Self-recognition based on atropoisomerism with new chiral bidentate ligands and copper(i)

Article abstract of DOI:10.1002/1521-3765(20001002)6:19<3595::AID-CHEM3595>3.0.CO;2-W

A new family of atropoisomeric bidentate ligands that have a dissymmetric benzimidazole-pyridine binding site has been synthesized. Aromatic rings, that is naphthyl, iolyl, and cumyl, were introduced in order to fine tune the complexation properties of th

Full text of DOI:10.1002/1521-3765(20001002)6:19<3595::AID-CHEM3595>3.0.CO;2-W

FULL PAPER
Self-Recognition Based on Atropoisomerism with New Chiral Bidentate
Ligands and Copper(i)
Jean-Marc Vincent,*[a] Christian Philouze,[b] Isabelle Pianet,[c] and
Jean-Baptiste Verlhac*[a]
Abstract: A new family of atropoiso-
meric bidentate ligands that have a
dissymmetric benzimidazole-pyridine
binding site has been synthesized. Aro-
matic rings, that is, naphthyl, tolyl, and
cumyl, were introduced in order to fine
tune the complexation properties of the
ligands. The tetrahedral copper(i) com-
plexes L2Cu were prepared and the
structure of the complex with the naph-
thyl-substituted ligand was established
by X-ray diffraction. The behavior of the
L2Cu complexes in solution was studied
by 1H NMR spectroscopy. With the most
crowded cumyl-derived ligand, ligand
self-recognition based on chirality oc-
cured: 95% of the complex was present
in solution as a racemate RRD/SSL, the
heterochiral RSD/SRL isomers repre-
sented only 5% of the mixture, and the
RRL/SSD isomers were not detected.
Owing to lower steric repulsions within
the other L2Cu complexes (i.e., with the
naphthyl- and tolyl-based ligands) the
homorecognition is less pronounced, as
diastereomeric excesses of 6 and 26%
were measured, respectively.
Keywords: atropoisomerism ´ chir-
ality ´ copper ´ N ligands ´ self-
assembly
Ligand recognition based on chirality was shown to play an
important role in the non-linear effects in asymmetric reac-
tions.[5] However, in such cases the heterorecognition is favored;
this leads to the formation of heterochiral, oligomeric metal
species. Recently, Stack and co-workers reported the first
example of ligand homorecognition based solely on chirali-
ty.[6] Starting from a racemic mixture of a tetradentate amino
ligand that contained a trans-1,2-diaminocyclohexane back-
bone, a racemic mixture of stereospecific, homochiral, binu-
clear complexes L,L-[{Cu(RRL)}2]2‡ and D,D-[{Cu(SSL)}2]2‡ was
formed selectively. The homochiral complexes produced a
more compact and stable cubic structure than the heterochiral
ones. A more intricate example of self-recognition based on chi-
rality was described in the self-assembly of [2 Â 2] grids from
ditopic ligands that possess terpyridine-like binding sites.[7]
We were interested in studying the self-recognition process
for a structure encountered widely in coordination chemistry:
two bidentate ligands arranged in a tetrahedral complex.
Here, we report that a racemate of an atropoisomeric
bidentate ligand can self-assemble stereoselectively in the
presence of copper(i) ions in such a way as to favor the
formation of homochiral complexes.
Introduction
In 1993, Lehn and co-workers[1] introduced the ligand self-
recognition process as a new paradigm for the selective
synthesis of supramolecular architectures.[2] The synthesis of a
well-defined superstructure can be programmed by using a
mixture of ªinstructedº components that contain steric and
electronic information. By treating copper(i) ions with a
mixture of oligo-2,2'-bipyridine strands, which differ in the
number of binding sites, self-assembly of the double helicates
with two di-, tri-, tetra-, or penta-topic [oligo(2,2')-bipyridine]
occured spontaneously. Other examples of ligand homorec-
ognition in helicates have been reported based on 1) the
preferred coordination geometry of the metal ions in the
presence of an appropriate structure of the ligand,[1] 2) the
differences in distances between the metal-chelating groups of
ligands,[3] or 3) the nature of the countercation that binds to
the tetra-anionic complexes formed.[4]
[a] Dr. J.-M. Vincent, Prof. J.-B. Verlhac
Â
Laboratoire de Chimie Organique et Organometallique
CNRS, UMR 5802
Â
351, cours de la Liberation, 33405 Talence Cedex (France)
Fax : (‡33)556-84-69-94
E-mail: j-b.verlhac@lcoo.u-bordeaux.fr
[b] Dr. C. Philouze
Results and Discussion
Institut de Chimie des Substances Naturelles
Ã
CNRS Bat. 27, Avenue de la Terasse
Preparation of the ligands: The ligands 4, 5, and 6, which
contain a dissymmetric binding site with a pyridine and a
benzimidazole ring, were prepared as racemic mixtures
(Scheme 1). The pyridine-benzimidazole moiety has proved
91198 Gif sur Yvette (France)
[c] Dr. I. Pianet
Â
Cesamo, UMR 5802, Universite Bordeaux I
351 Cours de la Liberation, 33405 Talence Cedex (France)
Â
Chem. Eur. J. 2000, 6, No. 19 ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0619-3595 $ 17.50+.50/0
3595

J.-M. Vincent, J.-B. Verlhac et al.
FULL PAPER
perpendicular to the axis that
contains the ligating atoms, that
favors steric interactions be-
tween the ligands within the
same complex, thus allowing
the isomers to be discriminated.
Moreover, an aryl ring intro-
duced at the 4-position will
experience a coplanar relation-
ship with the p system of the
second partner involved in a
tetrahedral complex. Depend-
ing on the atropoisomer, p
overlap will occur between the
aryl substituent and either the
pyridyl or the benzimidazole
part of the complementary li-
gand, which leads to a fine
tuning of the energy of the
system.
The ligands 4, 5, and 6 were
obtained through a three-step
procedure (Scheme 1) from the
previously synthesized ligand
1,[9] which was prepared accord-
Scheme 1. Reaction pathway for the synthesis of 4, 5, and 6.
ing to the method described by
to be a very efficient complexation unit for transition metals.[8]
The methyl groups at the 5- and 6-positions play a triple role:
1) they are very useful H NMR probes, 2) they protect the
most reactive positions of the benzimidazole ring, thereby
forcing the bromination reaction to occur at the less reactive
4-position of the benzyl derivative 2, and 3) they prevent
complete rotation around the chirality axis. In short, the
functionnalization at the 4-position of the benzimidazole
moiety is the key feature of this new class of atropoisomeric
ligands. They possess a rigid structure and a chirality axis,
Reed and co-workers.[10] The benzylated compound 2, ob-
tained by a standard procedure with NaH and benzylbromide,
was selectively monobrominated with N-bromosuccinimide
(NBS) in DMF. A Suzuki coupling reaction was used to
introduce the o-tolyl, a-naphthyl, or the o-cumyl groups.
Enantiomerization energies of 20.7, 23.4, and 29.1 kcalmolÀ1
for 4, 5, and 6 respectively, were found with AM1 semi-
empirical calculations (CAChe 3.1). These values are in
between those found for 1,1'-binaphthyl (19 kcalmolÀ1) and
2,2'-binaphthol (33.8 kcalmolÀ1) groups.
1
Preparation of the complexes 4a, 5a, and 6a: The copper(i)
complexes [Cu(4)2]PF6 (4a), [Cu(5)2]PF6 (5a), and
[Cu(6)2]PF6 (6a), were prepared by treating two equivalents
of the corresponding ligand with [Cu(CH3CN)4]PF6. Crystals
of 4a suitable for single-crystal X-ray diffraction studies were
obtained by slow diffusion of petroleum ether into a solution
of the complex in 1,2-dichloroethane. The asymmetric cell
shows the presence of a racemic mixture of stereospecific
Abstract in French: Une nouvelle famille de ligands bidentates
Á
Â
Â
atropoisomeres possedant un site de complexation dissyme-
 Â
Â
Â
trique (benzimidazole-pyridine) a ete synthetisee. Des substi-
Â
tuants aromatiques de tailles variees, naphtyle, tolyle ou
   Â
cumyle, ont ete introduits afin de modifier les proprietes
Â
Á
complexantes des ligands. Les complexes tetrahedriques de
 Â
Â
Â
cuivre(i) L2Cu ont ete synthetises et la structure du complexe
  Â
‡
‡
homochiral complexes D-[Cu(R4)2] , RRD, and L-[Cu(S4)2] ,
Â
avec le ligand substitue par un noyau naphtyle a ete resolue par
diffraction des rayons X. Une etude par RMN 1H de ces
Â
SSL(Figure 1).
Â
complexes en solution a permis de mettre en evidence un
The Cu ions are located in a strongly distorded tetrahedral
environment. As expected, in the primary structure, strong
intramolecular stacking interactions were observed and plane
to plane distances were measured at around 3.5 Š. It should
be noted that in the RRD/SSL enantiomers these p-p
interactions arise from the preferential overlaping of the
naphthalene ring of one ligand with the pyridine ring of the
other. Intermolecular p-stacking interactions (3.8 Š) were
also observed in the secondary structure (Figure 1).
Â
Â
processus dꢀauto-reconnaissance de ligand base sur la chiralite.
Â
Á
En solution dans le dichloromethane, lꢀespece L2Cu obtenue
Â
Â
avec le ligand substitue par le groupe cumyle se presente sous
Â
forme dꢀun complexe homochiral racemique RRD/SSL. Les
Á
 Â
Â
isomeres heterochiraux RSD/SRL ne representent que 5% du
Â
Á
Â
Â
melange alors que les isomeres RRL/SSD ne sont pas detectes.
Â
Â
Â
Les repulsions steriques etant moins fortes dans les complexes
Â
L2Cu dont les ligands sont substitues par les groupes naphtyle
Â
Á
ou tolyle, le phenomene dꢀauto-reconnaissance est moins
Ligand self-recognition process: The schematic representa-
tion of the possible isomers formed upon self-assembly of the
racemic ligands with copper(i) (Figure 2) clearly shows that
Á
Á
 Â
Â
important, conduisant a des exces diastereoisomeriques mesu-
Â
res de 6 et 26% respectivement.
3596
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0619-3596 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 19

Self Recognition of Chiral Ligands
3595±3599
spectrum of the complex generated in situ displays the
formation of a single metal species with C2 symmetry (Fig-
ure 2) which was attributed to the homochiral enantiomers
RRD/SSL (95%). The same spectrum was obtained by
dissolving crystals of 6a in [D2] dichloromethane (Figure 3).
The heterochiral isomers RSD/SRL were also present in small
amount (5%), whereas the other set of homochiral isomers
(RRL/SSD) was not observed.
The diastereomeric excess (de) is thus equal to 90%
(Table 1). With the less crowded ligands 4 and 5, the self-
recognition process is less pronounced with de values of 6%
and 26% for 4a and 5a, respectively.
The 1H NMR assignments for 6a were done by comparison
with the spectra obtained for the complexes 4a and 5a
1
(Figure 3). The H NMR spectrum of 4a displays eight well-
resolved CH3-benzimidazole resonances in agreement with a
mixture of the possible isomers, for example RRD/SSL, RRL/
SSD, and RSD/SRL.
The C2 symmetric homochiral RRD/SSL and RRL/SSD
complexes have only two distinct CH3-benzimidazole reso-
nances for each pair of enantiomers thus leading to four reso-
nances. In the heterochiral RSD/SRL isomers, four CH3-
benzimidazole were differenciated owing to a lower symmetry;
this led to another set of four resonances. With the tolyl deriv-
ative 5a, a similar pattern is obtained, except that four addi-
tional CH3-tolyl resonances were observed: one for each pair
of the homochiral isomers and two for the heterochiral one.
The schematic representation of the isomers formed upon
complexation with copper(i) allows the key structural features
that promote such a discrimination to be easily visualized
(Figure 2). In the RRL/SSD isomers, the steric repulsions
between the isopropyl groups preclude their formation. In the
heterochiral complexes, only the use of bulky isopropyl
groups can prevent their formation, since, in that case, the
steric hindrance is due to interaction between an isopropyl
group and a hydrogen atom from the o-cumyl ring of the other
ligand. Accordingly, extended MM2calculations (Cache 3.1)
have clearly shown that the homochiral RRD/SSL isomers are
much favored over RRL/SSD ones, with a difference in
minimization energy of 6 kcalmolÀ1 found in the case of 6a.
Interestingly, for all the ligands studied herein, the com-
plexation is diastereoselective, since the homochiral RR
complexes mainly display the D configuration, whereas the
SS isomers are L. Examples of stereoselective complexation
with two chiral bidentate ligands are scarce.[11] The selective
complexation of platinum(iv) or copper(i) has been achieved
by using thienylpyridine[12] and phenanthroline[13] ligands,
respectively, substituted with chiral terpene moieties.
‡
Figure 1. Top: Ortep representation of D-[Cu(R4)2] (4a). Bottom: Ball
and stick representation of the racemate (RRD/SSL) present in the cell
which reveals the intra- and intermolecular p-stacking interactions.
À
À
Selected bond lengths [Š] and angles [8]: Cu N1 2.037(12), Cu N2
À
À
2.027(11), Cu N4 2.059(11), Cu N5 2.013(10); N5-Cu-N2 143.6(4), N1-Cu-
N4 123.3(5), N4-Cu-N2 114.3(4), N5-Cu-N1 119.4(4), N1-Cu-N2 81.5(5),
N5-Cu-N4 80.4(4).
Figure 2. Schematic representation of the possible isomers formed upon
complexation of 6 with Cu .
As expected for copper(i) complexes associated with
bidentate ligands, the isomers are in equilibrium in solution.
Upon crystallization (slow diffusion of petroleum ether into a
solution of 4a or 5a in 1,2-dichloroethane) these equilibria
were displaced toward the selective precipitation of the RRD/
SSL racemate. Ground crystals of pure, recrystallized 4a or
5a were dissolved at À608C in [D2] dichloromethane and
‡
the methyl groups of the three pairs of enantiomers lie in very
different chemical environments; this allows the character-
1
ization of each couple of isomers by H NMR spectroscopy.
1
The H NMR spectra were recorded at À208C or À408C in
1
their H NMR spectra immediately recorded at À408C. In
order to decrease the resonance broadening due to ligand
exchange (Figure 3).
both cases, the major species found in solution were the RRD/
SSL complexes (85%), that is, those found in the crystal of 4a,
whereas the RRL/SSD and RSD/SRL isomers were present
‡
When two equivalents of 6 are treated with [CuI(MeCN)4]
(one equivalent) in [D2] dichloromethane, the 1H NMR
Chem. Eur. J. 2000, 6, No. 19
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0619-3597 $ 17.50+.50/0
3597

J.-M. Vincent, J.-B. Verlhac et al.
FULL PAPER
The o-tolylboronic acid was purchassed
commercially (Lancaster). The a-naph-
thylboronic-acid was obtained by using
the classical procedure from 1-bromo-
naphtalene, B(OMe)3, and nBu-Li.
The o-cumylboronic acid was prepared
according to Yamamoto and co-work-
ers.[14] The solvents were of commercial
analytical grade. Thin-layer chroma-
tography (TLC) was performed on
polyester sheets coated with silica gel
60 F254 (SDS 321334), or coated with
neutral alumina 60 F254 (Polygram
802023). After elution, the plates were
scrutinized under a UV lamp. Column
chromatography was carried out on
neutral alumina (Aldrich 19, 997-4,
150 mesh). Liquid secondary-ion-mass
spectrometry (LSIMS) was carried out
in the positive-ion mode with a cesium
atom beam on a CG AutoSpec-Q spec-
trometer. The 1H NMR spectra were
recorded on either a Bruker AM200
(200 MHz) or a DPX400 (400 MHz)
spectrometer. 1H NMR spectra for the
complexes
were
performed
at
400 MHz on aliquots of each complex
(10 mg) dissolved in CD2Cl2 (0.5 mL).
Cyclic voltammetry (CV) was carried
out with a Pt working electrode, a Pt
control electrode, and a saturated cal-
omel
electrode
as
reference.
Figure 3. 1H NMR at 400 MHz in CD2Cl2 (methyl region) of 4a (À408C, top), 5a (À208C, middle), and 6a
(nBu)4NPF6 was used as support elec-
trolyte.
~
*
^
(À408C, bottom). CH3-resonances for RRD/SSL, for RRL/SSD, for RSD/SRL, S ˆ residual water.
Preparation of 2: Ligand
1 (3.2g,
14.2mmol) was added in small portions
to a NaH suspension (60%, 0.64 g, 16.0 mmol) in dry DMF (70 mL) at 08C,
under an inert atmosphere. After one hour at ambient temperature,
PhCH2Br (1.98 mL, 16.7 mmol) was added dropwise and the solution was
stirred for 24 h. The mixture was poured into water (800 mL) that was
vigourously stirred. The precipitate was collected by filtration and washed
with water and diethyl ether. The resultant powder was dissolved in CH2Cl2
and dried over Mg(SO4)2. After evaporation, 2 was obtained as a yellowish
powder in 86% yield. 1H NMR (200 MHz, CDCl3, 2 58C): d ˆ 8.59 (d, 1H),
8.42 (d, 1H), 7.80 (t, 1H), 7.64 (s, 1H), 7.22 (m, 7H), 6.16 (s, 2H), 2.39 (s,
3H), 2.35 (s, 3H); 13C NMR (50 MHz, CDCl3, 2 58C): d ˆ 151.77, 150.12,
149.44, 142.37, 138.74, 137.62, 136.46, 133.88, 132.73, 129.46, 128.12, 127.64,
125.40, 124.42, 121.05, 111.70, 49.77, 21.63, 21.24; LSIMS-MS: m/z (%): 314
Table 1. Isomeric ratios for complexes 4a, 5a, and 6a.
Complex
RRD/SSL
RRL/SSD
RSD/SRL
de[a]
statistical ratio
25%
49%
60%
95%
25%
4%
3%
50%
47%
37%
5%
0%
6%
26%
90%
4a
5a
6a
< 1%
[a] Diastereomeric excess.
only in small amounts (4% and 11%, respectively). The time
course behavior of the ratio of complexes clearly shows that
RRD/SSL concentration decreases as RSD/SRL concentra-
tion increases and RRL/SSD remains constant, equilibrium
being reached after five hours at À408C.
‡
(100) [M‡1] .
Preparation of 3: A solution of NBS (1.25 g, 7.00 mmol) dissolved in dry
DMF (20 mL) was added dropwise to a solution of 2 (2g, 6.37 mmol) in dry
DMF (80 mL). After being stirred for 48 h at ambient temperature, the
solution was poured into water (1 L); this led to the precipitation of a pale
yellow powder. The powder was dissolved in CH2Cl2, dried over Mg(SO4)2,
and, after evaporation, the solid was washed with hot ethanol which gave 3
in 65% yield. 1H NMR (200 MHz, CDCl3, 2 58C): d ˆ 8.58 (d, 1H), 8.52(d,
1H), 7.80 (t, 1H), 7.21 (m, 7H), 6.13 (s, 2H), 2.50 (s, 3H), 2.40 (s, 3H);
13C NMR (50 MHz, CDCl3, 2 58C): d ˆ 150.35, 149.51, 148.39, 140.52,
137.32, 136.67, 134.97, 133.75, 130.81, 128.50, 127.25, 126.59, 125.09, 123.71,
Conclusion
The judicious design of the new, chiral, bidentate ligand 6
allowed the stereoselective self-assembly of the mononuclear,
homochiral copper(i) complex 6a RRD/SSL starting from the
racemic ligand. We are currently working on the resolution of
the ligands to further study their stereoselective complexation
properties and their ability to catalyze enantioselective
reactions in the presence of transition metals.
‡
115.73, 110.13, 49.00, 22.09, 18.99; MS (EI): m/z (%): 391 (100) [M À H] .
Preparation of 4 and 5: Bromo derivative 3 (2g, 5.1 mmol), Ba(OH) 2 (2.4 g,
7.65 mmol), [Pd(PPH3)4] (0.3 g, 25.5 Â 10À5 mol), and the corresponding
boronic acid (5.60 mmol) were heated under reflux in DME (90 mL) and
H2O (15 mL) under an inert atmosphere for six hours, dichloromethane
and water were added. The organic phase was then separated and dried
over Mg(SO4)2. After column chromatography with neutral alumina
(dichloromethane/petroleum ether, 50:50), ligands 4 and 5 were isolated
as white powders in 80% yield.
Experimental Section
General procedures: The chemicals were obtained commercially and were
used without further purification. The previously synthesized ligand 1[9] was
prepared according to the procedure described by Reed and co-workers.[10]
Ligand 4: 1H NMR (400 MHz, CD2Cl2, 2 58C): d ˆ 8.48 (d, 1H), 8.00 (d,
1H), 7.91 (d, 2H), 7.59 (t, 2H), 7.46-7.12 (m, 10H), 6.20 (d, 1H), 6.05 (d,
1H), 2.40 (s, 3H) 1.92 (s, 3H); 13C NMR (100 MHz, CD2Cl2, 2 58C): d ˆ
3598
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0619-3598 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 19

Self Recognition of Chiral Ligands
3595±3599
151.3, 149.7, 148.9, 141.9, 138.7, 137.5, 137.2, 135.5, 134.2, 133.8, 131.4, 130.7,
129.1, 128.7, 128.5, 127.9, 127.8, 127.4, 126.9, 126.4, 126.2, 126.1, 125.1, 124.0,
goodness of fit on F2 ˆ 1.003. Residual electron density between À0.719
and 0.721 e ŠÀ3. The whole cationic structure was refined anisotropically.
The solvent and anions show severe disorder, and the solvent molecules
were treated isotropically. The best refinement was obtained with
population factors of 0.99 for pentane and 0.79 for hexane. The anions
were refined on split positions; the major position was refined anisotropi-
cally and for the minor position isotropically. All hydrogen positions were
calculated. The complex crystallizes together with PF6 ions and solvent
molecules (two n-hexane and one n-pentane). The asymmetric cell shows
two complex molecules that are enantiomers. These two molecules are
effectively related by the following mirror plane: x2 ˆ x1 ‡ 0.50750, y2 ˆ
y1 ‡ 0.01602, z2 ˆ z1 ‡ 0.00322. This plane is not a plane of symmetry for the
cell because the solvent molecules and the PF6 ions do not respect this
symmetry and show severe disorder. As a result of this loss of symmetry,
one of the parameters of the cell is doubled, and, therefore, the required
number of data is also doubled and their intensities are weakened. Taken
together this explains the rough R factors observed for this structure.
Crystallographic data (excluding structure factors) for the structure
reported in this paper have been deposited at the Cambridge Crystallo-
graphic Data Center as supplementary publication no.CCDC-134368.
Copies of the data can be obtained free of charges on application to CCDC,
12, Union Road, Cambridge, CB2 1EZ (UK) (Fax: (‡44)1223-336-033;
e-mail: deposit@ccdc.cam.ac.uk).
‡
110.9, 49.3, 21.9, 17.1; LSIMS-MS: m/z (%): 440 (100) [M‡1] ; elemental
analysis calcd (%) for C31H25N3 (439.3): C 84.74, H 5.69, N 9.56; found C
84.26, H 5.74, N 9.49.
Ligand 5: 1H NMR (400 MHz, CD2Cl2, 2 58C): d ˆ 8.5 (d, 1H), 8.22 (d, 1H),
7.7 (t, 1H), 7.3 (m, 11H), 6.23 (d, 1H), 6.08 (d, 1H), 2.42 (s, 3H), 2.06 (s,
3H), 2.05 (s, 3H); 13C NMR (100 MHz, CD2Cl2, 2 58C): d ˆ 152.1, 150.8,
149.7, 140.4, 141.9, 139.6, 138.7, 137.9, 136.0, 134.6, 134.0, 131.6, 131.1, 130.5,
129.7, 128.5, 128.4, 128.0, 126.6, 125.8, 124.7, 111.2, 49.9, 22.6, 21.0, 17.4;
‡
LSIMS-MS: m/z (%): 404 (100) [M‡1] ; elemental analysis calcd (%) for
C28H25N3 (403.3): C 83.38, H 6.20, N 10.42; found C 82.90, H 6.30, N 10.36.
Preparation of 6: Bromo derivative 3 (1.4 g, 3.6 mmol), Cs(CO3)2 (3.47 g,
10.6 mmol), [Pd(PPH3)2Cl2] (0.5 g, 7.1 Â 10À4 mol), and cumylboronic acid
(1.2g, 7.3 mmol) were stirred in DMF (60 mL) at 95 8C for five hours under
an inert atmosphere. Dichloromethane was added, and the organic phase
was washed four times with water and then dried over Mg(SO4)2. After
column chromatography with neutral alumina (dichloromethane/petro-
leum ether, 50:50) an amorphous solid was precipitated in ethanol;
compound 6 was isolated as a white powder in 65% yield. 1H NMR
(400 MHz, CD2Cl2, 2 58C): d ˆ 8.41 (d, 1H), 8.18 (d, 1H), 7.61 (t, 1H), 7.30,
(m, 10H), 6.22 (d, 1H), 5.98 (d, 1H), 2.60 (sept, 1H), 2.38 (s, 3H), 2.04 (s,
3H), 1.1, (d, 3H), 1.05 (d, 3H); 13C NMR (100 MHz, CD2Cl2, 2 58C): d ˆ
149.79, 147.70, 145.30, 137.94, 137.30, 136.50, 136.07, 135.52, 135.47, 132.50,
132.29, 130.11, 129.77, 129.02, 126.99, 126.42, 126.04, 124.88, 123.92, 122.26,
110.29, 49.95, 30.56, 24.33, 23.83, 22.10, 17.13; LSIMS-MS: m/z (%): 432.2
Acknowledgement
‡
(100) [M‡1] .
We thank Dr. J. Ruiz for the redox potential measurements, Dr. C. Riche
for his help with the X-ray structure refinement and B. Barbe for his
assistance with the NMR spectroscopy recording.
Preparation of complexes [Cu(4)2]PF6, (4a), [Cu(5)2]PF6, (5a), and
[Cu(6)2]PF6, (6a): Ligand 4, 5, or 6 (two equivalents) in deoxygenated
CH2Cl2 was added to [Cu(CH3CN)4]PF6 (one equivalent) and the solution
taken to dryness. Recrystallization of 4a and 5a (1,2-dichloroethane/
petroleum ether) afforded black-brown crystals in 85% yield. Compound
6a was recrystallized by slow evaporation of a solution of the complex in
EtOH/CH2Cl2, 2:1.
[1] a) R. Krämer, J.-M. Lehn, A. Marquis-Rigault, Proc. Natl. Acad. Sci.
USA 1993, 90, 5394 ± 5398; b) D.-P. Funeriu, Y.-B. He, H.-J. Bister,
J.-M. Lehn, Bull. Soc. Chim. Fr. 1996, 133, 673 ± 678; c) R. Stiller, J.-M.
Lehn, Eur. J. Inorg. Chem. 1998, 977 ± 982.
[2] J.-M. Lehn, Supramolecular Chemistry: Concepts and Perspectives,
VCH, Weinheim, 1995.
[3] D. L. Caulder, K. N. Raymond, Angew. Chem. 1997, 109, 1508 ± 1510;
Angew. Chem. Int. Ed. Engl. 1997, 36, 1439 ± 1442.
[4] M. Albrecht, M. Schneider, H. Röttele, Angew. Chem. 1999, 111, 512±
515; Angew. Chem. Int. Ed. 1999, 38, 557 ± 559.
[5] a) R. Noyori, M. Kitamura, Angew. Chem. 1991, 103, 34 ± 54; Angew.
Chem. Int. Ed. Engl. 1991, 30, 49 ± 69; b) C. Girard, H. B. Kagan,
Angew. Chem. 1998, 110, 3088 ± 3127; Angew. Chem. Int. Ed. 1998, 37,
2922 ± 2959.
[6] M. A. Masood, E. J. Enemark, T. D. P. Stack, Angew. Chem. 1998, 110,
973 ± 977; Angew. Chem. Int. Ed. 1998, 37, 928 ± 931.
[7] D. M. Bassani, J.-M. Lehn, K. Fromm, D. Fenske Angew. Chem. 1998,
110, 2534 ± 2537; Angew. Chem. Int. Ed. 1998, 37, 2364 ± 2367.
[8] a) C. Piguet, B. Bocquet, E. Muller, A. F. Williams Helv. Chimica Acta
1989, 72, 323 ± 337; b) M. Haga, T. Ano, K. Kano, S. Yamabe Inorg.
Chem. 1991, 30, 3843 ± 3849; c) S. Rüttimann, C. Piguet, G. Bernardi-
nelli, B. Bocquet, A. F. Williams, J. Am. Chem. Soc. 1992, 114, 4230 ±
4237; d) L. J. Charbonniere, G. Bernardinelli, C. Piguet, A. M.
Sargeson, A. F. Williams, J. Chem. Soc. Chem. Commun. 1994,
1419 ± 1420.
Compound 4a: 1H NMR (400 MHz, CD2Cl2, À408C, TMS): homochiral
species RRD/SSL: d ˆ 1.87 (s, 6H), 2.51 (s, 6H), 5.84 (dd, 4H), 6.2 ± 8.2 (c,
34H); RRL/SSD: d ˆ 1.55 (s, 6H), 2.37 (s, 6H); heterochiral species RSD/
SRL: d ˆ 1.69 (s, 3H), 1.74 (s, 3H), 2.43 (s, 3H), 2.48 (s, 3H); LSIMS-MS:
‡
m/z (%): 941 (100) [M À PF6] ; elemental analysis calcd (%) for
C62H50N6PF6Cu ´ CH3CN (1128.2): C 68.14, H 4.70, N 8.70, P 2.74, F
10.10, Cu 5.62; found C 67.56, H 4.65, N 8.72, P 2.66, F 10.09, Cu 5.74; E1/2 vs.
SCE ˆ 400 mV.
Compound 5a: 1H NMR (400 MHz, CD2Cl2, À208C, TMS): homochiral
species RRD/SSL: d ˆ 1.55 (s, 6H), 1.85 (s, 6H), 2.43 (s, 6H), 5.85 (brs, 4H),
6.48 (t, 2H), 6.74 (d, 2H), 7.25-7.85 (c, 18H), 8.17 (d, 2H); RRL/SSD: d ˆ
1.13 (s, 6H), 1.77 (s, 6H), 2.37 (s, 6H); heterochiral species RSD/SRL: d ˆ
1.33 (s, 3H), 1.65 (s, 3H), 1.80 (s, 3H), 1.81 (s, 3H), 2.40 (s, 3H), 2.42 (s, 3H);
‡
LSIMS-MS: m/z (%): 869 (100) [M À PF6] ; elemental analysis calcd (%)
for C56H50N6PF6Cu ´ H2O (1033.1): C 65.10, H 5.03, N 8.13, P 3.00, F 11.01,
Cu 6.15; found C 64.50, H 5.00, N 7.71, P 3.10, F 10.45, Cu 6.47; E1/2 vs.
SCE ˆ 440 mV.
Compound 6a: 1H NMR (400 MHz, CD2Cl2, À408C, TMS): homochiral
species RRD/SSL: d ˆ 0.61 (d, 6H), 0.70 (d, 6H), 1.80 (s, 6H), 2.18 (m, 2H),
2.40 (s, 6H), 5.88 (c, 6H), 6.35 (d, 2H), 6.63 (d, 2H), 6.82(t, 2H), 7.2-7.6 (c,
16H), 7.76 (t, 2H), 8.00 (d, 2H); LSIMS-MS: m/z (%): 926 (100) [M À
‡
PF6] ; elemental analysis calcd (%) for C60H58N6PF6Cu (1071.1): C 67.27, H
5.41, N 7.84, P 2.89, F 10.64, Cu 5.93; found C 67.18, H 5.38, N 7.91, P 2.88, F
10.25, Cu 6.12.
[9] V. I. Cohen, J. Heterocycl. Chem. 1979, 16, 13 ± 16.
[10] V. McKee, M. Zvagulis, J. V. Dagdigian, M. G. Patch, C. A. Reed, J.
Am. Chem. Soc. 1984, 106, 4765 ± 4772.
[11] For a review on the stereoselective synthesis of metal centers see: A.
Knof, A. von Zelewsky Angew. Chem. 1999, 111, 312± 333; Angew.
Chem. Int. Ed. 1999, 308, 302 ± 322.
[12] M. Gianini, A. Forster, P. Haag, A. von Zelewsky Inorg. Chem. 1996,
35, 4889 ± 4895.
[13] E. C. Riesgo, A. Credi, L. De Cola, R. P. Thummel Inorg. Chem. 1998,
37, 2145 ± 2148.
X-ray crystal structure analysis: Crystal data for [Cu(C31H25N3)2]2 ´ (PF6)2 ´
(C5H12)0.99 ´ (C6H14)0.79 from a single red crystal of 0.51 Â 0.1 Â 0.12mm.
Data were collected on a Enraf Nonius CAD4 four-circle diffractometer at
293 K with graphite monochromated CuKa radiation (l ˆ 1.54180 Š).
Monoclinic space group P2(1)/n, Z ˆ 4, a ˆ 23.656(5), b ˆ 15.0676(10), c ˆ
31.163(4) Š, a ˆ 908, b ˆ 91.280(9)8, g ˆ 908, Vˆ 11105.1(28) Š3, 1calcd
ˆ
1.385 gcmÀ3, F(000) ˆ 4823, m ˆ 1.400. q range: 2.32 ± 67.89, index ranges:
À28 ꢀ h ꢀ 28, 0 ꢀ k ꢀ 18, 0 ꢀ l ꢀ 37, 18973 collected reflexions, 18973
unique, R(int) ˆ 0.0000, 6070 observed [I > 2s(I)], 18965 were used to
fit 79 restraints and 1486 parameters. The structure was refined by full-
matrix least-square calculation (SHELLX93), R values: R ˆ 0.1259 for
6070 observed reflexions, wR2 ˆ 0.5254 for 18973 unique reflexions,
[14] S. Saito, T. Kano, K. Hatanaka, H. Yamamoto, J. Org. Chem. 1997, 62,
5651 ± 5656.
Received: March 2, 2000 [F2334]
Chem. Eur. J. 2000, 6, No. 19
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0619-3599 $ 17.50+.50/0
3599