"Matched/mismatched" diastereomeric dirhodium(II) carboxamidate catalyst pairs. Structure-selectivity correlations in diazo decomposition and hetero-Diels-Alder reactions

Article abstract of DOI:10.1021/jo050609o

Homo-ligated dirhodium(II) carboxamidates provide well-defined structural frameworks with which to investigate catalyst-controlled multiple asymmetric induction ("match/mismatch" effects). Diastereomeric pairs of methyl 2-oxoimidazolidine-4(S)-carboxylate

Full text of DOI:10.1021/jo050609o

“Matched/Mismatched” Diastereomeric Dirhodium(II)
Carboxamidate Catalyst Pairs. Structure-Selectivity Correlations
in Diazo Decomposition and Hetero-Diels-Alder Reactions
Michael P. Doyle,*,† John P. Morgan,† James C. Fettinger,† Peter Y. Zavalij,† John T. Colyer,‡
Daren J. Timmons,‡ and Michael D. Carducci‡
Department of Chemistry and Biochemistry, The University of Maryland, College Park, Maryland 20742,
and Department of Chemistry, University of Arizona, Tucson, Arizona 85721
mdoyle3@umd.edu
Received March 29, 2005
Homo-ligated dirhodium(II) carboxamidates provide well-defined structural frameworks with which
to investigate catalyst-controlled multiple asymmetric induction (“match/mismatch” effects).
Diastereomeric pairs of methyl 2-oxoimidazolidine-4(S)-carboxylate ligands containing 2-phenyl-
cyclopropane (4S,2′S,3′S-HMCPIM and 4S,2′R,3′R-HMCPIM) and N-benzenesulfonylproline (4S,2′S-
HBSPIM and 4S,2′R-HBSPIM) attachments at the 1-N-acyl site have been prepared; the resulting
(cis-2,2)-Rh2L4 compounds have been produced in good yields, and the X-ray crystal structure of
each dirhodium(II) compound has been obtained. The incorporation of additional stereocenters into
the dirhodium(II) ligands leads to recognizable levels of double asymmetric induction for C-H
insertion, cyclopropanation, and hetero-Diels-Alder cycloaddition applications. The configurationally
“matched” cases provide modest increases in enantioselectivity for intramolecular C-H insertion
reactions relative to the model catalyst Rh2(MPPIM)4, but applications of the configurationally
mismatched catalysts result in significant lowering of enantioselectivity. The Rh2(BSPIM)4 catalysts
show the highest degree of differential selectivity. Hetero-Diels-Alder reactions show inverse
behavior from the configurationally matched and mismatched Rh2L4 catalysts to that found in the
metal carbene transformations.
Introduction
cases may lead to dramatically larger asymmetric induc-
tion in the products, whereas the mismatched cases often
result in significantly lower selectivity.3 Notable ex-
amples of this double asymmetric induction exist in
hydrogenation chemistry, focusing primarily on rhodium
or ruthenium with bidentate phosphine ligands contain-
ing multiple stereocenters.4 In particular, modifications
of DIOP,5 DuPHOS,6 and BINAP7 ligands, among others,8
have led to improvements in enantiocontrol in the
matched cases and reductions in enantiocontrol in the
mismatched cases. Similar effects have been reported by
Delineation of the stereochemical factors that allow a
chiral catalyst to influence product configuration remains
a central problem in asymmetric catalysis.1 With cata-
lysts whose ligand(s) have multiple stereogenic elements,
configurational and conformational complexities arise
that can be transmitted as stereoselectivity in product
formation. The ligand’s multiple stereocenters may be
situated favorably (“matched”) or unfavorably (“mis-
matched”) for stereocontrol when the catalyst interacts
with a prochiral substrate.2 Ultimately, the matched
* Corresponding author.
(2) (a) Masamune, S.; Choy, W.; Petersen, J. S.; Sita, L. R. Angew.
Chem., Int. Ed. Engl. 1985, 24, 1-30. (b) Heathcock, C. H.; White, C.
T. J. Am. Chem. Soc. 1979, 101, 7076-7079. (c) Horeau, A.; Kagan,
H.-B.; Vigeron, J.-P. Bull. Soc. Chim. Fr. 1968, 3795-3797.
(3) (a) Noyori, R. Asymmetric Catalysis in Organic Synthesis;
Wiley: New York, 1994. (b) Ojima, I.; Ed. Catalytic Asymmetric
Synthesis, 2nd ed.; Wiley: New York, 2000. (c) Stephenson, G. R.; Ed.
Advanced Asymmetric Synthesis; Chapman and Hall: London, 1996.
† University of Maryland.
‡ University of Arizona.
(1) (a) Jacobsen, E. N.; Pfaltz, A.; Yamamoto, H. Comprehensive
Asymmetric Catalysis; Springer: New York, 1999. (b) Vo¨gtle, F.;
Stoddart, J. F.; Shibasaki, M. Stimulating Concepts in Chemistry;
Wiley-VCH: New York, 2000. (c) Williams, J. M. J. Catalysis in
Asymmetric Synthesis; Sheffield Academic: Sheffield, 1999.
10.1021/jo050609o CCC: $30.25 © 2005 American Chemical Society
Published on Web 05/19/2005
J. Org. Chem. 2005, 70, 5291-5301
5291

Doyle et al.
SCHEME 1. Dirhodium Carboxamidate
Complexes in the (cis-2,2) Configuration
Katsuki for a broad selection of chemical transformations
by the introduction of a binaphthyl unit into the salen
ligand.9 The strategy employed is the design of multiple
stereocenters in the ligand to control the overall catalyst-
reactant orientation, which in turn influences product
stereochemistry.10,11 However, configurational changes in
ligands brought about by installing additional chirality
are generally unpredictable, and reaction selectivities are
determined empirically.4
There are no reliable predictive models to explain the
structural subtleties of catalysts with multiple chiral
centers that lead to dramatic changes in stereoselection,
although the “match/mismatch” strategy has been widely
applied over the past 20 years.1,5-9 Dirhodium(II) car-
boxamidates offer an appealing framework on which to
examine structural influences on reaction stereoselectiv-
ity because of their noted structural rigidity and their
suitability as catalysts for several chemical transforma-
tions.12 Their construction features four bridging ligands
around the dirhodium core so that two nitrogen and two
oxygen donor atoms are attached to each rhodium with
the composite arranged in a cis-2,2 fashion (Scheme 1).13
This ligand arrangement affords a relatively inflexible
structural framework that is amenable to configurational
modifications.
FIGURE 1. Diastereomeric dirhodium carboxamidate com-
plexes depicted with a view along the rhodium-rhodium bond
axis.
tional flexibility associated with diastereocontrol.15 In
structures such as Rh2(4S-MPPIM)4 (1), for example, the
3-phenylpropanoyl group provides optimum structural
advantage for control of diastereoselectivity and enantio-
selectivity in metal carbene carbon-hydrogen insertion
reactions15,16 and in hetero-Diels-Alder reactions.17 By
introducing chirality into the N-acyl substituent, two or
more dirhodium(II) carboxamidate diastereoisomers are
produced (A and B in Figure 1) whose configurational
orientation is either aligned with (A) or opposed to (B)
that of the 4-carboxylate attachment. In contrast, the
achiral N-acyl group, as is found in Rh2(4S-MPPIM)4, is
neither aligned with nor opposed to the configuration of
the 4-carboxylate attachment (see C). Because the elec-
tronic influence of the 1-acyl substituent is expected to
be largely invariant with its structure, steric effects on
reaction selectivity will govern the outcome. Herein we
describe the synthesis of diastereomeric pairs of catalysts
based on structural variations in the N-acyl substituent,
provide X-ray structural analyses that define their match/
mismatch relationships, and examine the stereochemical
outcomes from their application to metal carbene trans-
formations and Lewis acid catalyzed hetero-Diels-Alder
reactions.
1-Acyl-2-oxo-imidazolidine-4-carboxylates have a frame-
work on which additional stereocenters can be conve-
niently built.14 Whereas the configuration of the 4-car-
boxylate substituent affords enantiomer differentiation,12,13
the 1-acyl substituent is known to influence conforma-
(4) For additional reviews on double asymmetric induction, see: (a)
Kolodiazhnyi, O. I. Tetrahedron 2003, 59, 5953-6018 and references
therein. (b) Sharpless, K. B. Chem. Scr. 1985, 25, 71-77.
(5) (a) Kagan, H.-B.; Dang, T.-P. J. Am. Chem. Soc. 1972, 94, 6429-
6433. (b) Kagan, H.-B.; Fiaud, J. C.; Hoornaert, C.; Meyer, D.; Poulin,
J. C. Bull. Soc. Chim. Belg. 1979, 88, 923-931. (c) Li, W.; Zhang, X. J.
Org. Chem. 2000, 65, 5871-5874.
(6) Burk, M. J.; Pizzano, A.; Martin, J. A.; Liable-Sands, L. M.;
Rheingold, A. L. Organometallics 2000, 19, 250-260.
(7) Qiu, L.; Qi, J.; Pai, C.-C.; Chan, S.; Zhou, Z.; Choi, M. C. K.;
Chan, A. S. C. Org. Lett. 2002, 4, 4599-4602.
(8) (a) Komarov, I. V.; Borner, A. Angew. Chem., Int. Ed. 2001, 40,
1197-1200. (b) Huang, H.; Zheng, Z.; Luo, H.; Bai, C.; Hu, X.; Chen,
H. Org. Lett. 2003, 5, 4137-4139. (c) Hua, Z.; Vassar, V. C.; Ojima, I.
Org. Lett. 2003, 5, 3831-3834.
(14) Doyle, M. P.; Zhou, Q.-L.; Raab, C. E.; Roos, G. H. P.; Simonsen,
S. H.; Lynch, V. Inorg. Chem. 1996, 35, 6064-6073.
(9) Katsuki, T. Synlett 2003, 281-297.
(15) (a) Doyle, M. P.; Kalinin, A. V.; Ene, D. G. J. Am. Chem. Soc.
1996, 118, 8837-8846. (b) Doyle, M. P.; Tedrow, J. S.; Dyatkin, A. B.;
Spaans, C. J.; Ene, D. G. J. Org. Chem. 1999, 64, 8907-8915. (c) Doyle,
M. P.; Davies, S. B.; May, E. J. J. Org. Chem. 2001, 66, 8112-8119.
(16) (a) Bode, J. W.; Doyle, M. P.; Protopopova, M. N.; Zhou, Q.-L.
J. Org. Chem. 1996, 61, 9146-9155. (b) Doyle, M. P.; Hu, W.;
Valenzuela, M. V. J. Org. Chem. 2002, 67, 2954-2959. (c) Doyle, M.
P.; Hu, W. Chirality 2002, 14, 169-172.
(10) (a) Huang, H.; Zheng, Z.; Luo, H.; Bai, C.; Hu, X.; Chen, H.
Org. Lett. 2003, 5, 4137-4139. (b) Burgess, K.; Ohlmeyer, M. J.;
Whitmire, K. H. Organometallics 1992, 11, 3588-3600.
(11) For a review of secondary effects in stereoselective catalysis,
see: Sawamura, M.; Ito, Y. Chem. Rev. 1992, 92, 857-871.
(12) Doyle, M. P.; McKervey, M. A.; Ye, T. Modern Catalytic Methods
for Organic Synthesis with Diazo Compounds; Wiley: New York, 1998.
(13) (a) Doyle, M. P.; Ren, T. In Progress in Inorganic Chemistry;
Karlin, K., Ed.; Wiley: New York, 2001; Vol. 49, pg. 113-168. (b) Doyle,
M. P. In Catalysis by Di- and Polynuclear Metal Cluster Complexes;
Adams, R. D., Cotton, F. A., Eds.; VCH Publishers: New York, 1998;
Chapter 7.
(17) (a) Doyle, M. P.; Phillips, I. M.; Hu, W. H. J. Am. Chem. Soc.
2001, 123, 5366-5395. (b) Doyle, M. P.; Valenzuela, M.; Huang, P.
Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 5391-5395. (c) Valenzuela,
M.; Doyle, M. P.; Hedberg, C.; Hu, W.; Holmstrom, A. Synlett 2004,
2425-2428.
5292 J. Org. Chem., Vol. 70, No. 13, 2005

Diastereomeric Dirhodium(II) Carboxamidates
SCHEME 2. Syntheses of (cis-2,2)-Rh2(4S,2′S,3′S-MCPIM)4 (11) and (cis-2,2)-Rh2(4S,2′R,3′R-MCPIM)4 (12)
Results and Discussion
not been developed on the scale necessary for ligand
synthesis. Although highly enantioselective trans-selec-
tive cyclopropanation of styrene with menthyl diazoac-
etate is known,20 in our hands this route was inefficient,
requiring multiple recrystallizations to ensure high opti-
cal purity. On a small scale the trans-2-phenylcyclopro-
panecarboxylic acid racemate was separated by selective
recrystallization with brucine,21 and chromatographic
separation of enantiomers was also achieved. However,
the procedure outlined in Scheme 2 was found to be the
most effective and practical.
Diastereomeric imidazolidinones 9 were synthesized
from 8 and the racemate of trans-2-phenylcyclopropan-
ecarboxylic acid chloride (7). The product mixture was
then used without separation in the subsequent depro-
tection step to yield 10 as a mixture of diastereomers.
Using this optimized route, gram-scale synthesis of both
ligand diastereomers was achieved after chromatographic
separation [∆Rf (S,S,S-10:S,R,R-10) ) 0.1, 1:1 hexanes/
ethyl acetate]. Dirhodium(II) carboxamidates 11 and 12
were then synthesized through ligand exchange with
rhodium(II) acetate and were isolated as red crystals
(37% overall yield for Rh2(4S,2′S,3′S-MCPIM)4 (11); 29%
overall for Rh2(4S,2′R,3′R-MCPIM)4 (12), from acid chlo-
ride 7). The (3,1)-isomers were minor products from these
reactions and were not characterized.
Two sets of diastereomeric dirhodium(II) complexes
using the methyl 2-oxo-imidazolidine-4S-carboxylate
framework were prepared. In ligands 2 and 3, the N-acyl
attachment is derived from the enantiomers of trans-2-
phenylcyclopropanecarboxylic acid, whereas in 4 and 5
the N-acyl attachments are prepared from N-benzene-
sulfonylprolinate enantiomers. The former were em-
ployed because of their structural similarity to Rh2(4S-
MPPIM)4 (1), whereas the latter have a structural
relationship to the chiral dirhodium(II) prolinate cata-
lysts (6) of Davies18 and McKervey.19
Crystal structures for each MCPIM-ligated dirhodium
compound are shown in Figures 2 and 3. Selected bond
lengths and angles are reported in Tables 1 and 2,
respectively. Crystals suitable for X-ray diffraction were
obtained by slow evaporation of a solution in methanol
with trace acetonitrile (for 11) or without acetonitrile (for
(18) Davies, H. M. L. Eur. J. Org. Chem. 1999, 2459-2469 and
references therein.
(19) Kennedy, M.; McKervey, M. A.; Maguire, A. R.; Roos, G. H. P.
J. Chem. Soc., Chem. Commun. 1990, 361-362.
(20) Nishiyama, H.; Itoh, Y.; Matsumoto, H.; Park, S. B. J. Am.
Chem. Soc. 1994, 116, 2223-2224.
(21) The brucine procedure is detailed in Supporting Information:
(a) Overberger, C. G.; Shimokawa, Y. Macromolecules 1971, 4, 718-
725. (b) Bru¨ckner, C.; Reissig, H. U. Chem. Ber. 1987, 120, 627-629.
(cis-2,2)-Rh2(MCPIM)4. Prior to this work enan-
tiopure trans-2-phenylcyclopropanecarboxylic acids had
J. Org. Chem, Vol. 70, No. 13, 2005 5293

Doyle et al.
TABLE 1. Atom Labels and Selected Bond Lengths (Å)
for Imidazolidinone-Ligated Dirhodium(II)
Carboxamidatesa
FIGURE 2. ORTEP plot of (cis-2,2)-[Rh2(4S,2′S,3′S-MCPIM)4-
(CH3OH)(CH3CN)] (11) showing a partial atom labeling scheme.
The axial ligands have been removed for clarity. Thermal
ellipsoids are scaled to the 50% probability level.
bond
1
11
12
15
16
Rh1-Rh2 2.4637(8) 2.4566(6) 2.4475(4) 2.4394(8) 2.4585(4)
Rh1-N1A 2.012(2) 2.013(5) 2.017(3) 2.022(6) 2.026(4)
Rh1-N1B 2.014(4) 2.016(5) 1.997(3) 1.955(7) 2.014(4)
Rh2-O6A 2.063(2) 2.086(4) 2.059(2) 2.075(5) 2.049(3)
Rh2-O6B 2.055(4) 2.069(4) 2.088(2) 2.068(5) 2.077(3)
Rh1-X1
Rh2-Y1
2.219(4) 2.201(5) 2.291(2) 2.325(5) 2.205(5)
2.202(5) 2.194(5) 2.321(2) 2.325(5) 2.205(5)
N1A-C2A 1.311(4) 1.330(7) 1.311(4) 1.456(9) 1.315(7)
N1B-C2B 1.313(7) 1.305(8) 1.312(4) 1.339(10) 1.298(6)
N1A-C5A 1.463(3) 1.455(7) 1.460(4) 1.305(9) 1.444(6)
N1B-C5B 1.451(7) 1.455(7) 1.461(4) 1.458(10) 1.463(6)
C2A-O6A 1.267(3) 1.262(7) 1.251(4) 1.242(9) 1.265(6)
C2B-O6B 1.271(7) 1.266(7) 1.255(4) 1.244(9) 1.279(6)
C2A-N3A 1.408(3) 1.406(7) 1.426(4) 1.440(9) 1.408(6)
C2B-N3B 1.405(7) 1.420(7) 1.414(4) 1.415(10) 1.415(6)
N3A-C4A 1.454(4) 1.456(8) 1.463(4) 1.461(9) 1.463(6)
N3B-C4B 1.460(7) 1.463(8) 1.468(4) 1.456(11) 1.472(7)
C4A-C5A 1.543(4) 1.554(8) 1.537(5) 1.505(11) 1.566(7)
C4B-C5B 1.551(8) 1.527(8) 1.537(5) 1.490(12) 1.545(7)
a Data for Rh2(4S-MPPIM)4 (1) are provided for comparison.11
The values in parentheses are the estimated standard deviations
and were calculated by σav ) [1/(Σi(1/σi)]1/2, where σi is the
estimated standard deviation of each bond length contributing to
the average.
(cis-2,2)-Rh2(BSPIM)4. The synthesis of the diaster-
eomeric BSPIM catalysts is outlined in Scheme 3, focus-
ing on one of the ligand diastereomers. Dirhodium(II)
carboxamidates 15 and 16 were prepared by ligand
exchange with rhodium(II) acetate [42% overall yield
from (S)-proline for Rh2(4S,2′S-BSPIM)4 (15); 34% overall
for Rh2(4S,2′R-BSPIM)4 (16) from (R)-proline by the same
stepwise procedure]. The X-ray crystal structures for the
BSPIM pair are shown in Figures 5 and 6. Selected bond
lengths and angles are reported in Tables 1 and 2,
respectively.
The sterically large prolinyl (N-acyl) side chain in the
BSPIM structures is twisted nearly perpendicular to the
plane of the imidazolidinone ring (C9-C11-C12-C13
dihedral angles ) 80-100°). By comparison, the 2-phen-
ylcyclopropyl (N-acyl) side chains of 11 and 12 adopt a
more extended conformation (C9-C11-C12-C13 dihe-
dral angles ) 135-155°), as expected from the rigidity
of the cyclopropane ring. Therefore, the pendant phenyl-
sulfonyl groups of 15 and 16 are expected to have a
greater influence on the four spatial quadrants surround-
ing the bimetallic core than do the phenyl substituents
of 11 and 12.
FIGURE 3. ORTEP plot of (cis-2,2)-[Rh2(4S,2′R,3′R-MCPIM)4-
(CH3OH)2] (12) with a partial atom labeling scheme. The axial
ligands have been removed for clarity. Thermal ellipsoids are
scaled to the 50% probability level.
12). In both complexes, one of the faces of the cyclopropyl
ring in the N-acyl group lies over the dirhodium axial
site, placing its angular phenyl substituent in one of the
spatial quadrants around the bimetallic core.
Viewing the MCPIM catalysts down the rhodium-
rhodium bond axis (Figure 4) reveals that these catalysts
are indeed configured as shown in Figure 1. Specifically,
Rh2(4S,2′S,3′S-MCPIM)4 (11) has its pendant ester and
N-acyl side chains oriented in the same direction, forming
a counterclockwise spiral. This orientation is particularly
well suited to intramolecular reactions in which the
active site for reaction is tethered to the dirhodium(II)
axial coordination site. Conversely, the ester and N-acyl
side chains are in configurational opposition in Rh2-
(4S,2′R,3′R-MCPIM)4 (12), and this orientation is ex-
pected to be a barrier to stereoselectivity enhancement
in intramolecular transformations.
5294 J. Org. Chem., Vol. 70, No. 13, 2005

Diastereomeric Dirhodium(II) Carboxamidates
TABLE 2. Selected Bond Angles and Torsion Angles (in deg) for Imidazolidinone-Ligated Dirhodium(II)
Carboxamidatesa
bond angle
1
11
12
15
16
Rh2-Rh1-N1A
Rh2-Rh1-N1B
Rh1-Rh2-O6A
Rh1-Rh2-O6B
Rh2-Rh1-X1
85.98(5)
86.29(12)
89.45(5)
89.31(10)
172.55(7)
171.85(12)
92.3(1)
85.84(13)
85.24(14)
89.91(11)
89.93(11)
177.64(13)
177.73(13)
92.06(18)
87.56(16)
93.25(17)
93.93(17)
90.29(17)
87.81(17)
122.2(4)
121.1(4)
128.3(4)
128.8(4)
115.9(4)
114.4(4)
111.8(5)
111.7(5)
108.6(5)
107.8(5)
102.1(5)
102.4(4)
104.3(5)
104.7(5)
109.5(5)
109.9(5)
125.9(5)
127.8(5)
122.2(5)
120.4(5)
84.17(8)
85.40(8)
91.43(6)
89.57(6)
177.18(7)
178.42(7)
90.73(11)
85.93(9)
98.44(10)
95.18(10)
86.30(9)
90.03(9)
121.9(2)
122.1(2)
128.1(2)
126.7(2)
113.5(2)
114.7(2)
110.3(3)
110.7(3)
108.5(3)
109.3(3)
100.9(3)
102.6(3)
103.9(3)
104.5(3)
110.0(3)
111.1(3)
128.2(3)
126.7(3)
121.4(3)
122.5(3)
84.10(17)
86.42(19)
91.66(13)
90.11(14)
176.17(14)
176.17(14)
91.4(3)
84.64(12)
85.72(12)
91.15(10)
89.56(10)
174.55(12)
177.26(12)
92.18(17)
87.50(14)
92.71(16)
99.16(17)
86.57(15)
88.83(16)
122.1(3)
122.2(3)
127.6(4)
125.4(3)
115.1(3)
115.0(3)
111.6(4)
111.7(4)
108.1(4)
108.8(4)
101.0(4)
102.9(4)
103.6(4)
104.7(4)
110.3(4)
111.5(4)
127.1(5)
126.4(5)
121.3(5)
121.8(4)
Rh1-Rh2-Y1
N1A-Rh1-N1B
O6A-Rh2-O6B
N1A-Rh1-X1
N1B-Rh1-X1
O6A-Rh2-Y1
O6B-Rh2-Y1
Rh1-N1A-C2A
Rh1-N1B-C2B
Rh1-N1A-C5A
Rh1-N1B-C5B
Rh2-O6A-C2A
Rh2-O6B-C2B
N1A-C2A-N3A
N1B-C2B-N3B
C2A-N3A-C4A
C2B-N3B-C4B
N3A-C4A-C5A
N3B-C4B-C5B
C4A-C5A-N1A
C4B-C5B-N1B
C5A-N1A-C2A
C5B-N1B-C2B
N1A-C2A-O6A
N1B-C2B-O6B
N3A-C2A-O6A
N3B-C2B-O6B
90.20(7)
99.0(1)
88.2(2)
94.0(2)
93.4(2)
98.3(2)
86.4(1)
90.97(19)
87.25(18)
126.4(5)
121.3(5)
121.9(5)
127.2(5)
112.5(5)
113.9(5)
110.3(6)
109.7(7)
107.7(6)
108.7(7)
104.6(6)
104.7(7)
104.3(6)
105.3(7)
111.4(6)
110.7(7)
129.2(7)
128.3(7)
120.5(6)
128.3(7)
86.0(2)
121.3(2)
121.0(4)
127.9(2)
128.1(3)
115.9(2)
116.6(4)
111.8(2)
111.8(5)
108.7(2)
109.4(4)
102.5(2)
102.7(4)
104.1(2)
105.1(4)
109.8(2)
110.9(4)
126.9(2)
126.7(5)
121.2(2)
121.5(5)
C2A-N3A-C9A-C11A
C2B-N3B-C9B-C11B
N1A-C5A-C7A-O8A
N1B-C5B-C7B-O8B
-19.9(9)
-15.9(8)
-47.0(7)
-54.6(7)
8.2(9)
13.9(9)
-10.5(9)
176(3)
24.1(5)
9.8(5)
147.1(5)
93.8(4)
-1.5(12)
2.2(16)
23.0(8)
6.0(8)
-22.5(7)
136.8(5)
95.7(9)
-34.1(12)
a Refer to Table 1 for atom labels. Data for Rh2(4S-MPPIM)4 (1) are provided for comparison.11 The values in parentheses are the
estimated standard deviations and were calculated by σav ) [1/(Σi(1/σi)]1/2, where σi is the estimated standard deviation of each angle
contributing to the average.
When viewed along the rhodium-rhodium bond, the
BSPIM catalysts (Figure 4) also correspond to the depic-
tion offered in Figure 1. Although the twist of the
phenylsulfonyl groups on the ligands makes the deter-
mination less clear, Rh2(4S,2′S-BSPIM)4 (15) corresponds
to the “matched” case, with the ligands’ tetrahydropyrrole
rings and pendant methyl esters are oriented in the same
helical sense (M).22 In the diastereomeric Rh2(4S,2′R-
BSPIM)4 (16) the configuration of the pendant methyl
esters and N-prolinate attachment are opposed to each
other, making this catalyst the “mismatched” example.
Match/Mismatch Selectivity. Because dirhodium(II)
carboxamidates catalyze highly enantioselective and
diastereoselective intramolecular C-H insertion reac-
tions with diazoacetates, three model reactions were
surveyed: (1) diazo decomposition of cyclohexyl diazoac-
etate (Table 3), which exhibits both diastereoselectivity
and enantioselectivity;23 (2) a similar reaction with
cyclopentyl diazoacetate (Table 4), where enantioselec-
tivity in C-H insertion beyond 93% ee has not been
achieved with any chiral catalyst;24 and (3) cyclization
with the acyclic 2-methoxy-1-ethyl diazoacetate (Table
5),25 which has more degrees of freedom in its orientation
for reactions. A distinct “match/mismatch” phenomenon
is evident in the enantioselectivities obtained from these
reactions: the “matched” cases result in equal or greater
enantiomeric excesses in comparison to the ee’s obtained
with the known Rh2(4S-MPPIM)4. Conversely, the “mis-
matched” cases result in significantly lower enantiomeric
excesses. Particularly large match/mismatch effects are
noted in the cases of the BSPIM catalysts with a
maximum difference of 88% ee for the C-H insertion
reactions of 2-methoxy-1-ethyl diazoacetate. Presumably,
the closer proximity of the large N-acyl side chain in Rh2-
(BSPIM)4 to the axial active site gives these catalysts
greater enantiodiscriminating capabilities. However, di-
asteroselectivity (cis/trans ratio, Table 3) is not subject
to the same influences as is enantioselectivity, and 12
appears to be the only catalyst from which a significant
impact is realized.
In contrast, model intra- and intermolecular cyclopro-
panation reactions did not show the same selectivity
trends as those observed for the C-H insertions (Tables
6 and 7). Intramolecular cyclopropanation of methallyl
(22) For a discussion of helical chirality, see: Eliel, E. L.; Wilen, S.
H. Stereochemistry of Organic Compounds; Wiley: New York, 1994.
(23) Doyle, M. P.; Dyatkin, A. B.; Roos, G. H. P.; Canas, F.; Pierson,
D. A. J. Am. Chem. Soc. 1994, 116, 4507-4508.
(24) Doyle, M. P.; Zhou, Q.-L.; Dyatkin, A. B.; Ruppar, D. A.
Tetrahedron Lett. 1995, 36, 7579-7582.
(25) Doyle, M. P.; van Oeveren, A.; Westrum, L. J.; Protopopova,
M. N.; Clayton, T. W., Jr. J. Am. Chem. Soc. 1991, 113, 8982-8984.
J. Org. Chem, Vol. 70, No. 13, 2005 5295

Doyle et al.
FIGURE 4. Configurational differences with chiral N-acyl groups in imidazolidinone-ligated dirhodium(II) carboxamidate catalysts.
Each catalyst structure is viewed down the length of the rhodium-rhodium bond. Only the pendant groups around the front
rhodium atom are shown; axial ligands are removed for clarity. Thermal ellipsoids are scaled to 50% probability.
diazoacetate was chosen because use of Rh2(4S-MPPIM)4
gave the highest % ee value for this process of any chiral
catalyst yet examined.24 Although the catalysts described
here give rise to lower enantioselectivities than does 1,
the match/mismatch phenomenon continues to hold. The
intermolecular cyclopropanation reaction of styrene with
ethyl diazoacetate offers an overview of diastereoselec-
tivity as well as enantioselectivity, but this reaction has
generally resisted high levels of stereocontrol.26 In both
cases the % ee of the predominant cis product was lower
in comparison to the selectivity arising from Rh2(4S-
MPPIM)4, and diastereoselectivity was also low.
Lower enantioselectivities relative to Rh2(4S-MPPIM)4
are also observed for hetero-Diels-Alder (HDA) reactions
of the Danishefsky diene with either p-nitrobenzaldehyde
(Table 8) or 5-nitro-2-thiophenecarboxaldehyde (Table
9).17 These aldehydes were selected because of their high
reactivity with the Rh2(4S-MPPIM)4 catalyst. The high
product yields suggest that all cycloaddition product is
formed via the catalytic pathway.27 However, in neither
(26) For exceptions using the chiral azetidinone-ligated dirhodium-
(II) catalysts, see: (a) Doyle, M. P.; Davies, S. B.; Hu, W. Chem.
Commun. 2000, 867-868. (b) Doyle, M. P.; Zhou, Q.-L.; Simonsen, S.
H.; Lynch, V. Synlett 1996, 697-698.
(27) Estimated rates for these processes were 100-fold greater than
those for the background uncatalyzed reactions.
5296 J. Org. Chem., Vol. 70, No. 13, 2005

Diastereomeric Dirhodium(II) Carboxamidates
FIGURE 5. ORTEP plot of (cis-2,2)-[Rh2(4S,2′S-BSPIM)4(CH3-
OH)2] (15) showing a partial atom labeling scheme. The axial
ligands have been removed for clarity. Thermal ellipsoids are
scaled to the 50% probability level.
FIGURE 6. ORTEP plot of (cis-2,2)-[Rh2(4S,2′R-BSPIM)4-
(CH3CN)(PhCN)] (16) showing a partial atom labeling scheme.
The axial ligands have been removed for clarity. Thermal
ellipsoids are scaled to the 50% probability level.
SCHEME 3. Synthesis of
(cis-2,2)-Rh2(4S,2′S-BSPIM)4 (15). Complex 16,
(cis-2,2)-Rh2(4S,2′R-BSPIM)4, Is Produced
Analogously from R-Proline
TABLE 3. Carbon-Hydrogen Insertion Reaction of
Cyclohexyl Diazoacetatea
yield
(%)b
cis/
% ee
% ee
catalyst
transc
cisc
transc
Rh2(4S-MPPIM)4, 1
71
78
63
88
89
100/0
99/1
80/20
97/3
98/2
92
97
na
nd
Rh2(4S,2′S,3′S-MCPIM)4, 11
Rh2(4S,2′R,3′R-MCPIM)4, 12
Rh2(4S,2′S-BSPIM)4, 15
Rh2(4S,2′R-BSPIM)4, 16
72
13
>99
74
>99
33
a Reactions performed in refluxing CH2Cl2 using 1.0 mol %
catalyst. b Reported as isolated yields after column chromatogra-
phy. c cis/trans ratios and enantiomeric excesses were determined
by gas chromatography.
TABLE 4. Carbon-Hydrogen Insertion Reaction of
Cyclopentyl Diazoacetatea
case does enantioselectivity for product formation ap-
proach that achieved with the use of Rh2(4S-MPPIM)4.
Curiously, with the “matched” catalysts 11 and 15 there
is lower enantiocontrol than with the “mismatched”
catalysts 12 and 16. This selectivity change underscores
the fact that configurational changes in the catalyst affect
the HDA and metal carbene reactions in unique ways,
depending on their respective reaction mechanisms.
catalyst
yield (%)b
% eec
Rh2(4S-MPPIM)4, 1
67
81
75
78
62
93d
88
40
98
22
Rh2(4S,2′S,3′S-MCPIM)4, 11
Rh2(4S,2′R,3′R-MCPIM)4, 12
Rh2(4S,2′S-BSPIM)4, 15
Rh2(4S,2′R-BSPIM)4, 16
a Reactions performed in refluxing CH2Cl2 using 1.0 mol %
catalyst. b Reported as isolated yields after column chromatogra-
phy. c Enantiomeric excesses were determined by gas chromatog-
raphy. d Reference 24.
Implications for Stereoselectivity. The mecha-
nisms of both metal carbene C-H insertion28 and cyclo-
propanation29 reactions are reasonably well under-
stood.12,13,18 More recently the mechanism of the hetero-
Diels-Alder reaction catalyzed by dirhodium(II) carboxa-
midates has been unraveled.17b,c Accordingly, in reactions
catalyzed by dirhodium(II) compounds, the four bridging
ligands are assumed to remain bound to the dirhodium
(28) Nakamura, E.; Yoshikai, N.; Yamanaka, M. J. Am. Chem. Soc.
2002, 124, 7181-7192. (b) Davies, H. M. L.; Beckwith, R. E. J. Chem.
Rev. 2003, 103, 2861-2904. (c) Pirrung, M. C.; Liu, H.; Morehead, A.
T., Jr. J. Am. Chem. Soc. 2002, 124, 1014-1023.
J. Org. Chem, Vol. 70, No. 13, 2005 5297

Doyle et al.
TABLE 5. Carbon-Hydrogen Insertion Reaction of
TABLE 8. Hetero-Diels-Alder Reaction between
Danishefsky’s Diene and p-Nitrobenzaldehydea
2-Methoxyethyl Diazoacetatea
catalyst
yield (%)b
% eec
Rh2(4S-MPPIM)4, 1
50
69
64
72
72
92
91
46
94
6
catalyst
yield (%)b
% eec
Rh2(4S,2′S,3′S-MCPIM)4, 11
Rh2(4S,2′R,3′R-MCPIM)4, 12
Rh2(4S,2′S-BSPIM)4, 15
Rh2(4S,2′R-BSPIM)4, 16
Rh2(4S-MPPIM)4, 1
82
83
93
80
93
95d
48
82
73
59
Rh2(4S,2′S,3′S-MCPIM)4, 11
Rh2(4S,2′R,3′R-MCPIM)4, 12
Rh2(4S,2′S-BSPIM)4, 15
Rh2(4S,2′R-BSPIM)4, 16
a Reactions performed in refluxing CH2Cl2 using 1.0 mol %
catalyst. b Reported as isolated yields after column chromatogra-
phy. c Enantiomeric excesses were determined by gas chromatog-
raphy.
a Reactions performed at room temperature in CH2Cl2 with 1.0
mol % catalyst. The reaction was quenched after 48 h by adding
neat trifluoroacetic acid. b Reported as isolated yields after column
chromatography. c Enantiomeric excesses were determined by
HPLC (Chiracel OD column). d Reference 17a.
TABLE 6. Intramolecular Cyclopropanation with
3-Methylbutenyl Diazoacetatea
TABLE 9. Hetero-Diels-Alder Reaction between
Danishefsky’s Diene and
5-Nitro-2-thiophenecarboxaldehydea
catalyst
yield (%)b
% eec
Rh2(4S-MPPIM)4, 1
75
70
65
58
70
89d
80
52
80
16
Rh2(4S,2′S,3′S-MCPIM)4, 11
Rh2(4S,2′R,3′R-MCPIM)4, 12
Rh2(4S,2′S-BSPIM)4, 15
Rh2(4S,2′R-BSPIM)4, 16
catalyst
yield (%)b
% eec
a Reactions performed in refluxing CH2Cl2 using 1.0 mol %
catalyst. b Reported as isolated yields after column chromatogra-
phy. c Enantiomeric excesses were determined by gas chromatog-
raphy. d Reference 24.
Rh2(4S-MPPIM)4, 1
81
80
85
86
95
94d
55
75
77
55
Rh2(4S,2′S,3′S-MCPIM)4, 11
Rh2(4S,2′R,3′R-MCPIM)4, 12
Rh2(4S,2′S-BSPIM)4, 15
Rh2(4S,2′R-BSPIM)4, 16
TABLE 7. Intermolecular Cyclopropanation of Styrene
with Ethyl Diazoacetatea
a Reactions performed at room temperature in CH2Cl2 with 1.0
mol % catalyst. The reaction was quenched after 48 h by adding
neat trifluoroacetic acid. b Reported as isolated yields after column
chromatography. c Enantiomeric excesses were determined by
HPLC (Chiracel OD column). d Reference 17a.
lower) the rate for carboxamidate ligand exchange is
negligible. Furthermore, when a ligand is removed from
the dirhodium core, as occurs in catalytic reactions with
organosilanes,32 the dirhodium(II) framework does not
remain intact, and the reactant catalyst cannot be
recovered.
Results from the matched/mismatched cases presented
in Tables 3-7 show consistent behavior across a variety
of C-H insertion and cyclopropanation substrates, con-
sistent with the intact dirhodium(II) tetracarboxami-
dates. For these diazo decomposition reactions, the
matched cases exhibit higher enantiocontrol for all reac-
tions under consideration, and use of the mismatched
catalysts leads to lower stereoselectivity. Such consis-
tency across substrates with different steric and elec-
tronic requirements is unlikely if ligand dissociation and/
or rearrangement were to occur. With the Lewis acid
catalyzed reactions in Tables 8 and 9, where there is no
% ee % ee
catalyst
yield (%)b cis/transc cisc transc
Rh2(4S-MPPIM)4, 1
64
70
59
58
48
74/26
62/38
49/51
58/42
63/37
45
18
21
38
31
3
14
68
24
29
Rh2(4S,2′S,3′S-MCPIM)4, 11
Rh2(4S,2′R,3′R-MCPIM)4, 12
Rh2(4S,2′S-BSPIM)4, 15
Rh2(4S,2′R-BSPIM)4, 16
a Reactions performed in refluxing CH2Cl2 using 1.0 mol %
catalyst. b Reported as isolated yields after column chromatogra-
phy. c cis/trans ratios and enantiomeric excesses were determined
by gas chromatography.
core, and the catalyzed transformation takes place at the
axial coordination site without disturbing ligand binding
on rhodium. This presumption has been borne out by
diverse examples of catalyst stability through uniform
selectivity in multiple use-reuse for applications with
diazoacetates.30,31 Carboxamidate ligands undergo slow
exchange at 70 °C,31 so at the temperatures used to
evaluate the catalyst series in this study (40 °C and
(30) (a) Doyle, M. P.; Yan, M.; Gau, H.-M.; Blossey, E. C. Org. Lett.
2003, 5, 561-563. (b) Doyle, M. P.; Timmons, D. J.; Tumonis, J. S.;
Gau, H.-M.; Blossey, E. C. Organometallics 2002, 21, 1747-1749. (c)
Nagashima, T.; Davies, H. M. L. Org. Lett. 2002, 4, 1989-1992. (d)
Davies, H. M. L.; Walji, A. M.; Nagashima, T. J. Am. Chem. Soc. 2004,
126, 4271-4280.
(31) Doyle, M. P.; Eismont, M. Y.; Bergbreiter, D. E.; Gray, H. N. J.
Org. Chem. 1992, 57, 6103-6105.
(32) Doyle, M. P.: Shanklin, M. S. Organometallics 1994, 13, 1081-
1088.
(29) (a) Nowlan, D. T., III; Gregg, T. M.; Davies, H. M. L.; Singleton,
D. A. J. Am. Chem. Soc. 2003, 125, 15902-15911. (b) Fraile, J. M.;
Garcia, J. I.; Martinez-Merino, V.; Mayoral, J. A.; Salvatella, L. J. Am.
Chem. Soc. 2001, 123, 7616-7625. (c) Mu¨ller, P.; Bernardinelli, G.;
Allenbach, Y. F.; Ferri, M.; Flack, H. D. Org. Lett. 2004, 6, 1725-1728.
(d) Davies, H. M. L.; Antoulinakis, E. G. Org. React. 2001, 57, 1-326.
5298 J. Org. Chem., Vol. 70, No. 13, 2005

Diastereomeric Dirhodium(II) Carboxamidates
romethane wash and concentrated under reduced pressure to
produce an off-white solid that was purified by column
chromatography (ethyl acetate/hexanes 1:1) that separated the
two diastereoisomers of 10 and afforded 1.33 g of S,S,S-10 as
a white solid (4.60 mmol, 67% yield over two steps), Rf ) 0.40.
Compound S,S,S-10 was recrystallized by slow evaporation
of a dichloromethane/hexanes solution prior to use in further
reactions. 1H NMR (400 MHz, CDCl3) δ 7.30-7.24 (comp, 2
H), 7.21-7.13 (comp, 3 H), 5.65 (br s, 1 H), 4.28-4.24 (m, 1
H), 4.15 (d, J ) 7.6 Hz, 2 H), 3.79 (s, 3 H), 3.62 (ddd, J ) 8.3,
5.3, 4.4 Hz, 1 H), 2.65 (ddd, J ) 9.3, 6.6, 4.4 Hz, 1 H), 1.70
(ddd, J ) 9.3, 5.3, 4.4 Hz, 1 H), 1.38 (ddd, J ) 8.3, 6.6, 4.4 Hz,
1 H); 13C NMR (100 MHz, CDCl3) δ 172.8, 170.7, 155.6, 140.1,
128.4, 126.4, 126.2, 53.1, 49.4, 45.4, 28.2, 22.9, 18.7. Mp 167-
168 °C. [R]24D ) +223.5 (c 0.1, CH2Cl2). HRMS for (C15H17N2O4)+
theoretical 289.1188 [M + H]+; found 289.1179. Anal. Calcd
for C15H16N2O4: 62.49 C, 5.59 H, 9.72 N. Found: 62.57 C, 5.80
H, 9.32 N.
evidence of ligand disruption from extensive studies in
which carboxamidate catalysts are closely monitored,17b,c
uniform and reproducible results are obtained that are
fully consistent with intact dirhodium(II) exhibiting four
bridging carboxamidate ligands.
Conclusion. The structural rigidity of the two pairs
of diastereomeric rhodium(II) carboxamidate catalysts
aids in understanding their role in asymmetric catalysis.
Intramolecular reactions are more profoundly affected
than are intermolecular reactions, consistent with the
operational sense of a tethered substrate in a rotor. When
orientation of the two chiral attachments is complimen-
tary, stereochemical enhancement occurs; when opposed,
stereochemically destructive interference occurs. Inter-
molecular reactions have more degrees of freedom in their
orientations, and the outcome is a lower overall effect
on stereocontrol. This has been a characteristic of the (cis-
2,2)-dirhodium(II) carboxamidate catalysts; they are
more versatile (higher stereocontrol) toward intramo-
Methyl 1-[3′(R)-Phenyl-2′(R)-cyclopropanecarbonyl]-
2-oxo-imidazolidine-4(S)-carboxylate, (4S,2′R,3′R)-HM-
CPIM (S,R,R-10) was separated from the diastereomeric
mixture of 10, as described above, and purified by column
chromatography (hexanes/ethyl acetate 1:1, Rf ) 0.31) to
produce 1.19 g of a white solid (4.13 mmol, 60% yield over two
steps): 1H NMR (300 MHz, CDCl3) δ 7.29-7.14 (comp, 5 H),
5.87 (br s, 1 H), 4.25 (t, J ) 7.3 Hz, 1 H), 4.14 (d, J ) 7.3 Hz,
2 H), 3.77 (s, 3 H), 3.62 (ddd, J ) 8.4, 5.1, 4.4 Hz, 1 H), 2.64
(ddd, J ) 9.3, 6.6, 4.4 Hz, 1 H), 1.70 (ddd, J ) 9.3, 5.1, 4.2 Hz,
1 H), 1.38 (ddd, J ) 8.4, 6.6, 4.2 Hz, 1 H); 13C NMR (75 MHz,
CDCl3) δ 172.8, 170.7, 155.6, 140.1, 128.4, 126.5, 126.4, 53.0,
lecular reactions than intermolecular reactions.12,13
A
surprising outcome of this study is the demonstration
that Rh2(4S-MPPIM)4 operates very close to optimum
selectivity for overall stereocontrol. The placement of the
two carboxylate substituents around the axial coordina-
tion site is sufficient to reach high levels of enantiocon-
trol, and only incremental enhancements are achieved
by further manipulations.
49.4, 45.4, 28.1, 23.0, 18.8. Mp 91-92 °C. [R]24 ) -211.6 (c
D
0.1, CH2Cl2). HRMS for (C15H17N2O4)+ theoretical 289.1188 [M
Experimental Section
+ H]+; found 289.1194.
Dirhodium(II) Tetrakis{methyl 1-[3′(S)-phenyl-2′(S)-
cyclopropanecarbonyl]-2-oxo-imidazolidine-4(S)-carbox-
ylate}, (cis-2,2)-[Rh2(4S,2′S,3′S-MCPIM)4(CH3CN)(CH3OH)]
(11). A 25 mL two-neck round-bottom flask, stirbar, Soxhlet
extractor, and condenser were flame-dried and assembled
under a constant flow of argon while still warm, sealing all
ground glass joints with Teflon tape. A cellulose extraction
thimble filled with an oven-dried mixture of sodium carbonate
and sand (2:1) was capped with glass wool and placed in the
Soxhlet extractor during assembly. Dirhodium(II) acetate (200
mg, 0.452 mmol), S,S,S-10 (1.04 g, 3.62 mmol), and 15.0 mL
of anhydrous chlorobenzene were added though the open neck
of the flask, which was sealed with a septum and then heated
at reflux for 17 h with a 1 min solvent flush rate through the
Soxhlet extractor. The progress of the ligand exchange reaction
was monitored by HPLC using a 3.9 mm × 150 mm reverse-
phase µ-Bondapak-CN column. When conversion to product
was complete (by HPLC), the reaction mixture was cooled to
room temperature and concentrated to a purple oil. The crude
material was purified by column chromatography on reverse
phase Bakerbond CN resin (100% methanol to 99:1 methanol/
acetonitrile), and the first red band was collected. Solvent was
removed under reduced pressure to produce a red solid. This
(2,2-cis) isomer (11) was recrystallized by slow evaporation
from a warm methanol solution containing a small amount of
acetonitrile to afford 374 mg of red crystals (0.276 mmol, 61%
Methyl 1-[3′(S)-Phenyl-2′(S)-cyclopropanecarbonyl]-2-
oxo-imidazolidine-4(S)-carboxylate, (4S,2′S,3′S)-HMCPIM
(S,S,S-10). A 250 mL round-bottom flask was charged with
methyl 3-benzyloxycarbonyl-2-oxo-imidazolidine-4(S)-carboxyl-
ate 8 (4.50 g, 16.2 mmol), 4-(dimethylamino)pyridine (DMAP,
0.190 g, 1.60 mmol), pyridine (3.3 mL, 32.4 mmol), and 75 mL
of distilled dichloromethane. The reaction vessel was then
flushed with nitrogen and cooled to 0 °C. A solution of 7 (3.50
g, 19.4 mmol) in 20 mL of dichloromethane was added
dropwise over 30 min via syringe pump and the reaction was
allowed to stir at 0 °C for an additional 30 min. The solution
was heated to reflux, turning brown over 16 h. An additional
200 mL of dichloromethane was added to the solution, which
was subsequently washed with aqueous solutions of cold 1 M
HCl (2 × 50 mL), saturated NaHCO3 (1 × 50 mL), and
saturated brine (1 × 50 mL). The solution was then dried with
anhydrous MgSO4, filtered, and concentrated under reduced
pressure to afford an orange foam, which was subsequently
purified via column chromatography (SiO2; ethyl acetate/
hexanes 1:1) to yield 5.88 g of a 1:1 mixture of S,S,S-9 and
S,R,R-9 as a white solid (13.9 mmol, 86% yield) which was
used immediately in the next step. Characterization data for
compound S,R,R-9: 1H NMR (300 MHz, CDCl3) δ 7.38-7.13
(comp, 10 H), 5.38 (d, J ) 12.2 Hz, 1 H), 5.24 (d, J ) 12.2 Hz,
1 H), 4.71 (dd, J ) 10.0, 3.7 Hz, 1 H), 4.00 (dd, J ) 12.2, 10.0
Hz, 1 H), 3.90 (dd, J ) 12.2, 3.7 Hz, 1 H), 3.70 (s, 3 H), 3.59
(ddd, J ) 8.3, 5.1, 4.4 Hz, 1 H), 2.68 (ddd, J ) 9.3, 6.6, 4.4 Hz,
1 H), 1.71 (ddd, J ) 9.3, 5.1, 4.2 Hz, 1 H), 1.46 (ddd, J ) 8.3,
6.6, 4.2 Hz, 1 H); 13C NMR (75 MHz, CDCl3) δ 173.0, 169.4,
150.8, 149.5, 139.6, 134.6, 128.6, 128.4, 128.3, 126.6, 126.3,
1
yield). H NMR (400 MHz, CD3CN) δ 7.44-7.40 (comp, 4 H),
7.35-7.30 (comp, 2 H), 7.27-7.23 (comp, 8 H), 7.14-7.07
(comp, 6 H), 3.94 (dd, J ) 8.7, 2.8 Hz, 2 H), 3.88-3.77 (comp,
4 H), 3.75 (s, 6 H), 3.73-3.64 (comp, 4 H), 3.60 (s, 6 H), 3.57-
3.53 (comp, 2 H), 3.46 (dd, J ) 11.5, 4.8 Hz, 2 H), 3.35 (ddd, J
) 8.5, 5.2, 4.4 Hz, 2 H), 2.37 (ddd, J ) 9.5, 6.4, 4.4 Hz, 2 H),
2.31 (ddd, J ) 9.5, 6.4, 4.4 Hz, 2 H), 1.96 (s, 6 H), 1.61 (ddd,
J ) 9.1, 5.2, 3.6 Hz, 2 H), 1.40 (ddd, J ) 9.1, 5.2, 4.0 Hz, 2 H),
1.08 (ddd, J ) 8.3, 6.4, 3.6 Hz, 2 H), 0.96 (ddd, J ) 8.3, 6.4,
4.0 Hz, 2 H); 13C NMR (100 MHz, CD3CN) δ 175.0, 173.7,
171.1, 170.8, 166.5, 166.1, 142.2, 142.1, 129.4, 129.3, 128.3,
127.6, 127.4, 118.4, 60.5, 59.9, 53.1, 52.5, 48.5, 48.0, 28.8, 28.4,
68.9, 53.1, 52.1, 42.7, 28.8, 23.6, 19.5. Mp 101-102 °C. [R]24
D
) -143.6 (c 0.1, CH2Cl2). HRMS for (C23H23N2O6)+ theoretical
423.1551 [M + H]+; found 423.1562.
A round-bottom flask was charged with a 1:1 mixture of
S,S,S-9 and S,R,R-9 (5.00 g, 11.8 mmol) and 50 mL of dry
ethyl acetate and placed under argon. Palladium black (10 mg,
0.094 mmol) was added, and the solution was placed under
hydrogen (balloon pressure) and stirred for 90 min. The
resulting mixture was filtered through Celite with dichlo-
23.1, 22.2, 18.2, 16.9. [R]24 ) -248.4 (c 0.1, CH2Cl2). HRMS
D
J. Org. Chem, Vol. 70, No. 13, 2005 5299

Doyle et al.
for (CsRh2C60H60N8O16+) theoretical 1487.1292 [M + Cs]+;
found 1487.1329. Anal. Calcd for Rh2C65H70N10O17: 53.14 C,
4.80 H, 9.53 N (includes 1 acetonitrile molecule per catalyst
molecule in addition to axial MeCN and MeOH ligands).
Found: 53.12 C, 4.84 H, 9.56 N.
and convergence process. The final Rw(F2) ) 0.0856 with a
goodness of fit ) 1.001 for refining 897 parameters. The
conventional R(F) ) 0.0367 for 15,086 reflections for Fo > 2σ-
(Fo). Data reduction, decay correction, structure solution, and
refinement were done using the SHELXTL/PC software pack-
age.35 An absorption correction was applied with the SADABS
program.34 Tables of positional and thermal parameters, bond
lengths and angles, and torsion angles and figures are located
in a CIF file as Supporting Information.
X-ray Structure of 11. Crystals grew as red plates from
dry methanol/acetonitrile (99:1). The data crystal had ap-
proximately orthogonal dimensions 0.41 × 0.21 × 0.03 mm3.
Crystals of 11 (Rh2C66H78N9O21, containing one methanol and
one acetonitrile molecule coordinated in the axial sites) were
orthorhombic, space group P2(1)2(1)2(1), with a ) 10.6025-
(4), b ) 17.5485(6), c ) 37.3067(13) Å, V ) 6941.2(4) Å3, Z )
4, F(calc) ) 1.473 Mg m-3. The unit cell additionally contained
3 molecules of methanol and 1 molecule of water. Data were
collected out to 2θ ) 55.03° by the ω-scan technique at -100
°C using graphite-monochromatized Mo KR radiation (λ )
0.71073 Å); 78592 reflections were measured of which 15585
were unique [Rint(F2) ) 0.0714]. The structure was solved by
direct methods and refined by full-matrix least-squares on F2
with anisotropic displacement parameters for the non-
hydrogen atoms. The hydrogen atoms were calculated and
placed in idealized positions throughout the refinement and
convergence process. The final Rw(F2) ) 0.1327 with a goodness
of fit ) 1.055 for refining 839 parameters. The conventional
R(F) ) 0.1029 for 11057 reflections for Fo > 2σ(Fo). Data
reduction, decay correction, structure solution, and refinement
were done using the SHELXTL/PC software package.33 An
absorption correction was applied with the SADABS pro-
gram.34 Tables of positional and thermal parameters, bond
lengths and angles, and torsion angles and figures are located
in a CIF file as Supporting Information.
Methyl 1-[1-Benzenesulfonyl-pyrrolidine-2′(S)-carbo-
nyl]-3-benzyloxycarbonyl-2-oxo-imidazolidine-4(S)-car-
boxylate (S,S-13). A round-bottom flask was charged with
(S)-N-benzenesulfonylproline (1.0 g, 3.95 mmol) and anhydrous
THF (25 mL) and cooled to -78 °C. To this solution was added
triethylamine (0.44 g, 4.34 mmol, 1.1 equiv) and pivaloyl
chloride (0.52 g, 4.34 mmol, 1.1 equiv). The resulting milky
white mixture was allowed to stir at -78 °C for 30 min and
then at 0 °C for 90 min. The temperature was then lowered to
-78 °C, while the lithiated imidazolidine solution is prepared
as follows: in a separate flask, a solution of 8 (0.724 g, 2.60
mmol) in THF (25 mL) was treated with n-butyllithium (1.0
equiv, 2.5 M in hexanes) at -78 °C and allowed to stir for 30
min. The resulting solution of the lithium salt was added via
cannula to the proline acid chloride solution at -78 °C. After
stirring overnight at room temperature, the reaction mixture
was washed with saturated Na2CO3 and extracted five times
with ethyl acetate. The organic layers were combined, and the
solvent removed under reduced pressure. The resulting pale
yellow solid was purified on a silica plug yielding a white solid
that was used immediately in the next step (1.04 g, 2.02 mmol,
1
78% yield). H NMR (300 MHz, CDCl3) δ 7.82-7.87 (comp, 2
H), 7.55-7.42 (comp, 3 H), 7.35-7.26 (comp, 5 H), 5.68 (dd, J
) 8.8, 3.4 Hz, 1 H), 5.25 (d, J ) 12.2 Hz, 2 H), 4.68 (dd, J )
8.8, 5.0 Hz, 1 H), 3.97-3.86 (comp, 2 H), 3.66 (s, 3 H), 3.56-
3.49 (m, 1 H), 3.21-3.13 (m, 1 H), 2.15-2.02 (m, 1 H), 1.92-
1.76 (comp, 2 H), 1.69-1.57 (m, 1 H); 13C NMR (75 MHz,
CDCl3) δ 172.1, 168.9, 150.4, 149.3, 137.9, 134.5, 132.7, 129.1,
128.7, 128.6, 128.4, 127.4, 69.0, 60.8, 53.3, 52.8, 48.7, 42.3, 31.0,
Dirhodium(II) Tetrakis{methyl 1-[3′(R)-phenyl-2′(R)-
cyclopropanecarbonyl]-2-oxo-imidazolidine-4(S)-carbox-
ylate}, (cis-2,2)-[Rh2(4S,2′R,3′R-MCPIM)4(CH3OH)2] (12)
was prepared analogously to the procedure reported for
compound 11. The (2,2-cis) isomer 12 was recrystallized from
undried methanol with trace acetonitrile (405 mg, 0.30 mmol,
1
53% yield). H NMR (400 MHz, CD3CN) δ 7.37-7.19 (comp,
24.2. [R]24 ) -118.4(6) (c ) 1.0, CH2Cl2).
14 H), 7.17-7.10 (comp, 2 H), 7.06-7.01 (comp, 4 H), 4.22 (dd,
J ) 10.7, 5.2 Hz, 2 H), 3.99 (dd, J ) 11.1, 10.7 Hz, 2 H), 3.83
(dd, J ) 9.5, 2.2 Hz, 2 H), 3.80-3.63 (comp, 2 H), 3.76 (br s,
12 H), 3.60 (dd, J ) 10.7, 2.2 Hz, 2 H), 2.98 (s, 6 H), 2.46 (ddd,
J ) 9.5, 6.4, 4.4 Hz, 2 H), 2.37 (ddd, J ) 9.5, 6.4, 4.4 Hz, 2 H),
2.15 (s, 6 H), 1.62 (ddd, J ) 9.1, 5.2, 4.0 Hz, 2 H), 1.24 (ddd,
J ) 8.6, 6.4, 4.0 Hz, 2 H), 1.13 (ddd, J ) 9.1, 5.2, 4.0 Hz, 2 H),
0.69 (ddd, J ) 8.6, 6.4, 4.0 Hz, 2 H); 13C NMR (100 MHz, CD3-
CN) δ 175.1, 173.3, 170.8, 166.0, 142.4, 141.8, 129.4, 129.3,
127.5, 127.4, 127.3, 127.2, 60.5, 60.2, 53.3, 51.9, 50.0, 48.5, 48.2,
D
Methyl 1-[1-benzenesulfonyl-pyrrolidine-2′(R)-carbo-
nyl]-3-benzyloxycarbonyl-2-oxo-imidazolidine-4(S)-car-
boxylate (S,R-13) was prepared by the same procedure as
that described for S,S-13 (0.96 g, 1.9 mmol, 70% yield). 1H
NMR (300 MHz, CDCl3) δ 7.91-7.86 (comp, 2 H), 7.63-7.49
(comp, 3 H), 7.39 (br s, 5 H), 5.59 (dd, J ) 9.1, 3.1 Hz, 1 H),
5.40 (d, J ) 12.2 Hz, 1 H), 5.25 (d, J ) 12.2 Hz, 1 H), 4.76 (dd,
J ) 10.2, 3.7 Hz, 1 H), 4.09 (dd, J ) 12.0, 10.2 Hz, 1 H), 3.85
(dd, J ) 12.0, 3.7 Hz, 1 H), 3.70 (s, 3 H), 3.63-3.54 (m, 1 H),
3.27-3.17 (m, 1 H), 2.21-2.07 (m, 1 H), 1.97-1.85 (comp, 2
H), 1.78-1.64 (m, 1 H); 13C NMR (75 MHz, CDCl3) δ 172.2,
169.5, 150.2, 149.1, 138.0, 134.5, 132.7, 129.1, 128.7, 128.6,
128.3, 127.4, 69.0, 60.7, 53.1, 52.6, 48.6, 42.3, 31.1, 24.3.
Methyl 1-[1-Benzenesulfonyl-pyrrolidine-2′(S)-carbo-
nyl]-2-oxo-imidazolidine-4(S)-carboxylate, (4S,2′S)-HB-
SPIM (S,S-14). A mixture of S-13 (1.50 g, 2.91 mmol),
palladium black (10 mg, 0.094 mmol), and methanol (15 mL)
was stirred under a balloon pressure of hydrogen for 2 h. After
TLC analysis showed complete conversion to S-14, the reaction
mixture was filtered through Celite. The resulting solution was
concentrated under reduced pressure to a white solid, which
was purified by column chromatography (ethyl acetate) to
28.2, 27.0, 24.0, 23.9, 23.7, 19.4, 18.3. [R]24 ) -331.8 (c 0.1,
D
CH3CN). HRMS for (Rh2C60H60N8O16)+, theoretical 1354.2237
[M]+; found 1354.2337. Anal. Calcd for Rh2C60H64N8O18 (con-
taining 2 molecules of water as axial ligands): 51.81 C, 4.64
H, 8.06 N. Found: 51.85 C, 5.04 H, 8.27 N.
X-ray Structure of 12. Crystals grew as red prisms from
dry methanol/acetonitrile (98:2). The data crystal had ap-
proximately orthogonal dimensions 0.42 × 0.31 × 0.23 mm3.
Crystals of 12 (Rh2C62H68N8O18, containing 2 molecules of
methanol coordinated in the axial sites) were orthorhombic,
space group P2(1)2(1)2(1), with a ) 10.8222(14), b ) 21.851-
(3), c ) 30.510(4) Å, V ) 7214.9(16) Å3, Z ) 4, F(calc) ) 1.424
Mg m-3. Data were collected out to 2θ ) 56.10° by the ω-scan
technique at -100 °C using graphite-monochromatized Mo KR
radiation (λ ) 0.71073 Å); 90,349 reflections were measured
of which 16,592 were unique [Rint(F2) ) 0.0523]. The structure
was solved by direct methods and refined by full-matrix least-
squares on F2 with anisotropic displacement parameters for
the non-hydrogen atoms. The hydrogen atoms were calculated
and placed in idealized positions throughout the refinement
1
afford 0.94 g of S-14 (2.467 mmol, 85% yield). H NMR (300
MHz, CDCl3) δ 7.92-7.87 (comp, 2 H), 7.62-7.49 (comp, 3 H),
5.72 (s, 1 H), 5.60 (dd, J ) 8.7, 3.7 Hz, 1 H), 4.34 (dd, J )
10.3, 4.9 Hz, 1 H), 4.24-4.10 (comp, 2 H), 3.86 (s, 3 H), 3.59-
3.57 (m, 1 H), 3.32-3.23 (m, 1 H), 2.17-2.05 (m, 1 H), 2.01-
1.84 (comp, 2 H), 1.77-1.65 (m, 1 H); 13C NMR (75 MHz,
CDCl3) δ 172.0, 170.3, 154.8, 138.2, 132.7, 129.0, 127.5, 60.0,
53.2, 49.8, 48.7, 45.0, 31.2, 24.4. [R]20 ) -100.5(4) (c 0.93,
D
CH2Cl2).
(33) Sheldrick, G. M. SHELXTL/PC, version 5.03; Siemens Analyti-
cal X-ray Instruments, Inc.: Madison, WI, 1994.
(34) Sheldrick, G. M. SADABS Siemens Area Detector Absorption
Correction; Universita¨t Go¨ttingen: Go¨ttingen, Germany, 1996.
(35) Sheldrick, G. M. SHELXTL/PC, version 6.10; Bruker AXS,
Inc.: Madison, WI, 1997.
5300 J. Org. Chem., Vol. 70, No. 13, 2005

Diastereomeric Dirhodium(II) Carboxamidates
Methyl 1-[1-benzenesulfonyl-pyrrolidine-2′(R)-carbo-
nyl]-2-oxo-imidazolidine-4(S)-carboxylate, (4S,2′R)-HB-
SPIM (S,R-14), was prepared by the same procedure as that
described for the preparation of S-14 (0.763 g, 2.01 mmol, 83%
Dirhodium(II) Tetrakis{methyl 1-[1-benzenesulfonyl-
pyrrolidine-2′(R)-carbonyl]-2-oxo-imidazolidine-4(S)-car-
boxylate}, (cis-2,2)-[Rh2(4S,2′R-BSPIM)4(CH3CN)(PhCN)]
(16). Purple crystals of 16 were prepared analogously to the
procedure described for compound 15 (0.23 g, 0.13 mmol, 61%
yield): 1H NMR (300 MHz, CDCl3) δ 7.86-7.79 (comp, 8 H),
7.55-7.44 (comp, 12 H), 6.00-5.94 (comp, 4 H), 4.75 (br dd, J
) 10.1, 5.6 Hz, 2 H), 4.42-4.35 (comp, 2 H), 4.16 (br t, J )
11.7 Hz, 2 H), 3.85 (br s, 6 H), 3.80-3.64 (comp, 6 H), 3.73 (br
s, 6 H), 3.56-3.45 (comp, 4 H), 3.34-3.23 (comp, 4 H), 2.26-
2.07 (comp, 4 H), 2.15 (s, 6 H), 1.98-1.86 (comp, 4 H), 1.83-
1.63 (comp, 4 H), 1.59-1.47 (comp, 2 H), 1.33-1.22 (comp, 2
H); 13C NMR (75 MHz, CDCl3) δ 174.7, 173.3, 170.0, 169.9,
164.3, 163.1, 138.8, 138.6, 132.8, 132.6, 128.9, 128.8, 127.3,
58.6, 57.6, 57.0, 52.5, 52.6, 48.6, 48.0, 47.0, 46.3, 31.8, 30.9,
24.4, 24.3, 23.9, 23.8. [R]25D ) -97.1 (c 0.068, CH2Cl2). HRMS
for (C64H73N12O24Rh2S4)+ theoretical 1727.1848 [M + H]+;
found 1727.0175. Anal. Calcd for Rh2C64H80N12O28S4 (contain-
ing 2 molecules of water as axial ligands and 2 unassociated
water molecules): C, 42.72; H, 4.48; N, 9.34. Found: C, 42.84;
H, 4.18; N, 9.55.
X-ray Structure of 16. Crystals grew as red blocks from
dry methanol/benzonitrile/acetonitrile (96:3:1). The data crys-
tal had approximately orthogonal dimensions 0.28 × 0.23 ×
0.16 mm3. Crystals of 16 contained two dirhodium molecules
in the asymmetric unit (Rh4C151.50H164N28O48.50S8, each dirhod-
ium core containing one molecule of acetonitrile and one
molecule of benzonitrile coordinated in the axial sites). Crystals
were triclinic, space group P1, with a ) 14.4850(9), b )
15.0017(9), c ) 20.5200(12) Å, V ) 4121.0(4) Å3, Z ) 1, F(calc)
) 1.540 Mg m-3. The unit cell additionally contained half a
molecule of methanol. Data were collected out to 2θ ) 55.11°
by the ω-scan technique at -100 °C using graphite-monochro-
matized Mo KR radiation (λ ) 0.71073 Å); 65840 reflections
were measured of which 37331 were unique [Rint(F2) ) 0.0338].
The structure was solved by direct methods and refined by
full-matrix least-squares on F2 with anisotropic displacement
parameters for the non-hydrogen atoms. The hydrogen atoms
were calculated and placed in idealized positions throughout
the refinement and convergence process. The final Rw(F2) )
0.1221 with a goodness of fit ) 1.054 for refining 2105
parameters. The conventional R(F) ) 0.0496 for 30413 reflec-
tions for Fo > 2σ(Fo). Data reduction, decay correction,
structure solution, and refinement were done using the
SHELXTL/PC software package.33 An absorption correction
was applied with the SADABS program.34 Tables of positional
and thermal parameters, bond lengths and angles, and torsion
angles and figures are located in a CIF file as Supporting
Information.
1
yield). H NMR (300 MHz, CDCl3) δ 7.94-7.86 (comp, 2 H),
7.63-7.49 (comp, 3 H), 5.71 (s, 1 H), 5.66-5.62 (m, 1 H), 4.39
(dd, J ) 10.5, 4.9 Hz, 1 H), 4.26 (dd, J ) 11.7, 10.5 Hz, 1 H),
4.07 (dd, J ) 11.7, 4.9 Hz, 1 H), 3.82 (s, 3 H), 3.62-3.54 (m, 1
H), 3.31-3.21 (m, 1 H), 2.14-2.02 (m, 1 H), 2.02-1.84 (comp,
2 H), 1.76-1.64 (m, 1 H); 13C NMR (75 MHz, CDCl3) δ 172.1,
170.6, 154.8, 138.1, 132.7, 129.0, 127.5, 59.7, 53.1, 49.8, 48.7,
45.0, 31.2, 24.5.
Dirhodium(II) Tetrakis{methyl 1-[1-benzenesulfonyl-
pyrrolidine-2′(S)-carbonyl]-2-oxo-imidazolidine-4(S)-car-
boxylate}, (cis-2,2)-[Rh2(4S,2′S-BSPIM)4(CH3OH)2] (15). A
solution of compound S-14 (0.720 g, 1.9 mmol) and Rh2(OAc)4
(0.105 g, 0.24 mmol) in 10 mL of anhydrous chlorobenzene was
heated at reflux for 6 h, with rapid flushing (1 flush/min)
through a Soxhlet extractor containing a thimble filled with a
mixture of 2:1 sodium carbonate/sand. The progress of the
ligand exchange reaction was monitored by HPLC using a 3.9
mm × 150 mm reverse-phase µ-Bondapak-CN column. When
conversion to 15 was complete (by HPLC), the solvent was
removed under reduced pressure, and the purple solid was
purified by passing through reverse phase Bakerbond CN resin
(methanol/acetonitrile 99:1) to yield 0.21 g of a purple solid
(0.12 mmol, 52% yield): 1H NMR (300 MHz, CDCl3) δ 8.09-
8.00 (comp, 4 H), 7.83 (d, J ) 11 Hz, 4 H), 7.59-7.42 (comp,
12 H), 5.58-5.37 (comp, 4 H), 4.22-3.80 (comp, 6 H), 3.78-
3.67 (comp, 4 H), 3.85 (s, 6 H), 3.80 (s, 6 H), 3.44-3.30 (comp,
6 H), 2.21-2.07 (comp, 6 H), 1.98-1.86 (comp, 8 H), 1.83-
1.63 (comp, 8 H), 0.85-0.82 (comp, 4 H); 13C NMR (75 MHz,
CDCl3) δ 174.4, 173.6, 170.1, 170.0, 164.6, 163.0, 139.0, 138.8,
132.5, 132.3, 128.6, 129.1, 127.0, 58.9, 57.5, 57.0, 52.5, 52.6,
48.8, 48.0, 47.2, 45.0, 31.5, 31.1, 24.1, 24.0, 23.7, 23.6. [R]25
D
) -240(2) (c 0.06, CH2Cl2). HRMS for (C64H72N12O24Rh2S4)+
theoretical 1726.1775 [M]+; found 1726.1085. Anal. Calcd for
Rh2C64H80N12O28S4 (containing 2 molecules of water as axial
ligands and 2 unassociated water molecules): C, 42.72; H, 4.48;
N, 9.34. Found: C, 42.84; H, 4.51; N, 8.90.
X-ray Structure of 15. Crystals grew as red blocks from
dry methanol/dichloromethane/acetonitrile (90:9:1). The data
crystal had approximately orthogonal dimensions 0.46 × 0.39
× 0.25 mm3. Crystals of 15 (Rh2C71H80N12O32S4, containing 2
molecules of methanol coordinated in the axial sites) were
orthorhombic, space group P2(1)2(1)2(1), with a ) 15.934(3),
b ) 22.317(5), c ) 24.695(5) Å, V ) 8782(3) Å3, Z ) 4, F(calc)
) 1.473 Mg m-3. The unit cell contained 5 additional molecules
of methanol and 1 molecule of water. Data were collected out
to 2θ ) 60° by the ω-scan technique at 20 °C using graphite-
monochromatized Mo KR radiation (λ ) 0.71073 Å); 89210
Acknowledgment. The National Institutes of Health
(GM 46503) and the National Science Foundation
(CHE9610374, X-ray diffractometer) supported this
work. The authors thank Dr. Christopher Welch (Merck,
Inc.) for his generous donation of chromatographically
separated trans-2-phenylcyclopropanecarboxylic acid
enantiomers.
reflections were measured of which 18261 were unique [Rint
-
(F2) ) 0.0867]. The structure was solved by direct methods
and refined by full-matrix least-squares on F2 with anisotropic
displacement parameters for the non-hydrogen atoms. The
hydrogen atoms were calculated and placed in idealized
positions throughout the refinement and convergence process.
The final Rw(F2) ) 0.1846 with a goodness of fit ) 1.059 for
refining 958 parameters. The conventional R(F) ) 0.0705 for
6955 reflections for Fo > 3σ(Fo). Data reduction, decay correc-
tion, structure solution, and refinement were done using the
SHELXTL/PC software package.33 An absorption correction
was applied with the SADABS program.34 Tables of positional
and thermal parameters, bond lengths and angles, and torsion
angles and figures are located in a CIF file as Supporting
Information.
Supporting Information Available: Additional experi-
1
mental details, H and 13C NMR spectra, tables of positional
and thermal parameters, and X-ray crystallographic informa-
tion in CIF format. This material is available free of charge
via the Internet at http://pubs.acs.org.
JO050609O
J. Org. Chem, Vol. 70, No. 13, 2005 5301