Ruffling-Induced Chirality: Synthesis, Metalation, and Optical Resolution of Highly Nonplanar, Cyclic, Benzimidazole-Based Ligands

Article abstract of DOI:10.1021/ja030196d

Expedient five-step syntheses of a cyclic bis(benzimidazole)-based amide 5 and two sterically more hindered analogues 23-24 have been developed. These amides are chiral due to the inherent ruffling of the macrocyclic plane. Racemization of the optical ant

Full text of DOI:10.1021/ja030196d

Published on Web 12/16/2003
Ruffling-Induced Chirality: Synthesis, Metalation, and Optical
Resolution of Highly Nonplanar, Cyclic, Benzimidazole-Based
Ligands
Tomasz Fekner, Judith Gallucci, and Michael K. Chan*
Contribution from the Department of Chemistry, The Ohio State UniVersity,
100 West 18th AVenue, Columbus, Ohio 43210
Received March 31, 2003; E-mail: chan@chemistry.ohio-state.edu
Abstract: Expedient five-step syntheses of a cyclic bis(benzimidazole)-based amide 5 and two sterically
more hindered analogues 23-24 have been developed. These amides are chiral due to the inherent ruffling
of the macrocyclic plane. Racemization of the optical antipodes of these compounds has been studied
using dynamic chiral stationary phase HPLC. These studies reveal that, while the parent amide 5 racemizes
rapidly, for the sterically more hindered amides 23-24, the rate of racemization is significantly reduced.
Bis(benzimidazole)-based amides 5 and 23-24 form stable Ni(II) complexes 25-27, respectively. Like
their parent ligands, complexes 25-27 are chiral due to their highly ruffled geometry. Studies of these
complexes by chiral stationary phase HPLC reveal that metalation leads to a much lower rate of racemization.
Incorporation of a strap can slow racemization even further. A series of strapped cyclic amides 54-57,
along with their corresponding dimers 58-61, have been prepared. The rate of racemization for amides
54-57 is strongly dependent on the length of the strap. X-ray single-crystal structure analysis of the Ni(II)
complex of strapped amide 54 reveals that the bis(benzimidazole) core retains its highly ruffled shape,
with the two phenyl rings of the macrocycle located anti to the strap. Chiral separation of strapped ligands
54-57 and their corresponding Ni(II) complexes is shown to be facile by chiral stationary phase HPLC.
Introduction
Over the past decade, the study of nonplanar distortions in
polyazamacrocyclic (PAM) ligands and their metal complexes
has been vigorously pursued, and a remarkable plethora of out-
of-plane deformation modes have been unraveled.1,2 In particu-
lar, because protein bound porphyrinoid pigments play a key
role in many vital biological systems, the main focus of attention
has been on the synthesis, spectroscopy, and X-ray crystal-
lography of porphyrins with a high degree of peripheral
substitution.3 Structural analyses of meso-tetraalkylporphyrins
(1, R ) H, R1 ) alkyl; Figure 1) have received particular
Figure 1. “Traditional” polyazamacrocyclic ligands.
attention, because severe quasi-equatorial interactions between
the substituents R1 attached to the meso carbons and the
neighboring pyrrole rings lead to large nonplanar distortions.1,2h
Studies of these synthetic porphyrins have contributed signifi-
cantly to the growing awareness that nonplanar distortions have
a profound effect on the biochemical, catalytic, and spectro-
scopic properties of porphyrins and their associated metal
complexes.
Studies of other PAM systems have also been conducted to
understand the effect of nonplanar distortions on their properties.
For example, Kobayashi et al. have recently reported the
synthesis and structural characterization of highly distorted
phthalocyanines (e.g., 2, R ) Ph; Figure 1).4 These nonplanar
(1) For a comprehensive review on the synthesis and properties of highly
nonplanar porphyrins and related compounds, see: Senge, M. O. In The
Porphyrin Handbook; Kadish, K. M., Smith, K. M., Guilard, R., Eds.;
Academic Press: San Diego, 2000; Vol. 1, Chapter 6, pp 239-347.
(2) (a) Wasbotten, I. H.; Wondimagegn, T.; Ghosh, A. J. Am. Chem. Soc. 2002,
124, 8104-8116. (b) Ogura, H.; Yatsunyk, L.; Medforth, C. J.; Smith, K.
M.; Barkigia, K. M.; Renner, M. W.; Melamed, D.; Walker, F. A. J. Am.
Chem. Soc. 2001, 123, 6564-6578. (c) Wertsching, A. K.; Koch, A. S.;
DiMagno, S. G. J. Am. Chem. Soc. 2001, 123, 3932-3939. (d) Mizuno,
Y.; Aida, T.; Yamaguchi, K. J. Am. Chem. Soc. 2000, 122, 5278-5285.
(e) Wondimagegn, T.; Ghosh, A. J. Phys. Chem. A 2000, 104, 4606-4608.
(f) Song, X.-Z.; Jaquinod, L.; Jentzen, W.; Nurco, D. J.; Jia, S.-L.; Khoury,
R. G.; Ma, J.-G.; Medforth, C. J.; Smith, K. M.; Shelnutt, J. A. Inorg.
Chem. 1998, 37, 2009-2019. (g) Lin, C.-Y.; Hu, S.; Rush, T.; Spiro, T.
G. J. Am. Chem. Soc. 1996, 118, 9452-9453. (h) Jentzen, W.; Simpson,
M. C.; Hobbs, J. D.; Song, X.; Ema, T.; Nelson, N. Y.; Medforth, C. J.;
Smith, K. M.; Veyrat, M.; Mazzanti, M.; Rammasseul, R.; Marchon, J.-
C.; Takeuchi, T.; Goddard, W. A.; Shelnutt, J. A. J. Am. Chem. Soc. 1995,
117, 11085-11097. (i) Gentemann, S.; Medforth, C. J.; Forsyth, T. P.;
Nurco, D. J.; Smith, K. M.; Fajer, J.; Holten, D. J. Am. Chem. Soc. 1994,
116, 7363-7368. (j) Barkigia, K. M.; Renner, M. W.; Furenlid, L. R.;
Medforth, C. J.; Smith, K. M.; Fajer, J. J. Am. Chem. Soc. 1993, 115, 3627-
3635. (k) Medforth, C. J.; Senge, M. O.; Smith, K. M.; Sparks, L. D.;
Shelnutt, J. A. J. Am. Chem. Soc. 1992, 114, 9859-9869.
(3) For a review on nonplanar porphyrins in proteins, see: Shelnutt, J. A.;
Song, X.-Z.; Ma, J.-G.; Jia, S.-L.; Jentzen, W.; Medforth, C. J. Chem. Soc.
ReV. 1998, 27, 31-41.
9
10.1021/ja030196d CCC: $27.50 © 2004 American Chemical Society
J. AM. CHEM. SOC. 2004, 126, 223-236
223

A R T I C L E S
Fekner et al.
phthalocyanines represent yet another important class of PAM
ligands and their metal complexes.
One convention that has evolved as a result of the study of
nonplanar PAM ligands, particularly porphyrins, is the clas-
sification of their distortions into four main modes: ruffling,
saddling, doming, and waving, as proposed by Scheidt and Lee.5
These modes serve as the basic set for the various distortions
that are possible. Distortions not belonging to one particular
mode can be described as a combination of the four fundamental
modes.
Figure 2. The original bis(benzimidazole) ligand 3 and its complexes 4.
Our interest in this arena stems from our long-standing desire
to develop topologically novel chiral transition-metal complexes
of PAM ligands as catalysts for asymmetric transformations.
In particular, structural analysis of high-valent transition-metal
oxo complexes of ruffled PAM ligands indicates that a side-on
perpendicular approach of an (E)-alkene (or a similarly shaped
substrate) to the metal center of such a complex should not
encounter any significant steric obstacles.6 As unfunctionalized
prochiral (E)-alkenes are known to be “difficult” substrates for
standard catalytic asymmetric epoxidation protocols,7 the de-
velopment of chiral, nonplanar transition-metal complexes of
PAM ligands could, in principle, broaden the range and scope
of existing methodologies. Recently, Gilheany et al. reported
on highly enantioselective epoxidation of alkenes catalyzed by
stable chromium(V)-oxo salen complexes which, in stark
contrast to Jacobsen’s catalytic system, provide much better
enantioselectivity for (E)-alkenes as compared to their (Z)-
isomers.8,9 These results were attributed to an increased twist
of the ligand backbone for the chromium salen complexes,
which creates a favorable trajectory for the (E)-alkene to
approach the chromium-oxo bond.
Figure 3. The amide-based bis(benzimidazole) ligand 5 and its metal
complexes 6.
series of its transition-metal complexes 4 (M ) Mn, Fe, Co,
Ni, Cu) (Figure 2).11 Although some of these compounds were
demonstrated to catalyze epoxidation of styrene with promising
turnover numbers (∼100), the high polarity and low solubility
of ligand 3 and its complexes 4 precluded their optical
resolution. In particular, we were not able to estimate the barrier
to racemization for these compounds, so their stereointegrity
could not be unequivocally evaluated.
As one approach toward preparing a chirally stable BBZ
ligand, we decided to modify the structure of the parent BBZ
macrocycle 3.12 The amide C-N bond is known to possess a
partial double-bond character due to donation of the nonbonding
electron pair on the nitrogen.13 It was, therefore, envisaged that
isosteric replacement14 of the imine bonds in the parent BBZ
ligand 3 with amide bonds (as in ligand 5; Figure 3) would
lead to macrocycles with a similar, ruffled geometry.15 It was
Because the parent porphyrin 1 (R ) R1 ) H) and phthalo-
cyanine 2 (R ) H) nuclei (Figure 1) are highly symmetrical,
their uniformly fully R-substituted saddled or ruffled analogues
remain achiral. Cognizant of this, we focused our attention on
the novel non-porphyrin bis(benzimidazole)-based ligands and
their transition-metal complexes, which display a lower sym-
metry than the traditional PAMs (e.g., 1 and 2) and can be
rendered chiral by being locked into a ruffled conformation.10
(11) (a) Payra, P.; Hung, S.-C.; Kwok, W.-H.; Johnston, D.; Gallucci, J.; Chan,
M. K. Inorg. Chem. 2001, 40, 4036-4039. (b) Kwok, W.-H.; Zhang, H.;
Payra, P.; Duan, M.; Hung, S.-C.; Johnston, D. H.; Gallucci, J.; Skrzypczak-
Jankun, E.; Chan, M. K. Inorg. Chem. 2000, 39, 2367-2376. (c) Payra,
P.; Zhang, H.; Kwok, W.-H.; Duan, M.; Gallucci, J.; Chan, M. K. Inorg.
Chem. 2000, 39, 1076-1080.
Recently, we reported on the synthesis and structural char-
acterization of the bis(benzimidazole) (BBZ) ligand 3 and a
(4) Kobayashi, N.; Fukuda, T.; Ueno, K.; Ogino, H. J. Am. Chem. Soc. 2001,
123, 10740-10741.
(5) Scheidt, W. R.; Lee, Y.-J. Struct. Bonding (Berlin) 1987, 64, 1-70.
(6) (a) Collman, J. P.; Zhang, X.; Lee, V. J.; Uffelman, E. S.; Brauman, J. E.
Science 1993, 263, 1404-1411. (b) Groves, J. T.; Viski, P. J. Org. Chem.
1990, 55, 3628-3634. (c) Groves, J. T.; Nemo, T. E. J. Am. Chem. Soc.
1983, 105, 5786-5791.
(7) (a) Jacobsen, E. N.; Wu, M. H. In ComprehensiVe Asymmetric Catalysis;
Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer: New York,
1999; Chapter 18.2, pp 649-677. (b) Katsuki, T. In Catalytic Asymmetric
Synthesis, 2nd ed.; Ojima, I., Ed.; Wiley-WCH: New York, 2000; Chapter
6B.
(8) (a) O’Mahony, C. P.; McGarrigle, E. M.; Renehan, M. F.; Ryan, K. M.;
Kerrigan, N. J.; Bousquet, C.; Gilheany, D. G. Org. Lett. 2001, 3, 3435-
3438. (b) Daly, A. M.; Renehan, M. F.; Gilheany, D. G. Org. Lett. 2001,
3, 663-666. (c) Daly, A. M.; Dalton, C. T.; Renehan, M. F.; Gilheany, D.
G. Tetrahedron Lett. 1999, 40, 3617-3620. (d) Bousquet, C.; Gilheany,
D. G. Tetrahedron Lett. 1995, 36, 7739-7742.
(9) For a comprehensive overview of mechanistic aspects of transition-metal
catalyzed asymmetric epoxidation of alkenes, see: Dalton, C. T.; Ryan,
K. M.; Wall, V. M.; Bousquet, C.; Gilheany, D. G. Top. Catal. 1998, 5,
75-91.
(10) In porphyrins, the ruffled (ruf) distortion involves the alternate displacement
of the meso carbons above and below the mean 4N-plane. By analogy, for
BBZ ligands and their complexes, the meso carbons are those within the
16-atom inner core that are not adjacent to any of the four ligating nitrogen
atoms.
(12) Selected examples of amide-based ligands and complexes: (a) Moreira,
R. F.; When, P. M.; Sames, D. Angew. Chem., Int. Ed. 2000, 39, 1618-
1621. (b) Miller, C. G.; Gordon-Wylie, S. W.; Horwitz, C. P.; Strazisar, S.
A.; Peraino, D. K.; Clark, G. R.; Weintraub, S. T.; Collins, T. J. J. Am.
Chem. Soc. 1998, 120, 11540-11541. (c) Dangel, B.; Clarke, M.; Haley,
J.; Sames, D.; Polt, R. J. Am. Chem. Soc. 1997, 119, 10865-10866. (d)
Zhao, S.-H.; Ortiz, P. R.; Keys, B. A.; Davenport, K. G. Tetrahedron Lett.
1996, 37, 2725-2728. (e) Yoon, H.; Wagler, T. R.; O’Connor, K. J.;
Burrows, C. J. J. Am. Chem. Soc. 1990, 112, 4568-4570. (f) Wagler, T.
R.; Fang, Y.; Burrows, C. J. J. Org. Chem. 1989, 54, 1584-1589. (g)
Kimura, E.; Shionoya, M.; Okamoto, M.; Nada, H. J. Am. Chem. Soc. 1988,
110, 3680-3682. (h) Che, C.-M.; Cheng, W.-K. J. Chem. Soc., Chem.
Commun. 1986, 1443-1444. (i) Kimura, E.; Koike, T.; Machida, R.; Nagai,
R.; Kodama, M. Inorg. Chem. 1984, 23, 4181-4188. (j) Fabbrizzi, L.;
Perotti, A.; Poggi, A. Inorg. Chem. 1983, 22, 1411-1412.
(13) Wiberg, K. B. In The Amide Linkage; Greenberg, A., Breneman, C. M.,
Liebman, J. F., Eds.; Wiley & Sons: New York, 2000; Chapter 2, pp 33-
46.
(14) Patani, G. A.; LaVoie, E. J. Chem. ReV. 1996, 96, 3147-3176.
(15) A sterically restricted C-N bond seems to be a prerequisite for this class
of compounds to adopt a highly ruffled geometry. This was aptly
demonstrated in the case of the cyclic amine obtained by reduction of the
Schiff base 3. Single-crystal X-ray analysis revealed in this case that the
two benzimidazole rings are parallel to each other, and the ruffled geometry
of the parent imine 3 is not retained. Fekner, T.; Gallucci, J.; Chan, M. K.,
unpublished results.
9
224 J. AM. CHEM. SOC. VOL. 126, NO. 1, 2004

Ruffling-Induced Chirality
A R T I C L E S
Scheme 1. Retrosynthetic Analysis of Cyclic Amides 7
Figure 4. ORTEP drawing (50% probability thermal ellipsoids) of the
molecular structure of the cyclic amide 5. The oxygen and nitrogen atoms
are hatched. Hydrogen atoms are omitted for clarity.
nated (1 atm H2, Pd/C) to give amino acids 20-22. Cyclocon-
densation of these amino acids under high-dilution conditions
(20-23 mM in CH2Cl2) in the presence of the BOP Reagent/
N-methylmorpholine (BOP/NMM)18 gave the requisite cyclic
amides 5 and 23-24 in moderate (46% for 5) to excellent (78%
and 86%, respectively, for 23 and 24) yield. We attribute the
increased efficiency of cyclization of amino acids 21 and 22,
relative to amino acid 20, to the restricted rotation about the
amide C-N bond for the initial monamide intermediate.
Presumably, for the linear dimer (monoamide) derived from
amino acids 21 and 22, the free amino and activated carboxylic
groups are forced into closer proximity, which favors intramo-
lecular cyclodimerization over intermolecular oligomerization.
The parent cyclic amide 5 and its more sterically hindered
analogues 23-24 are high-melting (mp > 260 °C) white
powders, which in their purified forms are only sparingly soluble
in common organic solvents (CH2Cl2, CHCl3). The very low
solubility of amides 23 and 24 in EtOAc, combined with their
remarkably high-yielding preparation from amino acids 21 and
22, respectively, greatly facilitates their efficient isolation from
the crude reaction mixtures by simple reflux in EtOAc, followed
by separation of the desired products by filtration.
Single-crystal X-ray analysis of the parent cyclic amide 5
(Figure 4) unequivocally demonstrates that, as expected from
the molecular modeling studies, the molecule adopts a highly
distorted, ruffled, chiral structure. The four aromatic units
(benzimidazoles and o-substituted phenyl rings) are alternately
tilted above and below the mean 4N-plane (defined by the
ligating benzimidazole and amide nitrogen atoms) of the
molecule. These features result in the formation of mutually
perpendicular valleys above and below the macrocyclic plane
and give rise to a macrocycle with two distinct diastereotopic
faces. Dynamic CSP HPLC studies (Figure 5) confirm that the
nonplanar, chiral geometry found in the crystalline state of amide
5 is retained in solution.
also expected that the cyclic amide 5 and its analogues would
be more soluble in common organic solvents than the parent
BBZ ligand 3. Because diamide 5 was predicted to behave like
a typical tetradentate donor, and form neutral complexes 6 with
divalent cations (via metalation and concomitant loss of the
amide protons), such complexes should also be more soluble
in organic media. This, in turn, would be beneficial for their
potential resolution by chiral stationary phase (CSP) HPLC. As
the first step toward these goals, we highlight herein the
synthesis, structural analysis, and metalation of ligand 5 and a
series of its substituted as well as strapped analogues.
Results and Discussion
A Cyclic, Bis(benzimidazole)-Based Amide Ligand. Our
first goal was the preparation of the cyclic amide 5 and its more
substituted congeners. Because of the C2 symmetry of the target
molecules, their synthesis is greatly simplified. Thus, retrosyn-
thetic cleavage (Scheme 1) of amide 7 led to amino acid 8.
Disconnecting further, amine 9, available in multigram quantities
via a five-step synthesis from commercially available 3-nitro-
phthalic acid,16 and an appropriate o-nitrobenzoic acid 10 were
selected as substrates for the preparation of the requisite amino
acid 8.
The synthesis commenced with the generation of benzimid-
azoles 11-13 (Scheme 2), which were obtained from the known
diamine 916 by a simple, one-pot procedure that is amenable to
large-scale preparations. Thus, acylation of amine 9 with 1 equiv
of an appropriate 2-nitrobenzoyl chloride in CH2Cl2 in the
presence of Et3N gave, in each case, a mixture of acylated
products. No purification was necessary at this stage, and, after
the volatiles were removed under reduced pressure, the residue
was dissolved in glacial AcOH and heated to reflux in the
presence of AcONa.17 Chromatographic purification gave benz-
imidazoles 11-13 in high overall yield in each case (>90%
from 9). Deprotonation with NaH in THF, followed by
chemoselective N-alkylation with MeI, gave benzimidazoles
14-16 as single isomers in high yield. Saponification of the
ester group by boiling in 5% NaOH/MeOH furnished, after
neutralization with HClaq, acids 17-19, which were hydroge-
Although we were pleased to establish that the chirality of
amide 5 is preserved in solution, the presence of the temperature-
dependent plateau between the two peaks that correspond to
the optical antipodes 5A and 5B indicated that interconversion
of the enantiomers is fast even on the time-scale of the HPLC
experiment.19 Although the plateau-shaped CSP HPLC profiles
(16) Chapman, E.; Stephen, H. J. Chem. Soc. 1925, 127, 1791-1797.
(17) (a) Preston, P. N. In The Chemistry of Heterocyclic Compounds; Wiess-
berger, A., Taylor, E. C., Eds.; John Wiley & Sons: New York, 1981; pp
5-12. (b) Go¨ker, H.; O¨ lgen, S.; Ertan, R.; Akgu¨n, H.; O¨ zbey, S.; Kendi,
E.; Topcu, G. J. Heterocycl. Chem. 1995, 32, 1767-1773.
(18) BOP Reagent: Benzotriazol-1-yloxytris(dimethylamino)phosphonium hexa-
fluorophosphate.
9
J. AM. CHEM. SOC. VOL. 126, NO. 1, 2004 225

A R T I C L E S
Fekner et al.
Scheme 2. Synthesis of Cyclic Amides 5, 23-24, and Their Ni(II) Complexes 25-27a
a Reagents and conditions: (a) 3-R-substituted 2-nitrobenzoyl chloride, Et3N, CH2Cl2, 0 °C f room temperature, 3-6 h, then AcONa, AcOH, reflux,
15 h; (b) NaH, MeI, THF, 0 °C f room temperature, 12-13 h; (c) NaOH, MeOH, reflux, 25-50 min; (d) H2 (1 atm), Pd/C, MeOH, room temperature,
5-24 h; (e) BOP, NMM, CH2Cl2, room temperature, 6-7 days; (f) Ni(OAc)2‚4H2O, MeOH, reflux, 1.5-5 h.
were not computer-simulated, comparison with the available
data19a indicates that the half-life for racemization at room
temperature for cyclic amide 5 is far less than 1000 s (an
arbitrarily set threshold for feasibility of resolution of optical
antipodes or other fast-interconverting entities).20
Because metalation was expected to increase the barrier to
racemization for this class of ligands, the cyclic amide 5
(Scheme 2) was heated to reflux in MeOH in the presence of
Ni(OAc)2 to give a diamagnetic, bright-orange Ni(II) complex
25 in nearly quantitative yield. Complex 25 exhibits a carbonyl
stretch at 1562 cm-1 that corresponds to a 103 cm-1 batho-
chromic shift upon concomitant deprotonation and metalation
of ligand 5.
Single-crystal X-ray analysis (Figure 6) revealed that complex
25 retains the ruffled structure of its parent ligand 5 (Figure 4)
and, hence, its inherent chirality. Unlike the charged metal
complexes 4 formed with the original bis(benzimidazole) ligand
3, however, the neutral nickel complex is, as predicted, soluble
in common organic solvents (especially CH2Cl2/MeOH mix-
tures). This property enabled its successful optical resolution
by CSP HPLC (Figure S1).
two prominent representatives of other classes of PAMs.21 The
first of them is Senge’s meso-tetrasubstituted porphyrin complex
t
Zn(II)‚H-21‚(pyridine) (R ) H, R1 ) Bu)22 that, like bis-
(benzimidazole) amide ligand 5 and complex 25, exhibits a
distinct ruffling. With an average distortion of the central 16
atoms of 0.47 Å, and a maximum displacement of its meso
carbons by 1.01 Å, Senge’s complex has been reported to be
one of the most distorted porphyrins characterized to date.23
Similarly, Kobayashi’s saddled phthalocyanine 2 (R ) Ph)
is the most deformed phthalocyanine (regardless of the deforma-
tion mode) reported so far.4 For this compound, the average
nonplanar distortion of its 16 central atoms is 0.22 Å, with a
maximum out-of-N-plane distortion reaching 0.40 Å. In the
saddle mode of distortion, it is the pyrrole carbons that exhibit
the greatest nonplanar deviations. Thus, while a direct com-
parison of the individual distortions of phthalocyanine 2 with
those for bis(benzimidazole) amide ligand 5 and its Ni(II)
complex 25 is not appropriate, the absolute magnitude of their
average distortions can be compared.
The distortions exhibited by cyclic amide 5 and its Ni(II)
complex 25 (Figure 7) are significantly larger than the distortions
observed for either Senge’s porphyrin or Kobayashi’s phtha-
locyanine. The average out-of-4N-plane distortion for the central
16 atoms of ligand 5 is 0.54 Å, that is, 0.07 Å (or 13%) higher
than that for Senge’s complex, and more than twice of that
exhibited by Kobayashi’s phthalocyanine. Similarly, the maxi-
mum distortion of 1.29 Å for the meso carbons in the cyclic
amide 5 is considerably larger than that for Senge’s porphyrin
(1.01 Å). Metalation of amide 5 leads to increased ruffling of
the macrocycle - the average nonplanar distortion for the central
The unusually high propensity of bis(benzimidazole) ligands
and their metal complexes to accommodate high levels of
nonplanar distortions is best illustrated by their comparison with
(19) (a) Spivey, A. C.; Charbonneau, P.; Fekner, T.; Hochmuth, D. H.;
Maddaford, A.; Malardier-Jugroot, C.; Redgrave, A. J.; Whitehead, M. A.
J. Org. Chem. 2001, 66, 7394-7401. (b) Gasparrini, F.; Lunazzi, L.;
Mazzanti, A.; Pierini, M.; Pietrusiewicz, K. M.; Villani, C. J. Am. Chem.
Soc. 2000, 122, 4776-4780. (c) Spivey, A. C.; Fekner, T.; Spey, S. E.;
Adams, H. J. Org. Chem. 1999, 64, 9430-9443. (d) Oxelbark, J.;
Allenmark, S. J. Org. Chem. 1999, 64, 1483-1486. (e) Wolf, C.; Pirkle,
W. H.; Welch, C. J.; Hochmuth, D. H.; Ko¨nig, W. A.; Chee, G.-L.; Charlton,
J. L. J. Org. Chem. 1997, 62, 5208-5210. (f) Gasparini, F.; Lunazzi, L.;
Misiti, D.; Villani, C. Acc. Chem. Res. 1995, 28, 163-170. (g) Jung, M.;
Schurig, V. J. Am. Chem. Soc. 1992, 114, 529-534. (h) Veciana, J.; Crespo,
M. I. Angew. Chem., Int. Ed. Engl. 1991, 30, 74-76. (i) Mannschreck, A.;
Zinner, H.; Pustet, N. Chimia 1989, 43, 165-166. (j) Mannschreck, A.;
Andert, D.; Eiglsperger, A.; Gmahl, E.; Buchner, H. Chromatographia 1988,
25, 182-188. (k) Eiglsperger, A.; Kastner, F.; Mannschreck, A. J. Mol.
Struct. 1985, 126, 421-432. (l) Bu¨rkle, W.; Karfunkel, H.; Schurig, V.
Chromatography 1984, 288, 1-14.
(21) For a compilation of data used to calculate the nonplanar distortions, see
Table S21.
(22) Senge, M. O.; Ema, T.; Smith, K. M. J. Chem. Soc., Chem. Commun. 1995,
733-734.
(23) For other examples of highly distorted porphyrins (meso carbon displace-
ments up to 1.05 Å), see: (a) Senge, M. O.; Renner, M. W.; Kalisch, W.
W.; Fajer, J. J. Chem. Soc., Dalton Trans. 2000, 3, 381-385. (b) Nelson,
N. Y.; Medforth, C. J.; Nurco, D. J.; Jia, S.-L.; Shelnutt, J. A.; Smith, K.
M. Chem. Commun. 1999, 2071-2072.
(20) Oˆ ki, M. Top. Stereochem. 1983, 14, 1-81.
9
226 J. AM. CHEM. SOC. VOL. 126, NO. 1, 2004

Ruffling-Induced Chirality
A R T I C L E S
Figure 6. ORTEP drawing (50% probability thermal ellipsoids) of the
molecular structure of the Ni(II) complex 25. The nickel, oxygen, and
nitrogen atoms are hatched. Hydrogen atoms are omitted for clarity.
Figure 7. Deviation from planarity (Å) relative to the mean 4N-plane for
the core 16 atoms in amide 5 and its Ni(II) complex 25.
undergoing fast (see Figure 5) racemization before it re-
metalates. Facile dissociation of the nickel ion from its amide-
based complexes has been previously reported.12e
Sterically Hindered, Bis(benzimidazole)-Based Amide
Ligands. A close inspection of the crystal structure of ligand 5
(Figure 4) revealed that to undergo the macrocyclic inversion
(racemization), the amide carbonyl groups must move to the
opposite face of the neighboring o-substituted phenyl rings.
Thus, it seemed reasonable that increasing the steric bulk at
the C-3 position of both phenyl rings (see Scheme 1 for
numbering) would introduce a steric clash between these
substituents and the carbonyl groups that, in turn, would hinder
this racemization. Moreover, from the point of view of asym-
metric catalysis, introducing a steric bulk at this position would
further discriminate between the two possible trajectories, A
and B (Figure 8), of the side-on approach of an (E)-alkene to
the oxo group generated at the metal center.25 As these
trajectories require the opposite diastereotopic faces of the (E)-
alkene to be directed toward the metal center, excluding or, at
Figure 5. Macrocyclic inversion for cyclic amide 5 (4.2 µg) as determined
by CSP HPLC (Chiralcel OD column, 4.6 mm × 25 cm; hexanes/2-
propanol, 60/40; 1 mL min-1).
atoms of ligand 5 increases from 0.54 to 0.58 Å as observed
for the resulting Ni(II) complex 25. Similarly, the average
nonplanar displacement of the meso carbons increases from 0.94
to 1.03 Å.
The enantiomers of complex 25 (Figure S1) are significantly
more stable to racemization than those of ligand 5 (Figure 5).
Baseline separation of the racemic complex 25 can be achieved
by CSP HPLC even at 40 °C (the upper operational limit for
the column used). Nevertheless, an enantiomerically pure sample
of complex 25 still racemizes at room temperature at an initial
rate of ∼1% ee/day.24 Inasmuch as direct racemization of Ni-
(II)-complex 25 cannot be ruled out, it is also plausible that
complex 25 is in equilibrium with ligand 5, with the latter
(24) After storage in solution (2-propanol/haxanes) at room temperature for 6
days, an optically pure (ee > 99.9%) sample of complex 25 was determined
to have ee ) 93.6% by CSP HPLC. It can, therefore, be estimated that at
room temperature ligand 5 racemizes ∼104 times faster than its Ni(II)
complex 25.
9
J. AM. CHEM. SOC. VOL. 126, NO. 1, 2004 227

A R T I C L E S
Fekner et al.
Figure 8. The two possible side-on approaches of an (E)-alkene toward
the metal-oxo center of the ruffled complex. R1 is sterically less demanding
than R2, and the metal-oxo bond is directed above the macrocyclic plane.
Figure 9. ORTEP drawing (50% probability thermal ellipsoids) of the
molecular structure of the Ni(II) complex 26. The nickel, oxygen, and
nitrogen atoms are hatched. Hydrogen atoms are omitted for clarity.
least, introducing a significant bias against one of them would
result in a more efficient chirality transfer from the chiral catalyst
(a ruffled metal complex) to the substrate (an alkene). The
bulkier R groups would discriminate against trajectory A, thus
ensuring a clear preference for the alternative mode of approach,
B, which is not expected to be affected by the increased size of
the R group. The above reasoning prompted us to synthesize
cyclic amides 23 and 24 from amine 9 by a procedure analogous
to that used for the preparation of the parent amide 5 (Scheme
2).
Metalation of ligands 23 and 24 (Scheme 2) with Ni(OAc)2
in MeOH furnished orange complexes 26 and 27, respectively,
in high yield. X-ray single-crystal structure analysis of complex
26 (Figure 9) confirmed that the molecule adopts a shape similar
to that observed for the unsubstituted complex 25 (Figure 6).
The potential steric clash between the amide carbonyls and the
methoxy groups is readily apparent from the structure. Com-
plexes 26 and 27 could be conveniently resolved by CSP HPLC
(Figures S4 and S5, respectively).
Strapped, Bis(benzimidazole)-Based Amide Ligands. Faced
with the problem of relatively fast racemization of ligands 5
and 23-24, we decided to explore strategies that might prevent
macrocyclic inversion altogether. One appealing strategy was
to span the two external benzimidazole nitrogen atoms (which
are not involved in metal binding) with an appropriate strap
that would make racemization disallowed due to steric reasons.27
Moreover, as in other macrocyclic cryptands,28 such a strap
could potentially be beneficial for both the thermodynamic and
the kinetic stability of the complexes, as compared to their
monocyclic analogues 5 and 23-24. Retrosynthetic analysis
(Scheme 3) indicated that the strapped ligand 28 could be
prepared from amino acid 29 via double intramolecular amide
bond formation. Further disconnection led to the previously
prepared (see Scheme 2) benzimidazole 11 and a properly
functionalized strap 30 (X - a leaving group).
CSP HPLC studies revealed that, as for the parent ligand 5,
cyclic amides 23 and 24 adopt a nonplanar, chiral conformation
in solution. Baseline separation of racemic cyclic amides 23
and 24 could be achieved even at 40 °C (Figures S2 and S3,
respectively), demonstrating their superior stereointegrity relative
to that of the parent amide 5. Nevertheless, the barriers to
macrocyclic inversion were still not sufficiently high to enable
storage of optically pure samples of amides 23 and 24 for a
prolonged period of time. When an enantiomerically enriched
(ee ) 84%) sample of amide 23 was kept in solution at room
temperature for 7 h, its optical purity decreased to 42%. After
an additional 11 h at room temperature, the sample was nearly
racemic (ee ) 13%). Because o-methyl groups are more efficient
than o-methoxy groups in increasing the barrier to atropisomer-
ization in biphenyl derivatives,26 it was expected that a similar
pattern would be observed for the barrier to macrocyclic
inversion in cyclic amides 23 and 24. Indeed, when an optically
pure (ee > 99.9%) sample of cyclic amide 24 was kept in
solution at room temperature for 35 h, it was determined to be
of 97.1% optical purity. After 12.3 and 31.8 days at room
temperature, the optical purity decreased further to 71.3% and
38.9%, respectively. When kept in boiling CHCl3 (bp 61 °C),
an optically pure (ee > 99.9%) sample of cyclic amide 24 was
determined to have an optical purity of 23.6% after 14 h, and
5.5% after 25 h. Assuming reversible, first-order kinetics for
the macrocyclic inversion, amide 23 undergoes racemization
at room temperature about 90 times faster than its Me-flanked
counterpart 24.
In the design of these strapped ligands, the length of the strap
30 is crucial. The strap must be long enough to span the two
peripheral benzimidazole nitrogen atoms in ligand 28, but short
enough to preclude ligand racemization via rotation of the core
(27) For examples of strapped and capped PAM ligands, see: (a) Bencini, A.;
Bianchi, A.; Giorgi, C.; Fusi, V.; Masotti, A.; Paoletti, P. J. Org. Chem.
2000, 65, 7686-7689. (b) Wynn, T.; Hegedus, L. S. J. Am. Chem. Soc.
2000, 122, 5034-5042. (c) Hubin, T. J.; McCormick, J. M.; Collinson, S.
R.; Buchalova, M.; Perkins, C. M.; Alcock, N. W.; Kahol, P. K.;
Raghunathan, A.; Busch, D. H. J. Am. Chem. Soc. 2000, 122, 2512-2522.
(d) Puntener, K.; Hellman, M. D.; Kuester, E.; Hegedus, L. S. J. Org. Chem.
2000, 65, 8301-8306. (e) Brandes, S.; Denat, F.; Lacour, S.; Rabiet, F.;
Barbette, F.; Pallumbi, P.; Guilard, R. Eur. J. Org. Chem. 1998, 2349-
2360. (f) Lachkar, M.; Guilard, R.; Atmani, A.; De Cran, A.; Fissher, J.;
Weiss, R. Inorg. Chem. 1998, 37, 1575-1584. (g) Brandes, S.; Lacour,
S.; Denat, F.; Pallumbi, P.; Guilard, R. J. Chem. Soc., Perkin Trans. 1
1998, 639-641. (h) Denat, F.; Lacour, S.; Brandes, S.; Guilard, R.
Tetrahedron Lett. 1997, 38, 4417-4420. (i) Dapporto, P.; Paoli, P.;
Bazzicalupi, C.; Bencini, A.; Nardi, N.; Valtoncoli, B.; Fusi, V. Supramol.
Chem. 1996, 7, 195-200. (j) Springborg, J.; Olsen, C. E.; Søtofte, I. Acta
Chem. Scand. 1995, 49, 555-563. (k) Weisman, G. R.; Rogers, M. E.;
Wong, E. H.; Jasinski, J. P.; Paight, E. S. J. Am. Chem. Soc. 1990, 112,
8604-8605.
(25) Although two diastereomeric BBZ-metal-oxo complexes are possible (with
the oxo ligand either syn or anti to the phenyl rings), only one of them
(with the oxo ligand syn to the phenyl rings) is considered herein. It stems
from our finding (vide infra) that strapped BBZ ligands, that do not racemize
as rapidly as ligands 5, 23, and 24, and therefore can potentially be used
in asymmetric epoxidation, have the strap located anti to the phenyl rings.
As a result, only the opposite face of the macrocyclic plane is available
for the oxo ligand.
(26) Eliel, E. L.; Wilen, S. H. Stereochemistry of Organic Compounds; J.
Wiley: New York, 1994.
(28) Busch, D. H. Chem. ReV. 1993, 93, 847-860.
9
228 J. AM. CHEM. SOC. VOL. 126, NO. 1, 2004

Ruffling-Induced Chirality
A R T I C L E S
Scheme 4. Synthesis of Benzylic Bromides 37-41a
Scheme 3. Retrosynthetic Analysis of the Strapped Cyclic
Amide 28
cyclic amide framework underneath the strap, or simple mac-
rocyclic inversion without rotation (in this respect, ligand 5 can
be viewed as possessing a strap of the “infinite” length).
Moreover, because the strapped ligand 28 is, unlike ligand 5,
intrinsically chiral irrespective of any nonplanar distortion of
the cyclic amide core, CSP HPLC would not be a suitable means
to provide direct evidence for any kind of out-of-plane distortion
of the cyclic amide subunit (although it remains a valuable tool
in studying racemization of such ligands).
As N-alkylation of benzimidazoles with nonactivated alkyl
halides is known to be difficult,29 we opted to use benzylic
bromides as the electrophiles anchoring the strap onto the
benzimidazole ring. The design was convenient because the
length of the strap could be adjusted by selection of an
appropriate tether to join the two benzylic anchors. Thus
(Scheme 4), phenol 31 was O-alkylated with a series of R,ω-
dibromoalkanes in the presence of NaOH to give benzyl alcohols
33-36 in good yield (63-70%) and alcohol 32 in low, 29%
yield. Apparently, in the case of alkylation with 1,2-dibromo-
ethane, facile â-elimination of HBr from either the substrate or
the intermediate monoether becomes a serious side reaction.
When 1,2-dichloroethane was used as alkylating agent, the yield
of diol 32 was even lower (7%). Upon treatment with PBr3,
alcohols 32-36 were converted to the corresponding benzyl
bromides 37-41, respectively.
a Reagents and conditions: (a) 1,n-dibromoalkane (n ) 2, 4, 6, 8, 10),
NaOH, EtOH, H2O reflux, 10-48 h; (b) PBr3, CH2Cl2, 0 °C f room
temperature, 11-16 h.
amino acids 50-53 in good overall yield (87-90% over two
steps). High-dilution (4.2-9.4 mM in CH2Cl2) cyclocondensa-
tion of amino acids 50-53 in the presence of BOP/NMM gave
strapped monomers 54-57 and their corresponding dimers 58-
61 in low yield. Although pure samples of the monomers are
slightly soluble in CH2Cl2 and CHCl3 (but not EtOAc), the
solubility of analytically pure samples of the dimers in all of
these solvents is low. Additionally, as dimers 58 and 59 exhibit
a lower solubility than dimers 60 and 61 (a property that made
recording of NMR spectra of dimer 58 prohibitively difficult),
dimer solubility appears to be a function of the distance between
the two bis(benzimidazole) rings. Despite their poor solubility,
chromatographic purification (see Experimental Section) of the
crude mixtures from the cyclocondensation reaction of amino
acids 50-53 was never plagued by crystallization of the
products on the column.
Upon treatment with Ni(OAc)2 in boiling MeOH, the strapped
ligand 54 gave a deep-orange Ni(II) complex 62 in nearly
quantitative yield (Scheme 6). As seen for a series of the
nonstrapped ligands 5 and 23-24 and their Ni(II)-complexes
25-27, metalation leads to a bathochromic shift for the amide
carbonyl stretching by 102 cm-1 (from 1678 to 1576 cm-1) for
the strapped Ni(II)-complex 62 as compared to the free ligand
54. Single-crystal X-ray diffraction analysis of complex 62
(Figure 10) confirmed that the core cyclic amide adopts a ruffled
conformation, and its geometry appears to be controlled by the
position of the strap (vide infra, for discussion).
1H NMR analysis of the series of monomers 54-57 revealed
that significant differences in chemical shifts could be observed
only for the aromatic proton Ha and two pairs of diastereotopic
protons Hb/Hb′ and Hc/Hc′ (Table 1). Molecular modeling based
on the X-ray single-crystal structure of the Ni(II) complex 62
(Figure 10) indicates that these protons are located directly above
the 10-electron aromatic benzimidazoles and, thus, are within
the shielding cone of the aromatic ring current.32 Because the
intensity of the anisotropic field diminishes rapidly with distance,
the corresponding protons in the monomers possessing shorter
straps would be expected to experience a more pronounced
With the suitable set of straps in hand, the synthesis of the
geometrically restricted, strapped ligands could be undertaken.
Thus, deprotonation of benzimidazole 11 (Scheme 5) with NaH
in THF, followed by chemoselective alkylation with a series of
benzylic bromides 38-41, gave isomerically pure bis(benzim-
idazole)s 42-45, respectively, in good to excellent yield (86-
94%).30 The SnCl2-mediated reduction of the nitro groups31
furnished amines 46-49, which were then subjected to ester
hydrolysis with LiOH/THF/H2O to give, after acidic workup,
(29) Muller, G.; Bu¨nzil, J.-C. G.; Schenk, K. J.; Piguet, C.; Hopfgartner, G.
Inorg. Chem. 2001, 40, 2642-2651.
(30) Analytically pure samples of benzimidazoles 42-45 exhibit liquid-crystal-
like properties. When heated, they initially turn into hazy liquids (mp 170
°C for 42, 172 °C for 43, 82 °C for 44, and 100 °C for 45). The clearing
temperature (transition from mesophase to isotropic liquid) is significantly
higher (180 °C for 42, 186 °C for 43, 93 °C for 44, and 118 °C for 45).
The highly polar nitro groups located at each end of these elongated
molecules are most likely responsible for this phenomenon.
(31) Bellamy, F. D.; Ou, K. Tetrahedron Lett. 1984, 25, 839-842.
9
J. AM. CHEM. SOC. VOL. 126, NO. 1, 2004 229

A R T I C L E S
Fekner et al.
Scheme 5. Synthesis of Strapped Cyclic Amides 54-57 and Dimers 58-61a
a Reagents and conditions: (a) NaH, 38-41, THF, 0 °C f room temperature, 14-20 h; (b) SnCl2, EtOH, reflux, 45-50 min; (c) LiOH, THF, H2O, room
temperature, 15-20 h; (d) BOP, NMM, CH2Cl2, room temperature, 7-10 days.
Scheme 6. Metalation of the Strapped Cyclic Amide 54a
a Reagents and conditions: (a) Ni(OAc)2‚4H2O, MeOH, reflux, 2 h; 94%.
upfield shift due to the geometric constraints that force their
close spatial proximity with the aromatic system. As illustrated
in Table 1, this is found to be the case.
Figure 10. ORTEP drawing (50% probability thermal ellipsoids) of the
molecular structure of the Ni(II) complex 62. The nickel, oxygen, and
nitrogen atoms are hatched. Hydrogen atoms are omitted for clarity.
The length of the strap in compounds 54-57 not only affects
their NMR spectra but also, and more importantly, is directly
related to the ability of the strap to hinder racemization. Because
for the strapped amides 54-57 we did not observe any evidence,
either spectroscopic (1H and 13C NMR) or chromatographic
(CSP HPLC), that these compounds exist as mixtures of more
than one diastereomer, it appears that the act of racemization
for this class of compounds must involve concerted rotation of
the core cyclic amide under the strap (which results in inversion
of configuration at the two stereogenic benzimidazole nitrogen
atoms) accompanied by its macrocyclic inversion. Consistent
with molecular modeling studies, the X-ray single-crystal
structure of the Ni(II) complex 62 (Figure 10) of the strapped
cyclic amide 54 indicates that the geometry of the core cyclic
bis(benzimidazole) part of the molecule is governed by the
position of the strap, with the two benzimidazoles pointing
toward it, and the two o-substituted phenyl rings occupying the
opposite face of the 4N-plane. This preference stems from the
intrinsic puckering of the cyclic amide core (cf., Figures 4, 6,
(32) (a) Waugh, J. S.; Fessenden, R. W. J. Am. Chem. Soc. 1957, 79, 846-849.
(b) Pople, J. A.; Schneider, W. G.; Bernstein, H. J. High-Resolution Nuclear
Magnetic Resonance; McGraw-Hill: New York, 1959. (c) Gu¨nther, H. NMR
Spectroscopy; Wiley: New York, 1998. (d) Gomes, J. A. N. F.; Mallion,
R. B. Chem. ReV. 2001, 101, 1349-1384.
9
230 J. AM. CHEM. SOC. VOL. 126, NO. 1, 2004

Ruffling-Induced Chirality
A R T I C L E S
Scheme 7. Cyclization Studies of Amino Acids 63-65a
a Reagents and conditions: (a) BOP, NMM, CH2Cl2, room temperature, 7-9 days.
Table 1. Selected 1H NMR Signals for Strapped Cyclic Amides
at least for relatively short straps, the macrocyclic core is forced
to adopt the conformation for which the distance to be spanned
by the strap is the shortest (the same face as the benzimidazole
phenyl rings), and, consequently, the strain introduced is
minimized. For longer straps, however, the macrocyclic inver-
sion is expected to be relatively free (as it is for amide 5) and
independent from any rotation of the macrocyclic core under
the strap.
54-57a
As cyclodimerization of sterically hindered amino acids 21
and 22 was more efficient than that of the parent amino acid
20 (Scheme 2), we also investigated whether the presence of
the bulkier flanking substituent was beneficial for the cyclization
of tethered bis(amino acid)s, as well. Thus (Scheme 7), benz-
imidazole 13 was converted to amino acids 63-65 (Scheme
S1) by a synthetic route analogous to that described for the
preparation of amino acids 50-53 from benzimidazole 11
(Scheme 5). Cyclization of amino acid 63 under standard
conditions gave monomer 66 in 23% yield, along with dimer
69 in 10% yield. These yields are comparable to those obtained
for the cyclization of the “H-flanked” amino acid 55 (Scheme
5; monomer, 22%; dimer, 13%). Amino acid 64 proved a very
poor substrate for cyclization under the standard conditions due
to its extremely low solubility in CH2Cl2. Only a small amount
(1.6%) of monomer 67 was isolated in this case after a prolonged
reaction time (9 days). When subjected to the cyclization
protocol, amino acid 65, possessing an extremely short strap,
gave dimer 71 in 11% yield. The corresponding monomer was
not isolated, presumably because its formation is thermody-
b
δ/ppm
H (s)
H /H (m)
H /H (m)
H /H (ABq)
d d′
a
b
b′
c
c′
54
5.22
2.90-3.05
0.00-0.12
0.22-0.33
0.57-0.68
1.26-1.44
5.38/5.77
55
56
5.71
6.17
3.41 (t)
5.30/5.78
5.26/5.78
3.51-3.59
3.68-3.75
3.62-3.70
3.78-3.85
57
6.43
1.50-1.66
5.18/5.72
a Spectra were recorded in CDCl3 (400 MHz) and referenced to the
residual CHCl3 peak. b The Hc/Hc′ signals were correlated to the Hb/Hb′
signals via H-1H 2D COSY spectroscopy.
1
and 9), for which substituents attached to the external benz-
imidazole nitrogen atoms are situated on the same face of the
mean 4N-plane as the benzimidazole phenyl rings. As a result,
9
J. AM. CHEM. SOC. VOL. 126, NO. 1, 2004 231

A R T I C L E S
Fekner et al.
Table 2. Racemization of Strapped Cyclic Amides 54-56 and
66a,b
room temp (∼25 °C)c CHCl3 (bp 61 °C) PhH (bp 80 °C) PhMe (bp 111 °C)
compound time/h
ee/% time/h ee/% time/h ee/% time/h
ee/%
54
55
56
0
404
1751
99.4
99.4
99.4
0
>99.9
14.6 >99.9
304
>99.9
0
529
1873
99.5
99.5
99.5
0
74
217
>99.9
96.5
88.7
0
255
716
99.7
0
>99.9
97.5
95.2
99.7 143
99.7 255
99.7
2011
66
0
47
234
645
1964
>99.9
0
3.5
18.7
43.4
99.7
92.3 25
67.2
0
>99.9
99.3
97.3
93.0
80.0
0
39.4
a Racemization was performed in the indicated solvents at their boiling-
point temperatures. b Enantiomeric excess (ee) was determined by analytical
CSP HPLC. c Samples were kept in CHCl3 at ambient temperature.
namically disfavored due to the high strain energy associated
with spanning the bis(benzimidazole) diamide with such a short
strap.
Figure 11. The set of possible stereoisomers from dimerization of a C2-
symmetric, ruffled molecule.
To evaluate the effect of the strap length on the rate of
racemization for the series of monomeric strapped amides,
ligands 54-56 were optically resolved using CSP HPLC, and
the optical purity of the ligands was measured after a given
racemization period. As we were not able to satisfactorily
resolve compound 57 on a wide range of CSP HPLC columns
available to us, compound 66 (Scheme 7), which possesses the
identical strap and proved amenable to optical resolution, was
vicariously used in the racemization studies. These studies were
performed at various temperatures, ranging from ∼25 °C (room
temperature) to 111 °C (bp of PhMe), and their results are
summarized in Table 2. As can be noted, only compound 66,
possessing the longest strap, undergoes racemization at room
temperature to any significant degree, losing 0.7% ee after 47
h, and 20% ee after ∼82 days. The stereointegrity of the
compounds possessing shorter straps is sufficiently high to
prevent racemization at room temperature, although monomers
55 and 56 do racemize at elevated temperatures.
In a related fashion, dimers 58-61 can be thought of as
monomers with extremely long straps. As such, one ruffled
cyclic amide can be considered a part of a long strap spanning
the two external benzimidazole nitrogen atoms of the other
ruffled cyclic amide. Molecular modeling studies indicate that
for dimers 58-61 the two cyclic amides are held strictly in a
face-to-face orientation. In terms of stereochemistry, however,
this type of dimerization for the ruffled C2-symmetric molecules
is potentially considerably more complex than the simple
spanning discussed previously for monomers 54-57. As
mentioned previously for the cyclic amide monomers with
extremely long straps, macrocyclic inversion of the individual
monomers forming the dimer should be possible without the
need for rotation of the macrocycle under the strap. Moreover,
as the two cyclic amides of dimers 58-61 can face each other
in three different fashions (R-R/R-R, R-R/S-S, and S-S/S-S;
depending on the configuration generated at the two pairs of
the stereogenic benzimidazole nitrogens), the dimers can
potentially exist as mixtures of four pairs of enantiomers (rac
1-4) and two achiral forms (meso 1,2) (Figure 11), with each
compound being at least C2-symmetric.33 Any of the 10 possible
isomers can be formally converted into any other by macrocyclic
inversion of either cyclic amide, rotation of one macrocycle
under the “strap”, or a combination of these two basic
transformations.
The barrier to rotation under the strap depends on the distance
between the two cyclic amides, whereas the barrier to macro-
cyclic inversion is a function of steric constraints within the
cyclic amide core (vide supra). Spectroscopic data (1H and 13
C
NMR) indicate that the medium-length strap dimers 59-60
exist, at least on the NMR time-scale, either as nearly equimolar
mixtures of two diastereomers with either D2 or C2h symmetry
(Figure 11) or as diastereomerically homogeneous samples of
C2-symmetric molecules. For instance, the 1H NMR (400 MHz,
CDCl3) spectrum of dimer 60 shows two singlets (12.01 and
12.02 ppm) for the amide protons, as well as two singlets (6.34
and 6.37 ppm) for the Ha (see Table 1 for labeling system),
with many other peaks broadened or showing unusual multi-
plicities. The proton-decoupled 13C NMR (101 MHz, CDCl3)
spectrum of dimer 60 consists of 32 signals, whereas only 25
signals would be expected for a diastereomerically homogeneous
sample of either D2- or C2h-symmetric molecules. If dimers 59
and 60 are diastereomerically pure, they most likely exist as
rac 1 (Figure 11) for which the π-stacking interaction between
the two cyclic amide subunits is maximized.
Although CSP HPLC studies of dimers 58-61 were not very
successful due to their high affinity toward the CSPs used, it
was possible to partially resolve the related dimer 69 formed
from the dimethyl-substituted macrocycle. In this case, the
HPLC trace (Figure S11) shows the presence of five peaks,
(33) For the sake of simplicity, the side-on view (as in Figure 4) of the ruffled
cyclic amide is represented in Figure 11 by two crossed wedged lines, with
their sharp ends representing the o-substituted phenyl rings, and their wide
ends representing the benzimidazole parts of the molecule. The straps have
been omitted for clarity.
9
232 J. AM. CHEM. SOC. VOL. 126, NO. 1, 2004

Ruffling-Induced Chirality
A R T I C L E S
refluxing over CaH2. Anhydrous THF was obtained by distillation,
immediately before use, from sodium/benzophenone ketyl under an inert
atmosphere of nitrogen. Petroleum ether refers to the fraction of light
petroleum boiling between 40 and 60 °C. High-resolution mass
spectrometry (HRMS) measurements are valid to (5 ppm. Melting
points (mp) are quoted to the nearest 0.5 °C. Elemental analyses were
performed by Atlantic Microlab, Norcross, GA.
which could be attributed to two racemates and one meso form.
NMR studies performed on the related short-length strap dimer
1
71 reveal it to have much simpler H and 13C NMR spectra
than the medium-length strap dimers 59 and 60. Based on its
NMR, the two cyclic amide subunits of 71 appear to be in
chemically equivalent environments. Symmetry considerations
(Figure 11) indicate that at room temperature molecules of dimer
71 in the most thermodynamically stable geometry belong to
either point group D2 (rac 3, rac 4) or point group C2h (meso
1, meso 2). Apparently, the very short distance between the
two cyclic bis(benzimidazole) subunits fixes the macrocycle in
a well-defined conformation for which neither macrocyclic
inversion nor rotation of the cyclic amide under the strap can
take place. These symmetry considerations would rule out the
“ideal” π-stacking between the two cyclic amide subunits (as
in C2-symmetric rac 1), although the inverted π-stacking
interactions of the phenyl and benzimidazole units as observed
for rac 3 and rac 4 are still possible.
Ethyl 2-(2-Nitrophenyl)-1H-benzimidazole-4-carboxylate (11). To
a solution of oxalyl chloride (22.7 mL, 0.26 mol) in CH2Cl2 (250 mL)
were added 2-nitrobenzoic acid (33.4 g, 0.20 mol) and a catalytic
amount of DMF (∼2 drops). The reaction mixture was stirred at room
temperature for 12 h to give a clear solution. The volatiles were removed
in vacuo to afford the crude acyl chloride product as a pale brown oil
(37.2 g). A solution of the acyl chloride in CH2Cl2 (400 mL) was added
dropwise over 2 h at 0 °C to a solution of diamine 9 (34.9 g, 194
mmol) and Et3N (36 mL, 0.26 mol) in CH2Cl2 (1600 mL). After the
addition was complete, the reaction mixture was stirred at 0 °C for 1
h and at room temperature for 3 h. The volatiles were removed in vacuo
to give an off-white solid. Analysis of the crude product [1H NMR
(250 MHz, CDCl3)] revealed the presence of two isomeric monamides
in a ca. 5:1 ratio. The crude amide mixture was refluxed in glacial
AcOH (500 mL) in the presence of AcONa (16.4 g, 0.20 mol) for 15
h. The reaction mixture was cooled to room temperature and evaporated
in vacuo. The resulting brown oil was partitioned between CH2Cl2 and
water. After neutralization with solid K2CO3, the phases were separated,
and the extraction was completed with additional portions of CH2Cl2.
The combined organic extracts were dried (MgSO4) and evaporated in
vacuo to give a brown solid. Purification by flash chromatography (silica
gel, CH2Cl2 f EtOAc) gave the title compound 11 (56.1 g, 93% over
two steps from diamine 9) as a bright yellow solid: Rf ) 0.70 (EtOAc/
CH2Cl2, 1/1); mp 115.0-116.0 °C (EtOAc/petroleum ether). 1H NMR
(250 MHz, CDCl3): δ 1.26 (t, J ) 7.0 Hz, 3H), 4.23 (q, J ) 7.0 Hz,
2H), 7.17 (t, J ) 8.0 Hz, 1H), 7.36-7.49 (m, 2H), 7.73-7.84 (m, 4H),
and 11.0 (br s, 1H). 13C{1H} NMR (63 MHz, CDCl3): δ 14.72, 61.70,
114.6, 122.5, 125.0, 125.2, 125.6, 126.0, 131.2, 132.5, 133.2, 135.0,
144.6, 148.9, 149.1, and 166.6. IR (CHCl3): Vmax 1694, 1534, 1371,
1280, 1196, and 1145 cm-1. MS (ESI): m/z (rel intensity) 312 (30%,
MH+), 284 (100), and 266 (40). HRMS calcd for C16H13N3NaO4
(MNa+) 334.0804, found 334.0807. Anal. Calcd for C16H13N3O4: C,
61.73; H, 4.21; N, 13.50. Found: C, 61.89; H, 4.13; N, 13.43.
Conclusions
In summary, we have developed a practical synthesis of cyclic
bis(benzimidazole)-based amides 5 and 23-24. These ligands
were demonstrated to possess a highly distorted, ruffled, chiral
structure. Because of the greater solubility of these compounds
as compared to ligand 3, their optical resolution and racemiza-
tion could be studied by CSP HPLC. While the incorporation
of steric hindrance into amides 23-24 lowered the rate of
racemization, the effect was not sufficient to enable storage of
enantioenriched samples for a prolonged period of time without
significant loss of optical purity. Metalation with Ni(II) was
also demonstrated to increase the barrier to macrocyclic inver-
sion for this class of compounds. The problem of macrocyclic
inversion could be completely circumvented by spanning the
two external benzimidazole nitrogen atoms of the parent bis-
(benzimidazole) framework with the tailor-made straps of
various lengths. These compounds and their corresponding Ni-
(II) complexes were amenable to optical resolution by CSP
HPLC and were demonstrated to be chirally stable at ambient
temperature. Work is currently underway to prepare other
ligands of this class and their complexes with various transition
metals as potential catalysts for asymmetric transformations.
Ethyl 1-Methyl-2-(2-nitrophenyl)-1H-benzimidazole-4-carboxyl-
ate (14). To a solution of benzimidazole 11 (55.3 g, 178 mmol) in
THF (500 mL) was slowly added NaH (60% w/w dispersion in mineral
oil, 7.82 g, 0.20 mol) at 0 °C. After 30 min, the reaction mixture was
warmed to room temperature and stirred for an additional 40 min. The
resulting brown solution was recooled to 0 °C and quenched with MeI
(28 mL, 0.45 mol). After 13 h at room temperature, the volatiles were
removed in vacuo, and the residue was partitioned between CH2Cl2
and water. The phases were separated, and the extraction was completed
with additional portions of CH2Cl2. The combined organic extracts were
dried (MgSO4) and evaporated in vacuo to give a yellow solid.
Purification by flash chromatography (silica gel, CH2Cl2 f EtOAc)
gave the title compound 14 (53.2 g, 92%) as a pale yellow solid: Rf )
0.40 (EtOAc/CH2Cl2, 1/1); mp 163.0-164.0 °C (EtOAc/petroleum
ether). 1H NMR (250 MHz, CDCl3): δ 1.26 (t, J ) 7.0 Hz, 3H), 3.46
(s, 3H), 4.31 (q, J ) 7.0 Hz, 2H), 7.22 (t, J ) 8.0 Hz, 1H), 7.43 (dd,
J ) 8.0, 1.0 Hz, 1H), 7.51-7.64 (m, 3H), 7.82 (dd, J ) 7.5, 1.0 Hz,
1H), and 8.04-8.07 (m, 1H). 13C{1H} NMR (63 MHz, CDCl3): δ
14.81, 31.17, 61.23, 114.4, 122.4, 122.7, 125.2, 125.4, 126.4, 131.6,
133.6, 134.0, 137.2, 142.5, 149.0, 152.0, and 166.1. IR (CHCl3): Vmax
1715, 1535, 1460, 1347, 1281, 1251, and 1124 cm-1. MS (ESI): m/z
(rel intensity) 348 (5%, MNa+), 326 (55, MH+), and 298 (100). HRMS
calcd for C17H15N3NaO4 (MNa+) 348.0960, found 348.0973. Anal.
Calcd for C17H15N3O4: C, 62.76; H, 4.65; N, 12.92. Found: C, 62.77;
H, 4.65; N, 12.91.
Experimental Section
General Methods. All reactions were performed under anhydrous
conditions and in an inert atmosphere of argon using oven-dried
glassware. Yields refer to chromatographically and spectroscopically
(1H NMR) homogeneous materials, unless otherwise indicated. Reagents
were used as obtained from commercial sources or purified according
to known procedures.34 Flash chromatography was carried out using
Merck Kiesegel 60 F254 (230-400 mesh) silica gel following the method
of Still et al.35 Only distilled solvents were used as eluents. Thin-layer
chromatography (TLC) was performed on Merck DC-Alufolien or glass
plates precoated with silica gel 60 F254 which were visualized either
by quenching of ultraviolet fluorescence or by charring with 5% w/v
phosphomolybdic acid in 95% EtOH, 10% w/v ammonium molybdate
in 1 M H2SO4, or 10% KMnO4 in 1 M H2SO4. Observed retention
factors (Rf) are quoted to the nearest 0.05. All reaction solvents were
distilled before use and stored over activated 4 Å molecular sieves,
unless otherwise indicated. Anhydrous CH2Cl2 was obtained by
(34) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory Chemicals;
Pergamon Press: New York, 1988.
(35) Still, W. C.; Hahn, M.; Mitra, A. J. Org. Chem. 1978, 43, 2923-2925.
9
J. AM. CHEM. SOC. VOL. 126, NO. 1, 2004 233

A R T I C L E S
Fekner et al.
1
1-Methyl-2-(2-nitrophenyl)-1H-benzimidazole-4-carboxylic Acid
(17). To a solution of NaOH (10.0 g, 0.40 mol) in MeOH (250 mL)
was added ester 14 (19.7 g, 60.6 mmol), and the mixture was refluxed
for 25 min (TLC). The reaction mixture was cooled to room temperature
and evaporated in vacuo to give a yellow solid. The solid was dissolved
in water (600 mL) and neutralized at 0 °C with concentrated HCl. The
precipitate formed was removed by filtration, washed with a copious
amount of water, and dried in vacuo to give the title compound 17
(16.7 g, 93%) as a pale yellow solid. Acid 17 was used in the subsequent
step without further purification. An analytical sample of the product
was obtained by crystallization from EtOH: mp 234.0-235.0 °C
(EtOH). 1H NMR (250 MHz, d6-DMSO): δ 3.61 (s, 3H), 7.34 (t, J )
7.5 Hz, 1H), 7.72-7.90 (m, 5H), and 8.19 (∼d, J ) 8.0 Hz, 1H). 13C-
{1H} NMR (63 MHz, d6-DMSO): δ 31.83, 116.5, 121.6, 123.4, 125.2,
125.5, 125.8, 132.9, 133.5, 135.0, 137.1, 141.9, 149.5, 151.6, and 167.2.
IR (KBr): Vmax 1742, 1696, 1518, 1468, 1339, and 1244 cm-1. MS
(ESI): m/z (rel intensity) 298 (100%, MH+). HRMS calcd for
C15H12N3O4 (MH+) 298.0828, found 298.0838. Anal. Calcd for
C15H11N3O4: C, 60.61; H, 3.73; N, 14.14. Found: C, 60.69; H, 3.65;
N, 13.98.
°C (MeOH). H NMR (250 MHz, d4-MeOH + CDCl3): δ 4.14 (s,
3H), 7.20-7.30 (m, 1H), 7.37-7.54 (m, 3H), 7.60-7.72 (m, 2H), and
7.99 (d, J ) 7.5 Hz, 1H). 13C{1H} NMR (101 MHz, d4-MeOH +
CDCl3): δ 34.73, 115.5, 122.3, 125.2, 125.8, 126.2, 126.6, 129.1, 132.5,
132.6, 137.4, 138.3, 149.4, 151.9, and 169.1. IR (KBr): Vmax 1606,
1562, 1526, 1485, 1437, 1325, and 1292 cm-1. MS (ESI): m/z (rel
intensity) 499 (10%, MH+) and 499 (100). HRMS calcd for C30H21N6-
NiO2 (MH+) 555.1079, found 555.1090. Anal. Calcd for C30H21.5N6-
NiO2.75 (M‚0.75H2O): C, 63.36; H, 3.81; N, 15.14. Found: C, 63.28;
H, 3.89; N, 14.81. Single crystals of complex 25 were grown by slow
evaporation of a CH2Cl2/MeOH solution, and its structure was
unequivocally confirmed by single-crystal X-ray analysis. For crystal-
lographic data, see Supporting Information. Ni(II) complex 25 was
optically resolved by analytical CSP HPLC (Figure S1).
1,6-Bis(3-hydroxymethylphenoxy)hexane (34). To a solution of
phenol 31 (15.0 g, 121 mmol) in EtOH (180 mL) was added 10 M
NaOH (12.1 mL, 121 mmol), followed by 1,6-dibromohexane (14.8 g,
60.4 mmol). The reaction mixture was refluxed for 19 h, cooled to
room temperature, and diluted with water (300 mL). The brown mixture
was subsequently extracted with CH2Cl2, and the combined extracts
were dried (MgSO4) and concentrated in vacuo to give an off-white
solid. Purification by flash chromatography (CH2Cl2 f CH2Cl2/Me2-
CO, 4/1) gave the title compound 34 (12.9 g, 65%) as a slightly pink
solid: Rf ) 0.65 (EtOAc/CH2Cl2, 1/1); mp 93.0-94.0 °C (Me2CO/
2-(2-Aminophenyl)-1-methyl-1H-benzimidazole-4-carboxylic Acid
(20). A suspension of benzimidazole 17 (16.1 g, 54.2 mmol) in MeOH
(900 mL) was hydrogenated at normal pressure in the presence of Pd/C
(10% w/w, 1.6 g) for 24 h (TLC). The resulting mixture was passed
through a thin pad of Celite, and the filtrate was concentrated in vacuo
to give the title compound 20 (14.4 g, 100%) as a yellow solid that
could be used in the subsequent step without further purification. An
analytical sample of the product was obtained by crystallization from
EtOH: mp 214.0-215.0 °C (EtOH). 1H NMR (250 MHz, d6-DMSO):
δ 3.69 (s, 3H), 5.77 (br s, 2H), 6.58 (∼t, J ) 8.0 Hz, 1H), 6.75 (d,
J ) 8.0 Hz, 1H), 7.12 (dt, J ) 7.5, 1.5 Hz, 1H), 7.25 (dd, J ) 7.5, 1.5
Hz, 1H), 7.29 (t, J ) 8.0 Hz, 1H), 7.72 (∼d, J ) 7.5 Hz, 1H), and
7.77 (d, J ) 8.0 Hz, 1H). 13C{1H} NMR (63 MHz, d6-DMSO): δ 32.70,
112.2, 116.3, 116.4, 116.7, 120.3, 122.9, 125.0, 131.5, 132.0, 137.3,
142.0, 148.9, 154.6, and 167.3. IR (KBr): Vmax 1729, 1614, 1485, 1427,
and 1385 cm-1. MS (ESI): m/z (rel intensity) 268 (100%, MH+). HRMS
calcd for C15H14N3O2 (MH+) 268.1086, found 268.1083. Anal. Calcd
for C15H13N3O2: C, 67.40; H, 4.90; N, 15.72. Found: C, 67.12; H,
4.86; N, 15.66.
1
petroleum ether). H NMR (250 MHz, d4-MeOH): δ 1.39-1.45 (m,
2H), 1.65-1.70 (m, 2H), 3.85 (t, J ) 6.5 Hz, 2H), 4.44 (s, 2H), 6.67
(∼dd, J ) 8.0, 2.5 Hz, 1H), 6.75-6.79 (m, 2H), and 7.09 (t, J ) 8.0
Hz, 1H). 13C{1H} NMR (63 MHz, d4-MeOH): δ 25.99, 29.38, 64.16,
67.84, 113.0, 113.4, 119.1, 129.4, 143.3, and 159.7. IR (CHCl3): Vmax
2943, 2873, 1601, 1585, 1488, 1449, and 1264 cm-1. MS (ESI): m/z
(rel intensity) 353 (100%, MNa+) and 313 (20). HRMS calcd for C20H26-
NaO4 (MNa+) 353.1729, found 353.1742. Anal. Calcd for C20H26O4:
C, 72.70; H, 7.93. Found: C, 72.61; H, 8.00.
1,6-Bis(3-bromomethylphenoxy)hexane (39). To a suspension of
diol 34 (12.9 g, 39.0 mmol) in CH2Cl2 (400 mL) was added PBr3 (4.8
mL, 51 mmol) at 0 °C, and the resulting clear solution was stirred at
room temperature for 14 h. The reaction mixture was quenched with
saturated NaHCO3 (50 mL) and stirred at room temperature for an
additional 30 min. The resulting white suspension was extracted with
CH2Cl2, and the combined extracts were dried (MgSO4) and evaporated
in vacuo to give a white solid. Purification by flash chromatography
(CH2Cl2/petroleum ether, 1/1) gave the title compound 39 (15.0 g, 84%)
as a white solid: Rf ) 0.80 (CH2Cl2/petroleum ether, 1/1); mp 87.0-
Cyclic Amide (5). To a suspension of amino acid 20 (2.75 g, 10.3
mmol) in CH2Cl2 (450 mL) was added NMM (2.5 mL, 23 mmol),
followed by BOP (5.00 g, 11.3 mmol). The reaction mixture was stirred
at room temperature for 6 days and washed with saturated NaHCO3.
The organic layer was dried (MgSO4) and evaporated in vacuo to give
a yellow solid. Purification by flash chromatography (silica gel, CH2-
Cl2 f EtOAc) gave the title compound 5 (1.17 g, 46%) as a white
solid: Rf ) 0.60 (EtOAc); mp > 260 °C (EtOAc). 1H NMR (250 MHz,
CDCl3): δ 3.87 (s, 3H), 7.32-7.47 (m, 3H), 7.55-7.63 (m, 2H), 8.13
(d, J ) 8.0 Hz, 1H), 8.18 (d, J ) 7.5 Hz, 1H), and 12.1 (s, 1H). 13C-
{1H} NMR (101 MHz, CDCl3): δ 32.11, 113.8, 122.0, 122.9, 123.0,
124.7, 124.8, 128.0, 130.1, 130.8, 136.3, 137.1, 140.2, 152.0, and 164.1.
1
88.0 °C (EtOAc/petroleum ether). H NMR (250 MHz, CDCl3): δ
1.44-1.49 (m, 2H), 1.70-1.76 (m, 2H), 3.89 (t, J ) 6.5 Hz, 2H), 4.37
(s, 2H), 6.75 (ddd, J ) 8.0, 2.0, 1.0 Hz, 1H), 6.83-6.89 (m, 2H), and
7.15 (t, J ) 8.0 Hz, 1H). 13C{1H} NMR (63 MHz, CDCl3): δ 26.34,
29.65, 34.07, 68.30, 115.1, 115.6, 121.6, 130.2, 139.6, and 159.8. IR
(CHCl3): Vmax 2944, 2871, 1600, 1585, 1489, 1446, and 1266 cm-1
.
MS (ESI): m/z (rel intensity) 481 [50%, MNa+ (81Br/81Br)], 479 [100%,
MNa+ (79Br/81Br)], and 477 [50%, MNa+ (79Br/79Br)]. HRMS calcd
for C20H24(79Br)2NaO2 (MNa+) 477.0041, found 477.0027. Anal. Calcd
for C20H24Br2O2: C, 52.65; H, 5.30; Br, 35.03. Found: C, 52.87; H,
5.25; Br, 35.05.
IR (KBr): Vmax 1665, 1533, 1516, 1481, 1434, 1380, and 1303 cm-1
.
MS (ESI): m/z (rel intensity) 499 (100%, MH+). HRMS calcd for
C30H23N6O2 (MH+) 499.1882, found 499.1865. Anal. Calcd for
C30H22N6O2: C, 72.28; H, 4.45; N, 16.86. Found: C, 72.02; H, 4.39;
N, 16.76. Single crystals of the cyclic amide 5 were grown by slow
evaporation of a CH2Cl2/EtOAc solution, and its structure was
unequivocally confirmed by single-crystal X-ray analysis. For crystal-
lographic data, see Supporting Information.
1,6-Bis{3-[4-ethoxycarbonyl-2-(2-nitrophenyl)-benzimidazol-1-
ylmethyl]phenoxy}hexane (43). To a solution of benzimidazole 11
(7.00 g, 22.5 mmol) in THF (100 mL) was added NaH (60% w/w
dispersion in mineral oil, 990 mg, 25 mmol) at 0 °C. After 30 min, the
reaction mixture was warmed to room temperature and stirred for an
additional 40 min. The resulting brown solution was recooled to 0 °C
and treated with dibromide 39 (5.13 g, 11.3 mmol). After 16 h at room
temperature, the reaction mixture was quenched with water (10 mL),
the volatiles were removed in vacuo, and the residue was partitioned
between CH2Cl2 and water. The phases were separated, and the
extraction was completed with additional portions of CH2Cl2. The
combined organic extracts were dried (MgSO4) and evaporated in vacuo
Ni(II) Complex (25). To a suspension of cyclic amide 5 (809 mg,
1.62 mmol) in MeOH (70 mL) was added Ni(OAc)2‚4H2O (425 mg,
1.70 mmol), and the mixture was refluxed for 90 min (TLC) to give
an orange suspension. After being cooled to room temperature, the
suspension was separated by filtration. The collected solid was washed
with cold MeOH (30 mL) and dried to give the title compound 25
(875 mg, 97%) as a deep-orange solid: Rf ) 0.45 (MeOH); mp > 260
9
234 J. AM. CHEM. SOC. VOL. 126, NO. 1, 2004

Ruffling-Induced Chirality
A R T I C L E S
to give a yellow solid. Purification by flash chromatography (silica
gel, CH2Cl2 f EtOAc) gave the title compound 43 (8.83 g, 86%) as a
pale yellow solid: Rf ) 0.35 (EtOAc); mp 179.0-181.0 °C (EtOAc).
1H NMR (250 MHz, CD2Cl2): δ 1.30 (t, J ) 7.0 Hz, 3H), 1.25-1.42
(m, 2H), 1.52-1.62 (m, 2H), 3.73 (t, J ) 6.5 Hz, 2H), 4.31 (q, J ) 7.0
Hz, 2H), 5.13 (s, 2H), 6.45 (d, J ) 1.5 Hz, 1H), 6.50 (d, J ) 8.0 Hz,
1H), 6.67 (dd, J ) 8.0, 2.0 Hz, 1H), 7.06 (t, J ) 8.0 Hz, 1H), 7.19 (t,
J ) 7.5 Hz, 1H), 7.37 (dd, J ) 8.0, 1.0 Hz, 1H), 7.43-7.50 (m, 1H),
7.58-7.65 (m, 2H), 7.83 (dd, J ) 7.5, 1.0 Hz, 1H), and 8.07-8.14
(m, 1H). 13C{1H} NMR (101 MHz, CD2Cl2): δ 14.98, 26.49, 29.82,
49.21, 61.59, 68.64, 113.6, 114.7, 115.8, 119.4, 123.1, 123.2, 125.7,
125.9, 126.6, 130.7, 132.1, 133.5, 134.3, 137.1, 137.7, 142.8, 149.6,
152.2, 160.4, and 166.4. IR (CHCl3): Vmax 1716, 1604, 1534, 1453,
1429, 1346, and 1282 cm-1. MS (ESI): m/z (rel intensity) 939 (20%,
MNa+) and 917 (100). HRMS calcd for C52H49N6O10 (MH+) 917.3510,
found 917.3497. Anal. Calcd for C52H48N6O10: C, 68.11; H, 5.28; N,
9.16. Found: C, 68.05; H, 5.40; N, 9.07.
solution was stirred at room temperature for 10 days and quenched
with saturated NaHCO3 (100 mL). After 1 h at room temperature, the
phases were separated, and the extraction was completed with additional
portions of CH2Cl2. The combined organic extracts were dried (MgSO4)
and evaporated in vacuo to give a brown thick oil. Purification by flash
chromatography (silica gel, CH2Cl2 f CH2Cl2/MeOH, 20/1) gave
monomer 55 (288 mg, 10%) and dimer 59 (363 mg, 13%) as white
solids. Monomer 55: Rf ) 0.65 (CH2Cl2/EtOAc, 5/1); mp > 260 °C
1
(EtOAc). H NMR (400 MHz, CDCl3): δ 0.33-0.38 (m, 2H), 0.59-
0.64 (m, 2H), 3.41 (t, J ) 6.0 Hz, 2H), 5.30 and 5.78 (ABq, J ) 16.0
Hz, 2H), 5.71 (s, 1H), 6.62 (dd, J ) 8.0, 2.0 Hz, 1H), 6.79 (d, J ) 7.5
Hz, 1H), 7.14 (t, J ) 8.0 Hz, 1H), 7.27 (t, J ) 8.0 Hz, 1H), 7.59 (ddd,
J ) 9.0, 9.0, 1.5 Hz, 1H), 8.13 (dd, J ) 7.5, 1.0 Hz, 1H), 8.14 (d, J )
7.5 Hz, 1H), and 12.2 (s, 1H). 13C{1H} NMR (63 MHz, CDCl3): δ
25.03, 27.97, 49.48, 67.20, 110.8, 115.7, 117.1, 119.2, 122.3, 123.3,
123.5, 125.2, 125.4, 128.2, 130.1, 130.4, 131.4, 135.5, 136.7, 137.9,
140.9, 153.0, 159.6, and 163.9. IR (KBr): Vmax 1673, 1607, 1583, 1534,
1478, 1386, 1302, and 1248 cm-1. MS (ESI): m/z (rel intensity) 787
(100%, MNa+) and 765 (15). HRMS calcd for C48H40N6NaO4 (MNa+)
787.3009, found 787.2993. Anal. Calcd for C48H40N6O4: C, 75.37; H,
5.27; N, 10.99. Found: C, 75.08; H, 5.23; N, 10.93. Monomer 55 was
optically resolved by analytical CSP HPLC (Figure S7). Dimer 59:
1,6-Bis{3-[2-(2-aminophenyl)-4-ethoxycarbonylbenzimidazol-1-
ylmethyl]phenoxy}hexane (47). To a suspension of benzimidazole 43
(8.83 g, 9.63 mmol) in anhydrous EtOH (200 mL) was added SnCl2
(18.3 g, 96.3 mmol), and the mixture was refluxed for 50 min to give
a yellow solution. The reaction mixture was cooled in an ice bath,
diluted with CH2Cl2 (300 mL), and made basic by addition of saturated
NaHCO3. After 30 min at room temperature, the resulting white
suspension was filtered through a pad of Celite. The phases were
separated, and the extraction was completed with additional portions
of CH2Cl2. The combined organic extracts were dried (MgSO4) and
evaporated in vacuo to give a yellow solid. Purification by flash
chromatography (silica gel, CH2Cl2 f EtOAc) gave the title compound
47 (8.09 g, 98%) as a pale yellow solid: Rf ) 0.75 (EtOAc). 1H NMR
(250 MHz, CDCl3): δ 1.20-1.30 (m, 2H), 1.33 (t, J ) 7.0 Hz, 3H),
1.50-1.60 (m, 2H), 3.70 (t, J ) 6.5 Hz, 2H), 4.37 (q, J ) 7.0 Hz,
2H), 5.27 (s, 2H), 5.47 (br s, 2H), 6.41-6.55 (m, 3H), 6.65-6.69 (m,
2H), 6.99-7.14 (m, 4H), 7.22 (dd, J ) 8.0, 1.0 Hz, 1H), and 7.84 (dd,
J ) 7.5, 1.0 Hz, 1H). 13C{1H} NMR (63 MHz, CDCl3): δ 14.93, 26.21,
29.49, 49.14, 61.28, 68.21, 112.6, 112.7, 114.2, 115.3, 117.3 (2 × C),
118.5, 121.6, 122.5, 125.7, 129.9, 130.6, 131.1, 131.5, 137.0, 138.2,
142.2, 148.6, 154.9, 160.2, and 166.8. IR (CHCl3): Vmax 1707, 1617,
1602, 1488, 1428, 1290, 1265, and 1251 cm-1. MS (ESI): m/z (rel
intensity) 870 (50%, MNa+) and 857 (100). HRMS calcd for C52H53N6O6
(MH+) 857.4026, found 857.4015.
1
Rf ) 0.40 (CH2Cl2/EtOAc, 5/1). H NMR (400 MHz, d6-DMSO): δ
0.95 (br s, 4H), 1.25 (br s, 4H), 3.47-3.54 (m, 4H), 5.50-5.60 (m,
4H), 6.23 (s, 1H), 6.27 (s, 1H), 6.57 (d, J ) 7.5 Hz, 1H), 6.61 (d, J )
7.5 Hz, 1H), 6.65 (d, J ) 8.0 Hz, 2H), 7.04 (t, J ) 7.5 Hz, 1H), 7.06
(t, J ) 7.5 Hz, 1H), 7.31 (t, J ) 8.0 Hz, 1H), 7.35 (t, J ) 8.0 Hz, 1H),
7.40 (t, J ) 7.5 Hz, 2H), 7.63 (t, J ) 7.5 Hz, 2H), 7.69 (d, J ) 8.0 Hz,
1H), 7.73 (d, J ) 8.0 Hz, 1H), 7.75 (d, J ) 8.0 Hz, 2H), 7.95 (d, J )
7.5 Hz, 1H), 8.00 (dd, J ) 8.0, 3.0 Hz, 2H), 11.90 (s, 1H), and 11.91
(s, 1H). 13C{1H} NMR (63 MHz, CDCl3): δ 25.27, 28.52, 48.15, 67.24,
67.28, 112.1, 112.3, 114.1, 116.3, 118.8, 118.9, 122.1, 122.2, 122.5,
123.2, 124.4, 125.7, 127.5, 130.2, 130.5, 131.1, 131.2, 135.6 (2 × C),
136.8, 137.7, 139.8, 152.1 (2 × C), 159.1 (2 × C), and 162.8. IR
(KBr): Vmax 1672, 1607, 1583, 1533, 1487, 1427, 1385, 1302, and 1247
cm-1. MS (ESI): m/z (rel intensity) 1530 (100%, MH+). HRMS calcd
for C96H80N12NaO8 (MNa+) 1551.6120, found 1551.6127. Anal. Calcd
for C96H80N12O8: C, 75.37; H, 5.27; N, 10.99 or for C96H81N12O8.5
(M‚0.5H2O): C, 74.93; H, 5.31; N, 10.92. Found: C, 74.95; H, 5.37;
N, 10.96.
Optical Resolution of the Strapped Cyclic Amide (55). The
enantiomers of the strapped cyclic amide 55 were separated on a
semipreparative scale by CSP HPLC (Chiralcel OD column, 1.0 cm ×
25 cm; 2-propanol/hexanes, 40/60; 4 mL min-1, 40 °C). UV detection
was performed at 254 nm. Injections of ∼0.5 mg of the racemate in
50 µL of CH2Cl2 were made every 30 min. The fast-eluting enantiomer
was collected between 11.8 and 16.1 min, and the slow-eluting
enantiomer was collected between 19.0 and 26.5 min. The collected
products were enantiomerically pure (ee > 99.9% and 99.5%, respec-
tively) by analytical CSP HPLC and were used in the subsequent
racemization studies (see Table 1).
1,6-Bis{3-[2-(2-aminophenyl)-4-hydroxycarbonylbenzimidazol-
1-ylmethyl]phenoxy}hexane (51). To a solution of ester 47 (8.09 g,
9.44 mmol) in THF (230 mL) was added 1 M LiOH (58 mL, 58 mmol),
and the mixture was stirred at room temperature for 20 h. The reaction
mixture was evaporated in vacuo, and the residue was suspended in
water (300 mL) and brought to pH 5 by addition of 1 M HCl. The
resulting suspension was stirred at room temperature for 2 h and filtered.
The precipitate was collected, washed with a copious amount of water,
and dried in vacuo to give the title compound 51 (6.95 g, 92%) as a
pale yellow solid that could be used in the subsequent step without
1
further purification. Amino acid 51. H NMR (400 MHz, d6-DMSO):
Ni(II) Complex (62). To a suspension of strapped cyclic amide 54
(59 mg, 80 µmol) in CH2Cl2/MeOH (1/1, v/v, 5 mL) was added
Ni(OAc)2‚4H2O (22 mg, 88 µmol), and the mixture was refluxed for
15 h (TLC) to give a deep-orange suspension. After being cooled to
room temperature, the suspension was concentrated to ∼2 mL and
filtered off. The collected solid was washed with cold MeOH (2 mL)
and dried to give the title compound 62 (60 mg, 94%) as a deep-orange
solid: Rf ) 0.70 (MeOH). 1H NMR (250 MHz, CDCl3): δ 0.72-0.94
(m, 1H), 1.43-1.65 (m, 1H), 3.25-3.42 (m, 1H), 3.70-3.88 (m, 1H),
5.17 (s, 1H), 5.56 and 6.11 (ABq, J ) 16 Hz, 2H), 6.64 (dd, J ) 8.0,
2.0 Hz, 1H), 6.88 (d, J ) 7.5 Hz, 1H), 7.12-7.35 (m, 4H), 7.46 (dt,
J ) 7.5, 1.5 Hz, 1H), 7.57 (dd, J ) 7.5, 1.0 Hz, 1H), 7.63 (d, J )
8.0 Hz, 1H), and 8.08 (d, J ) 7.5 Hz, 1H). 13C{1H} NMR (101 MHz,
CDCl3): δ 26.10, 50.95, 66.71, 107.7, 115.3, 118.2, 118.3, 120.5, 123.9,
125.1, 125.2, 125.8, 127.5, 130.3, 131.8, 132.2, 135.1, 136.4, 138.0,
δ 1.29-1.36 (m, 2H), 1.54-1.67 (m, 2H), 3.83 (t, J ) 6.5 Hz, 2H),
5.49 (s, 2H), 5.89 (br s, 2H), 6.55-6.62 (m, 2H), 6.64 (t, J ) 7.0 Hz,
1H), 6.77 (dd, J ) 7.5, 2.0 Hz, 1H), 6.89 (d, J ) 8.0 Hz, 1H), 7.15 (t,
J ) 7.0 Hz, 1H), 7.23 (dt, J ) 7.0, 1.5 Hz, 1H), 7.28 (dd, J ) 7.5, 1.0
Hz, 1H), 7.35 (t, J ) 8.0 Hz, 1H), 7.76 (d, J ) 8.0 Hz, 1H), and 7.83
(d, J ) 7.5 Hz, 1H). 13C{1H} NMR (101 MHz, d6-DMSO): δ 25.57,
28.84, 48.11, 67.60, 112.0, 113.0, 114.1, 116.0, 116.3 (2 × C?), 118.8,
120.3, 122.7, 124.7, 130.2, 130.5, 131.6, 136.0, 138.3, 141.8, 148.3,
154.1, 159.2, and 166.8. IR (KBr): Vmax 1740, 1610, 1488, 1433, 1388,
and 1260 cm-1. MS (ESI): m/z (rel intensity) 801 (100%, MH+). HRMS
calcd for C48H44N6NaO6 (MNa+) 823.3220, found 823.3189.
Monomer (55) and Dimer (59). To a solution of amino acid 51
(3.00 g, 3.75 mmol) in CH2Cl2 (400 mL) was added NMM (1.8 mL,
16 mmol), followed by BOP (3.64 g, 8.24 mmol). The resulting yellow
9
J. AM. CHEM. SOC. VOL. 126, NO. 1, 2004 235

A R T I C L E S
Fekner et al.
149.3, 154.7, 158.9, and 165.9. IR (KBr): Vmax 1604, 1576, 1480,
1433,1414, 1320, 1292, and 1262 cm-1. MS (ESI): m/z (rel intensity)
793 (25%, M+) and 737 (100). HRMS calcd for C46H34N6NaNiO4
(MNa+) 815.1893, found 815.1887. Crystals suitable for X-ray dif-
fraction analysis were grown by slow evaporation of a CH2Cl2/MeOH
solution. For crystallographic data, see Supporting Information. The
Ni(II) complex 62 was optically resolved by analytical CSP HPLC
(Figure S9).
Supporting Information Available: Synthetic procedures for
the compounds not described in the Experimental Section;
selected CSP HPLC profiles; solutions and refinement, atomic
coordinates, bond lengths and angles, anisotropic thermal
1
parameters, and CIF files for 5, 25, 26, and 62; copies of H
and 13C NMR spectra for all new compounds (PDF). This
material is available free of charge via the Internet at
http://pubs.acs.org.
Acknowledgment. This work was supported by the National
Science Foundation (CAREER Award No. 9984071).
JA030196D
9
236 J. AM. CHEM. SOC. VOL. 126, NO. 1, 2004