Click-connected ligand scaffolds: macrocyclic chelates for asymmetric hydrogenation

Article abstract of DOI:10.1021/ol702890s

Click chemistry is used to construct ligand scaffolds for a series of chiral diphosphites. Enantioselectivity as high as 97% ee is obtained using these click ligands in rhodium-catalyzed asymmetric hydrogenation. Control experiments and spectroscopic data suggest that a 16-membered P,P-macrocyclic Rh(1) chelate is formed.

Full text of DOI:10.1021/ol702890s

ORGANIC
LETTERS
2008
Vol. 10, No. 4
545-548
Click-Connected Ligand Scaffolds:
Macrocyclic Chelates for Asymmetric
Hydrogenation
Qing Zhang and James M. Takacs*
Department of Chemistry, UniVersity of NebraskasLincoln,
Lincoln, Nebraska 68588-0304
jtakacs1@unl.edu
Received November 30, 2007
ABSTRACT
Click chemistry is used to construct ligand scaffolds for a series of chiral diphosphites. Enantioselectivity as high as 97% ee is obtained
using these click ligands in rhodium-catalyzed asymmetric hydrogenation. Control experiments and spectroscopic data suggest that a 16-
membered P,P-macrocyclic Rh(I) chelate is formed.
The goal of all chiral ligand designs is to (i) create a chiral
“catalytic pocket” (topography) around the metal to direct
stereochemistry and (ii) impart the appropriate electronic
characteristics at that metal center for efficient catalysis. Most
modular ligand designs start with one or a small set of
“interesting scaffolds” and sequentially append the ligating
groups.1 The challenge is to find the “correct ligating group”,
one appropriately oriented by the selected scaffold to form
an efficient catalyst.
asymmetric catalysis. Put to practice, the ligand scaffold is
constructed by connecting subunits, each typically a unique
structure and ligating group, via a facile chemical reaction
or suitable intermolecular association.3 Combining different
subunits creates a diverse set of scaffolds and ligating groups.
The alkyne-azide [3+2]-cycloaddition reaction, i.e., “click
chemistry”,4 has been applied to a wide range of problems
for which the advantages of a fast, clean coupling reaction
can be exploited;5 several applications to the preparation of
novel ligands are among them.6 For example, the preparation
and use of chelating P,N-ligands was recently described by
Zhang7 (e.g., 1) and Reek8 (e.g., 2). Kann and co-workers9
recently reported the use of click chemistry for the prepara-
tion of a library of P-chirogenic ligands, including a diphos-
phine derivative, 3.
Ligand scaffold optimization is an alternative design
strategy for improving catalyst efficiency.2 It relies upon the
ability to generate a collection of ligand scaffolds that define
a diverse set of topographies or chiral space to be explored.
The overall strategy is simple; use the nature of the ligating
group to define the electronic nature of the metal-ligand
interaction and a chiral surface, and then the ligand scaffold
is varied to find an optimal orientation of that surface for
(3) (a) Sandee, A. J.; Reek, J. N. H. Dalton Trans. 2006, 3385-3391.
(b) Reyes, S. J.; Burgess, K. Chem. Soc. ReV. 2006, 35, 416-423. (c)
Wilkinson, M. J.; van Leeuwen, P. W. N. M.; Reek, J. N. H. Org. Biomol.
Chem. 2005, 3, 2371-2383. (d) Breit, B. Angew. Chem., Int. Ed. 2005, 44,
6816-6825.
(4) (a) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem., Int.
Ed. 2001, 40, 2004-2021. (b) Rostovtsev, V. V.; Green, L. G.; Fokin, V.
V.; Sharpless, K. B. Angew. Chem., Int. Ed. 2002, 41, 2596-2599.
(5) Moses, J. E.; Moorhouse, A. D. Chem. Soc. ReV. 2007, 36, 1249-
1262.
(1) (a) Gennari, C.; Piarulli, U. Chem. ReV. 2003, 103, 3071-3100.
(2) (a) Moteki, S. A.; Takacs, J. M. Angew. Chem., Int. Ed. 2007, 46, in
press. (b) Takacs, J. M.; Chaiseeda, K.; Moteki, S. A.; Reddy, D. S.; Wu,
D.; Chandra, K. Pure Appl. Chem. 2006, 78, 501-509. (c) Takacs, J. M.;
Reddy, D. S.; Moteki, S. A.; Wu, D.; Palencia, H. J. Am. Chem. Soc. 2004,
126, 4494-4495.
10.1021/ol702890s CCC: $40.75
© 2008 American Chemical Society
Published on Web 01/12/2008

Table 1. Screening Click Ligands in the Rhodium-Catalyzed
Asymmetric Hydrogenation of Enamide 9
Our application of click chemistry to ligand scaffold
construction is shown in Figure 1. The 2-, 3-, and 4-hydroxy-
entry
CL8
% yield
% ee
1
2
3
4
5
6
8ia
8ib
8iia
8iib
8iiia
8iiib
99
99
97
83
30
52
88
96
95
89
91
85
run under a standard set of reaction conditions (30 psi H2,
DCM, rt, 20 h). Each click ligand affords the secondary
amine derivative (S)-10 with enantiomeric excesses ranging
from 85% (entry 6, CL8iiib) to 96% (entry 2, CL8ib). In
addition to the variation in enantiomeric excess, we find the
degree of conversion obtained under these standard reaction
conditions varies as a function of the ligand scaffold. The
chemical yield of (S)-10 varies from 30% (entry 5, CL8iiia)
to near-quantitative (entries 1 and 2, CL8ia and CL8ib).
Click ligands CL8ib (99%, 96% ee) and CL8iia (97% yield,
95% ee) give the overall most promising results.
Figure 1. Facile, modular route to novel ligand scaffolds via click
chemistry.
phenyl azides 4i-iii are coupled with the 3- and 4-(hy-
droxymethyl)phenyl acetylenes (5a,b) to afford the six
isomeric diols represented by structure 6. Coupling each of
the diols with phosphorus trichloride and (R)-tetramethyl-
2,2′-bisphenol10 (7) affords a set of isomeric click ligands
CL8. These chiral diphosphites differ only with respect to
the structure of the ligand scaffold.
There are four potential ligating sites in CL8ib (Figure
2) raising questions as to whether it functions as a P,N- or
Asymmetric hydrogenation is a common testing ground
for evaluating new ligand designs,11 and the six diphosphites
were used in conjunction with Rh(nbd)2(BF4) to effect the
hydrogenation of enamide 9 (Table 1).12 The reactions were
(6) (a) Suijkerbuijk, B. M. J. M.; Aerts, B. N. H.; Dijkstra, H. P.; Lutz,
M.; Spek, A. L.; van Koten, G.; Klein Gebbink, R. J. M. Dalton Trans.
2007, 1273-1276. (b) Bastero, A.; Font, D.; Pericas, M. A. J. Org. Chem.
2007, 72, 2460-2468. (c) Mindt, T. L.; Struthers, H.; Brans, L.; Anguelov,
T.; Schweinsberg, C.; Maes, V.; Tourwe, D.; Schibli, R. J. Am. Chem. Soc.
2006, 128, 15096-15097. (d) Chan, T. R.; Hilgraf, R.; Sharpless, K. B.;
Fokin, V. V. Org. Lett. 2004, 6, 2853-2855.
(7) (a) Liu, D.; Gao, W.; Dai, Q.; Zhang, X. Org. Lett. 2005, 7, 4907-
4910. (b) Dai, Q.; Gao, W.; Liu, D.; Kapes, L. M.; Zhang, X. J. Org. Chem.
2006, 71, 3928-3934. (c) See also: Fukuzawa, S.-i.; Oki, H.; Hosaka, M.;
Sugasawa, J.; Kikuchi, S. Org. Lett. 2007, 9, 5557-5560.
Figure 2. Diminished reactivity and enantioselectivity with triazole
monophosphites 11 and 12 suggests that both phosphite moieties
in CL8ib are involved in the active catalyst.
(8) Detz, R. J.; Arevalo Heras, S.; De Gelder, R.; Van Leeuwen, P. W.
N. M.; Hiemstra, H.; Reek, J. N. H.; Van Maarseveen, J. H. Org. Lett.
2006, 8, 3227-3230.
P,P-ligand, or perhaps a tri- or tetradentate ligand,13 if it
indeed chelates at all. If P,N-chelation is important, then it
seems likely that either or both of the truncated triazole
monophosphites 11 and 12 would give results comparable
to those obtained with CL8ib. Neither does. Two ligand to
metal ratios were examined for 11 and 12 in the hydrogena-
tion of 9. 11 affords a poor catalyst at either a 1:1 or 2:1
ratio. 12 exhibits poor reactivity at the 2:1 12/Rh(I) ratio.
The reactivity is improved at the 1:1 ratio, suggesting that
P,N-ligation may be relevant for 12, although it is not clear
how its structure can accommodate P,N-chelation within the
(9) Dolhem, F.; Johansson, M. J.; Antonsson, T.; Kann, N. J. Comb.
Chem. 2007, 9, 477-486.
(10) (a) Hua, Z.; Vassar, V. C.; Ojima, I. Org. Lett. 2003, 5, 3831-
3834. (b) Rubio, M.; Suarez, A.; Alvarez, E.; Pizzano, A. Chem. Commun.
2005, 628-630. (c) Rubio, M.; Vargas, S.; Suarez, A.; Alvarez, E.; Pizzano,
A. Chem. Eur. J. 2007, 13, 1821-1833.
(11) Blaser, H.-U.; Malan, C.; Pugin, B.; Spindler, F.; Steiner, H.; Studer,
M. AdV. Synth. Catal. 2003, 345, 103-151.
(12) For recent examples of the use of diphosphites in rhodium-catalyzed
asymmetric hydrogenation, see: (a) Reetz, M. T.; Mehler, G.; Bondarev,
O. Chem. Commun. 2006, 2292-2294. (b) Junge, K.; Hagemann, B.;
Enthaler, S.; Erre, G.; Beller, M. ARKIVOC 2007, 50-66. (c) Axet, M. R.;
Benet-Buchholz, J.; Claver, C.; Castillon, S. AdV. Synth. Catal. 2007, 349,
1983-1998. (d) Marson, A.; Freixa, Z.; Kamer, P. C. J.; van Leeuwen, P.
W. N. M. Eur. J. Inorg. Chem. 2007, 4587-4591.
546
Org. Lett., Vol. 10, No. 4, 2008

constraints of a 1:1 complex. Nevertheless, even using the
better ratio, the yield and enantioselectivity are still lower
than that observed using CL8ib. These results suggest that
P,P-chelation is the important mode for CL8ib; that is, both
phosphite moieties are involved in the active catalyst.14
It is tempting to assume that the active catalyst is a 1:1
CL8ib:Rh(I) complex, rather than 2:2 or some higher
multiple complex or an ensemble of oligomers. While many
macrocyclic metal-ligand chelates have been characterized
and others invoked to explain efficient asymmetric catalysis,15
the 1:1 complex with CL8ib requires a 16-membered P,P-
macrocyclic chelate that preliminary modeling studies indi-
cate should be quite strained.16 Nonetheless, other data are
consistent with its formation.
complex forms preferentially or the diastereomeric catalysts
exhibit significantly different rates, then nonlinear effects
are expected.
The rhodium-catalyzed hydrogenation of enamide 9 was
carried out using various ratios of the enantiomers of CL8ib.
As shown in Figure 4, we find a linear relation between the
The first piece of evidence derives from the work of Kagan
on nonlinear effects in asymmetric catalysis.17 As outlined
in Figure 3, if the active catalyst is the 1:1 complex, a mixture
Figure 4. Linear relation between the enantiomeric purity of CL8ib
and the enantioselectivity observed in the rhodium-catalyzed
asymmetric hydrogenation of enamide 9.
enantiomeric purity of the click ligand and the enantiose-
lectivity of the hydrogenation. This result, while not unam-
biguous proof, is consistent with the 1:1 16-membered P,P-
macrocyclic Rh(I) chelate being relevant to the efficient
asymmetric catalysis observed.
The 31P NMR spectrum of the free click ligand CL8ib
shows singlets at 130.6 and 135.3 ppm. As shown in Figure
5, the (CL8ib)Rh(nbd)(BF4) complex shows two sets of
resonances, 127.0 and 148.1 ppm, in a 1:1 ratio; each is a
doublet of doublets. The coupling constants JRh,P(1) ) 221
Hz, JRh,P(2) ) 234 Hz, and JP(1),P(2) ) 77 Hz are con-
sistent with the chelated structure shown. The mass spec-
trum shows a peak corresponding to the complex minus BF4;
its isotopic distribution pattern is in good agreement with
theory. High-resolution mass spectrometry (FAB, 3-NBA
matrix) determines the exact mass as m/z 1016.2417, a value
in close agreement with the 1016.2465 theoretical value for
C55H53N3O6P2Rh (M - BF4).
The phosphite moieties in CL8ib differ slightly, one a ben-
zyl phosphite, the other a phenyl phosphite. Several related
structures were prepared and evaluated (Figure 6). For
example, a derivative in which a methyl substituent is present
at the benzylic position was prepared. This added methyl
substituent makes the ligand scaffold chiral; the R- and
S-scaffolds afford the diastereomeric diphosphites CL13 and
CL14, respectively. The former click ligand shows somewhat
diminished enantioselectivity while the latter gives results
comparable to those obtained with the parent ligand CL8ib.
Substituting a phenyl for the benzyl phosphite moiety
markedly diminishes the efficiency of the ligand. CL15 ex-
hibits low turnover (7% yield, 72% ee). In contrast, replacing
the phenyl phosphite with a second benzyl phosphite (i.e.,
CL16) has relatively little effect (99% yield, 92% ee).
Slightly improved results are found with CL17 (99% yield,
97% ee). This diphosphite is isomeric with CL8ib and is
Figure 3. Predicted consequences of 1:1 or 2:2 ligand/metal
complexation of CL8ib with Rh(nbd)(BF4).
of enantiomer ligands will give a mixture of enantiomeric
1:1 complexes or, equivalently, a mixture of enantiomeric
catalysts. Thus, a linear relationship between the observed
enantioselectivity of the catalyzed reaction and the enantio-
meric purity of the click ligand used is expected; nonlinear
effects are precluded. However, if the 2:2 complex (or some
higher homologue or an oligomeric mixture) competes, then
at least three species are expected to contribute, the enan-
tiomeric all-R and all-S complexes and the diastereomeric
R/S-mixed complex. If, as is often the case, the R/S-mixed
(13) Tetradentate coordination to Rh(I) seems unlikely on the basis of
structural constraints and the fact that these ligands function efficiently in
asymmetric hydrogenation, a reaction in which tetradentate coordination is
unsuitable. See: Landis, C. R.; Halpern, J. J. Am. Chem. Soc. 1987, 109,
1746-1754.
(14) For CL8ib, the catalytic activity drops off substantially at higher
ligand to Rh(I) ratios.
(15) (a) Hattori, G.; Hori, T.; Miyake, Y.; Nishibayashi, Y. J. Am. Chem.
Soc. 2007, 129, 12930-12931. (b) Birkholz, M.-N.; Dubrovina, N. V.; Jiao,
H.; Michalik, D.; Holz, J.; Paciello, R.; Breit, B.; Boerner, A. Chem. Eur.
J. 2007, 13, 5896-5907. (c) Wassenaar, J.; Reek, J. N. H. Dalton Trans.
2007, 3750-3753. (d) Jiang, X.-B.; Lefort, L.; Goudriaan, P. E.; de Vries,
A. H. M.; van Leeuwen, P. W. N. M.; de Vries, J. G.; Reek, J. N. H. Angew.
Chem., Int. Ed. 2006, 45, 1223-1227. (e) Weis, M.; Waloch, C.; Seiche,
W.; Breit, B. J. Am. Chem. Soc. 2006, 128, 4188-4189.
(16) We thank the reviewer for bringing this to our attention.
(17) Kagan, H. B. Synlett 2001, 888-899.
Org. Lett., Vol. 10, No. 4, 2008
547

Figure 5. 31P NMR spectrum of (CL8ib)Rh(nbd)(BF4): JRh,P(1) ) 221 Hz, JRh,P(2) ) 234 Hz, and JP(1),P(2) ) 77 Hz.
prepared by swapping the alkyne and azide subunits; that
other common test substrates. Each of the substrates in Figure
is, the alkyne precursor incorporates the p-benzyl alcohol
7 gives 90% ee or greater (0.5 mmol substrate, 0.2 mol %
catalyst, DCM, 30 psi H2, 16-20 h, rt).
In summary, click chemistry lends itself to the ligand
scaffold optimization approach to chiral catalyst discovery
and development. The facile alkyne-azide [3 + 2]-cycload-
dition reaction was used to prepare a small library of chiral
triazole diphosphites. While these novel chiral ligands differ
only in the scaffold structure, their performance with respect
to yield and asymmetric induction varies significantly in a
common test case for asymmetric catalysts, rhodium-
catalyzed asymmetric hydrogenation. Enantioselectivity as
high as 97% ee is achieved. 31P NMR and mass spectral
analyses, as well as the results obtained using truncated
monophosphites and the lack of a nonlinear effect, suggest
an important role for a 16-membered P,P-macrocyclic
(CL8ib)Rh(I) complex in the reaction, although preliminary
modeling studies find this proposed complex to be quite
strained. The related CL8iia complex is less strained and
exhibits similar reactivity. Further studies into the nature of
these complexes and their role in catalysis are in progress.
Figure 6. Fine-tuning the click ligand scaffold structure.
subunit and the azide subunit incorporates the o-phenol sub-
unit. Click ligand CL17 was used in a larger scale hydroge-
nation of enamide 9 at a lower catalyst load (1 mmole of 9,
0.1 mol % catalyst) but no attempt to otherwise further opti-
mize the reaction conditions (DCM, 30 psi H2, 10 h, rt).
Under these conditions, CL17 gives 92% yield of (S)-10 with
96% ee. Click ligand CL17 is also effective with several
Acknowledgment. Financial support for this research
under grants from the Nebraska Research Initiative, ACS-
PRF (47257-AC1), and NSF (CHE-0316825) is gratefully
acknowledged. We thank T. A. George (UNL Chemistry)
for the loan of equipment, S. Smith (UNL Chemistry) for
preparing tetramethyl-2,2′-bisphenol, and the NSF (CHE-
0091975, MRI-0079750) and NIH (SIG-1-510-RR-06307)
for the NMR spectrometers used in these studies carried out
in facilities renovated under NIH RR016544.
Supporting Information Available: Experimental pro-
cedures and spectral characterization data for the click
ligands. This material is available free of charge via the
Internet at http://pubs.acs.org.
Figure 7. Efficient (CL17)Rh(nbd)(BF4)-catalyzed asymmetric
hydrogenation of related substrates.
OL702890S
548
Org. Lett., Vol. 10, No. 4, 2008