Racemic atropisomeric N,N-chelate ligands for recognizing chiral carboxylates via Zn(II) coordination: Structure, fluorescence, and circular dichroism

Article abstract of DOI:10.1021/ic801269z

Two racemic atropisomeric N,N′-chelate ligands, bis{3,3′-[N-Ph- 2-(2′-py)indolyl]} (1) and bis{3,3′-N-4-[N-2-(2′-py)indolyl] phenyl-2-(2′-py)indolyl} (2), have been found to be able to distinguish the enantiomers of Zn((R)-BrMeBu)₂ and Zn((S)-BrMeBu)₂ where BrMeBu = O₂CCH(Br)CHMe₂, with a distinct and intense CD spectral response at approximately the 10 μM concentration range. Computational studies established that the (R)-1-Zn((R)-BrMeBu)₂ or (S)-1-Zn((S)-BrMeBu)₂ diastereomer is more stable than (R)-1-Zn((S)-BrMeBu)₂ or (S)-1-Zn((R)-BrMeBu)₂. In addition, computational studies showed that the CD spectra of (S)-1-Zn((S)-BrMeBu)₂ and (S)-1-Zn((R)-BrMeBu)₂ are similar. ¹H NMR spectra confirmed that these two diastereomers exist in solution in about a 2:1 ratio for both complexes of 1 and 2. The distinct CD response of the racemic ligands 1 and 2 toward the chiral zinc(II) carboxylate is therefore attributed to the preferential formation of one diastereomer. The binding modes of the zinc(II) salt with ligands 1 and 2 were established by the crystal structures of the model compounds 1-Zn(tfa)₂ and 2-Zn(tfa)₂ (tfa = CF₃CO₂^-), where the Zn^II ion is chelated by the two central pyridyl groups in the ligand. Fluorescent titration experiments with various zinc(II) salts showed that the fluorescent spectrum of the atropisomeric ligand displays an anion-dependent change. The zinc(II) binding strength to the N,N′-chelate site of the atropisomeric ligand has been found to play a key role in the selective recognition of different chiral zinc(II) carboxylate derivatives by the racemic atropisomeric ligands.

Full text of DOI:10.1021/ic801269z

Inorg. Chem. 2008, 47, 10017-10024
Racemic Atropisomeric N,N-Chelate Ligands for Recognizing Chiral
Carboxylates via Zn(II) Coordination: Structure, Fluorescence, and
Circular Dichroism
Theresa M. McCormick and Suning Wang*
Department of Chemistry, Queen’s UniVersity, Kingston, Ontario K7L 3N6, Canada
Received July 7, 2008
Two racemic atropisomeric N,N′-chelate ligands, bis{3,3′-[N-Ph-2-(2′-py)indolyl]} (1) and bis{3,3′-N-4-[N-2-(2′-
py)indolyl]phenyl-2-(2′-py)indolyl} (2), have been found to be able to distinguish the enantiomers of Zn((R)-BrMeBu)2
and Zn((S)-BrMeBu)2 where BrMeBu ) O2CCH(Br)CHMe2, with a distinct and intense CD spectral response at
approximately the 10 µM concentration range. Computational studies established that the (R)-1-Zn((R)-BrMeBu)2
or (S)-1-Zn((S)-BrMeBu)2 diastereomer is more stable than (R)-1-Zn((S)-BrMeBu)2 or (S)-1-Zn((R)-BrMeBu)2. In
addition, computational studies showed that the CD spectra of (S)-1-Zn((S)-BrMeBu)2 and (S)-1-Zn((R)-BrMeBu)2
1
are similar. H NMR spectra confirmed that these two diastereomers exist in solution in about a 2:1 ratio for both
complexes of 1 and 2. The distinct CD response of the racemic ligands 1 and 2 toward the chiral zinc(II) carboxylate
is therefore attributed to the preferential formation of one diastereomer. The binding modes of the zinc(II) salt with
ligands 1 and 2 were established by the crystal structures of the model compounds 1-Zn(tfa)2 and 2-Zn(tfa)2 (tfa
) CF3CO2-), where the ZnII ion is chelated by the two central pyridyl groups in the ligand. Fluorescent titration
experiments with various zinc(II) salts showed that the fluorescent spectrum of the atropisomeric ligand displays an
anion-dependent change. The zinc(II) binding strength to the N,N′-chelate site of the atropisomeric ligand has
been found to play a key role in the selective recognition of different chiral zinc(II) carboxylate derivatives by the
racemic atropisomeric ligands.
Introduction
most commonly used one. In order for CD to work
effectively, however, either the chiral host or the chiral guest
must have strong absorption in the UV-vis region and the
resulting host-guest complexes must have distinct CD
spectra.2 Many examples of using enantiomerically pure host
molecules for chiral sensing by CD detection methods are
known in the literature.3 Supramolecular hosts that produce
a CD signal when interacting with a chiral guest have been
shown to be viable chiral sensors.4 Many achiral supramo-
lecular hosts such as porphyrin derivatives have been shown
Chiral sensing is an important and intensive research field
because of its broad applications in probing and understand-
ing chiral recognition events in biological systems, the
development of chiral catalytic systems, and the syntheses
of new chiral drugs.1 Among the various physical methods
for fast and sensitive chiral sensing and the determination
of enantiomeric excess (ee), circular dichroism (CD) is the
*
To whom correspondence should be addressed. E-mail:
Wangs@chem.queensu.ca.
(1) (a) Gottarelli, G.; Lena, S.; Masiero, S.; Pieraccini, S.; Spada, G. P.
Chirality 2008, 20, 471. (b) Mateos-Timoneda, M. A.; Crego-Calama,
M.; Reinhoudt, D. N. Chem. Soc. ReV. 2004, 33, 363. (c) Matsushita,
H.; Yamamoto, N.; Meijler, M. M.; Whirching, P.; Lerner, R. A.;
Matsushita, M.; Janda, K. D. Mol. BioSyst. 2005, 1, 303. (d) Takeuchi,
M.; Mizuno, T.; Shinkai, S.; Shirakami, S.; Itoh, T. Tetrahedron:
Asymmetry 2000, 11, 3311. (e) Lee, S. J.; Lin, W. B. J. Am. Chem.
Soc. 2002, 124, 4554. (f) Ma, L.; White, P. S.; Lin, W. B. J. Org.
Chem. 2002, 67, 7577. (g) Tan, H. Q.; Li, Y. G.; Zhang, Z. M.; Qin,
C.; Wang, X. L.; Wang, E. B.; Su, Z. M. J. Am. Chem. Soc. 2007,
129, 10066. (h) Dybtsev, N. N.; Nuzhdin, A. L.; Chun, H.; Bryliakov,
K. P.; Talsi, E. P.; Fedin, V. P.; Kim, K. Angew. Chem., Int. Ed. 2006,
45, 916.
(2) (a) Fischbeck, A.; Bartke, N.; Humpf, H. U. Monatsh. Chem. 2005,
136, 397, and references cited therein. (b) Hayashi, T.; Aya, T.;
Nonoguchi, M.; Mitzutani, T.; Hiseda, Y.; Kitagawa, S.; Ogoshi, H.
Tetrahedron 2002, 58, 2803.
(3) (a) Maeda, K.; Morioka, K.; Yashima, E. Macromolecules 2007, 40,
1349. (b) Kim, W.; So, S. M.; Chagal, L.; Lough, A. J.; Kim, B. M.;
Chin, J. J. Org. Chem. 2006, 71, 8966. (c) Li, Z.-B.; Pu, L. J. Mater.
Chem. 2005, 15, 2860. (d) Lin, J.; Rajaram, A. R.; Pu, L. Tetrahedron
2004, 11277. (e) Chen, Z.-H.; He, Y.-B.; Hu, C.-G.; Huang, X.-H.;
Hu, L. Aust. J. Chem. 2008, 61, 310. (f) Pu, L. Chem. ReV. 2004,
104, 1687. (g) Nieto, S.; Lynch, V. M.; Anslyn, E. V.; Kim, H.; Chin,
J. J. Am. Chem. Soc. 2008, 130, 9232.
10.1021/ic801269z CCC: $40.75  2008 American Chemical Society
Inorganic Chemistry, Vol. 47, No. 21, 2008 10017
Published on Web 10/03/2008

McCormick and Wang
Chart 1
Thin-layer chromatography was carried out on silica gel. Flash
chromatography was carried out on silica (silica gel 60, 70-230
mesh). All UV-vis spectra were collected by using an Ocean Optics
Inc. spectrometer and Spectra Suite software. 1H NMR spectra were
recorded on a Bruker Avance 500 MHz spectrometer. Excitation
and emission spectra were recorded on a Photon Technologies
International QuantaMaster model C-60 spectrometer. The CD
spectra were recorded on a Jasco 715 CD spectrometer with a 1
cm path length. Elemental analyses were performed by Canadian
Microanalytical Service Ltd., Delta, British Columbia, Canada, or
at the University of Toronto, Toronto, Ontario, Canada. Ligands 1
and 2 were prepared using our recently reported procedures.8
Synthesis of Zn(BrMeBu)2. A total of 1 mmol of ZnO and 2
mmol of (R)- or (S)-2-bromo-3-methylbutyric acid were added to
10 mL of degassed toluene. The mixture was stirred under N2
overnight, and the solvent was removed under vacuum. The residue
was dissolved in tetrahydrofuran (THF). After removal of the
insoluble solid, the solvent was evaporated to give the product.
The racemic salt can be made in the same manner by using racemic
2-bromo-3-methylbutyric acid.
Synthesis of 1-Zn(tfa)2. A solution of 82 mg (0.280 mmol) of
Zn(tfa)2 in ∼5 mL of THF was layered on top of a solution of 150
mg (0.280 mmol) of 1 in ∼5 mL of CH2Cl2. Slow diffusion of the
layers resulted in light-yellow crystals of the complex in 44% yield
in ca. 5 days. 1H NMR in CD2Cl2 (δ, ppm, 223 K): 6.97-7.00 (m,
3H), 7.22-7.25 (m, 1H), 7.29 (dd, J ) 7.5 and 7.5 Hz, 1H), 7.36
(d, J ) 5.0 Hz, 1H), 7.39-7.41 (m, 2H), 7.47 (dd, J ) 7.5 and 7.5
Hz, 1H), 7.70 (dd, J ) 7.5 and 7.5 Hz, 1H), 7.87 (dd, J ) 7.5 and
7.5 Hz, 1H), 7.92 (d, J ) 5.0 Hz, 1H), 8.68 (d, J ) 5.0 Hz, 1H).
Anal. Calcd for C42H26F6N4O4Zn: C, 60.77; H, 3.15; N, 6.75. Found:
C, 60.65; H, 3.16; N, 6.60.
to be able to act as chiral sensors via induced chirality upon
binding with chiral guest molecules.5 Certain racemates and
biological molecules such as DNA have been demonstrated
to be effective in chiral sensing using CD methods.6
Examples of chiral sensing of organic molecules mediated
by metal-ligand interactions, albeit relatively uncommon,
are also known in the literature.7 Recently, we reported a
new class of N,N-chelate atropisomeric ligands including 1
and 2 shown in Chart 1. Although we have not been able to
resolve the enantiomers of 1 and 2, we have shown that the
racemate 1 can bind to a CuI ion to form racemic homochiral
complexes that upon resolution by hand-picking of the
crystals display distinct CD spectra.8 Further investigation
on this system has led to the discovery that the racemic
atropisomers 1 and 2 can be used directly for sensing certain
chiral zinc(II) carboxylates such as (S)-2-bromo-3-methyl-
butyrate ((S)-BrMeBu) and (R)-2-bromo-3-methylbutyrate
((R)-BrMeBu) and determining the ee by CD methods. The
zinc(II) binding to the chelates 1 and 2 is inevitably affected
by the carboxylate binding ability to the zinc(II) center, which
has been found to play a key role in the CD response. The
details of our investigation are reported herein.
Synthesis of 2-Zn(tfa)2. A solution 75 mg (0.081mmol) of 2 in
∼10 mL of CH2Cl2 was layered with a solution of 24 mg (0.081
mmol) of Zn(tfa)2 in THF. The layers were allowed to slowly mix.
As the solvents evaporated, a yellow powder precipitated. The
solution was decanted, and the powder was recrystallized from
Experimental Section
All starting materials were purchased from Aldrich Chemical
Co. and used without further purification. Solvents were freshly
distilled over appropriate drying reagents under a N2 atmosphere.
1
CH2Cl2 and hexanes to afford light-yellow crystals (∼50%). H
NMR in CD2Cl2 (500 MHz, δ, ppm, 298 K): 8.76 (d, J ) 5 Hz,
1H), 8.48 (d, J ) 5 Hz, 1H), 7.95 (dd, J ) 7.5 and 7.5 Hz, 1H),
7.78 (d, J ) 5 Hz, 1H), 7.69 (dd, J ) 7.5 and 7.5 Hz, 1H), 7.62 (br
s, 2H), 7.51-7.44 (m, 6H), 7.38 (d, J ) 10 Hz, 1H), 7.34-7.30
(m, 2H), 7.26 (dd, J ) 7.5 and 7.5 Hz, 1H), 7.21 (dd, J ) 5 and
5 Hz, 1H), 7.17 (s, 1H), 7.04 (s, 2H). Anal. Calcd for
C68H42F6N4O4Zn·3.5CH2Cl2: C, 56.81; H, 3.27; N, 7.41. Found:
C, 56.43; H, 3.14; N, 7.59.
(4) (a) Hembury, G. A.; Borovkov, V. V.; Inoue, Y. Chem. ReV. 2008,
108, 1. (b) Bhyrappa, P.; Borokov, V. V.; Inoue, Y. Org. Lett. 2007,
9, 433. (c) Ikeda, H.; Li, Q.; Ueno, A. Bioorg. Med. Chem. Lett. 2006,
16, 5420. (d) Mateos-Timoneda, M. A.; Crego-Calama, M.; Reinhoudt,
D. N. Chem. Soc ReV. 2004, 33, 363. (e) Ishi-I, T.; Crego-Calama,
M.; Timmerman, P.; Reinhoudt, D. N.; Shinkai, S. Angew. Chem.,
Int. Ed. 2002, 41, 1924. (f) Huang, X.; Nakanishi, K.; Berova, N.
Chirality 2000, 12, 237. (g) Piguet, C.; Bernardinelli, G.; Hopfgartner,
G. Chem. ReV. 1997, 97, 2005.
Syntheses of 1-Zn((S)-BrMeBu)2 and 2-Zn((S)-BrMeBu)2.
These two compounds were synthesized on an NMR scale, by
combining 4.7 mg (8.8 µmol) of 1 with 3.8 mg (9.1 µmol) of Zn((S)-
BrMeBu)2 or 5.8 mg (6.3 µmol) of 2 with 2.8 mg (6.5 µmol) of
Zn((S)-BrMeBu)2 in the same manner as that for the Zn(tfa)2
complexes. Both compounds were characterized by 1H NMR
spectra. 1H NMR for 1-Zn((S)-BrMrBu)2 in CD2Cl2 (500 MHz, δ,
ppm, 203 K): 8.73 (d, J ) 5 Hz, 0.35H, o-H, py, minor
diastereomer), 8.70 (d J ) 5 Hz, 0.65H, o-H, py, major diastere-
omer), 7.88 (d, J ) 10 Hz, 1H), 7.77 (dd, J ) 7.5 and 7.5 Hz, 1H),
7.67 (dd, J ) 7.5 and 7.5 Hz 1H), 7.45-7.41 (m, 3H), 7.37 (dd, J
) 7.5 and 7.5 Hz, 1H), 7.34-7.31 (m, 2H), 7.27-7.18 (m, 2H),
6.97 (d, J ) 5 Hz, 1H), 6.92 (d, J ) 10 Hz, 1H), 4.12 (d, J ) 5
Hz, 0.65H, R-H, BrMeBu, major diasteromer), 3.91 (d, J ) 5 Hz,
0.35H, R-H, BrMeBu, minor diastereomer), 1.93 (m, 1H), 0.61 (s,
(5) (a) Li, W.-S.; Jiang, D.-L.; Suna, Y.; Aida, T. J. Am. Chem. Soc. 2005,
127, 7700. (b) Allenmark, S. Chirality 2003, 15, 409. (c) Proni, G.;
Pescitelli, G.; Huang, X.; Quraishi, N. Q.; Nakanishi, K.; Berova, N.
Chem. Commun. 2002, 1590. (d) Kurtan, T.; Nasri, N.; Li, Y.-Q.;
Hang, X.; Nakanishi, K.; Berova, N. J. Am. Chem. Soc. 2001, 123,
5962.
(6) (a) Maeda, K.; Yahima, E. Top. Curr. Chem. 2006, 265, 47. (b) Krois,
D.; Lehner, H. J. Chem. Soc., Perkin Trans. 2 1995, 489. (c) Barton,
J. K.; Dannenberg, J. J.; Raphael, A. L. J. Am. Chem. Soc. 1982, 104,
4967. (d) Barton, J. K.; Lolis, E. J. Am. Chem. Soc. 1985, 107, 708.
¨
(e) Onfelt, B.; Lincoln, P.; Norden, B. J. Am. Chem. Soc. 2001, 123,
3630.
(7) (a) Gao, F.; Ruan, W.-J.; Chen, J.-M.; Zhang, Y.-H.; Zhu, Z.-A.
Spectrochem. Acta, Part A 2005, 62, 886. (b) Anderberg, P. I.; Turner,
J. J.; Evans, K. J.; Hutchins, L. M.; Harding, M. M. Dalton Trans.
2004, 1708. (c) Tsukube, H.; Shinoda, S.; Tamiaki, H. Coord. Chem.
ReV. 2002, 227. (d) Tsukube, H.; Hosokubo, M.; Wada, M.; Shinoda,
S.; Tamiaki, H. Inorg. Chem. 2001, 40, 740. (e) Meskers, S. C. J.;
Dekkers, H. P. J. M. J. Phys. Chem. A 2001, 105, 4589. (f) Kurta´n,
T.; Nesnas, N.; Li, Y.-Q.; Huang, X.; Nakanishi, K.; Berova, N. J. Am.
Chem. Soc. 2001, 123, 5962. (g) Dai, L.-X.; Zhou, Z.-H.; Zhang, Y.-
Z.; Zhou, Y.-F. J. Chem. Soc., Chem. Commun. 1987, 1760.
(8) McCormick, T. M.; Liu, Q.; Wang, S. Org. Lett. 2007, 9, 4087.
1
6H). H NMR 2-Zn((S)-BrMrBu)2 in CD2Cl2 (500 MHz, δ, ppm,
298 K): 8.86 (d, J ) 5 Hz, 1 H), 8.52 (s, 1H), 7.88-7.77 (m, 3H),
7.68 (dd, J ) 7.5 and 7.5 Hz, 1H), 7.51 (d, J ) 10 Hz, 1H),
10018 Inorganic Chemistry, Vol. 47, No. 21, 2008

Racemic Atropisomeric N,N-Chelate Ligands
Table 1. Selected Bond Lengths (Å) and Angles (deg)
Compound 1-Zn(tfa)2
Zn(1)-O(1)
Zn1(1)-O(4)
Zn(1)-N(1)
Zn(1)-N(4)
1.9890(13)
1.9872(14)
2.0557(15)
2.0655(15)
O(4)-Zn(1)-O(1)
O(4)-Zn(1)-N(1)
O(1)-Zn(1)-N(1)
O(4)-Zn(1)-N(4)
O(1)-Zn(1)-N(4)
N(1)-Zn(1)-N(4)
106.11(6)
102.53(6)
95.57(6)
105.34(6)
105.34(6)
140.81(6)
Compound 2-Zn(tfa)2
Zn(1)-O(2)
Zn1(1)-N(1)
1.9716(17)
2.061(2)
O(2)-Zn(1)-O(2′)
O(2)-Zn(1)-N(1)
O(2)-Zn(1)-N(1′)
N(1)-Zn(1)-N(1′)
119.57(11)
102.60(7)
98.20(7)
137.99(11)
Figure 1. Structure of 1-Zn(tfa)2 with 50% thermal ellipsoids and labeling
schemes for the key atoms. Only one set of disordered F atoms are shown
for clarity.
7.46-7.39 (m, 6H), 7.33-7.24 (m, 4H), 7.21 (dd, J ) 5 and 5 Hz,
1H), 7.18 (s, 1H), 7.05 (d, J ) 10 Hz, 1H), 7.01 (dd, J ) 5 and 5
Hz, 1H), 3.99 (s, 1H), 1.71 (m, 1H), 0.32 (s, 6H).
and 2, we attempted to obtain single crystals of the complexes
for X-ray diffraction analysis. However, no crystals suitable
for X-ray diffraction were obtained for the complexes of
Zn(BrMeBu)2. We therefore focused our efforts on obtaining
single crystals of the nonchiral zinc(II) carboxylate, Zn(tfa)2,
Molecular Orbital Calculations. The ground-state molecular
geometry optimizations were performed for (S)-1-Zn((R)-BrMeBu)2
and (S)-1-Zn((S)-BrMeBu)2. The geometric parameters from X-ray
diffraction analysis of 1-Zn(tfa)2 were modified and used as the
starting point for the geometry optimization. The Gaussian suite
of programs (Gaussian 03)9 was used with the B3LYP/6-311G(d)
basis set employing a restricted Hartree-Fock level of computation.
CD spectra for both diastereomers were computed by time-
dependent density functional theory (TD-DFT) calculations using
the optimized geometrical parameters and the same basis set. The
first 25 singlet states were calculated.
-
tfa ) CF3CO2 with ligands 1 and 2. Fortunately, single
crystals of both complexes 1-Zn(tfa)2 and 2-Zn(tfa)2 were
obtained, and their structures were successfully determined
by X-ray diffraction. Both complexes were found to be 1:1
adducts of the ligand with Zn(tfa)2.
Structures of 1-Zn(tfa)2 and 2-Zn(tfa)2. As shown in
Figure 1, the molecule of 1-Zn(tfa)2 has an approximate C2
symmetry, with the Zn center being chelated by two pyridyl
groups of ligand 1. The dihedral angle between the two
indolyl rings is 118.1°, which is considerably smaller than
that of the free ligand (125.1°),8 indicating a greater twist
between the two rings due to zinc(II) coordination. The two
tfa anions are bound to the Zn center as terminal ligands.
The Zn-N [2.056(1)-2.065(1) Å] and Zn-O [1.987(1)-
1.989 (1) Å] bond lengths are normal;1g,h,10,11 however, the
geometry around the ZnII ion is a highly distorted tetrahedron
with a very large N-Zn-N angle, 140.81(6)°, resembling
that observed in the structure of [Cu(1)2]+.8 Consistent with
the behavior of the copper(I) complex reported earlier, the
phenyl group on ligand 1 in 1-Zn(tfa)2 displays a hindered
rotation around the C-N bond in solution due to nonbonding
hydrogen interactions between the phenyl, the indolyl, and
the pyridyl groups, as confirmed by variable-temperature 1H
NMR spectra of 1-Zn(tfa)2 (see the Supporting Information).
The crystals of 2-Zn(tfa)2 are very small and not very
uniform. Fortunately, some of the crystals produced sufficient
diffractions when used in X-ray diffraction experiments,
hence allowing the determination of the structure of 2-Zn(t-
fa)2 shown in Figure 2. The molecule of 2-Zn(tfa)2 has a
Titration Experiments. For typical CD and fluorescent titration
experiments, a 1 × 10-5 M solution of 1 or 2 in CH2Cl2 was
prepared. A total of 0.1 mol equiv of a 1 × 10-2 M solution of
zinc(II) salt in THF was added with a micropipette and mixed for
2 min before recording of the spectrum. The NMR titration
experiment was carried out with ∼5 mg of 1 or 2 in CD2Cl2 in an
NMR tube, and a 1.0 M solution of Zn((R)- BrMeBu)2 in CD3OD
was added incrementally with a micropipette.
X-ray Crystallographic Analysis. Single crystals of 1-Zn(tfa)2
and 2-Zn(tfa)2 were mounted on glass fibers for data collection.
Data were collected at ambient temperature for 1-Zn(tfa)2 at 180
K for 2-Zn(tfa)2 on a Bruker Apex II single-crystal X-ray diffrac-
tometer with graphite-monochromated Mo KR radiation, operating
at 50 kV and 30 mA. The crystals of 1-Zn(tfa)2 belong to the
j
triclinic space group P1, while the crystals of 2-Zn(tfa)2 belong to
the monoclinic space group C2/c. No significant decay was observed
for both samples. Data were processed on a PC using the Bruker
SHELXTL software package (version 5.10) and are corrected for
absorption effects. The structures were solved by direct methods.
One of the CF3 groups in 1-Zn(tfa)2 displays rotational disordering,
which was modeled and refined successfully. All non-H atoms were
refined anisotropically. Selected bond lengths and angles are
provided in Table 1. Complete crystal data can be found in the
Supporting Information.
Results and Discussion
Binding Modes and Structures of Complexes of
Zinc(II) Carboxylates with Ligands 1 and 2. The chiral
recognition study focused on two chiral zinc(II) carboxylates:
(S)-BrMeBu and (R)-BrMeBu. To establish the binding mode
of the chiral zinc(II) carboxylate with the chelate ligands 1
(10) Wu, Q.; Lavigne, J. A.; Tao, Y.; D’Iorio, M.; Wang, S. Inorg. Chem.
2000, 39, 5248.
(11) (a) Liu, S.-F.; Wu, Q.; Schmider, H. L.; Aziz, H.; Hu, N.-X.; Popovic,
Z.; Wang, S. J. Am. Chem. Soc. 2000, 122, 3671. (b) Prince, R. H. In
ComprehensiVe Coordination Chemistry; Wilkinson, G., Gillard, R. D.,
McCleverty, J. A., ed.; Pergamon Press: New York, 1987; Vol. 5,
Chapter 56.1. (c) Kerr, M. C.; Preston, H. S.; Ammon, H. L.; Huheey,
J. E.; Stewart, J. M. J. Coord. Chem. 1981, 11, 111. (d) Pang, J.;
Marcotte, E. J.-P.; Seward, C.; Brown, R. S.; Wang, S. Angew. Chem.,
Int. Ed. 2001, 40, 4042.
(9) Frisch,M. J.; et al. Gaussian 03, revision C.02; Gaussian, Inc.:
Wallingford, CT, 2004.
Inorganic Chemistry, Vol. 47, No. 21, 2008 10019

McCormick and Wang
Figure 2. Structure of 2-Zn(tfa)2 with 50% thermal ellipsoids and labeling schemes for the key atoms.
Chart 2
NMR data of 2-Zn(tfa)2 appear to indicate that there are two
dominating isomers in solution with about a 1:1 ratio (see
the Supporting Information), based on the appearance of two
distinct H1-py chemical shifts at low temperature. However,
it is possible that the chemical shifts of isomer C overlap
with those of A and B, making it difficult to determine the
precise distributions of the three isomers in solution.
Nonetheless, the structural isomerism observed in 2-Zn(tfa)2
is not surprising because of their similarity with that observed
in platinum(II) and copper(I) complexes based on 1,3- and
1,4-bis[2-(2-py)benzimidazolyl]benzene ligands.12
Structures
of
1-Zn((S)-BrMeBu)2
and
2-Zn((S)-BrMeBu)2 in Solution. The chiral zinc(II) salt
Zn((S)-BrMeBu)2 or Zn((R)-BrMeBu)2 was found to bind
to ligands 1 and 2 in the same manner as Zn(tfa)2, based on
1
the similarity of the H NMR data of 1-Zn((S)-BrMeBu)2
and 2-Zn((S)-BrMeBu)2 with those of 1-Zn(tfa)2 and 2-Zn(t-
fa)2. However, the presence of the chiral center on the
BrMeBu ligand can lead to the formation of diastereomers.
For example, diastereomers (R)-1-Zn((S)-BrMeBu)2 and (S)-
1-Zn((S)-BrMeBu)2 are expected when Zn((S)-BrMeBu)2 is
reacted with the racemate 1. 1H NMR titration experiments
indeed show that when less than 1 equiv of Zn((S)-BrMeBu)2
is added to the solution of 1 in CD2Cl2, two sets of well-
resolved chemical shifts for both H1-py and HR-BrMeBu
protons with a ∼2:1 ratio are observed (see Figure 3 and
the Supporting Information), which can be attributed to the
two diastereomers. 1H NMR titration experiments also
confirmed that Zn(BrMeBu)2 binds to 1 fairly strongly
because of the fact that the free ligand and the bound ligand
have distinct and well-resolved signals when the ZnII:1 ratio
is less than 1 (Figure 3). When more than 1 equiv of Zn((R)-
BrMeBu)2 is added, the diastereomer peaks are no longer
resolved at ambient temperature, which can be attributed to
the dynamic exchange between bound and nonbound Zn((R)-
BrMeBu)2. For the 1:1 complex 1-Zn((R)-BrMeBu)2, the
diastereomer peaks can be resolved at ∼203 K with the same
2:1 integrated intensity ratio as was observed in the NMR
titration experiments (see the Supporting Information). The
variable-temperature 1H NMR spectra of 2-Zn((R)-BrMeBu)2
is complex because of the presence of geometric isomers
and diastereoisomers associated with each geometric isomer.
Similar geometric isomeric peaks are observed in 2-Zn((R)-
crystallographically imposed C2 symmetry. The ZnII ion is
bound to the two central pyridyl rings of ligand 2 in the
same manner as that in 1-Zn(tfa)2 with similar Zn-O and
Zn-N bond lengths. One notable difference is the bond
angles of O-Zn-O [106.11(6)° for 1-Zn(tfa)2 and 119.57(11)°
for 2-Zn(tfa)2] and N-Zn-N [140.81(6)° for 1-Zn(tfa)2 and
139.99(11)° for 2-Zn(tfa)2]. The much smaller O-Zn-O
bond angle in 1-Zn(tfa)2 is likely caused by the relatively
short separation distance between the noncoordinating O(3)
atom and the ZnII ion [2.836(1) Å], which is much shorter
than that between O(2) and Zn(1) [3.003(1) Å] in 2-Zn(tfa)2
[106.11(6)°]. The dihedral angle between the two central
indolyl rings in 2-Zn(tfa)2 is 128.8°, much greater than that
in 1-Zn(tfa)2 but nearly identical with that of the free ligand
2 (128.5°). The inner indolyl ring and the outer indolyl ring
are not coplanar with the phenyl linker with dihedral angles
of 122.7° and 126.1°, respectively, with respect to the phenyl
plane, which is again caused by nonbonding interactions of
H atoms between the 2-(2-py)indolyl group and the phenyl.
Significantly, in the crystal structure of 2-Zn(tfa)2, the two
neighboring inner and outer 2-(2-py)indolyl groups are on
the same side with respect to the phenyl plane, i.e., having
syn-syn geometry, A, shown in Chart 2. The coordination
of the two central py rings to the ZnII ion forces the two
central 2-(2-py)indolyls to orient in opposite directions
(dihedral angle, 128.8°). As a result, the other possible
geometric isomers are anti-anti and syn-anti, shown as B
(12) (a) Liu, Q. D.; Jia, W. L.; Wang, S. Inorg. Chem. 2005, 44, 1332. (b)
Jia, W. L.; McCormick, T.; Tao, Y.; Lu, J. P.; Wang, S. Inorg. Chem.
2005, 44, 5706.
1
and C, respectively, in Chart 2. Variable-temperature H
10020 Inorganic Chemistry, Vol. 47, No. 21, 2008

Racemic Atropisomeric N,N-Chelate Ligands
distinguishing chiral zinc(II) carboxylates, the CD spectral
change of 1 and 2 toward Zn((R)-BrMeBu)2 and Zn((S)-
BrMeBu)2 was examined. As shown in Figure 5 (and the
titration diagram in the Supporting Information), the addition
of either Zn((S)-BrMeBu)2 or Zn((R)-BrMeBu)2 to the
solution of the racemate of 1 in CH2Cl2 (1.0 × 10-5 M)
causes the appearance of a distinct CD spectrum in the
225-400 nm region, which increases in intensity with the
amount of zinc(II) added. The CD spectrum of 1-Zn((S)-
BrMeBu)2 is approximately the mirror image of that of
1-Zn((R)-BrMeBu)2 in the 250-400 nm region, supporting
that the racemic 1 can indeed distinguish the enantiomers
of Zn(BrMeBu)2 by CD. Because Zn(BrMeBu)2 has no
absorption or a CD signal at λ > 250 nm, the intense CD
bands in the 250-450 nm region are clearly a consequence
of the electronic transitions of 1, as supported by the UV-vis
spectra of 1 and its zinc(II) complex (Figure 6). Significantly,
at a fixed concentration of the ligand 1, the intensity of the
CD band at λmax ) 320 nm changes with the ee of the chiral
zinc(II) carboxylate, as shown by Figure 5, an indication that
the N,N-chelate 1 has the potential for quantitative deter-
mination of ee of Zn(BrMeBu)2. The larger racemic chelate
ligand 2 was found to display a similar CD spectral response
upon addition of the chiral Zn(BrMeBu)2 except that the CD
band at 320 nm is much more intense and dominant,
compared to that of 1 (Figure 6), consistent with the fact
that the zinc(II) binding site in 2-Zn(BrMeBu)2 is similar to
that of 1-Zn(BrMeBu)2 as established by the crystal structures
1
Figure 3. Top: Partial H NMR spectrum of 1 with 0.5 equiv of Zn((S)-
BrMeBu)2 in CD2Cl2 at 298 K showing the H1-py peaks and the
Ha-BrMeBu peaks of the two diastereomers. Bottom: Partial 1H NMR
spectrum of 2-Zn((S)-BrMeBu)2 in CD2Cl2 at 203 K showing the
Ha-BrMeBu peaks of diastereomers (the two sets of peaks at 4.00-3.93
and 3.83-3.72 ppm) and geometrical isomers (the multiple peaks within
each set of diastereomer peaks).
1
1-Zn(tfa)2 and 2-Zn(tfa)2 and H NMR data. The relatively
intense CD band at 320 nm in the spectrum of 2 with
Zn(BrMeBu)2 is consistent with the greater extinction
coefficient of the corresponding absorption band, compared
to that of 1, as shown in Figure 6. Based on the NMR data
and the calculated structures, the CD response of the racemic
1 and 2 toward the chiral Zn(BrMeBu)2 shown in Figure 6
is likely a consequence of the CD spectral difference of the
diastereomers and the preferential formation of one diaste-
reomer.
Computational Study. To determine if the (S)-1-Zn((R)-
BrMeBu)2 and (S)-1-Zn((S)-BrMeBu)2 [or (R)-1-Zn((S)-
BrMeBu)2 and (R)-1-Zn((R)-BrMeBu)2] diastereomers have
different CD spectra, TD-DFT calculations on this pair of
diastereomers were performed using the optimized geometry
parameters.13 The first 25 singlet states were calculated and
a σ ) 0.2 eV line width was used.14 The computed CD
spectra of (S)-1-Zn((R)-BrMeBu)2 and (S)-1-Zn((S)-BrMe-
Bu)2 are shown in Figure 7. These CD spectra are similar
and match well with the experimental data in the 300-450
nm region, but a notable difference is evident in the short-
wavelength region because of mostly the chiral carboxylate
contributions (see the Supporting Information). Population
analysis for the first 25 singlet excited states further confirms
that most of the transitions with high oscillator strengths are
tfa)2 (see the Supporting Information), which complicates
the diastereomer peak assignments in the H1-py region.
However, the two Ha-BrMeBu diastereomer peaks in the
spectrum of 2-Zn((R)-BrMeBu)2 are well resolved at ∼223
K with a ∼2:1 ratio, similar to that of 1-Zn((R)-BrMeBu)2
(see Figure 3 and the Supporting Information). Hence, NMR
experiments established that for both ligands 1 and 2 there
is a preferential formation for one of the diastereomers with
Zn((R)-BrMeBu)2 [or Zn((S)-BrMeBu)2].
Calculated Structures of 1-Zn(BrMeBu)2. To determine
which diastereomer of 1-Zn(BrMeBu)2 is favored, we
performed DFT calculations using the Gaussian 03 suite of
programs at the B3LYP level with the 6-311G(d) basis set9
for the diastereomers (S)-1-Zn((R)-BrMeBu)2 and (S)-1-
Zn((S)-BrMeBu)2, assuming that a binding mode similar to
that of Zn(tfa)2 is adopted by the chiral zinc salt. Geometry
optimization was performed for both diastereomers; the
calculated structures for both diastereomers are shown in
Figure 4. Indeed, the structures of the two diastereomers
resemble that of 1-Zn(tfa)2. The computational results
indicated an energy difference of ∼8.50 kJ/mol between (S)-
1-Zn((R)-BrMeBu)2 and (S)-1-Zn((S)-BrMeBu)2, with the
(S)-1-Zn((S)-BrMeBu)2 complex [or (R)-1-Zn((R)-BrMeBu)2]
being the more stable diastereomer.
(13) (a) Crawford, T. D.; Tam, M. C.; Abrams, M. L. J. Phys. Chem. A
2007, 111, 12057. (b) Autschbach, J.; Jorge, F. E.; Ziegler, T. Inorg.
Chem. 2003, 42, 2867.
(14) (a) Coughlin, F. J.; Oyler, K. D.; Pascal, R. A.; Bernhard, S. Inorg.
Chem. 2008, 47, 974. (b) O’Boyle, N. M.; Tenderholt, A. L.; Langner,
K. M. J. Comput. Chem. 2007, 29, 839.
CD Spectral Response of Ligands 1 and 2 toward
Zn((R)-BrMeBu)2 and Zn((S)-BrMeBu)2. To establish
whether or not the racemic ligands 1 and 2 are capable of
Inorganic Chemistry, Vol. 47, No. 21, 2008 10021

McCormick and Wang
Figure 4. Calculated structures of (S)-1-Zn((S)-BrMeBu)2 (left) and (S)-1-Zn((R)-BrMeBu)2. Color code: C, gray; H, white; N, blue; O, red; Br, orange; Zn,
dark gray.
Figure 5. CD spectra of 1 (2.0 × 10-5 M) with 1 equiv of Zn(BrMeBu)2
of varying ee in CH2Cl2.
dominated by 2-(2-py)indolyl parts of the ligand, which is
consistent with the UV-vis data. The orbital diagrams for
the highest occupied molecular orbital (HOMO), second
HOMO, and three unoccupied orbitals for (S)-1-Zn((S)-
BrMeBu)2 shown in Figure 8 further illustrate that the low-
energy electronic transitions in the complex involve the
central part of ligand 1 with little contributions from zinc(II)
-5
Figure 6. Top: CD spectra of 1 and 2 (1.0 × 10
M) with 1 equiv of
or the carboxylate. The computational results also support
that using the pure (R)-1 or pure (S)-1 alone will not be able
to effectively distinguish the enantiomers of Zn(BrMeBu)2
because of the similarity of the CD spectra of diastereomers
(S)-1-Zn((R)-BrMeBu)2 and (S)-1-Zn((S)-BrMeBu)2 [or (R)-
1-Zn(R-BrMeBu)2 and (R)-1-Zn((S)-BrMeBu)2]. On the basis
of the computational results and NMR data, we can conclude
that the distinct CD spectral response of racemic 1 toward
Zn((R)-BrMeBu)2 and Zn((S)-BrMeBu)2 is due to the pref-
erential formation of diastereometers (R)-1-Zn((R)-BrMeBu)2
and (S)-1-Zn((S)-BrMeBu)2, respectively, versus (S)-1-Zn((R)-
BrMeBu)2 and (R)-1-Zn((S)-BrMeBu)2.
Selectivity and Fluorescent Response of 1 and 2
toward Zinc(II) Carboxylates. To find out if the atropiso-
meric N,N-chelate ligand 1 can discriminate other chiral
zinc(II) carboxylates, we also examined the CD response of
1 toward Zn((S)-MeBu)2, MeBu ) 2-methylbutyrate.10
Surprisingly, the addition of Zn((S)-MeBu)2 to a solution of
the racemic 1 under the same conditions as those used for
Zn(BrMeBu)2 did not yield any significant CD signal in the
chiral Zn(BrMeBu)2 in CH2Cl2. Bottom: UV-vis spectra of 1, 2, and their
complexes with Zn(BrMeBu)2.
250-400 nm region. One possible explanation for the lack
of CD response of 1 toward Zn((S)-MeBu)2 is that 2-meth-
ylbutyrate is a stronger donor because of the electron-
donating 2-methyl group, compared with 2-bromo-3-
methylbutyrate, thus competing more effectively for the ZnII
ion and weakening the ZnII binding with the chelate ligand
1. Because 1 and 2 are highly fluorescent molecules, the
relative binding strength of zinc(II) carboxylates can be
compared qualitatively by fluorescent titration experiments.
Indeed, we have observed that the addition of Zn(BrMeBu)2
and Zn(tfa)2 to a solution of the racemic 1 in CH2Cl2 causes
a red shift of the emission band of 1. Under the same
conditions, Zn(OAc)2 and Zn(MeBu)2 have little impact on
the emission spectrum of 1, unless a large excess of the
zinc(II) salt (>15 equiv) is added. For comparison, we also
performed fluorescent titration with Zn(ClO4)2, where the
ClO4- anion is a well-known poor donor and the ZnII ion is
10022 Inorganic Chemistry, Vol. 47, No. 21, 2008

Racemic Atropisomeric N,N-Chelate Ligands
Figure 7. Computed CD spectra of (S)-1-Zn((R)-BrMeBu)2 and (S)-1-
Zn((S)-BrMeBu)2.
Figure 9. Fluorescent titration diagram of the solution of 1 (1.0 × 10-5
M) in CH2Cl2 with Zn(ClO4)2 (0.0-2.0 equiv), Zn(tfa)2 (0.0-5.0 equiv),
and Zn(BrMeBu)2 (0.0-5.0 equiv), respectively.
Figure 10. Stern-Volmer plots of the fluorescent titration of ligand 1 by
zinc(II) salts.
ligand 1 is much stronger than that of Zn(tfa)2, Zn(MeBu)2,
and Zn(OAc)2 but much weaker than that of Zn(ClO4)2.
Hence, the fluorescent data support that the lack of a CD
response of ligand 1 toward the chiral Zn((S)-MeBu)2 is due
to the weak binding of this particular zinc(II) salt. The
fluorescent response of ligand 2 toward zinc(II) carboxylates
follows the same pattern as that of ligand 1, as shown by
data in Figure 11.
We also examined the CD response of 1 with the chiral
(S)-2-bromo-3-methylbutyric acid [or (R)-2-bromo-3-meth-
ylbutyric acid] because the acid can interact with the
atropisomeric ligand via hydrogen bonds. However, no
significant CD spectral change was observed with the free
acid. It is likely that the weak hydrogen bonds between the
racemic host and the chiral guest are not sufficient to
differentiate the diastereomers to produce significant CD
signals. Therefore, the zinc(II) coordination to the atropiso-
meric ligand is clearly a key for distinguishing the two
enantiomers of 2-bromo-3methylbutyrate via CD spectra. In
Figure 8. Diagram showing the frontier orbitals (orbital number and energy
are shown on the left) of (S)-1-Zn((S)-BrMeBu)2. The corresponding orbitals
for (S)-1-Zn((R)-BrMeBu)2 are similar with slight differences in the orbital
energies.
expected to bind strongly with the chelate ligand. The
addition of Zn(ClO4)2 to a solution of 1 causes a strong
fluorescent quenching and an emission peak red shift as well.
As shown in Figures 9 and 11, the fluorescent response of
Zn(BrMeBu)2 toward 1 and 2 is between those of Zn(tfa)2
and Zn(ClO4)2. One notable difference between the chiral
and nonchiral zinc(II) salts is the absence of isosbestic points
in the chiral zinc(II) salt titration diagrams, which is
consistent with the presence of diastereomers of the chiral
zinc(II) complexes. From the Stern-Volmer plots shown in
Figure 10, it is evident that the binding of Zn(BrMeBu)2 to
Inorganic Chemistry, Vol. 47, No. 21, 2008 10023

McCormick and Wang
been synthesized and structurally characterized. The ZnII ion
has been found to bind preferentially to the two central
pyridyl rings in both ligands. We have demonstrated that
the racemic ligands 1 and 2 are capable of distinguishing
the enantiomers of Zn(BrMeBu)2 and determining the ee by
using CD spectroscopy. Furthermore, we have established
that binding to the chelate site of ligands 1 and 2 by the
chiral zinc(II) carboxylate is critical for achieving CD
response. The binding strength of the carboxylate to the ZnII
ion was also found to play a key role in the CD response of
1 and 2 toward chiral zinc carboxylates, which may be used
to discriminate different chiral carboxylates in conjunction
with fluorescence spectra.
Acknowledgment. We thank the Natural Sciences and
Engineering Research Council of Canada for financial
support.
Supporting Information Available: Complete crystal data and
1
fully labeled structural diagrams for 1-Zn(tfa)2 and 2-Zn(tfa)2, H
Figure 11. Fluorescent titration diagrams and Stern-Volmer plots of ligand
2 (1.0 × 10-5 M) in CH2Cl2 by three different zinc(II) salts.
other words, the ZnII ion functions as a mediator to facilitate
the recognition event between atropisomeric ligand and the
chiral carboxylate.
NMR titration data for 1 with Zn(BrMeBu)2, variable-temperature
1H NMR spectra for 1-Zn(tfa)2, 2-Zn(tfa)2, 1-Zn((S)-BrMeBu)2, and
2-Zn((S)-BrMeBu)2, fluorescent titration spectra of 1 with Zn(OAc)2
and Zn((S)-MeBu)2, and DFT and TD-DFT computational details.
This material is available free of charge via the Internet at
http://pubs.acs.org.
In summary, two Zn(tfa)2 complexes with two 2-(2-
py)benzimidazolyl-based atropisomeric ligands, 1 and 2, have
IC801269Z
10024 Inorganic Chemistry, Vol. 47, No. 21, 2008