 297, 1676−1678.
10) For enantioselective hydroformylation of 2-butene, see: Sakai,
N.; Nozaki, K.; Takaya, H. J. Chem. Soc., Chem. Commun. 1994, 395−
96.
11) Nozaki, K.; Matsuo, T.; Shibahara, F.; Hiyama, T. Adv. Synth.
3
selectivity. In line with this, we observed that a 1:Rh ratio of
(
5
:1 was necessary for optimal selectivity (see SI, Chart S1 for
ligand ratio optimization). This excess ligand is not required to
retain a stable box structure, and according to the large
association constant involved in the formation of 1, the excess
ligand is present in solution as the metal-free box structure 1.
To confirm the stability of ligand 1, we investigated its
performance at elevated temperature. We performed the
catalysis at 70 °C and determined selectivities after 1 h, as
isomerization at higher conversions returned convoluted results
3
(
Catal. 2001, 343, 61−63.
(12) For regioselective hydroformylation of functionalized internal
̌
alkenes, see: (a) Smejkal, T.; Breit, B. Angew. Chem., Int. Ed. 2008, 47
(
218), 311−315. (b) Dydio, P.; Dzik, W. I..; Lutz, M.; de Bruin, B.;
Reek, J. N. H. Angew. Chem., Int. Ed. 2011, 50, 396−400.
̌
(
13) Smejkal, T.; Gribkov, D.; Geier, J.; Keller, M.; Breit, B. Chem.
Eur. J. 2010, 16, 2470−2478.
(
see SI, Table S2). Interestingly, despite a loss of the preference
(14) For enantioselective hydroformylation of functionalized internal
for product b due to substrate isomerization, the catalyst still
alkenes, see: Chikkali, S. H.; Bellini, R.; Berthon-Gelloz, G.; van der
Vlugt, J. I.; de Bruin, B.; Reek, J. N. H. Chem. Commun. 2010, 46,
displayed high enantiomeric ratios for the innermost aldehyde:
8
2:18 for trans and 88:12 for cis. Under the same conditions,
1
(
244−1246.
15) Grunanger, C. U.; Breit, B. Angew. Chem., Int. Ed. 2010, 49,
967−970.
(16) Gual, A.; Godard, C.; Castillon
2010, 352, 463−477.
17) McDonald, R. I.; Wong, G. W.; Neupane, R. P.; Stahl, S. S.;
ligand 2 gave essentially racemic mixtures. These results
indicate that our design strategy was sound, i.e., that the
multivalent binding of 1 results in a ligand robust enough to
remain intact at elevated temperatures.
̈
́
, S.; Claver, C. Adv. Synth. Catal.
(
In conclusion, we have demonstrated that the confinement
effects in the hydroformylation of unfunctionalized internal
Landis, C. R. J. Am. Chem. Soc. 2010, 132, 14027−14029.
(18) Worthy, A. D.; Joe, C. L.; Lightburn, T. E.; Tan, K. L. J. Am.
Chem. Soc. 2010, 132, 14757−14759.
3
0
alkenes can be extended to the asymmetric reaction. The
supramolecularly generated chiral ligand has afforded extra-
ordinary enantioselectivities for these difficult substrates,
especially for the cis-2-olefins. Additionally, the design strategy
calling for multiple points of attachment for each of the
structural components has resulted in a robust architecture,
allowing for continued selectivity at elevated temperatures.
(19) Ringe, D.; Petsko, G. A. Science 2008, 320, 1428−1429.
(20) Wiester, M. J.; Ulmann, P. A.; Mirkin, C. A. Angew. Chem., Int.
Ed. 2010, 50, 114−137.
(21) Koblenz, T. S.; Wassenaar, J.; Reek, J. N. H. Chem. Soc. Rev.
2
008, 37, 247−262.
(22) Merlau, M. L.; del Pilar Mejia, M.; Nguyen, S. T.; Hupp, J. T.
Angew. Chem., Int. Ed. 2001, 40, 4239−4242.
ASSOCIATED CONTENT
(23) Lee, S. J.; Cho, S.-H.; Mulfort, K. L.; Tiede, D. M.; Hupp, J. T.;
Nguyen, S. T. J. Am. Chem. Soc. 2008, 130, 16828−16829.
■
*
S
Supporting Information
(24) Leung, D. H.; Fiedler, D.; Bergman, R. G.; Raymond, K. N.
Synthetic protocols and characterization data for 1 and 1-Rh;
experimental procedures and results for all catalysis runs. This
material is available free of charge via the Internet at http://
pubs.acs.org.
Angew. Chem. 2004, 116, 981−984.
(25) Fiedler, D.; Leung, D. H.; Bergman, R. G.; Raymond, K. N. Acc.
Chem. Res. 2005, 38, 349−358.
(26) Slagt, V. F.; Reek, J. N. H.; Kamer, P. C. J.; van Leeuwen, P. W.
N. M. Angew. Chem., Int. Ed. 2001, 40, 4271−4274.
(
27) Lee, S. J.; Hu, A.; Lin, W. J. Am. Chem. Soc. 2002, 124, 12948−
AUTHOR INFORMATION
■
1
2949.
28) Gianneschi, N. C.; Cho, S.-H.; Nguyen, S. T.; Mirkin, C. A.
Angew. Chem., Int. Ed. 2004, 43, 5503−5507.
29) Cavarzan, A.; Scarso, A.; Sgarbossa, P.; Strukul, G.; Reek, J. N.
H. J. Am. Chem. Soc. 2010, 133, 2848−2851.
30) Kuil, M.; Soltner, T.; van Leeuwen, P. W. N. M.; Reek, J. N. H. J.
Corresponding Author
j.n.h.reek@uva.nl
(
(
Notes
The authors declare no competing financial interest.
(
Am. Chem. Soc. 2006, 128, 11344−11345.
ACKNOWLEDGMENTS
(
(
31) Breit, B. Angew. Chem., Int. Ed. 2005, 44, 6816−6825.
■
32) Carboni, S.; Gennari, C.; Pignataro, L.; Piarulli, U. Dalton Trans.
Financial support for this work was provided by the
Schlumberger Foundation’s Faculty for the Future (fellowship
for T.G.), the NRSCC, and the University of Amsterdam.
2
(
(
011, 40, 4355−4373.
33) Meeuwissen, J.; Reek, J. N. H. Nat. Chem. 2010, 2, 615−621.
34) Moteki, S. A.; Takacs, J. M. Angew. Chem., Int. Ed. 2008, 47,
8
94−897.
REFERENCES
■
(35) Kuil, M.; Goudriaan, P. E.; Kleij, A. W.; Tooke, D. M.; Spek, A.
(
(
(
1) Schurig, V.; Betschinger, F. Chem. Rev. 1992, 92, 873−888.
L.; van Leeuwen, P. W. N. M.; Reek, J. N. H. Dalton Trans. 2007,
2311−2320.
2) Wills, M. Science 2006, 311, 619−620.
3) van Leeuwen, P. W. N. M.; Claver, C. Rhodium Catalyzed
(36) Bellini, R.; Chikkali, S. H.; Berthon-Gelloz, G.; Reek, J. N. H.
Angew. Chem., Int. Ed. 2011, 50, 7342−7345.
Hydroformylation; Kluwer Academic Publishers: Dordrecht, 2000.
4) van Leeuwen, P. W. N. M.; Roobeek, C. F. J. Organomet. Chem.
983, 258, 343−350.
5) van der Veen, L. A.; Kamer, P. C. J.; van Leeuwen, P. W. N. M.
Angew. Chem., Int. Ed. 1999, 38, 336−338.
6) Klein, H.; Jackstell, R.; Wiese, K.-D.; Borgmann, C.; Beller, M.
Angew. Chem., Int. Ed. 2001, 40, 3408−3411.
(
1
(
(37) Kleij, A. W.; Kuil, M.; Tooke, D. M.; Lutz, M.; Spek, A. L.; Reek,
J. N. H. Chem.Eur. J. 2005, 11, 4743−4750.
(38) Kuil, M.; Puijk, I. M.; Kleij, A. W.; Tooke, D. M.; Spek, A. L.;
Reek, J. N. H. Chem. Asian J. 2009, 4, 50−57.
(
(39) The titration curve shape suggests that the dissociation constant
(reciprocal binding constant) is orders of magnitude greater than the
2
862
dx.doi.org/10.1021/ja211455j | J. Am. Chem. Soc. 2012, 134, 2860−2863

Journal of the American Chemical Society
Communication
5
host and guest concentrations, i.e., K ≫ 10 . This is much greater than
expected for the coordination polymer and consistent with
simultaneous binding within a discrete structure.
(
40) For the ligand-free reaction, the regiomeric ratio a:b is 62:38.
2
863
dx.doi.org/10.1021/ja211455j | J. Am. Chem. Soc. 2012, 134, 2860−2863