Simple Strategy for the Immobilization of Dirhodium Tetraprolinate Catalysts Using a Pyridine-Linked Solid Support

Article abstract of DOI:10.1021/ja0393067

Dirhodium tetracarboxylates are readily immobilized on agitation in the presence of highly cross-linked polystyrene resins with a pyridine attachment. A systematic study demonstrates that the polymer backbone, the linker, the terminal pyridine group, and

Full text of DOI:10.1021/ja0393067

Published on Web 03/12/2004
Simple Strategy for the Immobilization of Dirhodium
Tetraprolinate Catalysts Using a Pyridine-Linked Solid
Support
Huw M. L. Davies,* Abbas M. Walji, and Tadamichi Nagashima†
Contribution from the Department of Chemistry, UniVersity at Buffalo, The State UniVersity of
New York, Buffalo, New York 14260-3000
Received October 29, 2003; E-mail: hdavies@acsu.buffalo.edu
Abstract: Dirhodium tetracarboxylates are readily immobilized on agitation in the presence of highly cross-
linked polystyrene resins with a pyridine attachment. A systematic study demonstrates that the polymer
backbone, the linker, the terminal pyridine group, and the catalyst structure all contribute to the efficiency
of dirhodium catalyst immobilization. The immobilization is considered to be due to the combination of
ligand coordination and encapsulation. The dirhodium tetraprolinate catalysts, Rh2(S-DOSP)4 (1a),
Rh2(S-TBSP)4 (1b), and Rh2(S-biTISP)2 (2), are all efficiently immobilized. The resulting heterogeneous
complexes are very effective catalysts for asymmetric cyclopropanation between methyl phenyldiazoacetate
and styrene, and under optimized conditions they can be recycled five times with virtually no loss in
enantioselectivity. The three-phase test studies indicated that a very slow reaction occurs when both the
catalyst and the diazo compound were immobilized, but the slow rate precluded the likelihood that the
cyclopropanation was predominately occurring by a release-and-capture mechanism.
Introduction
the utilization of immobilized versions of these catalysts. The
design strategy for effective asymmetric induction with these
The challenges of chiral catalyst immobilization have attracted
the interest of a wide range of polymer and synthetic chemists.1
Over the past few decades, the field of carbenoid chemistry has
witnessed explosive growth in the development of chiral
catalysts.2,3 Some of the commonly used catalysts for asym-
metric carbenoid reactions derived from copper, ruthenium, and
rhodium(II) complexes have been immobilized on organic and
inorganic supports.4 Most immobilization strategies have relied
on covalent attachments between chiral ligands and the solid
supports.4 In general, the method of immobilization, choice of
polymer matrix, and point of attachment to the catalyst influence
the chiral environment of the immobilized species.1a,5
Dirhodium tetraprolinate catalysts Rh2(S-DOSP)4 (1a),
Rh2(S-TBSP)4 (1b), and Rh2(S-biTISP)2 (2) (Figure 1) developed
in our laboratories have proven to be exceptional catalysts for
homogeneous asymmetric intermolecular cyclopropanation and
C-H activation reactions of donor/acceptor substituted car-
benoids 3 (Figure 2).6 Consequently, we have begun to explore
catalysts is based on the coordination of several identical chiral
ligands of low symmetry to the central bimetallic core to produce
a chiral complex of higher symmetry.6d Therefore, immobiliza-
tion methods would be required that would avoid modification
of any of the ligands because this would destroy the high-
symmetry environment of the chiral catalyst.
We have communicated a novel approach for the immobiliza-
tion of the dirhodium tetraprolinates using the general strategy
shown in Scheme 1.7,8 In this method, an appropriate polymer
backbone functionalized with a pyridine group was used to
coordinate to one rhodium while the other rhodium continued
to be an active site for catalysis. This approach has been very
(4) (a) Bergbreiter, D. E.; Chen, B.; Morvant, M. Tetrahedron Lett. 1991, 32,
2731. (b) Doyle, M. P.; Eismont, M. Y.; Bergbreiter, D. E.; Gray, H. N. J.
Org. Chem. 1992, 57, 6103. (c) Burguette, M. I.; Fraile, J. M.; Garcia, J.
I.; Garcia-Verdugo, E.; Luis, S. V.; Mayoral, J. A. Org. Lett. 2000, 2, 3905.
(d) Burguette, M. I.; Fraile, J. M.; Garcia, J. I.; Garcia-Verdugo, E.; Luis,
S. V.; Mayoral, J. A. J. Org. Chem. 2001, 66, 8893. (e) Diaz-Requejo, M.
M.; Belderrain, T. R.; Nicasio, M. C.; Perez, P. J. Organometallics 2000,
19, 285. (f) Glos, M.; Reiser, O. Org. Lett. 2000, 2, 2045. (g) Annunziata,
R.; Benaglia, M.; Cinquini, M.; Cozzi, F.; Pitillo, M. J. Org. Chem. 2001,
66, 3160. (h) Doyle, M. P.; Timmons, D. J.; Tumonis, S. J.; Gau, H.;
Blossey, C. E. Organometallics 2002, 21, 1747. (i) Cornejo, A.; Fraile, J.
M.; Garcia, J. I.; Garcia-Verdugo, E.; Gil, M. J.; Legarreta, G.; Luis, V.
S.; Martinez-Merino, V.; Mayoral, J. A. Org. Lett. 2002, 4, 3927. For
reviews, see: (j) Rechavi, D.; Lemaire, M. Chem. ReV. 2002, 102, 3467.
(k) Bergbreiter, D. E. In Chiral Catalyst Immobilization and Recycling;
De Vos, D. E., Vankelecom, I. F. J., Jacobs, P. A., Eds.; Wiley-VCH:
Weinheim, Germany, 2000; p 43.
† Current address: Fluorous Technologies, 970 William Pitt Way, PA
15238.
(1) (a) Chiral Catalysts Immobilization and Recycling; De Vos, D. E.,
Vankelecom I. F. J., Jacobs, P. A., Eds.; Wiley-VCH: Weinheim, Germany,
2000. (b) Clapham, B.; Reger, T. S.; Janda, K. D. Tetrahedron 2001, 57,
4637. (c) Sherrington, D. C. J. Polym. Sci., Part A: Polym. Chem. 2001,
39, 2364. (d) Flynn, D. L.; Devraj, R. V.; Parlow, J. J. In Solid-Phase
Organic Synthesis; Burgess, K., Ed.; Wiley-Interscience: New York, 2000;
pp 149-194. (e) de Miguel, Y. D.; Bryle, E.; Margue, R. G. J. Chem.
Soc., Perkin Trans. 1 2001, 3085-3094.
(2) Davies, H. M. L.; Antoulinakis, E. G. Org. React. 2001, 57, 1-326.
(3) (a) Davies, H. M. L.; Beckwith, R. J. W. Chem. ReV. 2003, 103, 2861. (b)
Doyle, M. P.; McKervey, M. A.; Ye, T. In Modern Catalytic Methods for
Organic Synthesis with Diazo Compounds; Wiley-Interscience: New York,
1998.
(5) Altava, B.; Burguete, M. I.; Garcia-Verdugo, E.; Luis, S. V.; Vincent, M.
J.; Mayoral, J. A. React. Polym. 2001, 48, 25.
(6) (a) Reference 3a. (b) Davies, H. M. L.; Antoulinakis, E. G. J. Organomet.
Chem. 2001, 617-618, 47. (c) Davies, H. M. L. J. Mol. Catal 2002, 189,
125. (d) Davies, H. M. L. Eur. J. Org. Chem. 1999, 2459.
(7) Nagashima, T.; Davies, H. M. L. Org. Lett. 2002, 4, 1989.
(8) Davies, H. M. L.; Walji, A. M. Org. Lett. 2003, 5, 479.
9
10.1021/ja0393067 CCC: $27.50 © 2004 American Chemical Society
J. AM. CHEM. SOC. 2004, 126, 4271-4280
4271

A R T I C L E S
Davies et al.
Scheme 1. Strategy for Catalyst Immobilization
successful for asymmetric intermolecular cyclopropanation7 and
asymmetric intermolecular C-H activation.8 During the course
of these studies, a number of puzzling observations were made.
In the original studies7,8 the Argopore9 resin was chosen as
the polymer backbone due to its rigid, highly cross-linked,
macroporous polystyrene framework. Such a polymer would
allow easy access for reagents into reaction sites without having
to swell the beads.10 Consequently, hydrocarbon solvents, which
are ideal for high asymmetric induction in solution-phase
reactions of donor/acceptor substituted carbenoids would be
applicable for reactions with the immobilized catalysts.6a The
Wang linker11 was chosen because it should ensure sufficient
spacing between the catalyst and the polymer to minimize
unfavorable steric interactions. The basic approach to introduce
the pyridine terminal group is shown in eq 1. The hydroxyl
group of the Wang linker 6 was converted to the bromo
derivative 7 and then reacted with 4-pyridyl carbinol to form
the 4-benzyloxymethyl-pyridine resin 8. The pyridine loading
of the Argopore-Wang resin 8 was determined to be 0.30
mmol/g by cleavage of the 4-pyridyl carbinol using trifloroacetic
acid/dichloromethane (1:1).
Figure 1. Dirhodium tetraprolinate catalysts.
Figure 2. Donor/acceptor substituted carbenoids.
The first series of experiments were directed toward the
development of optimum reaction conditions for catalyst
immobilization. These test reactions were conducted using
Rh2(S-TBSP)4 as catalyst. In our original communication,7 the
immobilization of Rh2(S-TBSP)4 on the resin 8 was carried out
using dichloromethane as solvent. Gentle shaking of 8 in a
solution of Rh2(S-TBSP)4 (2 equiv) in dichloromethane for 3.5
h caused the color of the resin to change from a pale yellow to
purple. After extensive washing with dichloromethane (9×), the
8a-Rh2(S-TBSP)4 resin remained purple, indicating the coor-
dination of the catalyst to the pyridine and its immobilization
Pyridine coordination is known to deactivate dirhodium com-
plexes, and control experiments of homogeneous reactions
clearly revealed that the prolinate catalysts were deactivated by
pyridine coordination.7 Furthermore, replacement of the pyridine
linker with benzene did not block the immobilization of the
catalyst. Therefore, the exact mechanism of the immobilization
process remained very uncertain. A further concern with the
earlier studies was the utilization of an excess of the expensive
catalysts. Herein, we describe a systematic study of the immobi-
lized catalysts, which defines what features of the catalyst,
polymer, linker, and terminal group are optimum for im-
mobilization and catalytic activity. Furthermore, a more practical
and cost-effective method for catalyst immobilization is de-
scribed.
(9) Purchased from Argonaut Technologies (www.argotech.com).
(10) Dorwald, F. Z. In Organic Synthesis on Solid Phase; Wiley-VCH:
Weinheim, Germany, 2000.
(11) Wang, S.-S. J. Am. Chem. Soc. 1973, 95, 1328.
9
4272 J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004

Immobilizing Dirhodium Tetraprolinate Catalysts
A R T I C L E S
onto the solid support (eq 2). The change in mass after drying
the 8a-Rh2(S-TBSP)4 was used to determine catalyst loading
(0.18 mmol/g), and this has now been confirmed by ICP-AES
analysis (0.20 mmol/g). A concern with this original method
was the use of dichloromethane as solvent because dichlo-
romethane is not the ideal solvent for high asymmetric induction
with Rh2(S-TBSP)4.6d A modified immobilization method was
developed in which toluene was used as the solvent in place of
dichloromethane. This gave 8b-Rh2(S-TBSP)4 with an ICP-AES
determined catalyst loading of 0.23 mmol/g (eq 3). For
immobilized catalysts 8a-Rh2(S-TBSP)4 and 8b-Rh2(S-TBSP)4,
an excess amount of catalyst to resin was used in the im-
mobilization step. To further improve this technology, the
immobilization was attempted with the catalyst as the limiting
reagent in both solvents. This process was also very effective
as 8c-Rh2(S-TBSP)4 (eq 4) and 8d-Rh2(S-TBSP)4 (eq 5) were
formed with catalyst loadings of 0.12-0.16 mmol/g, which
corresponds to 72-86% catalyst immobilization.
In the original communication,7 the evaluation of 8a-
Rh2(S-TBSP)4 was conducted in standard glassware, using the
reaction of methyl phenyldiazoacetate with styrene as the stand-
ard reaction. All the work described in the current study has
been conducted using an Argonaut Quest-210 parallel synthe-
sizer. This allowed the agitation to be electronically controlled
at a standard rate throughout all the experiments. The immobi-
lized catalysts were recycled up to five times, and the yield,
enantioselectivity, and reaction time were determined for every
other cycle.12 The rate of the reaction was easily monitored vis-
ually because the starting methyl phenyldiazoacetate is yellow
and when the reaction reaches completion the solution becomes
colorless.
Under these new reaction conditions, comparison of the new
variants 8b-d-Rh2(S-TBSP)4 with 8a-Rh2(S-TBSP)4 revealed
some very interesting trends (Table 1). The asymmetric induc-
tion in the reaction catalyzed by 8a-Rh2(S-TBSP)4 was lower
(82% ee) than the corresponding homogeneous reaction cata-
lyzed by Rh2(S-TBSP)4 (87-90% ee). Furthermore, the enan-
tioselectivity dropped to 70% ee in the fourth cycle using
recycled 8a-Rh2(S-TBSP)4. In contrast, 8b-Rh2(S-TBSP)4 re-
sulted in higher enantioselectivity in the first cycle (87% ee)
and retained the value reasonably well over five cycles (87 to
81% ee). Catalysts 8c-Rh2(S-TBSP)4 and 8d-Rh2(S-TBSP)4
behaved very similarly to 8b-Rh2(S-TBSP)4, which demonstrates
that an effective immobilized catalyst is formed when Rh2(S-
TBSP)4 is the limiting agent. Rhodium analysis by ICP-AES
of the initial reaction cycles before product isolation was
conducted to determine the extent of catalyst leaching between
the systems. The reaction mixture (first cycle) from catalyst 8b-
Rh2(S-TBSP)4 contained 1000 ppm Rh, which corresponds to
30% of the initial catalyst charge. The reaction mixture (first
cycle) with catalyst 8d-Rh2(S-TBSP)4 contained 100 ppm Rh,
which is 3% of the initial catalyst charge. The improved
efficiency of the 8d-Rh2(S-TBSP)4 is evident, which provides
reduced product contamination without compromising catalyst
activity and stereoselectivity.
The immobilization of the catalysts and the asymmetric
cyclopropanation are very efficient procedures as illustrated in
Figure 3. The creamy polymer (Figure 3a) rapidly became purple
on exposure to the green solution of Rh2(S-TBSP)4 (Figure 3b).
After being stirred for 40 min, the solution became colorless
(Figure 3c), indicating that the catalyst was efficiently absorbed.
Addition of methyl phenyldiazoacetate formed a yellow solution,
which on gentle stirring caused rapid evolution of nitrogen
(Figure 3d). Within minutes, the solution became colorless
(Figure 3e), and the purple 8d-Rh2(S-TBSP)4 was ready for
filtering and recycling.
(12) Asymmetric cyclopropanations of aryldiazoacetates are routinely highly
diastereoselective. Throughout these studies >92% de was observed and
is therefore not reported for each reaction. (a) Davies, H. M. L.; Bruzinski,
P. R.; Fall, M. J. Tetrahedron Lett. 1996, 37, 4133. (b) Doyle, M. P.; Zhou,
Q.-L.; Charnsangavej, C.; Longoria, M. A.; McKervey, M. A.; Garcia, C.
F. Tetrahedron Lett. 1996, 37, 4129.
One of the most surprising aspects of the original im-
mobilization studies was the discovery that effective im-
9
J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004 4273

A R T I C L E S
Davies et al.
Table 1. Evaluation of Optimum Catalyst Immobilization Procedurea
8a-Rh2(S-TBSP)4
timeb
8b-Rh2(S-TBSP)4
timeb
8c-Rh2(S-TBSP)4
timeb
8d-Rh2(S-TBSP)4
cycle
eec
eec
eec
timeb
eec
1
3
5
10
82
10
10
15
87
83
81
10
10
15
86
84
82
10
10
15
86
83
81
14
73
14d
70d
a
Reaction yields determined by 1H NMR using DMAP as internal standard or isolated yields; in all cases the yields were greater than 74%. See Supporting
b
c
d
Information. In minutes. In percent. Fourth reaction cycle.
(S-TBSP)4 resulted in slightly lower enantioselectivities (84 to
80% ee) in the initial cycle compared to 8d-Rh2(S-TBSP)4 (86%
ee). These catalysts could be recycled five times, but there was
a greater drop in enantioselectivity (71 to 68% ee) for the fifth
cycle compared to 8d-Rh2(S-TBSP)4 (81% ee). The reactions
with non-Wang-based supported catalyst 14d-Rh2(S-TBSP)4
resulted in a severe drop in enantioselectivity even after the
second cycle (83 to 53% ee). Argopore pyridine-supported
catalyst 15d-Rh2(S-TBSP)4 regained activity and was the only
system comparable to the Argopore-Wang pyridine catalyst 8d-
Rh2(S-TBSP)4. The catalyst could efficiently be recycled, with
only a 9% drop in enantioselectivity between the first and fifth
reaction cycles (86 to 77% ee). In comparison, the Argopore
phenyl-supported catalyst 16d-Rh2(S-TBSP)4 displayed a similar
drop in cyclopropane enantioselectivity to the Argopore-Wang
phenyl catalyst 11d-Rh2(S-TBSP)4. Pyridine-functionalized solid
supports 8d-Rh2(S-TBSP)4 and 15d-Rh2(S-TBSP)4 provided
higher catalyst loading and appeared to retain more of the
catalyst on repeated cycles. The positive effect of the pyridine
terminal group and the Wang linker is evident from these
experiments, since both elements are advantageous for maintain-
ing stereoselectivity and high catalytic activity.
Figure 3. Visualization of the cyclopropanation chemistry of 8d-Rh2(S-
TBSP)4. (a) Resin 8 in toluene. (b) Resin in toluene immediately after
addition of Rh2(S-TBSP)4. (c) 8d-Rh2(S-TBSP)4 in toluene after being
shaken for 40 min. (d) 8d-Rh2(S-TBSP)4 in toluene immediately after
addition of methyl phenyldiazoacetate. (e) 8d-Rh2(S-TBSP)4/methyl phe-
nyldiazoacetate after being shaken for 10 min.
mobilization occurred when the terminal pyridine group was
replaced by benzene.7 Therefore, we have undertaken an
extensive study to determine what characteristics of the polymer
backbone, the linker, and the terminal group are required for
effective immobilization of the dirhodium catalysts. First, the
effect of non-pyridine-based linkers was tested, with and without
the Wang linker (Table 2). Argopore-Wang phenyl resin 11 was
included in the study with the Argopore-Wang resin 12 and
Argopore-Wang Br resin 13. The amount of catalyst im-
mobilization was compared with the Argopore Cl resin 14
lacking the Wang linker. The presence of the Wang linker had
a drastic effect on catalyst immobilization, as evident from the
difference in catalyst loading obtained between catalysts 11d-
Rh2(S-TBSP)4 (75%)/12d-Rh2(S-TBSP)4 (71%)/13d-Rh2(S-
TBSP)4 (70%) and 14d-Rh2(S-TBSP)4 (12%). The positive effect
of a pyridine group on catalyst immobilization was verified using
the non-Wang-based pyridine linker 15. Catalyst immobilization
was observed by disappearance of the green color from the
reaction mixture, giving high catalyst immobilization (92%).
The corresponding phenyl linker 16 gave lower catalyst im-
mobilization (29%) compared to its Argopore-Wang phenyl
counterpart 11 (75%), which shows that the Wang linker has a
positive effect on catalyst immobilization.
At this stage, all test systems employed the highly cross-
linked Argopore9 polymer matrix. Previously,7 we proposed a
microencapsulation effect1a,13 due to this polymer, which would
permit the bulky prolinate catalysts to be entrapped within the
macroporous polystyrene framework. Therefore, the next series
of experiments were designed to investigate the effect of the
Argopore resin by substituting the polymer backbone with
various solid supports (Table 4). Initially, immobilization of Rh2-
(S-TBSP)4 on 1% cross-linked polystyrene PS-Wang14 pyridine
resin 17 was attempted. This resin gave very similar catalyst
immobilization (78%) to its Argopore variant 8 (86%), and a
similar catalyst color disappearance was observed in the
experiment. The difference between the Argopore resin and the
1% cross-linked polystyrene resin was more evident with the
phenyl-substituted resin 18. A considerably lower catalyst
loading (46%) compared to the Argopore-Wang phenyl resin
11 (76%) was obtained. For further comparison, commercially
available pyridine functionalized solid supports were tested.
Ortho-substituted pyridine-functionalized silica gel 1914 gave
To determine catalyst efficiency and stereoselectivity, im-
mobilized catalysts 11d-16d-Rh2(S-TBSP)4 were tested in the
standard asymmetric cyclopropanation reaction using 0.5 mol
% catalyst (Table 3). In all cases in which the reaction went to
completion the cyclopropane 10 was formed in >70% yield
and >92% de favoring the E isomer. The Wang-based supported
catalysts 11d-Rh2(S-TBSP)4, 12d-Rh2(S-TBSP)4, and 13d-Rh2-
(13) (a) Kobayashi, S.; Endo, M.; Nagayama, S. Org. Lett. 2001, 3, 2649. (b)
Kobayashi, S.; Endo, M.; Nagayama, S. J. Am. Chem. Soc. 1999, 121,
11229. (c) Nagayama, S.; Endo, M.; Kobayashi, S. J. Org. Chem. 1998,
63, 6094. (d) Kobayashi, S.; Nagayama, S. J. Am. Chem. Soc. 1998, 120,
2985.
(14) Purchased from Aldrich Chemical Co. (www.sigma-aldrich.com).
9
4274 J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004

Immobilizing Dirhodium Tetraprolinate Catalysts
A R T I C L E S
Table 2. Catalyst Immobilization Studies on Argopore-Based Resins
Table 3. Evaluation of Immobilized Catalysts on Argopore Resins
11-16a
minimal catalyst immobilization (8%) while the 1% PS-DMAP
resin 209 and polypyridine JandaJel resin 2114 were not effective
(<1%). To confirm whether the JandaJel polymer15 support is
a viable system for this immobilization method, the JandaJel
Wang14-pyridine resin 22 was prepared and tested. With the
presence of the Wang linker and the pyridine group, the system
is now very efficient at catalyst immobilization (97%). The lack
in catalyst immobilization displayed by the non-Wang pyridine
polymers is very striking and would indicate that the chemical
environment around the pyridine group is crucial for effective
immobilization.
Evaluating these catalysts under the standard conditions
quickly revealed that subtle factors were involved in the
efficiency of the various solid supports (Table 5). Immobilized
catalysts derived from the solid phase containing a terminal
pyridine group on the Wang linker (17d-Rh2(S-TBSP)4 and 22d-
11d-Rh2(S-TBSP)4
12d-Rh2(S-TBSP)4
13d-Rh2(S-TBSP)4
cycle
timeb
eec
timeb
eec
timeb
eec
1
3
5
10
10
10
84
78
71
10
10
14
80
75
68
10
17
14
82
77
68
14d-Rh2(S-TBSP)4
15d-Rh2(S-TBSP)4
16d-Rh2(S-TBSP)4
cycle
timeb
eec
timeb
eec
timeb
eec
1
3
5
10
20d
nde
83
53
nde
10
20
20
86
79
77
10
22
30
82
72
67
(15) (a) Toy, P. H.; Janda, K. D. Tetrahedron Lett. 1999, 40, 6329. (b) Reger,
T. S.; Janda, K. D. J. Am. Chem. Soc. 2000, 122, 6929. (c) Garibay, P.;
Toy, P. H.; Hoeg-Jensen, T.; Janda, K. D. Synlett 1999, 9, 1438. (d)
Takamura, M.; Funabashi, K.; Kanai, M.; Shibasaki, M. J. Am. Chem. Soc.
2001, 123, 6801. (e) Toy, P. H.; Reger, T. S.; Janda, K. D. Aldrichimica
Acta 2000, 33, 87. (f) Toy, P. H.; Reger, T. S.; Garibay, P.; Garno, J. C.;
Malikayil, J. A.; Liu, G.; Janda, K. D. J. Comb. Chem. 2001, 3, 117.
a
Reaction yields determined by 1H NMR using DMAP as internal
standard or isolated yields; in all cases yields were greater than 74%. See
b
c
d
Supporting Information. In minutes. In percent. Second reaction cycle.
e
nd ) not determined (reaction incomplete overnight).
9
J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004 4275

A R T I C L E S
Davies et al.
Table 4. Catalyst Immobilization Studies on Various Solid Supports
a
b
Loading determined by ICP analysis for Rh (mmol/g). ICP-AES analysis of the resin beads gave 1.04 ppm Rh, which corresponds to less than 1%
immobilization.
Table 5. Evaluation of Immobilized Rh2(S-TBSP)4 on Various Polymer Supportsa
17d-Rh2(S-TBSP)4
timeb
18d-Rh2(S-TBSP)4
timeb
19d-Rh2(S-TBSP)4
timeb
22d-Rh2(S-TBSP)4
timeb
cycle
eec
eec
eec
eec
1
3
5
20
20
35
82
81
74
25
87
77
76
10
120
overnightf
78
66
54
15
15
20
85
79
77
45d
overnighte
a
Reaction yields determined by 1H NMR using DMAP as internal standard or isolated yields; in all cases yields were greater than 70%. See Supporting
b
c
d
e
f
Information. In minutes. In percent. Second reaction cycle. Third reaction cycle. Fourth reaction cycle.
Rh2(S-TBSP)4) displayed the best catalytic activity. These
catalysts could be recycled five times, and the enantioselectivity
is very comparable to the Argopore-Wang pyridine catalyst 8d-
Rh2(S-TBSP)4. Although catalyst 18d-Rh2(S-TBSP)4 function-
9
4276 J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004

Immobilizing Dirhodium Tetraprolinate Catalysts
A R T I C L E S
Table 6. Immobilization of Achiral Catalysts on Argopore-Wang Pyridine 8
polymer
(mmol)
catalyst
(mmol)
loading
(mmol/g)a
immobilization
(%)
catalyst
catalyst
Rh2(OAc)4
Rh2(acam)4
Rh2(Oct)4
Rh2(OPiv)4
Rh2(TFA)4
Rh2(TPA)4
0.018
0.018
0.018
0.018
0.018
0.018
0.014
0.014
0.014
0.014
0.014
0.014
0.08
0.06
0.17
0.16
0.16
0.13
37
29
86
79
80
71
8c-Rh2(OAc)4
8c-Rh2(acam)4
8c-Rh2(Oct)4
8c-Rh2(OPiv)4
8c-Rh2(TFA)4
8c-Rh2(TPA)4
a
Loading determined by ICP analysis for Rh (mmol/g).
Table 7. Evaluation of Achiral Immobilized Catalysts
alized with a phenyl group on the Wang linker furnished high
enantioselectivity in the initial cycle, catalytic activity and
enantioselectivity diminished by the third reaction cycle. The
silica gel-based immobilized catalyst 19d-Rh2(S-TBSP)4 was
catalytically active but was not effectively recycled. The
commercial pyridine-linked resins, PS-DMAP (20) and poly-
pyridine JandaJel resin (21), were not effective at immobilizing
Rh2(S-TBSP)4 and did not have sufficient rhodium catalyst
immobilized for even one reaction cycle. The results from these
catalyst immobilization and catalyst evaluation studies demon-
strate that the Argopore-Wang pyridine resin 8 is ideally suited
for immobilization and recovery of Rh2(S-TBSP)4.
catalyst
time (min)
yield (%)
8c-Rh2(OAc)4
8c-Rh2(acam)4
8c-Rh2(Oct)4
240
240
20
77
76
82
85
75
80
8c-Rh2(OPiv)4
8c-Rh2(TFA)4
8c-Rh2(TPA)4
20
40
20
From these studies, we have established that even though a
pyridine group enhances catalyst immobilization, other factors
are also involved. One possibility would be microencapsulation1a,13
of the rhodium prolinate, which could explain why a highly
cross-linked polymer such as the Argopore resin and specific
linkers such as the Wang linker improve catalyst immobilization
and recycling. The microencapsulation may be due to the large
size of Rh2(S-TBSP)4. To test this possibility, the immobilization
of a series of dirhodium catalysts of varying size was explored
(Table 6). The catalysts were immobilized and evaluated for
one reaction cycle in parallel. The small catalysts, Rh2(OAc)4
and Rh2(acam)4, show lower immobilization (29-37%) com-
pared to the larger catalysts Rh2(Oct)4, Rh2(OPiv)4, and Rh2-
(TPA)4 (71-86%). Rh2(TFA)4 is the exception, giving better
immobilization (80%) than the smaller catalysts probably due
to its Lewis acidity. The immobilized small catalysts 8c-Rh2-
(OAc)4 and 8c-Rh2(acam)4 are not very effective catalysts
because they require long reaction times to complete one
reaction cycle (240 min; Table 7). Catalysts 8c-Rh2(Oct)4, 8c-
Rh2(OPiv)4, and 8c-Rh2(TPA)4 show better activity because the
cyclopropanation reactions are complete within 20 min. Catalyst
8c-Rh2(TFA)4 shows better reactivity than the other small
catalysts (40 min), but this may be because this catalyst is more
electrophilic than the others.
was prepared by exposing 8 to an excess of the catalyst in
toluene. 8d-Rh2(R-DOSP)4 was a very effective catalyst for
asymmetric cyclopropanation, capable of being recycled five
times with virtually no drop in the enantioselectivity of the
cyclopropanation. 8d-Rh2(S-biTISP)2 was also effective at
maintaining enantioselectivity over five cycles; however, higher
catalyst loading (1 mol %) was necessary. Even better results
were obtained with 8b-Rh2(S-biTISP)2, which can be recycled
up to 15 times without any drop in yield or enantioselectivity
in the cyclopropanation (Table 9).7
Pyridine coordination is generally considered to cause
deactivation of dirhodium tetracarboxylates and so it is conceiv-
able that either the polymer bound dirhodium complex or
liberated dirhodium complex might be the catalytically active
species. The release-and-capture mechanism17a is a distinct
possibility in this system because it has been established that
the dirhodium tetraprolinates are capable of catalyzing cyclo-
propanations with very high turnover numbers.16 Furthermore,
homogeneous control experiments with 4-alkyl-pyridine 23
confirm the adverse effect of pyridine coordination.7 Under
standard conditions, 1.5 equiv of 4-alkyl-pyridine 23 added to
the reaction mixtures (Table 10) deactivates the rhodium
catalysts. Product yields are lower in both cases, and the reaction
rate is greatly affected in the Rh2(S-biTISP)2/23 coordinated
system.
The detailed studies with Rh2(S-TBSP)4 indicated that Ar-
gopore-Wang pyridine solid phase 8 was optimum for catalyst
immobilization. In the previous communications, Rh2(S-biTISP)2
was shown to be the most robust recoverable catalyst for
asymmetric cyclopropanation7 and Rh2(S-DOSP)4 was shown
to be the most effective catalyst for asymmetric C-H activa-
tion.8 The immobilized versions 8d-Rh2(R-DOSP)4 and 8d-
Rh2(S-biTISP)2 were prepared by exposing 8 to the catalyst as
the limiting agent in toluene (Table 8). Also, 8b-Rh2(S-biTISP)2
To determine the phase of the active catalytic species, a
Rebek-type three-phase test was conducted.17 Our experimental
(16) Davies, H. M. L.; Venkataramani, C. Org. Lett. 2003, 5, 1403.
(17) (a) Davies, I. W.; Matty, L.; Hughes, D. L.; Reider, P. J. J. Am. Chem.
Soc. 2001, 123, 10139. (b) Rebek, J.; Gavina, F. J. Am. Chem. Soc. 1974,
96, 7112. (c) Rebek, J.; Brown, D.; Zimmerman, S. J. Am. Chem. Soc.
1975, 97, 454. (d) Rebek, J., Jr. Tetrahedron 1979, 35, 723.
9
J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004 4277

A R T I C L E S
Davies et al.
Table 8. Immobilization of Chiral Dirhodium Tetraprolinates Rh2(R-DOSP)4 and Rh2(S-biTISP)2
polymer
(mmol)
catalyst
(mmol)
loading
(mmol/g)a
immobilization
(%)
catalyst
catalyst
Rh2(R-DOSP)4
Rh2(S-biTISP)2
Rh2(S-biTISP)2
0.018
0.018
0.023
0.014
0.014
0.030
0.12
0.11
0.17b
74
68
c
8d-Rh2(R-DOSP)4
8d-Rh2(S-biTISP)2
8b-Rh2(S-biTISP)2
a
b
c
Loading determined by ICP analysis for Rh (mmol/g). Loading determined by mass gain. Excess catalyst used for immobilization.
Scheme 2. Preparation of Phenyl Diazoacetate 26 on Solid Support
Table 9. Evaluation of Immobilized Rh2(R-DOSP)4 and
Rh2(S-biTISP)2
Table 10. Control Experiments with Homogeneous
4-Alkyl-pyridine 237
a
b
c
8d-Rh2(R-DOSP)4
8d-Rh2(S-biTISP)2
8d-Rh2(S-biTISP)2
Rh2(prolinate)
time, min
ee, %
yield, %
cycle
timed
eee
timed
eee
timed
eee
Rh2(S-TBSP)4/23
Rh2(S-biTISP)2/23
10
720
81
88
43
1
3
5
15
15
20
45
84
85
86
40
75
140
86
86
86
18
26
30
92
85
87
86
88
18a
a
Enantiomer of 10 is formed.
a
Reaction yields determined by 1H NMR using DMAP as internal
decomposition. The reaction can be qualitatively monitored by
FT-IR (KBr), by comparing the appearance of the cyclopropane
ester signal (∼1730 cm-1) and disappearance of the diazo signal
(∼2100 cm-1). The three-phase test reaction was conducted
using 1 mol % 8d-Rh2(S-TBSP)4 and immobilized phenyl
diazoacetate 26 and was allowed to agitate for 180 min (eq 6).
Under these reaction conditions, no decomposition of the
immobilized diazo compound 26 was observed by FT-IR (KBr).
A 40% conversion of the immobilized diazo compound 26 to
cyclopropane 27 was achieved by using higher catalyst loading
(3 mol % 8d-Rh2(S-TBSP)4) and prolonged reaction time (12
h) (eq 7). As a further test to determine if the immobilized
catalyst system was operating via a release-and-capture mech-
anism, methyl phenyldiazoacetate 9 was added to the reaction
mixture of 26 and 8d-Rh2(S-TBSP)4 (eq 8). The homogeneous
diazo compound 9 immediately reacted under these conditions
(20 min), and the cyclopropane product was isolated in 87%
yield and 81% ee. However, there was no obvious rate difference
in the decomposition of the immobilized diazo compound 26
(40% conversion). On the basis of these results, we conclude
that at least the majority of the reactivity of the immobilized
catalyst is not by a release-and-capture mechanism. To further
study the cooperative effect of pyridine coordination and
standard or isolated yields; in all cases yields were greater than 80%. See
Supporting Information. 1 mol % catalyst. Reference 7. In minutes
In percent.
b
c
d
e
design was to place the diazo compound on solid support18 and
add it to a reaction mixture of styrene and immobilized Rh2-
(S-TBSP)4. Using the 1% cross-linked polystyrene resin PS-
Wang 24, we prepared solid supported phenyl diazoacetate 26
in two steps (Scheme 2). Carbodiimide coupling of phenyl acetic
acid and PS-Wang, followed by standard diazo transfer with
DBU and tosyl azide, gave the diazo compound 26, which was
characterized by FT-IR (2100 and 1700 cm-1) and elemental
analysis (0.95 mmol/g).
A typical asymmetric cyclopropanation of styrene using
methyl phenyldiazoacetate catalyzed by immobilized catalyst
8d-Rh2(S-TBSP)4 was complete within 10-15 min after diazo
compound addition.6d The control solid-phase reaction with
immobilized phenyldiazoacetate 26 catalyzed by homogeneous
Rh2(S-TBSP)4 (1 mol %) required 60 min for complete diazo
(18) (a) Clapham, B.; Spanka, C.; Janda, K. D. Org. Lett. 2001, 3, 2173. (b)
Iso, Y.; Shindo, H.; Hamana, H. Tetrahedron 2000, 56, 5353-5361. (c)
Gowravaram, M. R.; Gallop, M. A. Tetrahedron Lett. 1997, 38, 6973. (d)
Zaragoza, F.; Petersen, S. V. Tetrahedron 1996, 52, 5000. (e) Cano, M.;
Camps, F.; Joglar, J. Tetrahedron Lett. 1998, 39, 9819.
9
4278 J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004

Immobilizing Dirhodium Tetraprolinate Catalysts
A R T I C L E S
Experimental Section
microencapsulation, the phenyl-substituted catalyst 11d-Rh2(S-
TBSP)4 was evaluated in the three-phase test (eq 9). As
expected, this polymer was more active than 8d-Rh2(S-TBSP)4
(70% conversion, 3 mol % catalyst) although the rate difference
was relatively small. This shows that even in the case of the
phenyl-substituted catalyst 11d-Rh2(S-TBSP)4, the release-and-
capture mechanism is not likely to be playing a major role.
General Method for the Immobilization of Rhodium(II) Cata-
lysts. General procedure using Quest-210 parallel synthesizer: proce-
dure for homogeneous catalyst as limiting agent.
To a 5-mL reaction vessel with stir bar was added resin (0.020 mmol,
1.2 equiv) and toluene (3.5 mL). The resin was agitated for 10 min,
before the addition of rhodium(II) catalyst (0.017 mmol, 1 equiv). Extra
toluene (1 mL) was used to wash down the sides of the reaction vessel.
The mixture was agitated for 2-4 h, after which the solution was
drained and the purple-green beads were washed with toluene (5×).
Air was used to dry the beads before transferring them to a vial and
dried under vacuum to give the solid supported catalyst which was
analyzed by inductively coupled plasma emission spectroscopy (ICP-
AES).
In summary, the systematic studies described herein demon-
strate that the polymer backbone, the linker, the terminal
pyridine group, and the catalyst structure all contribute to the
efficiency of dirhodium catalyst immobilization. The im-
mobilization is considered to be due to a combination of ligand
coordination and an encapsulation effect. The three-phase test
studies indicated that a very slow reaction occurs when both
the catalyst and the diazo compound were immobilized, but the
slow rate precluded the likelihood that the cyclopropanation was
predominately occurring by a release-and-capture mechanism.
The most attractive feature of this immobilization strategy is
the fact that the immobilized catalysts are generated by simply
mixing the homogeneous catalysts with the polymer. Currently,
we are studying the potential use of resin 8 as a universal solid
support for immobilization of all the standard chiral dirhodium
catalysts.
Standard Asymmetric Cyclopropanation of Styrene Using Solid
Supported Catalysts and Methyl Phenyldiazoacetate.
To a 5-mL Quest 210 reaction vessel was added solid supported
catalyst (0.0005 mmol, 0.5 mol %) and magnetic stirrer bar. The
reaction vessel was inserted into the Quest 210 and purged with argon
for 10 min. Toluene (3 mL), followed by styrene (0.2 mmol), was added
to the reaction mixture and agitated for 10 min. A 1-mL toluene solution
of methyl phenyldiazoacetate (0.1 mmol) was prepared and added
dropwise via a syringe over 10 min to the agitating mixture. Endpoint
of the reaction mixture was determined by cessation of nitrogen gas
evolution and by disappearance of the yellow color from the diazo
compound. The reaction mixture was agitated for 10 min after reaching
the endpoint and drained. The resin was washed with toluene three to
9
J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004 4279

A R T I C L E S
Davies et al.
four times with toluene and dried by pushing air through the reaction
vessel. The reaction vessel was then purged with argon and filled with
toluene (3 mL) for the next cycle. Excess styrene from drained reaction
mixture was removed by evaporation under vacuum, and product yield
Acknowledgment. Financial support of this work by the
National Science Foundation (CHE-0092490) is gratefully
acknowledged.
1
was determined by H NMR using DMAP as internal standard or by
Supporting Information Available: Full experimental details
for the preparation of resins 8, 11, 13, 15-18, 22, 25, 26,
detailed experimental procedures, and tables with enantiose-
lectivities and yields (PDF). This material is available free of
charge via the Internet at http://pubs.acs.org.
passing the crude reaction mixture through a short plug of silica gel
(5:1 pentane/ether). HPLC, Whelk column, 2% 2-PrOH in hexanes,
1.0 mL/min, 1 mg/mL, tR ) 9.05 (minor) and 10.19 (major) min, UV
254 nm. 1H NMR (500 MHz, CDCl3): δ 7.12 (m, 3 H), 7.04 (m, 5 H),
6.76 (m, 2 H), 3.65 (s, 3 H), 3.12-3.09 (dd, J ) 4.0, 9.0 Hz, 1 H),
2.14-2.11 (dd, J ) 5.0, 9.5 Hz, 1 H), 1.88-1.86 (dd, J ) 5.0, 7.5 Hz,
1 H). The spectroscopic data are consistent with previously reported
data.12a
JA0393067
9
4280 J. AM. CHEM. SOC. VOL. 126, NO. 13, 2004