Stereoselective coordination of C5-symmetric corannulene derivatives with an enantiomerically pure [RhI(nbd*)] metal complex

Article abstract of DOI:10.1002/anie.201006877

Catching the bowl! Enantiopure Rh^I dimethylnorbornadiene fragments selectively catch interconverting enantiomers of C₅-sym- pentasubstituted corannulene derivatives in one bowl form and allow the observation and isolation of enantiopure metal-buckybowl complexes. A quantum mechanical model (see picture) predicts the mechanism of molecular dynamics and degree of stereoselective recognition.

Full text of DOI:10.1002/anie.201006877

DOI: 10.1002/anie.201006877
Dynamic Resolution
Stereoselective Coordination of C5-Symmetric Corannulene
Derivatives with an Enantiomerically Pure [RhI(nbd*)] Metal
Complex**
Davide Bandera, Kim K. Baldridge,* Anthony Linden, Reto Dorta, and Jay S. Siegel*
Dedicated to Professor Franco Cozzi on the occasion of his 60th birthday
Numerous stereoselective catalytic processes rely on the
differentiation of internally enantiotopic faces of planar p-
systems by metal complexation.[1] Chiral bowl-shaped mole-
cules like 1,3,5,7,9-pentasubstituted corannulene, which
invert rapidly on the reaction timescale, present themselves
as achiral due to tautomeric equilibration of enantiomers,[2]
and thereby display externally enantiotopic p-faces.[3] Stereo-
selective complexation of one such p-face shifts the tauto-
meric equilibrium and offers the possiblity to effect a
Figure 1. Equilibrating enantiomers (P and M) of C5-symmetric coran-
nulene derivatives.
complete dynamic resolution.[4] Thus, such systems provide
a special platform for studying chiral ligand–metal molecular
recognition and metal–arene complex dynamics.
The hydrogens in one bowl form of corannulene (1) are
chirotopic and segregate into two internally enantiotopic sets
of five homotopic hydrogens; the endo and exo faces of the
bowl are diastereotopic. Bowl inversion renders all hydro-
gens, and the two faces, respectively homotopic. Under
conditions of bowl inversion, simple metal coordination to
either face of corannulene is not an event wherein stereo-
selection can occur (the products are homomeric). In
contrast, sym-pentasubstituted corannulenes, such as
1,3,5,7,9-pentamethylcorannulene (2), oscillate between
enantiomeric C5-symmetric bowl conformations (Figure 1).
The hydrogens of these conformers are internally homotopic
but externally enantiotopic; the faces are diastereotopic in the
static bowl, and rendered enantiotopic by inversion. Simple
metal complexation of 2 yields enantiomeric products,
whereby enantioselection becomes possible.[5]
metal fragment, and 3) complexation of 2 with an enantio-
merically pure metal fragment should lead to diastereoselec-
tive chiral complex formation that effectively resolves the
bowl forms of 2 into 100% of a single chiral diastereomer.
Each of these complexes provides an unambiguous way to
follow the dynamics of metal–arene rotation, migration, and
transfer from a single complex.[6]
The corannulene partners in this study include 1, 2,[7] and
1,3,5,7,9-penta-tert-butylcorannulene (3).[8] On the metal side,
the fragments include {Rh(nbd)}+ (nbd = bicyclo[2.2.1]hepta-
2,5-diene), in reagent form as [{(nbd)RhCl}2] (4), and {Rh-
(nbd*)}+ (nbd* = C2-symmetric (R,R)-2,5-dimethyl-bicyclo-
[2.2.1]hepta-2,5-diene), in reagent form as [{(nbd*)RhCl}2]
(5).[9] Reaction of a corannulene and a [{(nbd)RhCl}2]
derivative activated by silver(I) in dichloromethane at room
temperature produces the expected complexes 6–9
(Scheme 1).
From this stereochemical analysis, 1) complexation of 1
with a racemic metal fragment should yield two enantiomeric
products with diastereotopic hydrogen NMR signals from
corannulene, 2) complexation of 2 with a metal bearing
enantiotopic hydrogens should yield enantiomeric products
with diastereotopic hydrogen NMR signals coming from the
[*] Dr. D. Bandera, Prof. Dr. K. K. Baldridge, Priv.-Doz. Dr. A. Linden,
Prof. Dr. R. Dorta, Prof. Dr. J. S. Siegel
Organic Chemistry Institute, University of Zurich
Winterthurerstrasse 190, 8057 Zurich (Switzerland)
Fax: (+41)44-635-6888
Scheme 1. Synthesis of compounds 6–9.
Complexation of 1 with 5 to yield 6 exemplifies the first
case from the above stereochemical analysis. At the static
limit, the {Rh(nbd*)} fragment is located over one ring and is
not rotating. The molecular symmetry is C1 and all protons of
the corannulene and the nbd* are evident as unique signals.
Rotation of the metal fragment without migration increases
E-mail: kimb@oci.uzh.ch
jss@oci.uzh.ch
[**] We gratefully acknowledge the University of Zꢀrich and the Swiss
National Science Foundation for financial support.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201006877.
Angew. Chem. Int. Ed. 2011, 50, 865 –867
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
865

Communications
the symmetry to effectively dynamic C2 for the nbd*, but
leaves the corannulene in site symmetry C1. Migration of the
metal fragment creates a dynamic C5 symmetry for the
corannulene fragment but leaves it chirotopic. Under such
conditions, two diastereotopic sets of five exchanging hydro-
gens are observed either by 2D EXSY[10] directly on the
distinguished hydrogens (Figure 2) or by line-shape coales-
Figure 4. Left: X-ray molecular structure of 7; right: schematic repre-
sentation of the steric differentiation hypothesis.
result. This hypothesis was modeled quantum mechanically to
get a quantitative prediction of the effect.
M06/cc-pVDZ + (SDD + f) density functional theory cal-
culations on 7 reproduced the X-ray structure well, and
further showed that it would be possible to predict the
diastereoselectivity of the complexation of 2 with 5 to yield
8a/b or of 3 with 5 to yield 9a/b. Further computations carried
out on 8a/b and 9a/b predicted a difference of 1.88 kcalmolÀ1
and 3.54 kcalmolÀ1, respectively. These predicted energy
differences in the product isomers would correspond to a
product ratio of 25:1 for the former and 400:1 for the latter,
were the full energy difference expressed in the transition
state to complexation.
Figure 2. 1H EXSY NMR spectrum for 6 (273 K in CD2Cl2 at tm =1.2 s).
Reaction of 2 with 5 yields 8a/b (2.5:1) and reaction of 3
with 5 produces 9a with an undetectable amount of 9b.
Compared to the computations, one can see that the product
path is locked in before the full energy difference is felt. Such
a result is reasonable given that transition state structures are
sterically looser than the product structures. The molecular
recognition on the basis of the computational results matches
the P configuration at corannulene with the R,R configuration
of nbd*.
Additional experimental evidence for the chiral nature of
these compounds exists in the Cotton effects observed in the
circular dichroism (CD) spectra (Figure 5). In the case of
nbd*, a negative Cotton effect is observed at 255 nm and only
a part of the curve is measurable, as previously observed for
non-conjugated norbornadienes.[12] Compound 6 displays a
positive Cotton effect at 355 nm, whereas 9 has a maximum at
347 nm. The shifted Cotton effects towards longer wave-
cence. The symmetry argument excludes external metal–
arene exchange and supports a dynamic character where the
metal wanders the aromatic surface without dissociation.
Complexation of 2 with 4 to yield 7 exemplifies the second
case from the above stereochemical analysis. Here also the
static limiting symmetry is C1 with all signals uniquely present
in the NMR spectrum. Rotation of the metal fragment effects
only a C2 symmetry for the nbd fragment, and as such, A and
B signals remain after coalescence of A/C and B/D pairs.
Migration again creates a dynamic C5 symmetry for the
corannulene fragment, but does not change the chirotopic
character for the nbd unit, and A and B signals persist
(Figure 3).
In the case of 7, crystals suitable for X-ray crystallographic
analysis revealed
a structure of the static complex
(Figure 4).[11] The structure showed the four different stereo-
electronic environments of the nbd vinyl hydrogens. From this
structure one sees that replacement of homotopic protons by
methyl groups could create a significant steric interaction
such that matched and mismatched diastereomers would
Figure 3. Dynamic symmetry due to rotation of the {M(nbd)} frag-
ment. st: static, all diastereotopic; dy: dynamic, two homotopic pairs.
Figure 5. CD spectra of (R,R)-2,5-dimethyl-bicyclo[2.2.1]hepta-2,5-
diene, 6, and 9 ([nbd*]=5ꢁ10À4 m; [6] and [9]=1ꢁ10À4 m).
866
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 865 –867

lengths for 6 and 9, compared to the free diene ligand nbd*,
point to a probable electron density transfer from the ligand
to the corannulene core.[13]
Keywords: corannulenes · dynamic NMR spectroscopy · metal–
arene complexes · stereochemistry
.
In conclusion, a dynamic resolution of the equilibrating
enantiomers of sym-pentasubstituted corannulenes by com-
plexation with a chiral, enantiomerically pure metal–NBD*
reagent is understood from a matched–mismatched analysis
of the stereoelectronic environment of the product diaste-
reomers.[14] Because the bowl forms are equilibrating faster
than the time scale for complexation, it is possible to drive the
reaction completely to one isomer in the case of 9, where
differences in the energy of the diastereomeric transition
states is high. Quantum mechanical computations predict this
selectivity well. CD analysis of the various derivatives gives a
first glimpse into the possible CD of sym-pentasubstituted
corannulenes frozen into one bowl form. These complexes
also presage stereochemical ways to study the structure and
dynamics of a future family of metal complexes of “imper-
fect” graphenes.[15]
[1] a) W. S. Knowles, Angew. Chem. 2002, 114, 2096 – 2107; Angew.
Chem. Int. Ed. 2002, 41, 1998 – 2027; b) R. Noyori, Angew.
Chem. 2002, 114, 2108 – 2123; Angew. Chem. Int. Ed. 2002, 41,
2008 – 2022; c) K. B. Sharpless, Angew. Chem. 2002, 114, 2126 –
2135; Angew. Chem. Int. Ed. 2002, 41, 2024 – 2032.
[2] L. Pauling in The Nature of the Chemical Bond, Cornell
University Press, Ithaca, 1960, pp. 564.
[3] K. Mislow, M. Raban in Topics in Stereochemistry, Vol. 1 (Eds.:
N. L. Allinger, E. L. Eliel), Wiley, Hoboken, 1967, pp. 1 – 38.
[4] a) P. Pfeiffer, K. Quehl, Ber. Dtsch. Chem. Ges. 1931, 64, 2667 –
2671; b) R. Noyori, M. Tokunaga, M. Kitamura, Bull. Chem. Soc.
Jpn. 1995, 68, 36 – 56.
[5] For reference studies on corannulene–metal complexation and
dynamics, see: a) T. J. Seiders, K. K. Baldridge, J. M. OꢁConnor,
J. S. Siegel, J. Am. Chem. Soc. 1997, 119, 4781 – 4782; b) J. S.
Siegel, K. K. Baldridge, A. Linden, R. Dorta, J. Am. Chem. Soc.
2006, 128, 10644 – 10645.
[6] a) S. Brydges, L. E. Harrington, M. J. McGlinchey, Coord. Chem.
Rev. 2002, 233–234, 75 – 105; b) S. Brydges, N. Reginato, L. P.
Cuffe, C. M. Seward, M. J. McGlinchey, C. R. Chim. 2005, 8,
1497 – 1505; c) T. G. Traylor, M. J. Goldberg, Organometallics
1987, 6, 2413 – 2416; d) T. G. Traylor, M. J. Goldberg, Organo-
metallics 1987, 6, 2531 – 2536.
Experimental/Computational Section
9: A solution of AgPF6 (10.1 mg, 0.04 mmol) in CH2Cl2 (0.2 mL) was
added dropwise to a vial containing 1,3,5,7,9-penta(tert-butyl)coran-
nulene (21.2 mg, 0.04 mmol) and [{Rh(nbd*)Cl}2] (10.0 mg,
0.02 mmol) in CH2Cl2 (0.4 mL) and the mixture (yellow) was stirred
in the dark for 30 min. The resulting suspension was filtered through a
pad of cotton/celite to eliminate the AgCl and the filtrate was
evaporated to dryness. The resulting yellow solid was washed with
pentane (3 ꢀ 2 mL) and dried in vacuo, yielding 33.6 mg (92%) of the
desired compound.
[7] a) T. J. Seiders, E. L. Elliot, G. H. Grube, J. S. Siegel, J. Am.
Chem. Soc. 1999, 121, 7804 – 7813; b) G. Grube, E. Elliot, R.
Stephens, J. S. Siegel, Org. Lett. 2003, 5, 713 – 716.
[8] Y. Sevryugina, A. Y. Rogachev, E. A. Jackson, L. T. Scott, M. A.
Petrukhina, J. Org. Chem. 2006, 71, 6615 – 6618.
[9] G. Berthon-Gelloz, T. Hayashi, J. Org. Chem. 2006, 71, 8957 –
8960.
1H NMR (500 MHz, CD2Cl2, 300 K): d = 8.58 (s, 1H), 8.48 (s,
1H), 8.28 (s, 1H), 8.17 (s, 1H), 8.08 (s, 1H), 7.00 (s, 1H), 3.85–3.80 (br,
2H), 3.18–3.14 (m, 2H), 1.92 (s, 9H), 1.77 (s, 9H), 1.73 (s, 18H), 1.63
(s, 9H), 1.51–1.47 (br, 6H), 1.24–1.22 ppm (t, J = 1.4 Hz, 2H).
13C NMR (125 MHz, CD2Cl2, 215 K): d = 157.2, 151.8, 150.0,
144.8, 136.9, 135.3, 134.4, 130.4, 129.9, 129.2, 128.7, 128.5, 124.7, 124.0,
121.7, 116.8, 116.1, 112.5, 101.0, 88.7, 78.6 (d, JRh-C = 5.1 Hz), 74.3,
59.6, 53.7 (DEPT 135), 53.3 (DEPT 135), 49.0 (d, JRh-C = 7.88 Hz), 46.4
(d, JRh-C = 7.9 Hz), 37.9, 37.3, 37.19, 37.16, 36.9, 32.0, 31.9. 31.6 (2C),
31.5, 23.1, 18.2 ppm. MS-ESI: m/z 753.6 [RhC49H62]+ in chloroform.
HRMS-ESI: calcd for RhC49H6: 753.3907; found: 753.3902.
The conformational analyses of the molecular systems described
in this study, including structural and orbital arrangements and other
property calculations, were carried out using Truhlarꢁs module
revision D.01 of Gaussian03[16] running on the Minnesota super-
computers (MSI), and the GAMESS[17] software packages. The M06
density functional of Truhlar and Zhao,[18] was used together with a
basis set consisting of the relativistic effective Stuttgart/Dresden-
double-z (6d, 10f) (SDD)[19] core potential for the transition metals
with state-averaged optimum f exponents (0.567;1.989) of Martin and
Sundermann,[20] and the Dunning correlation consistent basis set,[21]
cc-pVDZ, a [3s2p1d] contraction of a (9s4p1d) primitive set for all
other atoms. Full geometry optimizations were performed and
uniquely characterized via second derivatives (Hessian) analysis to
determine the number of imaginary frequencies (0 = minima; 1 =
transition state), and zero point contributions. Molecular orbital
contour plots, used as an aid in the analysis of results, were generated
and depicted using WEBMO,[22] and QMView.[23]
[10] 2D EXSY NMR = 2D exchange NMR spectroscopy, for review
see: a) C. L. Perrin, T. J. Dwyer, Chem. Rev. 1990, 90, 935 – 967;
b) J. Sandstrꢂm, Dynamic NMR Spectroscopy, Academic Press,
New York, 1982.
[11] CCDC-797434 (7) contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
from the Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
[12] K. Mislow, Ann. N. Y. Acad. Sci. 1962, 93, 459 – 484, and
references therein.
[13] V. V. Bashilov, F. M. Dolgushin, P. V. Petrovskii, V. I. Sokolov,
M. Sada, T. Benicori, G. Zotti, J. Organomet. Chem. 2005, 690,
4330 – 4336.
[14] S. Masamune, W. Choy, J. S. Petersen, L. R. Sita, Angew. Chem.
1985, 97, 1 – 31; Angew. Chem. Int. Ed. Engl. 1985, 24, 1 – 30.
[15] A. K. Geim, K. S. Novoselov, Nat. Mater. 2007, 6, 183.
[16] Truhlar/Zhang local module revision D.01 Gaussian 03, Gaus-
sian Inc., Pittsburgh PA, 2003.
[17] M. W. Schmidt, K. K. Baldridge, J. A. Boatz, S. T. Elbert, M. S.
Gordon, J. H. Jensen, S. Koseki, N. Matsunaga, K. A. Nguyen, S.
Su, T. L. Windus, S. T. Elbert, J. Comput. Chem. 1993, 14, 1347 –
1363.
[18] Y. Zhao, D. G. Truhlar, J. Chem. Phys. 2006, 125, 1 – 18.
[19] P. Fuentealba, H. Preuss, H. Stoll, L. von Szentpꢃly, Chem. Phys.
Lett. 1982, 89, 418 – 422.
[20] J. M. L. Martin, A. Sundermann, J. Chem. Phys. 2001, 114, 3408 –
3420.
[21] T. H. Dunning, J. Chem. Phys. 1989, 90, 1007 – 1023.
[22] WEBMO: T. Cundari, J. R. Schmidt, www.webmo.net.
[23] K. K. Baldridge, J. P. Greenberg, J. Mol. Graphics 1995, 13, 63 –
66.
Received: November 2, 2010
Published online: January 5, 2011
Angew. Chem. Int. Ed. 2011, 50, 865 –867
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
867