Surprising substituent effects on the self-assembly of helicates from bis(bipyridyl) BINOL ligands

Article abstract of DOI:10.1021/jo900254r

(Figure Presented) A number of different bis(bipyridyl) BINOL ligands were prepared using a convergent building block approach. These were studied with regard to their ability to undergo self-assembly to dinuclear helicates upon coordination to suitable late-transition-metal ions. Surprisingly, the substituents at the periphery of the ligand structure were found to have a marked influence on the outcome of the self-assembly processes with regard to the helicates composition, the stereoselectivity of the helicate formation, their redox reactivity, and their electronical properties as scrutinized by NMR- and CD-spectroscopic methods as well as ESI-mass spectrometric methods.

Full text of DOI:10.1021/jo900254r

pubs.acs.org/joc
Surprising Substituent Effects on the Self-Assembly of Helicates from
Bis(bipyridyl) BINOL Ligands
€
Jens Bunzen, Rainer Hovorka, and Arne Lutzen*
ꢀ
Kekule Institute of Organic Chemistry und Biochemistry, University of Bonn, Gerhard-Domagk-Str. 1, 53121
Bonn, Germany
arne.luetzen@uni-bonn.de
Received February 6, 2009
A number of different bis(bipyridyl) BINOL ligands were prepared using a convergent building block
approach. These were studied with regard to their ability to undergo self-assembly to dinuclear
helicates upon coordination to suitable late-transition-metal ions. Surprisingly, the substituents at
the periphery of the ligand structure were found to have a marked influence on the outcome of the
self-assembly processes with regard to the helicates composition, the stereoselectivity of the helicate
formation, their redox reactivity, and their electronical properties as scrutinized by NMR- and
CD-spectroscopic methods as well as ESI-mass spectrometric methods.
Introduction
degree of selectivity of these processes provided by a given
ligand structure and even more difficult to predict the
configuration of newly formed stereochemical elements in
the self-assembled supramolecular aggregates.
Since the beginning of supramolecular chemistry, chemists
have tried to build up complex structures via self-assembly,
thereby mimicking nature.1 One of the most challenging
aspects of these efforts is the implementation of information
into the different moieties supposed to form the supramole-
cular structures. Chirality is one of the most prominent
features in this context, and there have been quite a number
of beautiful examples where (dia-)stereoselective self-assem-
bly could be achieved.2 However, the field is far from being
mature because it is still almost impossible to predict the
(1) Some general reviews: (a) Lindsey, J. S. New J. Chem. 1991, 15, 153–
180. (b) Lawrence, D. S.; Jiang, T.; Levett, M. Chem. Rev. 1995, 95, 2229–
2260. (c) Fyfe, M. C. T.; Stoddart, J. F. Acc. Chem. Res. 1997, 30, 393–401.
(d) Greig, L. M.; Philp, D. Chem. Soc. Rev. 2001, 30, 287–302.
(2) Some recent reviews about stereoselective self-assembly: (a) Spector,
M. S.; Selinger, J. V.; Schnur, J. M. Top. Stereochem. 2003, 24, 281–372.
(b) Hamilton, T. D.; MacGillivray, L. R. Cryst. Growth Design 2004, 4, 419–
430. (c) Mateos-Timoneda, M. A.; Crego-Calama, M.; Reinhoudt, D. N.
Chem. Soc. Rev. 2004, 33, 363–372. (d) Crego-Calama, M., Reinhoudt, D. N.,
Eds. Supramolecular Chirality. Top. Curr. Chem. 2006, 265. (e) He, C.; Zhao,
Y.; Guo, D.; Lin, Z.; Duan, C. Eur. J. Inorg. Chem. 2007, 3451–3463. (f) Lee, S.
L.; Lin, W. Acc. Chem. Res. 2008, 41, 521–537. For a more comprehensive list
of references on the diastereoselective self-assembly of helicates see references
listed in ref 19.
In 2002, we reported on the synthesis and self-assembly
behavior of chiral ligand 1.3 This ligand is able to form
€
(3) Lutzen, A.; Hapke, M.; Griep-Raming, J.; Haase, D.; Saak, W.
Angew. Chem., Int. Ed. 2002, 41, 2086–2089.
5228 J. Org. Chem. 2009, 74, 5228–5236
Published on Web 06/11/2009
DOI: 10.1021/jo900254r
r
2009 American Chemical Society

Bunzen et al.
JOCArticle
enantiomerically pure double- and triple-stranded helicates
with transition-metal ions in a completely diastereoselective
self-assembly process. This encouraging result was only
dimmed by the low solubility of the helicate obtained with
the OH-free ligand, which precipitated from organic solu-
tions within minutes to hours after formation depending on
the concentration of the initial mixture. We therefore decided
to introduce substituents into the ligand structure in order to
enhance the solubility and to further elaborate our concept.
The positioning of each substituent was a crucial part of
the design of the modified ligands. We chose the 50-posi-
tion on the 2,20-bipyridine moiety and the 6-position on the
1,10-binaphthyl (BINOL) part because interactions with the
self-assembly process should be at a minimum using these
positions.
SCHEME 1. Synthesis of BINOL Building Block 13
Results and Discussion
Synthesis. The main and last step in the synthesis of ligands
(P)- and (M)-2-6 is a 2-fold Sonogashira cross-coupling of
the BINOL and the bipyridine moieties, thereby connecting
the two main building blocks, which have to be prepared
prior to these reactions. By using this convergent strategy, we
were able to build up the ligand structures in a concise
manner with a minimum amount of time and resources,
thereby minimizing the number of steps which would other-
wise lie between 10 and 14 steps per ligand.
SCHEME 2. Synthesis of Ligands 2-6 via 2-Fold Sonogashira
Cross-Coupling
The 2,20-bipyridine building blocks 7-10 were synthesized
according to known procedures starting from (functionalized)
2-chloro- or 2-bromopyridines,4-10 which were coupled via
our modified Negishi cross-coupling procedure10,11 (see the
Supporting Information for details).
Two different BINOL moieties were prepared. Starting from
enantiomerically pure BINOL, 3,30-diiodo-2,20-di(methoxy-
methoxy)-1,10-binaphthyl (11) could be generated by introdu-
cing the acetal protecting group,12 followed by ortho-lithiation
and subsequent quenching with an iodine solution.13 In the case
of BINOL 13, the first step of the reaction sequence was a
bromination followed by the introduction of the MOM protect-
ing group to give 12 (Scheme 1).12,14 Then, a Kumada protocol
was applied in order to introduce the n-hexyl chains in 12,15
(4) (a) Hama, Y.; Nobuhara, Y.; Aso, Y.; Otsubo, T.; Ogura, F. Bull.
Chem. Soc. Jpn. 1988, 61, 1683–1686. (b) Dolci, L.; Dolle, F.; Valette, H.;
Vaufrey, F.; Fuseau, C.; Bottlaender, M.; Crouzel, C. Bioorg. Med. Chem.
1999, 7, 467–479.
(5) Daab, J. C.; Bracher, F. Monatsh. Chem. 2003, 134, 573–583.
(6) Bouillon, A.; Lancelot, J.-C.; Collot, V.; Bovy, P. R.; Rault, S.
Tetrahedron 2002, 58, 2885–2890.
(7) Magidson, O.; Menchikov, G. Chem. Ber. 1925, 58B, 113–118.
(8) Baxter, P. N. W. J. Org. Chem. 2000, 65, 1257–1272.
(9) Grave, C.; Lentz, D.; Schaefer, A.; Samori, P.; Rabe, J. P.; Franke, P.;
which was followed by ortho-lithiation and subsequent iodina-
tion to afford the desired product 13 (Scheme 1).13b Following
these sequences, we obtained 11 and 13 in both enantiomeric
forms each. These compounds proved to be configurationally
very stable because boiling in toluene for 1 week did not result in
racemization.
€
Schluter, A. D. J. Am. Chem. Soc. 2003, 125, 6907–6918.
€
(10) Kiehne, U.; Bunzen, J.; Staats, H.; Lutzen, A. Synthesis 2007, 1061–
1069.
(11) (a) Lutzen, A.; Hapke, M. Eur. J. Org. Chem. 2002, 2292–2297.
€
€
(b) Lutzen, A.; Hapke, M.; Staats, H.; Bunzen, J. Eur. J. Org. Chem. 2003,
3948–3957. (c) Hapke, M.; Staats, H.; Wallmann, I.; Lutzen, A. Synthesis
€
€
2007, 2711–2719. (d) Hapke, M.; Brandt, L.; Lutzen, A. Chem. Soc. Rev.
2008, 37, 2782–2797.
(12) (a) Ishitani, H.; Ueno, M.; Kobayashi, S. J. Am. Chem. Soc. 2000,
122, 8180–8186. (b) Yang, X.-W.; Sheng, J.-H.; Da, C.-S.; Wang, H.-S.;
Su, W.; Wang, R.; Chan, A. S. C. J. Org. Chem. 2000, 65, 295–296.
As mentioned before, the 2,20-bipyridine and BINOL
building blocks were finally connected in usually very good
yields by a 2-fold Sonogashira coupling procedure. Scheme 2
depicts these last reactions in the transformation sequences
of the ligands 2-6, which were prepared in both enantio-
meric forms each.
Metal Coordination Studies. After having finished the
syntheses, we started to investigate the complexation beha-
vior of each ligand. For our studies, we chose silver(I),
€
€
(13) (a) Bahr, A.; Droz, A. S.; Puntener, M.; Neidlein, U.; Anderson, S.;
Seiler, P.; Diederich, F. Helv. Chim. Acta 1998, 81, 1931–1963. (b) Yamashita,
Y.; Ishitani, H.; Shimizu, H.; Kobayashi, S. J. Am. Chem. Soc. 2002, 124,
3292–3302.
(14) Cui, Y.; Evans, O. R.; Ngo, H. L.; White, P. S.; Lin, W. Angew.
Chem., Int. Ed. 2002, 41, 1159–1162.
(15) Xiao, D.; Zhang, Z.; Zhang, X. Org. Lett. 1999, 1, 1679–1681.
J. Org. Chem. Vol. 74, No. 15, 2009 5229

Bunzen et al.
JOCArticle
copper(I), zinc(II), and iron(II) ions because these showed
remarkable results with our initial ligand 1: silver and
copper ions were demonstrated to lead to double-stranded
helicates whereas zinc and iron ions lead to triple-stranded
helicates in a completely diastereoselective manner. As in
our previous study, tetrafluoroborate salts of silver(I),
copper(I), iron(II), and zinc(II) were chosen because this
counterion has a weak coordination ability, thus making
sure that they do not compete with the bipyridine ligands in
the complex formation.16 When acetonitrile solutions of the
salts were mixed with dichloromethane solutions of the
ligands, silver(I) always gave pale yellow to clear solutions,
copper(I) was yellow to red brown but became green in a
period of 24 h in the case of ligands 2 and 4 to 6, and zinc(II)
was pale to strong yellow. Iron(II) gave different tinges of
red, depending on the ligand used and the concentration of
the complex in solution, thus indicating the expected for-
mation of iron(II) low-spin complexes.
ESI process than species that are present in solution since we
cannot observe these with any other spectroscopic tool (see,
e.g., NMR characterization below). Tandem MS experi-
ments of the metal complexes of ligand 6 corroborate this:
they revealed that the most stable dinuclear complexes are
formed with iron(II) ions. In fact, the binding of the metal
ions is so strong that the ion of the intact dinuclear complex
(m/z=850) loses the acetal groups and the alkyl chains by
breaking C-O and C-C bonds upon collision-induced
dissociation (CID) first instead of falling apart by losing
bipyridine ligands or mononuclear iron complexes (or frag-
ments thereof). The zinc complex (m/z=854) also loses acetal
groups upon CID; however, it also fragments into a doubly
charged species containing one zinc atom and either one or
two ligands at m/z=579 and m/z=1128 by losing a ligand
and a zinc ion, respectively. It is important to note, however,
that we never observed a fragment containing two zinc ions
and two ligand strands which would result from the loss of a
single ligand. This also means that the observation of an
[Zn2L2]4+ ion indicates that this dinuclear double-stranded
species is actually present in solution and not a fragment of a
triple-stranded one. Very similarly, the doubly charged
double-stranded silver(I) complex at m/z=1204 fragments
into a singly charged species containing one silver ion and
one ligand as can be observed from the isotopic pattern.
In the case of copper(I), the ESI mass spectra reveal
another interesting feature of ligands 2, 4, 5, and 6 that we
did not observe in the earlier studies with ligand 1: in the case
of ligand 6, e.g., there is a signal at m/z=1159 that results
This first encouraging hint at a successful complex forma-
tion was examined by ESI MS in order to check the correct
stoichiometry of the coordination compounds. Therefore,
mixtures of the ligands and the individual metal salts were
prepared that would correspond to 5ꢀ10-5 mol/L solutions
of the expected dinuclear double- or triple-stranded heli-
cates. These were inserted directly into the mass spectro-
meter. We discuss the mass spectra obtained for the iron(II),
zinc(II), and silver(I) complexes of ligand 6 here first, which
are also indicative for the behavior of ligands 2, 4, and 5 and
which are very similar to the ones we obtained earlier for
ligand 1. The mass spectra show strong signals for the
expected dinuclear triple-stranded aggregates in the case of
the iron(II) and zinc(II) complexes: the spectrum of the iron
complex shows the expected signal at m/z = 849 which
compares to the correct stoichiometry of a triple-stranded
complex [Fe2(6)3]4+ without any counterions. The sole other
signal at m/z=1170 is an adduct of two formate anions and
one sodium ion {Na[Fe2(6)3](HCO2)2}3+, which derives
from the internal standard used in our mass spectrometric
analyses. A very similar spectrum was obtained for the zinc
complex: Besides a signal at m/z=854 for the quadruply
charged triple-stranded dinuclear complex [Zn2(6)3]4+, two
smaller signals at m/z=1168 and m/z=1177 can be observed.
The former stands for the complex with one tetrafluorobo-
rate counterion {[Zn2(6)3]BF4}3+ and the latter gives the
mass for the above-mentioned sodium formate adduct {Na-
from the expected double-stranded helicate [Cu2(6)2]2+
;
however, an even stronger signal is found at m/z=854. This
unexpected new species can be assigned to a triple-stranded
complex of copper(II) [Cu2(6)3]4+. Obviously, making the
bipyridine more electron-rich by substitution with an aryl or
an alkyl group leads to an easier oxidation of the copper(I),
and the same seems to be true for introducing steric bulk in
the BINOL core which might cause a larger dihedral angle
between the two naphthyl groups of the BINOL that induces
some steric strain and hence destabilizes the double-stranded
aggregate but favors the triple-stranded arrangement and
thus facilitates the oxidation.
Interestingly, ligand 3 as the most electron-deficient ligand
of this series behaves quite differently: similar to the behavior
of ligand 1, the copper(I) complexes of 3 proved to be much
less prone to oxidation and thus only signals arising from the
expected dinuclear double-stranded complex [Cu2(3)2]
(BF4)2 but no signals of a triple-stranded dinuclear copper
(II) could be detected even after several days. A second
remarkable and unique feature of this ligand compared to
its derivatives is the complexation behavior toward zinc(II)
ions. Besides a signal that can be assigned to the expec-
ted triple-stranded coordination compound [Zn2(3)3]4+ at
m/z=668, another signal of very small intensity can be found
that arises from a dinuclear species that carries only two
ligand strands [Zn2(3)2]4+ at m/z=455. Also, two other
signals were observed that could be assigned to the 1- and
2-fold protonated ligand 3 at m/z=847 and 424, respectively.
This is surprising because zinc(II) complexes of this class of
complexes are usually very stable under the ESI MS condi-
tions in our hands and, hence, give very clean spectra that
only contain signals that arise from one dinuclear species.
This can have two reasons: either this dinuclear zinc(II)
[Zn2(6)3](HCO2)2}3+
.
The spectrum of the silver(I) complex of 6, however, does
contain some more signals. Besides the expected one at
m/z=1204 that results from the doubly charged double-
stranded dinuclear helicate, we also observed some signals
that can be assigned to a monomeric species [Ag(6)]+ and the
protonated ligand. This behavior, which is most pronounced
in this series in the case of the complexes of 6, is rather typical
for silver(I) helicates in our hands,3,17,18 and these ions are
rather fragments of the helicate that are formed during the
(16) Please note that our equipment does not allow us to measure
elemental analyses of fluorine-containing samples. Thus, we could not
measure these for our metal complexes.
€
(17) (a) Kiehne, U.; Weilandt, T.; Lutzen, A. Org. Lett. 2007, 9, 1283–
1286. (b) Kiehne, U.; Lutzen, A. Org. Lett. 2007, 9, 5333–5336. (c) Kiehne,
€
€
U.; Weilandt, T.; Lutzen, A. Eur. J. Org. Chem. 2008, 2056–2064.
€
(18) Bunzen, J.; Bruhn, T.; Bringmann, G.; Lutzen, A. J. Am. Chem. Soc.
2009, 131, 3621–3630.
5230 J. Org. Chem. Vol. 74, No. 15, 2009

Bunzen et al.
JOCArticle
FIGURE 1. PM3-TM (Spartan)-minimized structures of the three possible diastereomers of (P)-ligand 4 with copper(I) (top row) and zinc(II)
(bottom row) (methoxymethoxy groups omitted and heptyl chains reduced to methyl groups for viewing clarity, chirality descriptors for the
metal complexes follow the oriented skew line system).
coordination compound is not as stable as the others due to
the electron-withdrawing substituents on the bipyridines,
which make it more prone to fragmentation, or the self-
assembly process of the coordination compound formed with
zinc(II) ions and ligand 3 is not selective with regard to the
stoichiometry. This could be caused by the steric effects since
the steric crowding is definitely more pronounced in a tris
(bipyridyl) complex where the ester groups are in very close
proximity than in a bis(bipyridyl) complex. Thus, the double-
stranded helicate might become similarly favored.17a,19 To
distinguish between these two possibilities, we performed a
tandem MS study. Since we again did not find the [Zn2(3)2]4+
ion as a fragment of [Zn2(3)3]4+ in an MS/MS experiment, we
believe that the latter explanation is true and the self-assembly
is not selective with regard to the stoichiometry of the
assembly. Only the spectra of the silver(I) and iron(II) com-
plexes, however, do show similar ions like the ones obtained
with ligands 2, 4, 5, and 6 with the expected stoichiometry
corresponding to the double- and triple-stranded species
[Ag2(3)2]2+ and [Fe2(3)3]4+, respectively. Again, tandem ms
experimentswiththesilver(I) and the iron(II)complexshowed
the same fragmentation as the complexes of ligand 6.
proven by ligand-exchange experiments with isotopically
labeled ligands in the case of 1,3 we could still measure
NMR spectra of the copper(I) complexes of 1 and 3. How-
ever, only these could be studied in detail by NMR since the
oxidation to paramagnetic copper(II) in all other systems
severely hampered long-term experiments. The spectra show
sharp signals directly after mixing of the copper(I) salt and
the ligands and prove instant formation of the dinuclear
complexes. However, only a few hours later they undergo
oxidation, finally resulting in a total breakdown of the
signals due to the presence of the paramagnetic metal centers
and thus preventing experiments that require longer times.
Hence, we rather discuss the results of our NMR studies with
the silver(I) and zinc(II) complex solutions here first. Both silver(I)
and zinc(II) complex solutions usually give rise to sharp signals.
Figure 2 depicts the spectra obtained from the complexes of
ligand 4. The shifts of the signals, especially the ones of the
bipyridine moiety, are easy to recognize and exceptionally
strong. This is another indication for the successful formation
of defined metal coordination compounds with the particular
metal ion, thus corroborating the results obtained by mass
spectrometry. In principle, three different diastereomers of a
dinuclear helicate can be formed from an enantiomerically pure
dissymmetric bis(bipyridine) ligand as shown in Figure 1. In
almost all cases, we observed only a single set of signals indicating
that the self-assembly process is completely diastereoselective.20
The observation of specific and (in most cases) expected
stoichiometry of the complexes is a good proof for the success-
ful self-assembly behavior of our ligands. But mass spectro-
metry is of course very limited when it comes to specifying the
stereochemical arrangement of the complexes (Figure 1).
Therefore, we turned to NMR studies next. Despite the
fact that the kinetic lability of copper(I) helicates could be
(20) We also performed variable-temperature NMR studies and found
that the spectra did not change when we lowered the temperature (until the
complexes precipitated from solution). Due to the use of dichloromethane,
however, we could not perform any high-temperature experiments to observe
thermal dissociation of our helicates.
(19) Sigel, H.; Martin, R. B. Chem. Soc. Rev. 1994, 23, 83–91.
J. Org. Chem. Vol. 74, No. 15, 2009 5231

Bunzen et al.
JOCArticle
FIGURE 2. 1H NMR spectra of ligand 4 (500.1MHz, 4.5 mM; CD2-
Cl2/CD3CN 4:1 at 300 K): (a) free ligand 4, (b) 1:1 mixture of 4 and
[Ag(CH3CN)4]BF4, (c) 3:2 mixture of 4 and Zn(BF4)2 (note that there
are traces of free ligand visible in the spectrum).
FIGURE 3. 1H NMR spectra of ligands 3and 6(500.1 MHz, 1.4 mM;
CD2Cl2/CD3CN 3:1 at 298 K): (a) ligand 3, (b) 3:2 mixture of ligand 3
and Zn(BF4)2, (c) ligand 6, (d) 3:2 mixture of ligand 6 and Zn(BF4)2.
However, there are three exceptions to this rule: the first one is
the zinc(II) complex of ligand 3, which already gave unexpected
mass spectra that indicated the formation of more than one type
of aggregate upon coordination to zinc(II) ions.
paramagnetic contamination by iron(III) ions. Again, we
performed dilution studies down to concentrations close to
the ones used for the ESI MS experiments where we only
found signals arising from the discrete dinuclear complexes
but the spectra did not change, thus also ruling out the
formation of coordination polymers. As observed in pre-
vious cases, this is rather due to the dynamic nature of
these aggregates because of intramolecular rearrangements
(ligand exchange can also not be ruled out). Since the iron
complexes tend to precipitate from these solutions at tem-
peratures around 0 °C, we were not able to slow these
processes enough to obtain sharp signals by temperature-
dependent experiments. However, as demonstrated for a
similar ligand,18 these dynamic processes could be slowed
considerably by changing the solvent to a mixture of deut-
erated dichloromethane and DMSO. Therefore, we also
examined the iron complexes in this solvent. After proving
the formation of dinuclear triple-stranded coordination
compounds by ESI MS, we recorded NMR spectra of these
helicates. Actually, the ligand exchange in these solutions
was found to be so slow that we could observe the initial
formation of a number of coordination compounds directly
after mixing as they slowly equilibrate to finally give the
most stable D-symmetric diastereomer (almost) exclusively
(Figure 4).
However, the spectra of the zinc complexes of 5 and 6
also reveal the presence of at least one more species besides
the expected D-symmetric main component (Figure 3). In
the cases of 5 and 6, however, we could only observe
dinuclear triple-stranded zinc coordination compounds
by mass spectrometry. To rule out the formation of poly-
meric species, we made dilution experiments in order to
make sure that we are below the concentration that would
favor the formation of polymeric rather than discrete
species, but the ratio of the main species relative to the
secondary one remained unchanged. Thus, we conclude
that these ligands do undergo selective self-assembly in
terms of the stoichiometric ratio of metal ions and ligands,
but they are not selective with regard to the stereochemical
outcome of the assembly.
Nevertheless, simple 1H NMR experiments do already
allow us to make a first assignment of the stereochemical
arrangement of our helicates: the dominant (and in most
cases the exclusively formed) dinuclear coordination com-
pounds in solution cannot be (Δ,Λ)-configured because this
would result in much more complex spectra due to the lower
symmetry (Cn vs Dn) compared to the dissymmetric diaster-
eomers whose metal centers have the same configuration
((Δ,Δ) or (Λ,Λ)) and hence consist of two homomorphous
portions as the ligand does.
Interestingly, ligand 3 behaves different again: whereas all
iron complexes of the other ligands gave red solutions as
expected for iron(II) low-spin tris(bipyridine) complexes, the
dinuclear complexes of 3 were found to give green solutions
(in CD2Cl2/CD3CN 3:1 the complex solution is red). This
indicates that the iron should not be a low-spin iron(II) but
Unfortunately, the spectra of the iron solutions in mix-
tures of deuterated dichloromethane and acetonitrile could
not be interpreted because of an exceptional strong broad-
ening of the peaks, despite ruling out that this is a result of
5232 J. Org. Chem. Vol. 74, No. 15, 2009

Bunzen et al.
JOCArticle
6: D = 2.29 ꢀ 10-10 m2/s) except for the fact that these
experiments again confirm that the triple-stranded helicates
of ligand 6 are larger than the ones of ligand 3 with almost
the same ratio of the D values of D(6)/D(3) ≈ 0.6.
To finally discern between the two Dn-symmetric possibi-
lities, we measured ROESY-NMR spectra, but the distances
between the BINOL and the bipyridine moieties are too large
to get ROE-contacts strong enough to allow an unambig-
uous assignment of one or the other diastereomer. There-
fore, we applied chiral dichroism spectroscopy (CD spectro-
scopy). This powerful chirooptical method is the best known
approach to investigate chiral metal complexes in solution.21
In particular, chiral 2,20-bipyridine complexes were exam-
ined thoroughly in recent years.22
Since our ligands proved to be configurationally stable
when we heated them for longer periods in toluene, the CD
spectra prove that the optical antipodes of ligands 2-6 give
rise to enantiomeric dinuclear helicates. We discuss the
results obtained with ligand 4 exemplarily here. Free ligand
4 and its silver and copper helicate solutions give very similar
spectra. This behavior can be understood if one takes into
account the exiton theory: opposite cotton effects cancel
each other to zero, when the angle between the dipoles of the
ligands is 90° as it is (nearly) found in a tetrahedral coordina-
tion of a metal ion by two 2,20-bipyridine ligands that adopt
an (almost) perpendicular orientation in a double-stranded
helicate. Thus, CD-spectroscopy cannot be used alone to
discriminate between the two possible isomers of our silver
and copper complexes. However, as we found out earlier
with ligand 1 and other dissymmetric bis(bipyridine) BI-
NOL-based ligands,3,18 the stereochemical information of
the ligands usually induces the same configuration of the
metal centers regardless if they are double- or triple-
stranded. Thus, the stereochemical assignment of the tri-
ple-stranded helicates can usually be transferred to the
double-stranded analogues (as long as the self-assembly
process is highly stereoselective, of course). Hence, we turned
to the CD spectroscopic analysis of the triple-stranded
aggregates. As expected, the spectra of the iron and zinc
helicates show great differences compared to the one of free
ligand 4, thus corroborating the before mentioned theore-
tical considerations.21 Compared with CD data from the
literature there is clear evidence that the iron complexes
with (M)-configured ligands give (Λ,Λ)-configured heli-
cates and a (P)-configuration of the ligands results in a
(Δ,Δ)-configuration of the metal centers, respectively. Of
course, the spectroscopic behavior of all of these com-
plexes is in accordance with the exiton theory: the negative
exiton couplet in the region from 300 to 425 nm of the (P)-
ligands gives an unambiguous proof for a (Δ,Δ)-config-
uration at the stereogenic metal centers.23 Therefore, we
can assign a (Δ,Δ)-configuration for iron(II) complexes
FIGURE 4. 1H NMR spectra of iron(II) complexes of ligand 6
(500.1 MHz, 1.4 mM; CD2Cl2/DMSO-d6 3:1 at 298 K): (a) ligand 6
(400 MHz), (b)3:2 mixtureof 6and Fe(BF4)2 6H2O 1 h after mixing,
3
(c) the same mixture after 24 h, (d) the same mixture after 3 d.
has a different electron configuration resulting in a para-
magnetic species. This assumption is also corroborated by
the NMR spectra of this complex because the signals do
show significant broadening compared to the iron complexes
of the other ligands. Also, an ESI MS of this green
iron(II) complex showed only one signal at m/z = 663 resul-
ting from the expected dinuclear triple-stranded species
[Fe2(3)3]4+, that readily fragments at 20 eV, which reveals
a lower stability of the complex in comparison to the red
iron(II) complexes of, e.g., ligand 6 as was expected for
changing the electron configuration from diamagnetic low-
spin to some paramagnetic species.
Finally, we also performed DOSY NMR experiments to
study the aggregates’ size and composition. For the silver(I),
zinc(II), and iron(II) complexes of ligands 3 and 6, DOSY
experiments showed just one species in solution (the minor
species of the zinc(II) complexes were undetectable) with the
diffusion constants growing from the triple-stranded zinc
complexes (3: D=7.34ꢀ10-10 m2/s; 6: D=4.55ꢀ10-10 m2/s)
to the double-stranded silver complexes (3: D=7.65 ꢀ
10-10 m2/s; 6: D = 5.33 ꢀ 10-10 m2/s) and the free ligand
(6: D=7.03ꢀ 10-10 m2/s). Although the lack of any reliable
data for the viscosity of the solvent mixture used for these
experiments did not allow the calculation of any hydrodynamic
radii of the aggregates, the relative differences already corrobo-
rate the conclusions drawn from the ESI MS and NMR experi-
ments, since triple-stranded complexes contain the longest axis
and should therefore have the smallest diffusion constant. Also,
the constants are larger for the corresponding complexes of
ligands 3 compared to the ones of 6, which was expected, taking
into account that the complexes of 3 with the ester groups are
considerably smaller than those of 6 carrying the long alkyl
chains. Unfortunately, the iron(II) complexes were measured in
another solvent mixture and could therefore not really be
compared with the other complexes (3: D=4.04ꢀ10-10 m2/s;
(21) (a) von Zelewsky, A. Stereochemistry of Coordination Compounds;
John Wiley & Sons: Chichester, 1996. (b) Ziegler, M.; von Zelewsky, A.
Coord. Chem. Rev. 1998, 177, 257–300.
(22) (a) Quinodoz, B.; Labat, G.; Stoeckli-Evans, H.; von Zelewsky, A.
Inorg. Chem. 2004, 43, 7994–8004. (b) Murner, H.; von Zelewsky, A.;
Hopfgartner, G. Inorg. Chim. Acta 1998, 271, 36–39. (c) Prabaharan, R.;
Fletcher, N. C.; Nieuwenhuyzen, M. J. Chem. Soc., Dalton Trans. 2002, 602–
608. (d) Prabaharan, R.; Fletcher, N. C. Inorg. Chim. Acta 2003, 355, 449–
453.
(23) Berova, N., Nakanishi, K., Woody, R. W., Eds. Circular Dichroism:
Principles and Applications, 2nd ed.; Wiley-VCH: Weinheim, 2000.
J. Org. Chem. Vol. 74, No. 15, 2009 5233

Bunzen et al.
JOCArticle
based on (P)-configured ligands and a (Λ,Λ)-configura-
tion based on (M)-configured ligands. The sole exception
to this rule is observed for the green iron(II) complex in
CD2Cl2/DMSO of ligand 3. The CD spectrum of this
complex shows no complex specific cotton effects revealing
just the spectrum of the free ligand. In the case of octahe-
dral coordination of metal centers this behavior would be
possible in three scenarios where two opposite complex
specific cotton effects would compensate each other: either
a Δ,Λ-species is formed selectively, which would be quite
unexpected given the results of all other ligands, or a 1:1
mixture of the Δ,Δ- and Λ,Λ-species would be formed by
accident, or probably the most likely scenario a (statistical)
mixture of all three possible diastereomeric complexes is
generated. Unfortunately, the broadening of the NMR
signals of this complex make these spectra uninterpretable
and therefore no further information about the selectivity
of the self-assembly process could be obtained in this case.
The spectra of the zinc complexes are very similar to the
ones of the red low-spin iron(II) helicates. Every ligand
exerts strong cotton effects-even the solutions of ligands 3,
5, and 6 that do not self-assemble in a completely selective
manner although they each form one clearly dominant
species that should rule the spectrum compared to the minor
(in most cases diastereomeric) byproducts. Even though
there is not as much comparable circular dichroism data
for zinc complexes to be found in the literature than for the
respective iron complexes,22b,22c the cotton effects never-
theless clearly hint at a (Λ,Λ) configuration of the stereo-
genic metal centers in the helicates based on ligands with an
(M)-configured ligand and vice versa. This is of course again
in accordance with the exiton theory.
In most cases, these processes happen in a completely or at
least highly diastereoselective manner: the triple-stranded
helicates formed by iron(II) and zinc(II) ions and (P)-ligands
were found to have a (Δ,Δ)-configuration of the metal
centers, and the (M)-ligands were found to induce a (Λ,Λ)-
configuration. Unexpectedly, however, the introduction of
n-hexyl chains at the remote 6- and 60-positions on the
BINOL core reduces the diastereoselectivity of the self-
assembly processes leading to triple-stranded zinc(II) heli-
cates. Other astonishing effects were observed when elec-
tron-withdrawing groups like acyl groups were introduced in
the 50-position of the 2,20-bipyridine units like in ligand 3: in
this case, the self-assembly process of dinuclear zinc(II)
complexes was found to be unselective with regard to the
stoichiometry of the resulting helicate since both double- and
triple-stranded species were found. Furthermore, 3 as
the most electron-deficient ligand of this series obviously
induces a different electron configuration of iron(II) centers
in the helicates in a mixture of dichloromethane/DMSO
because we observed the formation of a paramagnetic, green
triple-stranded dinuclear complex (most probably as a mix-
ture of all possible diastereomers) instead of the expected
diamagnetic, red low-spin iron(II) complexes which are
formed with all other ligands. In contrast, electron-donating
alkyl or phenyl groups in the 50-position of the 2,20-bipyr-
idine groups, however, do not influence the stereoselectivity
of these processes but were found to have a remarkable effect
on the redox chemistry of the dinuclear coordination com-
pounds since dinuclear double-stranded copper(I) helicates
were found to undergo rapid oxidation (and rearrangement)
to dinuclear triple-stranded paramagnetic copper(II) heli-
cates. Interestingly, this behavior was not observed for lead
structure 1 or the acylated ligand 3. These results clearly
demonstrate how sensitive these processes are to subtle
changes in the ligand structure. Even the introduction of
substituents in remote positions of the ligand structures was
proven to have a marked and not always obvious influence
on the selectivity of the self-assembly of triple-stranded
helicates. This is by far less pronounced in case of the
formation of double-stranded helicates. Another interesting
feature is the surprisingly strong effect of substituents at the
bipyridines on the redox behavior of copper helicates and
the electron configuration of iron helicates which will be
explored in future studies.
Furthermore, the bands point into the same direction as in
the case of the respective iron helicates, indicating the same
absolute configuration at the metal centers. This result is also
corroborated by our previous work where we assigned a
(Δ,Δ) configuration to helicates derived from (P)-ligand 1
based on ROESY NMR experiments and an X-ray diffrac-
tion analysis of the molecular structure of the (Δ,Δ)-
{Zn2[(P)-1]3}(BF4)4 helicate.3
Conclusion
We prepared a series of bis(bipyridine)BINOL ligands in a
concise manner using a convergent building block approach.
These ligands are “programmed” to form dinuclear helicates
upon coordination to suitable late-transition-metal ions and
were designed to enhance the solubility of the ligands and the
metallosupramolecular aggregates compared to the one of
previously reported lead structure 1. This objective was fully
achieved; the solubility of the ligands and the resulting
helicates could be enhanced by the introduction of substit-
uents having a maximum with ligand 6. However, the
different solubilizing substituents used for these purposes
turned out to have some surprising effects on the self-
assembly of these metallosupramolecular entities which
was investigated in solution using NMR and CD spectros-
copy as well as ESI mass spectrometry.24
Experimental Section
(M)- or (P)-6,60-Di(n-hexyl)-2,20-di(methoxymethoxy)-1,10-
binaphthyl (12). An 82 mg (3.38 mmol; 7.2 equiv) portion of
magnesium powder (activated with 1,2-dibromoethane) was
suspended in 20 mL of dry diethyl ether. To this mixture was
added 416 μL (598 mg, 2.82 mmol; 6 equiv) of n-hexyl iodide
dropwise at rt. The solution was refluxed for 2 h. After cooling,
the solution was added to a solution of 250 mg (0.47 mmol;
1 equiv) of (M)- or (P)-6,60-dibromo-2,20-di(methoxymethoxy)-
1,10-binaphthyl and 25 mg (4.7ꢀ10-5 mol; 10 mol %) of (dppp)-
NiCl2 in 30 mL of dry diethyl ether at 0 °C. The reaction mixture
was refluxed for 18 h. After that time, the reaction was stopped
by quenching with diluted hydrochloric acid, followed by
extraction with dichloromethane. The organic layer was dried
with sodium sulfate. After removal of the solvent, the crude
product was purified by flash column chromatography on silica
gel (hexane/ethyl acetate+triethyl amine 10:1+5%): Rf 0.47;
yield 226 mg (89%); pale yellow oil; 1H NMR (400 MHz,
(24) Please note that other techniques like vapor pressure osmometry
(vpo) or dynamic light scattering (dls) are not really applicable to establish
the composition of this kind of ionic aggregates.
5234 J. Org. Chem. Vol. 74, No. 15, 2009

Bunzen et al.
JOCArticle
3
CDCl3) δ 0.92 (t, 6H, CH3CH2-, J=7.0 Hz), 1.29-1.44 (m,
12H, alkyl chain), 1.65-1.75 (m, 4H, alkyl chain), 2.74 (t, 4H,
-CH2-BINOL, 3J=7.8 Hz), 3.17 (s, 6H, CH3O-), 4.98, (d, 2H,
-OCH2O-, 2J=6.7 Hz), 5.08 (d, 2H, -OCH2O-, 2J=6.7 Hz),
7.09-7.15 (m, 4H, BINOL), 7.56 (d, 2H, BINOL, 3J=9.1 Hz),
in vacuo to afford the crude product. This was purified by
flash column chromatography on silica gel (eluent: hexane/ethyl
acetate/triethyl amine 3:1:1): Rf 0.38; yield 153 mg (91%); white
solid; mp=184-188 °C; 1H NMR (500 MHz, CDCl3) δ 2.60 (s,
6H, -OCH3), 4.98 (d, 2H, -OCH2O-, 2J=6.2 Hz), 5.19 (d, 2H,
-OCH2O-, 2J=6.2 Hz), 7.27 (d, 2H, BINOL, 3J=8.4 Hz), 7.34
(ddd, 2H, BINOL, 3J=8.4 Hz, 3J=7.0 Hz, 4J=0.9 Hz), 7.40-
7.52 (m, 8H, BINOL, phenyl), 7.66 (m, 4H, phenyl), 7.89 (d, 2H,
H-6, 3J=8.2 Hz), 7.98(dd, 2H, bipy, 3J=8.2 Hz, 4J=2.0 Hz), 8.03
(dd, 2H, bipy, 3J=8.3 Hz, 4J=2.1 Hz), 8.29 (s, 2H, BINOL), 8.50
(m, 4H, bipy), 8.87 (s, 2H, bipy), 8.93 (s, 2H, bipy) ppm; 13C
NMR (125.8 MHz, CDCl3) δ 56.2 (-OCH3), 90.7 (-CC-), 90.8
(-CC-), 99.1 (-OCH2O-), 116.8 (BINOL), 120.2 (bipy), 120.5
(bipy), 121.4 (bipy), 125.7 (BINOL), 125.9 (BINOL), 126.6
(BINOL), 127.1 (phenyl), 127.6 (BINOL), 127.7 (BINOL),
128.3 (phenyl), 129.1 (phenyl), 130.3 (BINOL), 134.0 (BINOL),
134.6 (BINOL), 135.3 (bipy), 136.8 (bipy), 137.4 (phenyl), 139.3
(pipy), 147.6 (bipy), 151.6 (bipy), 153.1 (BINOL), 154.1 (bipy),
154.6 (bipy) ppm; MS (ESI, pos. mode) m/z 883.3 (100) [C60H42-
N4O4+H]+; RP (M) [R]D20 =-345.6 (c=0.725, CH3Cl), (P)
[R]2D0=+352.3 (c=0.66, CH3Cl); CD (λ (Δε)) (M)=235 (-5.8),
271 (5.0), 322 (3.6), 362 (-10.3); (P)=236 (6.4), 271 (-5.2), 323
7.67 (s, 2H, BINOL), 7.89 (d, 2H, BINOL, 3J=9.1 Hz) ppm; 13
C
NMR (100.6 MHz, CDCl3) δ 14.2 (CH3CH2-), 22.7 (alkyl
chain), 29.2 (alkyl chain), 31.4 (alkyl chain), 31.9 (alkyl chain),
36.0 (-CH2-BINOL), 55.9 (CH3O-), 95.6 (-OCH2O-), 117.6
(BINOL), 121.7 (BINOL), 125.7 (BINOL), 126.3 (BINOL),
128.0 (BINOL), 128.9 (BINOL), 130.2 (BINOL), 132.6 (BI-
NOL), 138.7 (BINOL), 152.2 (BINOL) ppm; MS (EI) m/z 542.4
(100) [C36H46O4]+; HRMS (EI) calcd for [C36H46O4]+
542.3396, found 542.3390; RP (M) [R]2D0 =+25.4 (c=0.5075,
CH2Cl2), (P) [R]2D3.5=-24.9 (c = 0.495, CH2Cl2). Anal. Calcd
for C36H46O4 1/5H2O: C, 79.14; H, 8.56. Found: C, 79.16;
3
H, 8.30.
(M)- and (P)-6,60-Di(n-hexyl)-3,30-diiodo-2,20-di(methoxy-
methoxy)-1,10-binaphthyl (13). A 233 mg (0.43 mmol, 1 equiv)
portion of (M)- or (P)-12 was dissolved in 30 mL of dry THF,
and 1.465 mL (1.76 mmol; 1.2 M in cyclohexane, 4.1 equiv) of
s-butyllithium solution was added dropwise at -78 °C. After
the mixture was stirred for 1.5 h at this temperature, 652 mg
(2.57 mmol, 6 equiv) of iodine in 5 mL of dry THF was added.
The resulting mixture was stirred for 2 h at -78 °C. After that
time, the reaction mixture was quenched with methanol and
water and the solution was allowed to warm to rt. After
extraction with dichloromethane, the organic layer was dried
with sodium sulfate and concentrated in vacuo. The crude
product was purified by flash column chromatography on silica
gel (hexane/ethyl acetate + triethylamine 10:1 + 5%): Rf 0.7;
yield 229 mg (67%); yellow oil; 1H NMR (400 MHz, CDCl3) δ
0.90 (t, 6H, CH3CH2-, 3J=7.1 Hz), 1.25-1.39 (m, 12H, alkyl
chain), 1.61-1.71 (m, 4H, alkyl chain), 2.61 (s, 6H, CH3O-),
(-3.4), 363 (12.0). Anal. Calcd for C60H42N4O4 CH3COOC2H5:
3
C, 79.16; H, 5.19; N, 5.77. Found: C, 79.56; H, 5.34; N, 5.92.
Procedure for the Generation of Helicates Exemplified for the
Synthesis of {Ag2[(P)-2]2}(BF4)2. A 6.25 mg (7.09 μmol) por-
tion of (P)-2 was dissolved in 0.6 mL of CD2Cl2, and 2.594 mg
(7.09 μmol) of [Ag(CH3CN)4]BF4 was dissolved in 0.2 mL of
CD3CN. The metal salt solution was added to the solution of the
ligand resulting in a pale yellow solution of the helicate. This
solution was studied by NMR. Likewise, a solution was gener-
ated with a concentration of 5ꢀ10-5 mol/L in nondeuterated
dichloromethane/acetonitrile 1:1. This solution was directly
used for the MS analysis. The same solution was also used to
measure the CD spectra.
3
2.72 (t, 4H, -CH2-BINOL, J=7.9 Hz), 4.67 (d, 2H, -OC-
{Ag2[(M)-2]2}(BF4)2/{Ag2[(P)-2]2}(BF4)2: 1H NMR (500 MHz,
CD2Cl2/CD3CN; 4:1) δ 2.45 (s, 12H, -OCH3), 4.86 (d, 4H, -OC-
H2O-, 2J=5.8 Hz), 4.95 (d, 4H, -OCH2O-, 2J=5.8 Hz), 7.11 (d,
4H, BINOL, 3J=8.5 Hz), 7.28 (ddd, 4H, BINOL, 3J=8.5 Hz, 3J=
7.4 Hz, 4J=0.7 Hz), 7.40-7.49 (m, 16H, BINOL, phenyl), 7.64 (m,
8H, phenyl), 7.86 (d, 4H, BINOL, 3J=8.2 Hz), 8.18 (dd, 4H, bipy,
3J=8.4 Hz, 4J=1.8 Hz), 8.24 (s, 4H, BINOL), 8.27 (dd, 4H, bipy,
3J=8.4 Hz, 4J=2.1 Hz), 8.35 (d, 4H, bipy, 3J=8.4 Hz), 8.38 (d, 4H,
bipy, 3J=8.4 Hz), 8.81 (d, 4H, bipy, 4J=1.8 Hz), 8.91 (d, 4H, bipy,
4J=2.1 Hz) ppm; 13C NMR (125.8 MHz, CD2Cl2/CD3CN; 4:1) δ
55.8 (-OCH3), 89.1 (ethynyl), 92.4 (ethynyl), 98.9 (-OCH2O-),
116.0 (BINOL), 121.9 (bipy), 122.1 (bipy), 122.8 (bipy), 125.7 (bipy),
125.9 (bipy), 126.1 (BINOL), 127.0 (phenyl), 127.8 (BINOL), 127.9
(BINOL), 129.1 (phenyl), 129.3 (phenyl), 130.2 (BINOL), 134.1
(BINOL), 134.9 (BINOL), 135.8 (phenyl), 137.0 (bipy), 138.4 (bipy),
140.9 (bipy), 149.2 (bipy), 150.1 (bipy), 150.8 (bipy), 152.6 (bipy),
H2O-, 2J=5.7 Hz), 4.79 (d, 2H, -OCH2O-, 2J=5.7 Hz), 7.09
(d, 2H, BINOL, 3J = 8.7 Hz), 7.15 (dd, 2H, BINOL, 3J =
4
4
8.7 Hz, J=1.9 Hz), 7.52 (d, 2H, BINOL, J=1.9 Hz), 8.46
(s, 2H, BINOL) ppm; 13C NMR (100.6 MHz, CDCl3) δ 14.2
(CH3CH2-), 22.7 (alkyl chain), 29.1 (alkyl chain), 31.2 (alkyl
chain), 31.8 (alkyl chain), 35.9 (alkyl chain), 56.6 (CH3O-), 92.4
(BINOL), 99.5 (-OCH2O-), 125.1 (BINOL), 126.3 (BINOL),
126.5 (BINOL), 128.9 (BINOL), 132.4 (BINOL), 132.6 (BI-
NOL), 139.5 (BINOL), 140.7 (BINOL), 151.5 (BINOL) ppm;
MS (EI) m/z 794.2 (30) [C36H44I2O4]+, 718.1 (100) [C36H44I2O4
- C3H8O2]+, 591.2 (90) [C36H44I2O4 - C3H8IO2]+, 465.3 (20)
[C36H44I2O4 - C3H8I2O2]+; HRMS (EI) calcd for [C36H44-
I2O4]+ 794.1329, found 794.1319; RP (M) [R]2D2.7=-12.1 (c=
1.4725, CH2Cl2), (P): [R]2D5.1=+12.2 (c=0.625, CH2Cl2). Anal.
Calcd for C36H44I2O4 1/2C6H14: C, 55.92; H, 6.14. Found: C,
3
56.25; H, 5.88.
152.7 (BINOL) ppm; MS (ESI, pos. mode) m/z 991.3 ([Ag2(2)2]2+
,
General Procedure 1 for the Synthesis of Ligands 2-6 Exemp-
lified for the Synthesis of (M)- or (P)-2,20-Di(methoxymethoxy)-
3,30-di(5-ethynediyl-50-phenyl-2,20-bipyridyl)-1,10-binaphthyl (2).
A 116 mg (0.19 mmol, 1 equiv) portion of (M)- or (P)-2,20-di
(methoxymethoxy)-3,30-diiodo-1,10-binaphthyl (11), 100 mg
(0.39 mmol, 2.1 equiv) of 5-ethynyl-50-phenyl-2,20-bipyridine
[2+Ag]+, 100), 883.4 ([2+H]+, 15); CD (λ (Δε)) (M)=235 (-11.2),
270 (8.9), 324 (6.8), 364 (-20.0); (P)=235 (12.8), 271 (-10.0), 326
(-6.7), 362 (23.5).
{Zn2[(M)-2]3}(BF4)4/{Zn2[(P)-2]3}(BF4)4: 1H NMR (500 MHz,
CD2Cl2/CD3CN; 4:1) δ 2.26 (s, 18H, -OCH3), 4.71 (m, 12H,
3
(7), 8 mg (7.6 μmol, 4 mol %) of [Pd2dba3 CHCl3], 8.5 mg
-OCH2O-), 7.01 (d, 6H, BINOL, J=8.4 Hz), 7.31 (ddd, 6H,
3
(15.2 μmol, 8 mol %) of 1,10-diphenylphosphinoferrocene (dppf),
and 6 mg (30.4 μmol, 16mol %) ofCuI was thoroughly evacuated
and flushed with argon. Eight milliliters of dry triethylamine and
3 mL of dry THF were added, and the reaction mixture was
heated to 45 °C and stirred at this temperature for 70 h. After the
mixture was cooled to room temperature, saturated ethylene
diaminetetraacetic acid (EDTA) and satd sodium carbonate
solutions were added, and the mixture was stirred for 15 min.
Then the aqueous solution was extracted with dichloromethane.
The organic layer was dried with sodium sulfate and concentrated
BINOL, 3J=7.7 Hz, 3J=8.4 Hz, 4J=0.9 Hz), 7.34-7.40 (m, 30H,
3
3
4
phenyl), 7.49 (ddd, 6H, BINOL, J=8.2 Hz, J=7.7 Hz, J=
0.7 Hz), 7.88 (d, 6H, BINOL, 3J=8.2 Hz), 7.98 (d, 6H, bipy, 4J=
1.7Hz), 8.07(d, 6H, bipy,4J=2.1 Hz), 8.18 (m, 12H, BINOL, bipy),
3
4
8.46 (dd, 6H, bipy, J=8.6 Hz, J=2.1 Hz), 8.49 (d, 6H, bipy,
3J=8.4 Hz), 8.57 (d, 6H, bipy, 3J=8.6 Hz) ppm; 13C NMR
(125.8 MHz, CD2Cl2/CD3CN; 4:1) δ55.5 (-OCH3), 88.4 (ethynyl),
94.8 (ethynyl), 98.8 (-OCH2O-), 115.3 (BINOL), 123.6 (bipy),
123.9 (bipy), 124.6 (bipy), 126.1 (BINOL), 126.4 (BINOL), 127.2
(phenyl), 128.2 (BINOL), 128.5 (BINOL), 129.7 (phenyl), 130.1
J. Org. Chem. Vol. 74, No. 15, 2009 5235

Bunzen et al.
JOCArticle
(phenyl), 130.4 (BINOL), 134.6 (phenyl), 134.7 (BINOL), 136.2
(BINOL), 140.2 (bipy), 140.6 (bipy), 143.6 (bipy), 146.0 (bipy),
147.2 (bipy), 147.6 (bipy), 150.0 (bipy), 152.4 (BINOL) ppm; MS
(ESI, pos. mode) m/z 694.8 ([Zn2(2)3]4+, 100); CD (λ (Δε)) (M)=
233 (-23.9), 260 (8.3), 293 (-7.8), 354 (-12.3), 390 (38.4); (P)=233
(26.8), 260 (-9.4), 294 (8.4), 351 (13.0), 391 (-39.7).
Acknowledgment. Financial support from the Deutsche
Forschungsgemeinschaft (SFB 624) is gratefully acknowl-
edged. We thank Mrs. K. Peters-Pflaumbaum for measuring
a number of ESI mass spectra.
Supporting Information Available: Experimental details for
ligands 3-6 and their metal complexes, schematic representations
of the synthesis of the pyridines and 2,20-bipyridines, as well as
spectroscopic data (NMR, CD, and ESI-MS) of ligands 2-6 and
their metal complexes not depicted in Figures 2-4. Coordinates
for the metal complex structures shown in Figure 1 that were
obtained from molecular modeling studies. This material is
available free of charge via the Internet at http://pubs.acs.org.
{Cu2[(M)-2]2}(BF4)2/{Cu2[(P)-2]2}(BF4)2//{Cu2[(M)-2]3}(BF4)4/
{Cu2[(P)-2]3}(BF4)4: MS (ESI, pos. mode) m/z 946.2 ([Cu2(2)2]2+
,
[2+Cu)]+ (minimal amount), 100), 694.0 ([Cu2(2)3]4+, 90).
{Fe2[(M)-2]3}(BF4)4/{Fe2[(P)-2]3}(BF4)4: MS (ESI, pos. mode)
m/z 690.0 ([Fe2(2)3]4+, 100); CD (λ(Δε)) (M)=239 (-27.2), 265
(9.1), 314 (-21.5), 399 (38.9), 497 (3.3), 567 (-5.1); (P)=38 (30.1),
265 (-10.9), 312 (23.5), 398 (-41.2), 500 (-2.9), 573 (6.4).
5236 J. Org. Chem. Vol. 74, No. 15, 2009