Messages in molecules: Ligand/cation coding and self-recognition in a constitutionally dynamic system of heterometallic double helicates

Article abstract of DOI:10.1002/chem.200600143

Double helicates are known to exhibit self-recognition characteristics determined by the coordination geometry of the metal involved as well as by the topicity of the ligands. Combining tridentate (terpyridine, T) or bidentate (bipyridine, B) subunits in a tritopic strand affords a set of ligands able to assemble by pairs to form double heli cates, homo- or heterostranded, homo-or heterotopic, depending on the coordination properties of the metals in volved. The four ligand strands, BBB, TTT, BBT, and TBT form constitutionally dynamic sets of double helicates with the metal ions Cu^I, Cu^II, and Zn^II; these helicates correspond to the correct coding of the BB, BT, and TT pairs for tetra-, penta-, and hexacoordinate Cu^I, Cu ^II, and Zn" cations, respectively.

Full text of DOI:10.1002/chem.200600143

DOI: 10.1002/chem.200600143
Messages in Molecules: Ligand/Cation Coding and Self-Recognition in a
Constitutionally Dynamic System of Heterometallic Double Helicates
[
a]
[a]
[a]
[a]
Annie Marquis, Virginia Smith, Jack Harrowfield, Jean-Marie Lehn,*
[
b]
[b]
[b, c]
Haiko Herschbach, Rossella Sanvito, Emmanuelle Leize-Wagner,
Alain Van Dorsselaer
and
[
b]
Abstract: Double helicates are known
to exhibit self-recognition characteris-
tics determined by the coordination ge-
ometry of the metal involved as well as
by the topicity of the ligands. Combin-
ing tridentate (terpyridine, T) or biden-
tate (bipyridine, B) subunits in a tritop-
ic strand affords a set of ligands able to
assemble by pairs to form double heli-
cates, homo- or heterostranded, homo-
or heterotopic, depending on the coor-
dination properties of the metals in-
volved. The four ligand strands, BBB,
TTT, BBT, and TBT form constitution-
ally dynamic sets of double helicates
I
II
II
with the metal ions Cu , Cu , and Zn ;
these helicates correspond to the cor-
rect coding of the BB, BT, and TT
pairs for tetra-, penta-, and hexacoordi-
Keywords: helical structures · mo-
lecular information · self-assembly ·
self-recognition
chemistry
·
supramolecular
I
II
II
nate Cu , Cu , and Zn cations, respec-
tively.
Introduction
These features are prevalent in arrays of hydrogen-bond-
ing sites as present in nucleic acid strands, but they also
apply to metallosupramolecular chemistry. Here, the instruc-
tions are given by the nature, number, and arrangement of
the coordination subunits and their reading/processing is
performed by means of the recognition algorithm defined
Supramolecular chemistry has been instrumental in the per-
ception of chemistry as an information science. Indeed, the
controlled assembly of well-defined supramolecular entities
rests on the storage of molecular information within the co-
valently bound components as appropriate arrays of interac-
tion sites, and its processing through recognition algorithms
based on specific noncovalent interactional patterns. The
program is thus molecular and its operation is supramolecu-
[1–3]
by the coordination properties of the metal ion(s).
Indeed a great variety of metallosupramolecular architec-
tures has been generated by self-organization processes im-
[4]
plementing these programming principles.
Among them, helical metal complexes, the helicates, have
[
1–3]
lar.
[
4a–c]
served as benchmarks for such studies,
description.
since their initial
[5a]
They are based on polytopic ligand strands
[
a]Dr. A. Marquis, Dr. V. Smith, Dr. J. Harrowfield,
Prof. Dr. J.-M. Lehn
Institut de Science et dꢀIngØnierie SupramolØculaires (ISIS)
, allØe Gaspard-Monge, BP 70028, 67083 Strasbourg cedex (France)
Fax : (+33)3-90-24-51-40
that entwine themselves around a one-dimensional array of
metal cations, double helicates being particularly well-stud-
ied systems. Depending on the nature of the coordination
8
[5]
[6]
subunits and of the metal ions, homo- or heterostranded
double helicates are obtained, and combined subunit/cation
E-mail: lehn@isis-ulp.org
[
b]H. Herschbach, Dr. R. Sanvito, Dr. E. Leize-Wagner,
Dr. A. Van Dorsselaer
[7–9]
selectivity
is observed. Of special interest is the occur-
[9]
rence of self-recognition (self-sorting), in which mixtures
of ligands and metal ions in dynamic coordination equilibri-
um undergo spontaneous selection of the correct strands
and metal ions, yielding the helicates expected on the basis
of the coordination program and indicating the robustness
of the instructions.
Laboratoire de SpectromØtrie de Masse Bio-Organique
Ecole de Chimie, Polym›res et MatØriaux (ECPM)
UniversitØ Louis Pasteur and CNRS
2
5, rue Becquerel, 67087 Strasbourg cedex 2 (France)
[
c]Dr. E. Leize-Wagner
Institut de Chimie LC3, Institut de Science et
dꢀIngØnierie SupramolØculaires (ISIS)
Such self-selection amounts to a dynamic exploration of
the diversity space of the various possible combinations, ulti-
mately generating the thermodynamically favored one. It
8
, allØe Gaspard Monge, BP 70028, 67083 Strasbourg cedex (France)
Supporting information for this article is available on the WWW
under http://www.chemistry.org or from the author.
5632
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 5632 – 5641

FULL PAPER
displays self-assembly by a reversible process performing se-
lection of the correct partner(s). As such, it has been of spe-
cial significance in the emergence of dynamic combinatorial
[10]
chemistry, the reversible covalent sector of constitutional
[11]
dynamic chemistry, which also encompasses the supramo-
lecular domain, the entities of which are dynamic by essence
due to the lability of noncovalent connections, as in metallo-
supramolecular species such as the helicates.
Coding and selection in helicate self-assembly: Coding and
selection in the self-organization of metallosupramolecular
species are based upon 1) information storage at the molec-
ular level, in the coordination subunits (number, nature, po-
sitioning of the interaction sites); 2) information processing
at the supramolecular level, by metal-ion binding through
specific algorithms defined by the coordination features of
the cations (tetra-, penta-, hexacoordinate; tetrahedral,
trigonal-bipyramidal, square-pyramidal, octahedral, etc.);
and 3) correspondence between the subunits and the metal
cations, which determines the mode of ligand/cation assem-
bly. In the case of helicates, the order of the helicate
Figure 1. Coordination patterns and corresponding pair-coding with bi-
dentate (B; bipyridine) and tridentate (T; terpyridine) ligand subunits;
I
II
II
tetra- (e.g., Cu ), penta- (e.g., Cu ), and hexacoordinating (e.g., Zn ,
II
Fe ) metal ions are represented for simplicity by a square, a pentagon
and a hexagon, respectively.
[12]. They formally involve pairing between the two B and
T units through three patterns, tetra-, penta- or hexacoordi-
nation, akin to the pairing between the nucleobases A, G, T,
and C through two interaction patterns, that is, two A=T or
three GꢀC hydrogen bonds, in the nucleic acid double helix.
(
double, triple, or quadruple stranded) as well as its homo-
A pair of units, a “base pair”, provides a code for metal-ion
recognition; that is, B·B, T·T, and B·T coding for metal ions
with tetra-, hexa-, and pentacoordination, respectively. A se-
quence of B and T units represents a codon, corresponding
to an identical anticodon when the pairing is through even
coordination (tetra, hexa) or to a complementary anticodon
for odd coordination (penta). The two cases amount to oper-
ations of duplication (copying B into B, T into T) or transla-
tion (B into T or conversely).
or heterostranded nature depend on the denticity of the co-
ordination subunit (bidentate or tridentate) and the coordi-
nation number (four, five, or six) of the metal ion; these fac-
tors define the number of subunits surrounding a given coor-
dination center (Figure 1). A sequence of coordination sub-
A
C
H
T
R
E
U
N
G
units may be considered as a codon, a ligand strand consist-
ing of one or several such codons.
The helicates may be of different types depending on
whether the strands and coordination centers are the same
or different. Thus, double helicates may be homostranded–
homotopic, homostranded–heterotopic, heterostranded–ho-
motopic, or heterostranded–heterotopic.
Specifically, pairing between B and T subunits and metal
I
I
cations of tetrahedral (such as Cu or Ag ) or octahedral
II
II
II
II
(such as Fe , Co , Ni , or Zn ) coordination geometry,
tet
oct
leads to the formation of [BBM ]and [TTM ]coordina-
tion centers as well as [BTCu ]centers
II
[6]
II
Coordination subunits of various structures have been im-
with the Cu
[
4–8]
plemented for the generation of helicates,
being based on the bidentate 2,2’-bipyridine (B) and on the
our own work
cation, which presents preferential (B,T) pentacoordination,
although with limited robustness (see below). Thus, homotri-
topic ligands containing either B or T units, that is, BBB and
[
5–7,9a,12]
tridentate 2,2’:6’2’’-terpyridine (T) groups.
Self-assem-
I[5]
II [12]
bly modes of trinuclear double helicates have been repre-
sented schematically in Figure 1 of both references [3]and
TTT strands, form double helicates with Cu and Fe , re-
spectively, while a mixture of BBB and TTT gives a hetero-
II
[6]
stranded double helicate with Cu cations. Crystallograph-
ic studies have provided characterization of species such as
I
[5]
II
[12]
II
Abstract in French: Les mØtallohØlicates à double brin sont
connus pour manifester des propriØtØs dꢀauto-reconnaissance
codØes à la fois par la gØomØtrie des mØtaux impliquØs et par
la topicitØ des ligands. En combinant dans un mÞme brin tri-
m›re des sous-unitØs tridentØes (terpyridine, T) et/ou biden-
tØes (bipyridine, B), on gØn›re un ensemble de ligands sꢀap-
pariant en double hØlices pouvant Þtre homo- ou hØtØroas-
semblØes, homo- ou hØtØrotopiques, selon les propriØtØs de
coordination des mØtaux impliquØs. Ainsi, les quatre ligands
[M
3
A
T
E
G
2
3
A
T
E
N
(TTT’) ],
2
and [Cu -
3
[6]
A
H
R
U
G
designates ligands in which the terpyridine units are linked
by ethylene bridges.
Heterotritopic ligands, as represented by BTB and TBT
strands, give rise to particularly interesting systems with re-
spect to control of binding selectivity and formation of spe-
[
7]
I
II
I
cific heterometallic helicates. In [Cu Fe
A
H
R
U
G
2
2
II
Fe are localized in the terminal four-coordinate B·B sites
and in the central six-coordinate site, respectively, while in
I
II
BBB, TTT, BBT et TBT forment avec les ions Cu , Cu et
II
II
II
Zn une biblioth›que dynamique de double hØlicates, corres-
[Cu
3
A
T
E
N
A
H
R
U
G
pondant au codage correct entre les paires BB, BT et TT et
quinquedentate T·B “base pair”, as in the [Cu (BBB)-
I
II
II
[6]
les cations tØtra-, penta- et hexacoordinØs Cu , Cu et Zn
respectivement.
ACHTREUNG
Chem. Eur. J. 2006, 12, 5632 – 5641
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5633

J.-M. Lehn et al.
Results and Discussion
To obtain reference data, ES-
MS studies were first per-
formed on the species formed
I
from the three cations Cu ,
II
II
Cu , and Zn and the monotop-
ic B and T (bi- and ter-pyri-
dine) ligands. In a mixture 2B,
I
II
2
T, 1Cu , and 1Zn , only B-
I
II
Cu -B and T-Zn -T species
were formed, while a mixture
of B, T, and Cu gave both pen-
tacoordinate B-Cu -T and hexa-
coordinate T-Cu -T complexes
II
II
II
Figure 2. Left: The four tritopic ligand strands, based on a sequence of bipyridine and terpyridine coordination
subunits, used in the present work for combined strand/metal ion recognition. Right: schematic representation
together with remaining ligand
II
of the ligand strands. The two spacers in TTT’ are -CH
2
-CH
2
- groups.
B. A mixture of 2B, 2T, 1Cu ,
II
and 1Zn gave predominantly
II
II
B-Cu -T and T-Zn -T species.
I
II
II
(
Figure 2) for the metal ions Cu , Cu , and Zn . In an initial
Given its preference for hexacoordination similar to that
of Fe (results not shown), Zn may be considered as not
only interesting in its own right, but also as a convenient
[7]
II
II
II
II
study, in which Fe was used instead of Zn , the results
were complicated apparently by the kinetic inertness of
II
II
II
some of the Fe intermediates. The more labile Zn has
proven preferable, as it provides systems that equilibrate
more rapidly and in which the speciation is simpler. The fact
model for what may be expected with Fe under forcing
conditions to activate inert intermediates. Evidently, subtle
differences between these two metal ions must arise and
might well prove useful in exploring such factors as the dif-
ference between inner and outer sites of homostranded
double helicates, differences that have been quantified in
II
II
that overall the behavior of Zn and Fe is similar points to
the generality of the coding, as it is the type of coordination
not the nature of the cation that determines the outcome.
Importantly, for the first transition-metal series, despite
the fact that each metal exhibits various coordination num-
bers and geometries (even in a single oxidation state) in a
broad range of complexes, polypyridine ligand units appear
to evoke quite strong individual preferences for tetra-,
penta-, or hexacoordination.
[8a]
triple-helix systems.
The four tritopic ligands employed (Figure 2) all involved
linkage of the B or T sites by 2-oxapropyl bridges; that is,
[14]
[7]
[7]
BBB, BBT, TBT and TTT (see Experimental Section).
I
II
II
Assuming the codings B·B=Cu , B·T=Cu and T·T=Zn
to be valid, a mixture of all four ligands could give rise to a
total of 11 different double-helical complexes in the pres-
ence of a mixture of the three cations Cu , Cu , and Zn
(Figure 3).
Of these species, only [Cu (BBB) ] has been structural-
In conducting the present study of the species distribution
arising when various mixtures of four tritopic polypyridine
ligands (Figure 2) were added to solutions containing differ-
ent mixtures of Zn , Cu , and Cu cations, it was anticipated
that these would be labile systems, representing dynamic
combinatorial libraries distributed in accord with thermo-
dynamic preferences, resulting (in part) from the nature of
the coordination environments. Our aim was thus to estab-
I
II
II
II
II
I
I
3+
A
H
R
U
G
3 2
ly characterized as a double helicate, though both an ana-
[10]
II
6+
[6]
II
logue of 8, [Cu
logue of 5, [Fe
A
H
R
U
A
H
R
U
3
II
6+ [12]
A
H
R
U
G
have been similarly authenti-
3
2
cated. Whether both helicate and nonhelicate species might
[11]
lish in a complex system of constitutionally dynamic met-
allosupramolecular species, that is, the double helicates, and
the occurrence of information-directed self-assembly by self-
selection of the correct double strand/metal cation partners
from an equilibrating set of polytopic strands and metal cat-
A
C
H
T
R
E
U
N
G
ions. Electrospray mass spectrometry (ES-MS) was used as
the primary tool for this work, its utility having been estab-
lished earlier in a study of the self-assembly of pentanuclear
[13]
double helicates.
However, it is clear that this method,
while allowing the identification of the species present, does
not give accurate quantitative information.
Figure 3. The 11 possible trinuclear complexes that may be formed with
the four tritopic ligands BBB, BBT, TBT, and TTT (Figure 2) and a set
of metal ions of tetra-, penta- and hexacoordination algorithms.
5634
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 5632 – 5641

Helical Structures
FULL PAPER
1
be present in solution mixtures is a matter requiring investi-
gation by systematic characterization of the products from
particular mixtures. In the present systems involving para-
results support the indications from H NMR and electronic
I
absorption spectroscopic studies that the BBB/Cu system is
rather simple.
II
magnetic Cu , such analysis was heavily dependent on the
use of ES-MS, which establishes composition but not spatial
structure. Thus, H NMR spectroscopy was an indispensable
Strand BBT: The case of the ligand BBT is special, since its
symmetry is such that two double helicate forms, involving
parallel (2) and antiparallel ligand strands (3) (Figure 3), are
possible and are associated with different base-pairings.
Consistent with the coding anticipated for a parallel
(head-to-head) arrangement, the electrospray mass spec-
1
accompaniment for all diamagnetic species (involving only
I
II
Cu and Zn ). It is of interest to note, however, that al-
though trinuclear species of heterostrand composition were
found in the present work to give ES-MS peaks for several
À
I
ion associates (with [BF ] ), in no case did they provide in-
trum for a mixture of BBT with one molar equivalent of Cu
4
II
tense solvate peaks, a characteristic resulting probably from
the absence in the helicate species of molecular cavities or
clefts that might provide sites for holding solvent molecules.
It was expected that the nature of the species present in
any mixture of the selected ligands and metal ions would
depend not only on the ratios of reactants, but also on their
concentrations and thus the concentration chosen for the
and half that amount of Zn shows a major peak for a bis-
I
II
4+
A
H
R
U
G
A
H
R
U
G
2
(2). Direct synthesis of this diamagnetic material and re-
1
cording of its H NMR spectrum revealed the diastereoto-
picity of the methylene protons expected for a helical (or a
[4c]
mesocate) structure. Further, COSY experiments enabled
the complete assignment of all 30 inequivalent proton sig-
À4
À1
present measurements was set at ~10 I molL , a value
nals (aromatic and methylene) expected for C symmetry,
2
known in the well-characterized BBB/Cu system to lead to
the double helicate [Cu (BBB) ] , in which >90% of the
total amount of metal is complexed. The overall results of
ES-MS measurements showed that, for all systems investi-
while ROESY measurements (See Supporting Information,
Figure S2) showed the close proximity of the terminal B
units of the two strands, reflected in the approach of the
methyl group of one strand to the methylene group close to
I
3+
A
H
R
U
G
3 2
[5d]
gated, this choice led to the prominence of bis
A
H
R
U
G
the terminal pyridine of the other strand, as anticipated
I
metal species (associated with various numbers of counter-
ions), indicating that while no doubt the stability constants
for all species are not the same, all are of a value at least
from the structure of [Cu
A
H
R
U
G
3
2
I
3+
comparable to that of [Cu (BBB) ] . Care was also taken
A
T
E
N
3 2
to achieve experimental conditions leading to thermody-
namic equilibrium in the solvent used. To validate any
ligand coding, it is not essential that the particular complex
of interest be the sole or even the dominant species present,
but that it should have the correct composition, be readily
detectable, and that extensive coding errors be absent.
Generation of double helicates from single ligand strands
3
Figure 4. Proximity of the protons of the terminal CH group of one BBT
strand to the methylene protons on the second strand in the parallel-
I
II
4+
Strand BBB: Considering first the addition of the separate
stranded double helicate [Cu Zn
A
H
R
U
(BBT) ] (2).
2
2
I
II
II
ligands to Cu /Cu /Zn mixtures, the behavior of BBB in
I
strongly favoring exclusive binding of Cu is well establish-
[
5]
ed, so only the ES-MS of a solution containing BBB and
the ROESY spectrum provided no evidence for proximity
of protons associated with B and T units of separate strands.
Addition of BBT to a 2:1 mixture of Cu and Cu also
provided a solution for which most of the principal peaks in
the high-resolution electrospray mass spectrum could be as-
(ligand) species, containing three Cu
ions and with an overall charge of 5+, associated with vari-
ous numbers of either [BF ] or F ions. Thus, both isotopic
I
Cu was investigated here. The most intense peak in the
I
3+
II
I
spectrum is that assigned to [Cu
A
T
E
N
(BBB) ] (m/z 451), with
3
2
a much less intense peak corresponding to the ion-pair
I
2+
{
[Cu (BBB) ]·BF } (m/z 719). The isotopical simulation of
A
C
H
T
R
E
U
N
G
3 2 4
I
the ES-MS spectrum of the signal of triply charged [Cu -
(BBB) ] is in perfect agreement with the experimental
signed to a single bis
A
H
R
U
G
3
3
+
A
C
H
T
R
E
U
N
G
2
À
À
spectrum, thus allowing a unambiguous identification of the
4
peak (See Supporting Information, Figure S1c,d). Interest-
ingly, a peak of intermediate intensity corresponds to the
peak separations and patterns were consistent with a peak
II
I
5+
at m/z 295 being due to [Cu Cu
A
H
R
U
G
2
2
I
+
II
I
À
4+
II
I
species [Cu
A
C
H
T
R
E
U
N
G
(BBB)] (m/z 643), for which it is possible that
{[Cu Cu
A
H
R
U
G
2
À
2
4+
II
the ligand is not in its extended form but instead is folded
[BF ] } , at m/z 550 to {[Cu Cu
4
I
back and has wrapped itself about Cu to give tetrahedral
m/z 869 to {[Cu Cu
II
coordination by binding with the first and second or the first
and third bipyridine units. This species is not observed at
to {[Cu Cu
A
H
R
U
G
A
H
R
U
G
2
with the formation of the antiparallel (head-to-tail; 3)
À2
II
I
II
5+
high concentrations (10 m). Molecular modeling indicates
double helicate species [Cu Cu Cu
A
H
R
U
A
H
R
U
G
that such entities can indeed form, but generally the ES-MS
ed for the coding that this array of B, T pairs implies. Al-
Chem. Eur. J. 2006, 12, 5632 – 5641
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5635

J.-M. Lehn et al.
1
though the paramagnetism of 3 excludes the use of H NMR
ly controlled metal ion composition and with specificity of
binding, these are not necessarily of equal likelihood (i.e. as-
suming the metal ion composition controls the ligand pair-
ing), but a 1:1 mixture of 1 and 3, for example, would have
the same overall composition as 6. Indeed, a 1:1 mixture of
spectroscopy to establish its symmetry, it seems reasonable
to assume a double-helical form given that this has been es-
I
3+[5a]
tablished crystallographically for both [Cu
A
T
E
N
(BBB) ]
and
(Note that the presence of the
ride ion in the m/z 374 species indicates that some
3
2
II
6+ [12]
[Cu
A
C
H
T
R
E
U
N
G
(BBB)(TTT’)] .
A
C
H
T
R
E
U
N
G
3
I
II
fluo
A
C
H
T
R
E
U
N
G
BBB and BBT with 2Cu +Cu provides a mass spectrum
showing significant peaks for all three species 1, 3, and 6, all
three corresponding to correct (2strand+3cation) combina-
tions. Entropic factors may favor the mixed-ligand species 6,
though they do not seem to be dominant, since the mixture
À
degree of hydrolysis of BF4 may have occurred in solu-
tion.)
The formation of parallel- or antiparallel-stranded double
helicates from the same pair of ligands and different sets of
metal ions corresponds to a double expression of molecular
information: processing the same molecular information by
means of two different coordination algorithms (sets of
metal ions) generates two different output species in a
I
II
generated from a solution of BBB+TBT+Cu +2Cu con-
tains in addition to 7 as the major species, also detectable
amounts of 1.
Evidence that satisfaction of the local metal-ion require-
ments is the major factor influencing the helicate composi-
tion, that is, that the strength of the binding of a given metal
ion to a given site (the “local” information) is not markedly
dependent upon the nature of the adjoining sites, is provid-
ed in the system BBT+TBT, which is unique in that it may
provide a ligand pair able to bind all three metal cations
considered to give 9. An equimolar mixture of BBT and
TBT provides the three pairing combinations (BBT, TBT),
(TBT) and (BBT) in which the strands may be either par-
[3]
single-code/two-product fashion.
Strands TBT and TTT: For the symmetrical ligands TBT and
TTT, previous investigations of their individual interac-
tions with Cu /Fe mixtures have demonstrated the forma-
tion of species analogous to 4 and 5, again consistent with
the coding implicit in the helical bis(ligand) array. In the
case of the analogue of 4, the electrospray mass spectrum
showed the predominance of a 1:1 Fe /TBT species, perhaps
[7]
I
II
A
H
R
U
G
II
2
2
as a consequence of the ligand folding to envelop the cation
and of the expected kinetic inertness of this helicate precur-
allel or antiparallel. Treatment with an equimolar mixture
I
II
II
Cu +Cu +Zn yields indeed a mixture of helicates giving a
mass spectrum dominated by the peaks corresponding to all
four “correct” species 2, 3, 4, and 9 (Figure 5).
II
II
sor. Replacement of Fe by Zn appears in all cases to
II
I
result in rapid equilibration, with TBT+Zn +0.5Cu giving
II
predominantly 4 and TTT+1.5Zn giving 5. Considering
2) Similarly, in progressing to a ternary ligand mixture but
II
[12]
I
II
the known structure of the Fe analogue of 5,
it is as-
restricting the choice of metal ions to Cu and Zn , only the
homoligand species 1, 2, and 5 are seen to be present in the
product mixture, consistent with the fact that the coding in a
II
sumed that the Zn complex is also double-helical. The ES-
MS signal of the sixfold charged [Zn (TTT) ] is in excel-
II
6+
A
H
R
U
G
2
II
lent agreement with the isotopical simulation (See Support-
ing Information, Figure S1a,b).
heteroligand pair would require the presence of Cu for
binding (Figure 3).
3
) Proceeding further to a quaternary ligand mixture
Strand/cation self-selection in the assembly of double heli-
cates from mixtures of ligands and metal ions: The goal of
this work was to analyze the occurrence of complex recipro-
cal self-selection between ligand strands and metal cations
as a function of the coordination subunits of the ligands and
the coordination number of the ions.
(BBB+BBT+TBT+TTT) and adding all three metal ions,
the mass spectrum displays peaks due only to trimetallic–
bis(ligand) species (given isobaric overlaps) incorporating
A
H
R
U
1) For the various binary mixtures possible with the pres-
ent ligands, it is expected that any (heteroligand) double-
II
helical array would involve at least one site for Cu
(
Figure 3) and indeed combination of the different ligand
pairs with the appropriate mixtures of metal ions led, in all
cases, to the predominance of the species 6–11 in the elec-
trospray mass spectra. The presence of the paramagnetic
II
Cu center renders the ES-MS technique of particular im-
portance in these instances. Noting that there are two iso-
baric pairs 3 and 7 within the full range of complexes possi-
ble, some ambiguities may be anticipated in the analysis of
systems based on mixtures of more than two ligands (see
below). Even with binary ligand mixtures and suitable
choice of the nature and composition of the metal-ion mix-
ture, various complications must be considered, the most
evident being that both homo- and heteroligand-pairs may
be present in any given double-helical species. For a precise-
I
II
II
Figure 5. ES/MS spectrum of a mixture of BBT+TBT+Cu +Cu +Zn
showing the formation of all four possible double helicates corresponding
to the correct (2 subunits/1 cation) combinations: 2, 3, 4 and 9.
5636
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 5632 – 5641

Helical Structures
FULL PAPER
the correct selection of ligands and metal cations corre-
sponding to the “base pair” coding provided by each pair of
ligand strands (Supporting information, Figure S3).
multiple species, this may either result from an intrinsic sta-
II
bility of a given species, here [Cu
A
H
R
U
G
A
H
R
U
G
3
intrinsic destabilization of another member of the equilibrat-
II
ing system, that is, [Cu (TTT’)], due to the lack of flexibili-
A
H
R
U
G
3
Five-point reading—pentacoordination and the special case
of Cu : Five coordination is required for the generation of
ty of the ethylene bridges.
II
Other factors, such as differences in rotational energy
within the bridges or in stacking interactions between the ar-
omatic units of the ligands, must play a part. Although one
may wish to explore other [(metal ion) (bidentate site)(tri-
dentate site)]combinations, the [Cu , B, T]unit, despite lim-
heterotopic coordination centers from a combination of a bi-
dentate and a tridentate coordination subunit. Similarly, het-
erostranded double helicates as well as translation, by B/T
correspondence, also result from pentacoordination. Al-
though there are some other cases of pentacoordination
n+
II
ited robustness, nevertheless offers good information-pro-
II
[16]
with other metal ions and binding units, Cu appears to
A
H
R
U
G
occupy a unique place, by being the metal cation among all
of them that fits best the requirements for the present pur-
poses. As has been seen from experiments with the mono-
tinuclear systems presenting multiple expression of molecu-
lar information and dominant/recessive behavior.
[17]
[18]
II
topic ligands B and T, Cu forms both B·T and T·T pairs so
II
that (B,T) instructions and their reading by Cu appear to
Conclusions
be less “robust” than are (B,B) and (T,T) four-point and six-
point interactions with tetrahedrally and octahedrally coor-
dinating cations, respectively. With the present tritopic li-
While it has not been shown that in every case studied here
the trimetallic–bis(ligand) species detected by ES-MS has a
double helical structure, this is a plausible rationalization of
all the present results, consistent with both the spectroscopic
and crystallographic observations for several species.
If, as in the present work, the binding of all metal ions in-
volved is labile, it becomes possible to rapidly establish
equilibrium mixtures of (double) helicate complexes involv-
AHCTREUNG
II
gands, the preferential coding of a B·T site for Cu appears
to be obeyed, except in the particular case of BBB/TTT
6
+
mixtures, for which species 8 ([Cu
A
T
E
G
A
T
E
N
3
6
+
detected in competition with [Cu
A
H
R
U
G
3
2
Information, Figure S4). This contrasts markedly with the
BBB/TTT’ system (the TTT’ ligand having ethylene rather
than oxapropylene bridges), in which the ES-MS spectrum
exhibits predominantly signals due to anion associates of
ing poly(B/T) ligands in which the particular strand pairings
A
H
E
G
are determined by specific coordination preferences of the
cations. Provision of the appropriate metal-ion mixture may
thus be used to control ligand pairing by means of the infor-
II
6+
[Cu
A
C
H
T
R
E
U
N
G
(BBB)
A
C
H
T
R
E
U
N
G
(TTT’)]
(See Supporting information, Fig-
3
ure S5), the structure of which has also been confirmed in
[6]
[19]
the solid state by X-ray crystallography. Thus, in any hetero-
strand species, a common bridge between donor units is not
mation contained in the mixture of metal ions and ligands.
A
C
H
T
R
E
U
N
G
Clearly, complications may arise when a broader range of
I
II
II
necessarily ideal.
metal ions than Cu , Cu , and Zn is considered. Also, the
I
II
In addition to the consequences of the choice of the B
and T subunits, the detailed structure of the strands appears
to play an important role, in particularly their commensura-
differences between Cu and Cu indicate that there may be
many useful subtleties to be explored in the use of redox-
active transition-metal ions to control ligand assembly, as is
the case in the electrochemical interconversion between two
highly intertwined multinuclear architectures or between
two states of a catenate.
II
bility and flexibility. As already pointed out, the Cu coordi-
[17]
nation instructions are not fully “robust” with respect to co-
ordination number five or six. Furthermore, a pentacoordi-
[
20]
II
nate BT·Cu center may be have either trigonal bipyramidal
One of the remarkable features of helicate ligand binding
is the occurrence of “self-recognition”, meaning that for a
mixture of strands of differing topicity, addition of a metal
ion compatible with helicate formation results preferentially
or square pyramidal geometry. In the crystal structure,
II
[15a,b]
BT·Cu has a square pyramidal geometry
and in the
II
[Cu
A
C
H
T
R
E
U
N
G
(BBB)(TTT’)]double helicate, the terminal centers
A
C
H
T
R
E
U
N
G
3
[9a]
are square pyramidal, whereas the central one is trigonal bi-
in the pairing of ligands of the same topicity. This suggests
the possibility of devising an analogue of the polymerase
[6,15c]
pyramidal.
Assuming that binding of the tridentate units
[21]
would be energetically dominant, it may be that it is the
array of bidentate sites that must accommodate most to any
mismatch. Hence, if the oxapropylene bridges of BBB
cannot adapt sufficiently to match the requirements of a
chain reaction in which coordination subunits, functional-
ized so as to enable their reversible linkage (for instance,
[9d]
through imine groups ), would first be assembled on a
primer strand under the control of metal-ion coordination
“base-pair” coding; the newly assembled strand would
thereafter be irreversibly connected by post-assembly modi-
fication (e.g. reduction of imine groups), and finally strand
separation by sequestering the metal ions would prepare the
II
6+
[
Cu (TTT)] unit, this might lead to the relative destabili-
A
C
H
T
R
E
U
N
G
3
6
+
zation of [Cu
A
C
H
T
R
E
U
N
G
(BBB)(TTT)] . Shortening the links between
A
H
R
U
G
3
tridentate units, as in TTT’, would make the ligand commen-
surable with the BBB strand, which could explain the return
6
+
[22]
to favoring of pentacoordination, as in [Cu3
On the other hand, bridge matching in [Cu
A
T
E
N
A
H
R
U
G
.
stage for the next round. Given the observed effect of dif-
6
+
A
H
R
U
G
(TTT’)]
ferences in the linking units between donor sites on the sta-
bility of mixed-ligand species, an appealing prospect would
3
also leads to the dominance of the heterostrand species. As
in all systems involving thermodynamic equilibria between
Chem. Eur. J. 2006, 12, 5632 – 5641
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5637

J.-M. Lehn et al.
be that the linking reaction might automatically lead to the
ejection of the newly generated ligand.
Spectrometric analysis of the helicate mixtures
Mass spectrometry: Spectra for the solution equilibrium studies were ob-
tained using electrospray ionization on three different instruments, con-
siderable time being devoted to establishing the optimum conditions for
detection of the helicates on each. Signal attributions were verified by
comparing the high-resolution ES-MS data to isotopical simulations.
Metal and ligand mixtures were prepared from stock solutions of the sep-
To read the message considered to be conveyed by the
formation of a given helicate, it must, of course, be possible
to detect that helicate. This does not mean that it must be
the only assembly present on the supramolecular “page”,
but simply that it must be spatially, energetically, or tempo-
rally separable from the other species present (which may
themselves carry other messages, perhaps more inscrutable).
In the systems currently described, not only are the species
I
II
II
ACHTREUNG
arate ligands and the metal salts Cu BF
4
A
T
E
N
4 2 4 2
, Cu [BF ] , and Zn [BF ] in
acetonitrile. The solutions finally injected into the spectrometers were di-
luted so that the total ligand concentration was 10 m.
À4
Calibration: (for all) Calibration was performed using the multiply charg-
ed peaks of protonated horse heart myoglobin (2 mm, in water/acetonitrile
[MM’M’’LL’]readily isolated and detected by ES-MS, but
5
0/50 (v/v) acidified with 1% HCOOH. For calibration in the high-reso-
+ +
they are generally major components of the reaction mix-
tures. Of much interest in the perspective of the present
work is the exploration of coordination units other than B
and T and of their ability to undergo complementary pairing
with metal cations. Triple-stranded helicates might be con-
sidered, based for instance on the combination of B and T
with strands containing monodentate pyridine groups.
One of the prospects for the use of ligands described
herein is in the construction of a more sophisticated analyti-
cal dictionary so as to be able to decipher the information
contained in the ligand/metal-ion coordinative language. In
the conventional analytical technique of inductively coupled
plasma mass spectrometry (ICP-MS), for example, a single
signal (ion current) identifies a single element. For ES-MS
of solutions containing polytopic ligands, a single signal may
identify several elements (as a particular mixture), depen-
lution mode, the singly charged clusters of [H PO ) H ] (n=2–20) were
3
4 n
used by injecting a solution of 0.01% H
v/v).
3
PO
4
in water/acetonitrile 50/50
(
MicroTof: High-Resolution ESI mass spectra in positive mode were ac-
quired on a time-of-flight mass spectrometer (microTof, Bruker Dalton-
ics, Bremen, Germany). The ESI-source was heated to 2008C. Sample
solutions were introduced into the mass spectrometer source with a sy-
ringe pump (Cole Parmer, Vernon Hills, Illinois, USA) with a flow rate
À1
of 3 mLmin . Scanning was performed on an m/z range from 50–3000,
data was averaged for 1 min and then smoothed using the Gaussian algo-
rithm.
Quattro II: ESI mass spectra in positive mode were acquired on a triple
quadrupole mass spectrometer (Quattro II, Micromass, Altrincham, UK)
in the positive mode. The ESI-source was heated to 808C. Sample solu-
tions were introduced into the mass spectrometer source with a syringe
pump (Harvard type 55 1111: Harvard Apparatus Inc., South Natick,
À1
USA) with a flow rate of 3 mLmin . Scanning was performed on an m/z
range from 200–3000 in 28 s; five scans were summed to obtain the final
spectrum.
A
C
H
T
R
E
U
N
G
ding upon the ligand topicity. There is still much work to be
MSD: To get comparative results, selected mixtures were also analyzed
with a second quadrupole mass spectrometer (Mass Selective Detector,
Series 1100mSD, Hewlett Packard Inc, Toronto, Canada) in the positive
mode. The desolvation gas temperature was set to 3008C, sample solu-
tions were introduced into the mass spectrometer source with a syringe
pump (Harvard type 55 1111: Harvard Apparatus Inc., South Natick,
done, however, to establish ways of converting the qualita-
tive information of the nature of the species present into the
quantitative information of how much there is of each.
In summary, the present systems involving mixtures of
several ligand strands and metal ions have enabled conduct
of a dynamic exploration of the supramolecular constitution-
À1
USA) with a flow rate of 3 mLmin . Scanning was performed on an m/z
range from 200–1200 in 10 seconds; five scans were summed to obtain
the final spectrum. Spectra were smoothed using the Gaussian algorithm.
[11]
al space under the control of the processing of molecular
information, driven by recognition between complementary
coordination subunit/metal-cation pairing involving specific
Synthesis: Syntheses of functionalized bipyridine units and tritopic li-
gands derived from them were described previously, as were those of
ligands closely related to those used that contained terpyridine units.
Hence, only a selection is presented herein as illustrative of the general
synthetic procedures.
[
14]
[
7,10]
[3]
interactional algorithms, and resulting in a process of self-
[11]
organization by selection of the correct components.
[
12]
5
-Bromomethyl-2,2’-6’,2’’-terpyridine:
In rapid succession, N-bromo-
succinimide (1.07 g, 6 mmol) and benzoylperoxide (0.23 g, 0.33 mmol)
were added to a solution of 5,5’’-dimethyl-2,2’-6’,2’’-terpyridine (1 g,
Experimental Section
4
4 mmol) in CCl (100 mL). The resulting yellow solution was heated
under Ar at 958C for 55 min before being filtered while hot, and the sol-
vent removed in vacuo. Purification was achieved by column chromatog-
Materials and general methods: The following compounds were prepared
as previously described: BBB, BBT, TBT. The following reagents
[
14]
[7]
[7]
raphy (silica; CH
(0.891 g, 68% yield). M.p. 224–2258C; H NMR (200 MHz; CDCl
2
Cl
2
)
and by recrystallization from CHCl
3
/hexane
): d=
1
II
II
3
were purchased from commercial sources: Cu
A
H
R
U
G
4 2
] (Aldrich), Zn -
1
8.69 (m, 2H; H6, H6’’), 8.60 (brd, J=7.8 Hz, 2H; H3, H3’’), 8.46 (dd, J=
7.8, 0.9 Hz, 1H; H3’), 8.44 (dd, J=7.8, 0.9 Hz, 1H; H5’), 7.96 (t, J=
7.8 Hz, 1H; H4’), 7.86 (m, 2H; H4, H4’’), 7.33 (ddd, J=7.5, 4.8, 1.2 Hz,
A
C
H
T
R
E
U
N
G
4 2
[BF ]
[
23]
Ultrashield Avance 400 spectrometer. The solvent residual signal was
used as an internal reference for H NMR spectra. The following nota-
tion is used for the H NMR spectral splitting patterns: singlet (s), dou-
1
1
1H; H5’’), 4.56 ppm (s, 2H; CH Br); MS (EI): m/z (%): 327 (6)
2
8
1
+
79
+
+
[
M+Br ] , 325 (6) [M+Br ] , 247 (38), 246 (100) [MÀBr] . Polybromi-
nated products could be reduced to monobrominated products by the use
of DIBAL (DIBAL=diisobutylaluminum hydride) in CH Cl
at À788C.
In rapid succession, N-
blet (d), triplet (t), multiplet (m). The two-dimensional NMR experi-
ments used were: COSY (correlation spectroscopy), NOESY (nuclear
Overhauser enhancement spectroscopy or nuclear Overhauser and ex-
change spectroscopy), ROESY (rotating-frame Overhauser enhancement
2
2
[
12]
5,5’’-Bis(bromomethyl)-2,2’-6’,2’’-terpyridine:
(
effect) spectroscopy); they were done on 500 MHz Bruker spectrome-
bromosuccinimide (5.55 g, 31.19 mmol) and benzoylperoxide (0.23 g,
ters. FAB-MS, EI-MS, and ES-MS measurements were performed by the
Service de SpectromØtrie de Masse, UniversitØ Louis Pasteur. Melting
points were recorded on a Büchi Melting Point B-540 apparatus and are
uncorrected.
0.94 mmol) were added to a solution of 5,5’’-dimethyl-2,2’-6’,2’’-terpyri-
dine (1.63 g, 6.24 mmol) in CCl
was heated under Ar at 958C for 55 min before being filtered while hot.
Insoluble material present was washed with CCl (2”75 mL), the wash-
4
(150 mL). The resulting yellow solution
4
5638
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 5632 – 5641

Helical Structures
FULL PAPER
˚
1
ings combined with the initial filtrate, and the total evaporated under
white crystalline solid (0.50 g, 80% yield). M.p. 154–156 C; H NMR
vacuum to give a yellow paste. This paste was dissolved in CH
300 mL) and washed with aqueous Na (0.5m; 3”100 mL). The
combined aqueous washings were back-extracted with CH Cl (100 mL),
and the combined organic phases were then dried over Na SO before
the solvent was removed in vacuo. The yellow residue obtained was re-
crystallized twice from CCl to obtain pure 5,5’’-bis(bromomethyl)-2,2’-
’,2’’-terpyridine (0.27 g; 10%). Considerable additional material was
subsequently obtained by reduction of higher bromination products with
DIBAL. Thus, the CCl -soluble material was recovered by evaporation of
the solvent, then dissolved in CH Cl
(120 mL) and cooled to À788C.
DIBAL (1m in CH Cl ) was added until TLC indicated that only dibro-
mide was present, the solution then being green. Saturated aqueous
NH Cl (70 mL) was added and the mixture allowed warm to room tem-
perature. Water (150 mL) was added and the mixture filtered through
Celite, the filter pad being subsequently washed with CH Cl . The total
organic phase was washed with water (2”150 mL), dried over Na SO
and the solvent removed in vacuo. Recrystallization of the residue from
CH Cl provided more product as a very pale yellow powder (1.20 g;
6% yield). M.p. 189–1918C; H NMR (200 MHz; CDCl
J=2.2 Hz, 2H; H6, H6’’), 8.61 (d, J=8.1 Hz, 2H; H3, H3’’), 8.47 (d, J=
2
Cl
2
(200 MHz; CD OD): d=8.59 (d, J=1.7 Hz, 2H; H6, H6’’), 8.50 (d, J=
8.1 Hz, 2H; H3, H3’’), 8.27 (d, J=7.8 Hz, 2H; H3’,H5’), 7.93 (t, J=
3
(
2
S
2
O
3
2
2
7.8 Hz, 1H; H4’), 7.89 (dd, J=8.1, 1.7 Hz, 2H; H4, H4’’), 4.70 ppm (s,
+
+
2
4
4H; 2CH OH); MS (EI): m/z (%): 293 (100) [M] , 292 (24) [MÀH] ,
2
+
+
276 (15) [MÀOH] , 264 (52), 262 (11) [MÀCH OH] .
2
4
Ligand strand TTT: Under Ar, KOtBu (0.112 g, 1 mmol) was added to a
6
solution
of
5,5’’-di(hydroxymethyl)-2,2’-6’,2’’-terpyridine
(0.147 g,
0
.5 mmol) in THF (20 mL) turning the colorless solution to pale orange.
4
The mixture was stirred for 1 h at room temperature before adding 5-bro-
momethyl-2,2’-6’,2’’-terpyridine (0.326 g, 1 mmol) and stirring for 17 h.
The precipitate was obtained upon filtration and washed with water,
2
2
2
2
MeOH, and diethyl ether to give TTT as a white powder (0.288 g, 74%
4
1
yield). M.p. 199–2018C; H NMR (500 MHz; 10% CF
3
SO
3
D
in
[
6
D ]acetone): d=9.54 (br s, 2H; Ar), 9.47 (brs, 2H; Ar), 9.37 (d, J=
2
2
6
8
1
.0 Hz, 2H; Ar), 9.15 (m, 6H; Ar), 9.04 (m, 6H; Ar), 8.91 (m, 6H; Ar),
.61 (t, J=7.9 Hz, 2H; Ar), 8.61 (m, 1H; Ar), 8.43 (ddd, J=7.4, 6.0,
2
4
.1 Hz, 2H; Ar), 5.30 ppm (s, 8H; 4CH O); MS (FAB): m/z (%): 784
2
2
2
+
+
(
31) [M+H] , 522 (10) [MÀterpyCH
2
O] , 345 (25), 262 (28) [terpy-
+
1
4
3
): d=8.72 (d,
+
2
CH O] , 247 (33), 192 (100); HS-MS (MicroTOF): m/z: 784.31 [M+H] ,
+
8
06.29 [M+Na] .
7
2
4
.8 Hz, 2H; H3’, H5’), 7.97 (t, J=7.8 Hz, 1H; H4’), 7.91 (dd, J=8.1,
.2 Hz, 2H; H4,H4’’), 4.57 ppm (s, 4H; 2CH Br); MS (EI): m/z (%):
19.9 (100), [M+H] with expected isotopic pattern), 391 (49), 340/338
Ligand strand TBT: Following the same procedure as for the synthesis of
TTT, 6,6’-di(hydroxymethyl)-2,2’-bipyridine (0.173 g, 0.8 mmol) and 5-
bromomethyl-2,2’-6’,2’’-terpyridine (0.522 g, 1.6 mmol) gave TBT as a
2
+
+
(
22/22) [MÀBr] .
-Acetoxymethyl-2,2’-6’,2’’-terpyridine: CH
added to solution of 5,5’’-bis(bromomethyl)-2,2’-6’,2’’-terpyridine
2.60 g, 8.0 mmol) in DMF (120 mL) and the mixture heated at 1408C for
1
white powder (0.311 g, 55% yield). M.p. 196–1988C; H NMR (400 MHz;
5
3 2
CO Na (3.92 g, 48 mmol) was
CDCl
.8 MHz, 2H; terpy H6), 8.63 (d, J=8.1 Hz, 2H; terpy H3’’), 8.62 (brd,
J=7.7 Hz, 2H; terpy H3), 8.46 (d, J=8.0 Hz, 4H; terpy H3’ H5’), 8.34
d, J=7.8 Hz, 2H; bpy H5), 7.96 (t, J=8.0 Hz, 2H; terpy H4’), 7.92 (dd,
J=8.1, 2.0 Hz, 2H; terpy H4’’), 7.86 (td, J=7.4, 1.8 Hz, 2H; terpy H4),
3
): d=8.73 (d, J=2.0 Hz, 2H; terpy H6’’), 8.70 (brdd, J=4.7,
a
1
(
2
2 3
0 h. The organic fractions were dried over Na SO and the solvent re-
(
moved in vacuo to yield the methylacetate as a white solid (2.10 g, 86%
1
yield) sufficiently pure for direct use in the next step. H NMR
7
.85 (t, J=7.8 Hz, 2H; bpy H3), 7.53 (ddd, J=7.4, 4.7, 1.1 Hz, 2H; terpy
(
200 MHz; CDCl
Ar), 8.40 (brd, J=7.7 Hz, 2H; Ar), 7.96–7.77 (m, 3H; Ar), 7.96–7.77 (m,
H; Ar), 7.32–7.25 (m, 1H; Ar), 5.16 (s, 2H; CH OAc), 2.09 ppm (s, 3H;
CH ).
,5’’-Bis(acetoxymethyl)-2,2’-6’,2’’-terpyridine: CH
solution of 5,5’’-bis(bromomethyl)-2,2’-6’,2’’-terpyridine
0.90 g) in DMF (40 mL) and the mixture was heated, under Ar, at
3
): d=8.65 (brs, 2H; Ar), 8.56 (brd, J=8.0 Hz, 2H;
1
3
2 2
H5), 4.84 (s, 4H; 2CH O), 4.80 ppm (s, 4H; 2CH O); C NMR
(
50.33 MHz; CDCl
N), 149.2, 148.7 (Ar-CH, o-N), 137.9, 137.6, 136.9, 136.5 (Ar-CH, p-N),
33.6 (Ar-CCH O, m-N), 123.8, 121.5, 121.2, 121.0, 120.9, 120.0 (Ar-CH,
m-N), 73.6 (CH ), 70.3 ppm (CH ); MS (FAB): m/z (%): 707 (75)
M+H] , 491 (43), 247 (100).
3
): d=157.7, 156.3, 155.9, 155.6, 155.4, 155.2 (Ar-C, o-
3
2
-
O
2
3
1
2
5
3 2
CO Na (2.12 g) was
2
2
added to
(
a
+
[
Ligand strand BBT: Following the same procedure as for the synthesis of
TTT, 5-hydroxymethyl-2,2’-6’,2’’-terpyridine (0.171 g, 0.65 mmol) and 6’-
1
408C for 19 h. Removal of the solvent in vacuo gave a pale brown solid,
and filtered through Celite to remove
which was extracted with CHCl
3
[
14]
(
bromomethyl)-6’’’-methyl-6,6’’-[oxybix(methylene)]bis(2,2’-bipyridine)
NaBr. Evaporation of the solvent provided material (0.80 g, 99% yield)
sufficiently pure for direct use in the next step.
(
0.300 g, 0.65 mmol) gave BBT as a white powder (0.251 g, 60% yield).
1
M.p. 206–2088C; H NMR (200 MHz; CDCl3): d=8.72 (m, 2H; Ar), 8.63
5
2
-Hydroxymethyl-2,2’-6’,2’’-terpyridine: Solutions of 5-acetoxymethyl-
,2’-6’,2’’-terpyridine (1.435 g, 4.7 mmol) in CH OH (100 mL) and NaOH
(
m, 2H; Ar), 8.46 (d, J=7.8 Hz, 2H; Ar), 8.33 (m, 3H; Ar), 8.19 (d, J=
3
7
7
.9 Hz, 1H; Ar), 8.0–7.8 (m, 6H; Ar), 7.68 (t, J=7.7 Hz, 1H; Ar), 7.60–
.51 (m, 3H; Ar), 7.34 (ddd, J=7.5, 4.8, 1.2 Hz, 1H; terpy H5’’), 7.16 (d
(
1.20 g) in water (20 mL) were mixed and heated at reflux for 17 h. On
cooling, the solution was concentrated in vacuo and then diluted with
water (90 mL). The product, which formed a suspension, was extracted
J=7.6 Hz, 1H; Ar), 4.91 (s, 4H; 2CH
2 2
O), 4.83 (s, 4H; 2CH O), 4.80 (s,
2
H; CH O), 2.63 ppm (s, 3H; Me); MS (FAB): m/z (%): 646, 645, 644
2
into CH
The combined extracts were dried over Na
ness, giving the product as a white crystalline solid (1.095 g, 82% yield).
2
Cl
2
until no terpyridine could be detected in the aqueous phase.
+
+
(
100:48:14) [M+H] , 455 (29), 383 (34), 262 (31) [terpyCH
2
O] , 246 (50)
O] , 184 (60), 165 (38); HS-MS (Micro-
2
SO then evaporated to dry-
4
+
+
[
terpyCH
2
] , 199 (73) [bpyCH
2
+
+
1
TOF): m/z: 644.27 [M+H] , 666.25 [M+Na] .
M.p. 1318C; H NMR (200 MHz; CDCl
3
): d=8.65–8.61 (m, 1H; Ar),
Preparation of the complexes 1–11: Stock solution of the ligands BBB,
8
2
7
5
.53–8.49 (m, 2H; Ar), 8.44 (d, J=8.2 Hz, 1H; Ar), 8.33 (d, J=7.9 Hz,
À2
I
BBT, TBT, and TTT in CH
2
Cl
2
(10 m, 100 mL) and of [Cu -
H; Ar), 7.87 (t, J=7.9 Hz, 1H; Ar), 7.80 (td, J=7.7, 1.8 Hz, 1H; Ar),
.73 (dd J=8.1, 2.1 Hz, 1H; Ar), 7.29 (ddd, J=7.5, 4.9, 1.2 Hz, 1H; Ar
II
II
ACHTREUNG
A
H
E
G
3
CN)
4
]BF
4
,
[Cu
A
T
E
N
(CH
3
CN)
4
]
A
H
R
U
G
4
]
2
,
Zn
[BF
4
]
2
·8H
2
O
3
in CH CN
1
3
À2
’’), 4.69 (brs, 2H; CH
): d=156.1, 155.3, 155.1, 154.9 (Ar-C, o-N), 148.9,
47.7 (Ar-CH, o-N), 137.8, 136.9 (Ar-CH, p-N), 136.5 (Ar-CCH OH, m-
2
OH), 3.86 ppm (br, 1H; OH); C NMR
(10 m, 100 mL) were prepared. The appropriate amounts of ligands
(
50.33 MHz; CDCl
3
strands and metallic cations were mixed, the resulting solution evaporat-
1
2
ed and redissolved in pure CH
in complex would be 5.10 m, and injected in the spectrometer without
3
CN (0.500 mL) so that the concentration
À5
N), 135.7 (Ar-CH, p-N), 123.7, 121.3, 120.9 (Ar-CH, m-N), 62.3 ppm
+
+
(
2
CH
2
OH); MS (EI): m/z (%): 264, 263 (19:100) [M] , 262 (35) [MÀH] ,
46 (20) [MÀOH] , 234 (57) [MÀCHO] , 149 (36), 78 (38).
,5’’-Bis(hydroxymethyl)-2,2’-6’,2’’-terpyridine: Solutions of 5,5’’-bis(ace-
OH (50 mL) and NaOH
1.20 g) in water (20 mL) were mixed and heated at reflux for 17 h. On
any further purification.
+
+
[5a]1
Data for 1:
3
H NMR (400 MHz; CD CN): d=8.29 (d, J=9 Hz, 4H;
5
Ar), 8.26 (d, J=9 Hz, 4H; Ar),8.26 (d, J=9 Hz, 4H; Ar), 8.17 (d, J=
9 Hz, 4H; Ar), 8.08 (t, J=9 Hz, 4H; Ar), 7.98 (t, J=9 Hz, 4H; Ar), 7.75
(t, J=9 Hz,4H; Ar), 7.51 (d, J=9 Hz, 4H; Ar), 6.95 (d, J=9 Hz,4H;
toxymethyl)-2,2’-6’,2’’-terpyridine (0.81 g) in CH
(
3
cooling, the solution was concentrated in vacuo and then diluted with
water (50 mL). The product, which formed a suspension, was extracted
Ar), 6.77 (d, J=9 Hz, 4H; Ar), 3.87 (AB, J=12 Hz, 4H; CH
(AB, J=12 Hz, 4H; CH O), 3.68 (AB, J=12 Hz, 4H; CH O), 3.61 (AB,
J=12 Hz, 4H; CH O), 2.16 ppm (s, 12H; CH ); ES-MS: m/z calcd for
2
O), 3.85
2
2
into CHCl
could be detected in the aqueous phase. The combined extracts were
dried over Na SO then evaporated to dryness, giving the product as a
3
(50 mL), the extractions being repeated until no terpyridine
2
3
2+
I
2+
{[Cu
3
A
H
R
U
G
2
]
A
T
U
G
4
)]} =[C72
H
64BCu
3
F
4
N
12
O
4
3+
]
: 719.40; found: 719.16;
450.66; found: 450.47
I
3+
2
4
calcd for [Cu
3
A
H
R
U
G
2
]
=[C72H
64Cu
3
N
12
O
4
]
:
Chem. Eur. J. 2006, 12, 5632 – 5641
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5639

J.-M. Lehn et al.
I
II
(
see Supporting information); additional peak: [Cu
A
H
R
N
A
T
E
N
(BF
4
)]: calcd
Data for 8: ES-MS: no peaks found for the complex [Cu
3
A
H
R
U
G
A
C
H
T
R
E
U
N
G
I
+
+
II
3
ACHTREUNG
for [Cu
A
C
H
T
R
E
U
N
G
(BBB)] =[C36
H
32CuN
6
O
2
A
H
R
U
G
4
]
6
(see text and Supporting Information, Figure S3). [Cu
(TTT)
2
]-
=
II
2+
1
A
H
R
U
G
ES-MS:
Cu
m/z
calcd
for
[Cu
A
H
R
U
G
A
H
T
R
E
U
N
G
(BF ) ]
Data for 2: H NMR (400 MHz; CD
3
4
6
3
2
4 4
2
+
II
3
ACHTREUNG
[
B
4
+
3
F
16
N
18
O
4
]
: 1052.80; found: 1052.80; calcd for [Cu
(TTT)
2
-
H5’), 8.54 (d, J=6.4 Hz, 2H; terpy H3’’), 8.53 (d, J=4.2 Hz, 2H; bpy
H3’’’), 8.41 (d, J=6.4 Hz, 2H; bpy H5’’), 8.34 (d, J=8.4 Hz,2H; bpy H3),
.32 (t, J=8 Hz,2H; bpy H4’’’), 8.31 (d, J=(:! Hz, 2H; bpy H3’), 8.17 (d,
J=6.5 Hz, 2H; terpy H3), 8.15 (t, J=6.3 Hz, 2H; bpy H4), 8.13 (t, J=
.3 Hz, 2H; terpy H4’’’), 7.99 (t, J=6.4 Hz, 2H; bpy H4’’’), 7.92 (t, J=
.4 Hz, 2H; bpy H4’), 7.74 (s, 2H; terpy H6), 7.56 (d, J=5.9 Hz, 4H;
terpy H6’’’, bpy H5), 7.42 (d, J=6 Hz, 2H; bpy H5’’’), 7.37 (dd, J=
.2 Hz, 4.1 Hz, 2H; terpy H5’’’), 7.30 (d, J=6.4 Hz, 2H; terpy H4), 6.99
d, J=6 Hz, 2H; bpy H5’), 6.95 (d, J=6.2 Hz, 2H; bpy H5’’), 4.49 (d, J=
3.33 Hz, 2H; bpy-CH OCHH-tpy), 3.99 (d, J=13.63 Hz, 2H; bpy-
CHHOCH -bpy), 3.95 (d, J=13.67 Hz, 2H; bpy-CH OCHH-bpy), 3.79
d, J=13.30 Hz, 2H; bpy-CH OCHH-terpy), 3.76 (d, J=13.63 Hz, 4H;
bpy-CHHOCH -bpy bpy-CHHOCH -terpy,), 3.66 (d, J=13.67 Hz, 2H;
bpy-CH OCHH-bpy), 3.22 (d, J=13.63 Hz, 2H; bpy-CHHOCH -terpy),
3
3+
A
H
R
U
G
=[C98
H
74
B
3
4+
Cu
3
F
12
N
18
O
4
]
:
672.93; found: 672.80; calcd for
3+
[
2
A
H
R
U
(BF
4
)
2
]
=[C98
H
74
B
2
Cu
3
F
8
N
18
O
4
]
: 483.00; found: 483.10;
: 355.48; found: 355.28;
: 293.06; found: 293.07.
8
II
5+
5+
calcd for [Cu
calcd for [Cu
3
A
H
R
U
G
2
F] =[C98
H
74Cu
3
FN18
O
4
]
II
6+
3+
3
A
H
R
U
G
2
]
=[C98
H
74Cu
3
N
18
II
O
4
]
6
6
I
II
2+
Data for 9: ES-MS: m/z calcd for [Cu Cu Zn
[
A
H
R
U
G
A
H
E
N
(TBT)
A
H
R
E
U
N
G
(BF
4
)
3
I
]
=
II
2
+
II
B
3
2 12 15 4
F N O Zn] : 901.71; found: 901.25; calcd for [Cu Cu Zn -
3
+
3+
A
H
R
U
G
A
H
R
U
G
4
)
2
]
=[C84
H
67
B
2
Cu
2
F
8
N
15
II
O
4
Zn]
:
572.21;
(TBT)
found:
4
(
1
I
II
4+
5
[
for
[Cu Cu Zn
A
H
R
U
G
A
H
R
U
G
A
C
H
T
R
E
U
N
G
(BF
4
)]
=
II
4
+
II
I
2
F
4
5
N
15
O
4
Zn] : 406.58; found: 406.90; calcd for [Cu Cu Zn -
2
+
5+
A
H
R
U
G
H
67Cu
2
N
15
O
4
Zn] : 308.60; found: 308.45. Addi-
2
2
I
II
II
I
ACHTREUNG
tional peaks: [Cu
2
Zn
[BF
Data for 10: ES-MS: m/z calcd for [Cu
A
H
R
U
G
2
]
A
H
R
U
G
4
]
4
(2); [Cu
2
Cu
2 4 5
(BBT) ][BF ] (3);
A
H
R
U
G
(
2
II
2
ACHTREUNG
[
(TBT)
2
]
A
H
R
U
G
4 5
]
2
2
II
II
2+
2
2
2
Zn
Zn] : 982.19; found: 982.9; calcd for [Cu
A
H
R
N
(BBT)
A
H
R
U
G
A
H
T
R
E
U
N
G
(BF
4
)
4
]
=
II
I
II
2+
II
2
.18 ppm (s, 6H; CH
3
-bpy); ES-MS: m/z calcd for [(BBT)
2
Cu
2
Zn -
[C89
B
4
Cu
2
F
16
N
16
O
4
2
Zn -
found:
2
+
2+
3+
2+
A
C
H
T
R
E
U
N
G
(BF
4
I
)
2
]
=[C80
(BBT)
H
66
B
2
Cu
)] =[C80
2
F
8
N
14
O
H
(BBT)
4
Zn] : 826.78; found: 826.17; calcd for
A
H
R
U
A
H
R
U
G
4
)
3
]
=[C89
H
70
B
3
Cu
2
F
12
N
16
II
O
4
Zn]
:
626.83;
(TTT)(BF
II
3+
3+
II
3+
[
Cu
2
Zn
A
C
H
T
R
E
U
N
G
2
A
C
H
T
R
E
U
N
G
(BF
4
66BCu
2
F
4
N
14
O
4
Zn]
:
N
522.25; found:
Zn]
for
[Cu
2
Zn
A
H
R
U
G
A
H
R
U
G
A
H
E
G
4
)
2
II
F]
=
I
2
4+
3+
2+
II
5
21.78; calcd for [Cu
Zn
A
H
R
U
G
2
]
=[C80H
66Cu
2
14
O
4
:
369.98;
[C89
B
2
Cu
2
F
9
N
16
O
4
4
Zn] : 604.23; found: 604.1; calcd for [Cu
2
Zn -
+
2+
A
H
R
U
G
A
H
R
U
G
4
)F] =[C89
H
70BCu
2
F
5
N
16
O
4
Zn] : 431.37; found: 431.3;
found: 369.60, 445.12.
II
II
4+
2+
II
I
+
A
H
R
U
G
A
H
R
U
G
2
]
=[C89
H
70Cu
2 2 16 4
F N O Zn] : 414.52;
Data for 3: ES-MS: m/z calcd for [Cu
2
Cu
A
H
R
U
2
A
H
R
U
G
4
)
4
]
=
II
II
5+
+
for
[Cu
2
Zn
A
H
R
U
G
A
H
R
U
G
=
[
[
C
Cu
80
II
H
66
B
4
Cu
(BBT)
3
F
16
N
14
O
4
] : 1825.33; found: 1825.85; 1187.85; calcd for
2
+
II
II
ACHTREUNG
I
2+
2+
[C89
2
FN16
II
O
H
(TTT)
4
Zn] : 327.81; found: 327.6; calcd for [Cu
2
Zn
(BBT)-
2
Cu
A
C
H
T
R
E
U
N
G
2
A
C
H
T
R
E
U
N
G
(BF
4
)
3
]
=[C80
H
66
B
3
Cu
(BF
(BBT) (BF
calcd
: 374.27; found: 374.08; calcd for [Cu
3
F
12
N
14
3+
O
4
]
:
868.17;
Cu
66BCu
found:
6
+
6+
II
I
3+
A
H
R
U
G
70Cu
2
N
16
O
4
Zn]
:
270.01; found: 269.9; additional
8
5
3
68.85; calcd for [Cu
2
II
Cu
A
H
R
U
G
2
A
T
E
N
4
)
2
]
=[C80
H
66
B
2
3
F
F
8
4
N
N
14
O
O
F]
4
]
]
:
:
II
II
ACHTREUNG
peaks: [Zn
3
A
H
E
N
2
]
A
H
R
U
G
4
]
6
(5); [Cu
A
H
R
U
G
A
H
R
U
G
4 2
[BF ] : calcd for [Cu (BBT)-
I
4+
4+
50.51; found: 550.28, [Cu
2
Cu
A
H
R
U
G
2
A
H
R
U
4
)] =[C80
H
3
14
I
2
4
+
II
A
H
R
N
H
H
33BCuF
33CuN
4
N
7 2
O
II
4+
91.22; found: 391.07;
for
[Cu
2
Cu
A
H
R
U
G
=
2
+
2+
A
H
E
N
7
O
2
]
: 353.64; found: 353.45.
4
+
II
I
5+
[
C
80
H
66Cu Cu
3
FN14
O
4
]
2
A
H
R
U
G
2
(BBT) ]
II
II
ACHTREUNG
5
+
II
2
(TBT)-
=
[C80 66Cu N O ] : 295.62; found: 295.38; additional peaks: [Cu -
H
3 14 4
II
II
II
II
2+
2+
A
H
R
U
G
A
H
R
U
G
4
]
6
. [Zn
H
H
3
A
T
E
N
(TTT)
2
]
A
H
R
N
[BF
4
]
6
(6); [Zn
A
H
E
N
(TBT)]
A
T
E
N
4 2
[BF ] : calcd for [Zn -
Zn] : 857.21; found: 857.2; calcd for [Zn -
A
C
H
T
R
E
U
N
G
(BBT)]
found: 353.45.
Data for 4: ES-MS: m/z calcd for [Zn
A
C
H
T
R
E
U
N
G
[BF
4
]
2
:
calcd for [Cu
A
H
R
U
G
H
33CuN
7
O
2
]
:
353.64;
2
+
+
II
A
H
R
U
G
34BF
34ZnN
4
N
8
O
2
2+
A
H
R
U
G
8
O
2
]
: 385.10; found: 385.2.
II
I
2+
2
Cu
A
H
R
U
G
2
A
H
R
U
G
4
)
3
II
]
=
2
+
I
[
C
88
H
68
B
3
CuF12
N
=[C88
16
O
4
Zn
H
2
]
: 934.16; found: 933.66; calcd for [Zn
2
Cu -
3
+
2+
A
C
H
T
R
E
U
N
G
(TBT)
2
A
C
H
T
R
E
U
N
G
(BF
4
)
2
]
68
B
2
CuF
8
N
16
4+
O
4
Zn
2
]
:
593.84; found: 593.25;
II
I
4+
calcd for [Zn
found: 423.44; calcd for [Zn
3
2
Cu
A
C
H
T
R
E
U
N
G
(TBT)
2
A
H
R
U
G
4
)] =[C88
H
68BCuF
4
N
16
O
H
4
Zn
68CuN16
2
]
:
423.68;
II
I
5+
4+
2
Cu
A
H
R
U
G
2
]
=[C88
O
4
Zn
2
]
:
Acknowledgements
II
21.58; found: 321.33, 257.10 additional peaks: [Zn
A
H
R
U
G
A
H
R
U
G
4
]
2
: calcd
II
+
+
for [Zn
A
C
H
T
R
E
U
N
G
(TBT)
A
C
H
T
R
E
U
N
G
(BF
4
)] =[C44
H
34BF
4
N
8
O
2
Zn]
857.21; found: 857.21;
385.10; found: 385.40;
We are grateful to Dr. J.-P. Kintzinger and Dr. R. Graff for their help
with the NMR investigations. This work was supported by a doctoral fel-
lowship from the European Postgraduate School (GRK 532) to H.H.
II
2+
2+
calcd for [Zn
454.0]
Data for 5: ES-MS: m/z calcd for [Zn
A
C
H
T
R
E
U
N
G
34 8 2
(TBT)] =[C44H N O Zn]
[
II
2+
3
A
H
R
N
2
A
H
R
U
G
4
)
4
]
=
2
+
II
[
C
98
H
74
B
4
+
F
16
N
18
O
4
Zn
3
]
3
ACHTREUNG
2
-
3
3+
A
C
H
T
R
E
U
N
G
(BF
4
II
)
3
]
=[C98
H
74
B
]
3
4+
F
12
N
18
O
H
4
Zn
3
]
:
[1]J.-M. Lehn, Angew. Chem. 1990, 102, 1347; Angew. Chem. Int. Ed.
Engl. 1990, 29, 1304.
2]J.-M. Lehn, Supramolecular Chemistry—Concepts and Perspectives,
VCH, Weinheim, 1995, Chapter 9.
4+
[
Zn
3
A
C
H
T
R
E
U
N
G
(TTT)
2
A
C
H
T
R
E
U
N
G
(BF
4
)
2
=[C98
74
5+
B
2
F
8
N
18 4 3
O Zn ] : 484.38; found: 484.20;
II
5+
[
calcd for [Zn
3
A
C
H
T
R
E
U
N
G
((TTT)
2
BF
4
)] =[C98
H
74BF
H
4
N
18
O
O
4
Zn
3
]
: 370.14; found:
: 356.58; found:
: 293.99; found:
II
5+
5+
3
3
2
69.88; calcd for [Zn
3
A
C
H
T
R
E
U
N
G
(TTT)
2
F] =[C98
74FN18
4
Zn
3
6+
]
II
6+
[3]J.-M. Lehn, Chem. Eur. J. 2000, 6, 2097.
[
56.68; calcd for [Zn
93.74.
3
A
H
R
U
G
2
]
=[C98
H
74
N
18
O
4
Zn
3
]
4]For selected references, see: a) C. Piguet, G. Bernardinelli, G. Hopf-
gartner, Chem. Rev. 1997, 97, 2005; b) E. C. Constable, Tetrahedron
I
II
2+
Data for 6: ES-MS: m/z calcd for [Cu
2
Cu
A
H
R
U
G
A
T
U
G
A
H
R
U
G
4
)
2
]
=
1
992, 48, 10013; c) M. Albrecht, Chem. Rev. 2001, 101, 3457; d) D. S.
2
+
I
II
ACHTREUNG
[
C
76
H
65
B
2
Cu
3
3
F
8
+
N
13
O
4
]
: 793.16; found: 793.30; calcd for [Cu
H65BCu F N O ] : 500.62; found: 500.45; calcd for
2
Cu
(BBB)-
Lawrence, T. Jiang, M. Levett, Chem. Rev. 1995, 95, 2229; e) M.
Fujita, Chem. Soc. Rev. 1998, 27, 417; f) D. L. Caulder, K. N. Ray-
mond, Acc. Chem. Res. 1999, 32, 975; g) R. W. Saalfrank, B. Dem-
leitner, Transition Metals in Supramolecular Chemistry (Ed.: J.-P.
Sauvage), Wiley, New York, 1999, Chapter 1, pp. 1–51; h) S. Lei-
ninger, B. Olenyuk, P. J. Stang, Chem. Rev. 2000, 100, 853; i) G. F.
Swiegers, T. J. Malefetse, Chem. Rev. 2000, 100, 3483; j) M. Al-
brecht, J. Incl. Phenom. Macro. Chem. 2000, 36, 127; k) W.-Y. Sun,
M. Yoshizawa, T. Kusukawa, M. Fujita, Curr. Opin. Chem. Biol.
2002, 6, 757; l) S. R. Seidel, P. J. Stang, Acc. Chem. Res. 2002, 35,
972; m) M. D. Ward, Annu. Rep. Prog. Chem. Sect. A 2002, 98, 285;
n) F. Würthner, C.-C. You, C. R. Saha-Mçller, Chem. Soc. Rev. 2004,
33, 133; o) M. Ruben, J. Rojo, F. J. Romero-Salguero, L. H. Uppa-
dine, J.-M. Lehn, Angew. Chem. 2004, 116, 3728; Angew. Chem. Int.
Ed. 2004, 43, 3644.
3+
A
C
H
T
R
E
U
N
G
(BBT)
Cu
A
C
H
T
R
E
U
N
G
(BF
4
)] =[C76
(BBB)(BBT)] =[C76
ditional peaks: [Cu
3 4 13 4
I
2
II
4+
+
[
Cu
A
C
H
T
R
E
U
N
G
A
C
H
T
R
E
U
N
G
3 13 4
H65Cu N O ] : 353.76; found: 353.60; ad-
I
3
I
II
ACHTREUNG
A
C
H
T
R
E
U
N
G
(BBB)
2
]
A
H
R
U
G
4
]
3
(1); [Cu
2
Cu
(BBT)
2
]
A
H
R
U
G
4
]
4
: calcd for
I
2
II
2+
2+
[
Cu
Cu
A
C
H
T
R
E
U
N
G
(BBT)
2
A
C
H
T
R
E
U
N
G
(BF
4
)
2
]
=[C80
Cu
H
66
B
2
Cu
3
F
8
N
14
3+
O
4
]
:
found:
I
II
3+
8
5
3
25.05; calcd for [Cu
21.64; found: 521.45; calcd for [Cu
69.53; found: 369.43.
2
A
H
R
U
G
2
A
H
R
U
G
4
)] =[C80
H
66BCu
3
F
4
3
N
N
14
O
O
4
]
]
:
:
I
II
3+
3+
2
Cu
A
H
R
U
G
2
]
=[C80H66Cu
14
4
II
2
I
2+
Data for 7: ES-MS: m/z calcd for [Cu
Cu
A
H
R
U
G
A
T
E
N
(TBT)
A
H
R
U
G
4
)
3
II
]
=
2
+
I
[
C
80
H
66
B
3
Cu
(TBT)
for
3
F
12
N
14
O
4
2
]
:
869.26; found: 869.20; calcd for [Cu
2
Cu -
+
2+
A
C
H
T
R
E
U
N
G
(BBB)
A
C
H
T
R
E
U
N
G
A
C
H
T
R
E
U
N
G
(BF
4
)
2
F] =[C80
H
66
B
2
Cu
3 9 14 4
F N O ] : 835.36; found: 835.30;
II
I
3+
3+
calcd
5
[Cu
2
Cu
A
C
H
T
R
E
U
N
G
(BBB)
A
H
R
U
G
A
H
R
U
G
4
)
2
II
]
=[C80
H
(BBB)
66
B
2
Cu
3
F
8
N
14
O
)F]
4
]
:
I
3+
50.57; found: 550.60; calcd for [Cu
2
Cu
A
H
R
U
G
A
H
E
N
A
H
R
U
G
4
I
=
3
+
II
[
C
80
H
66BCu
3
F
5
N
14
O
4
]
: 527.90; found: 527.90; calcd for [Cu
(TBT)F] =[C80H66Cu FN14O ] : 374.17; found: 374.20; calcd for
3 4
A
H
R
U
G
4
+
4+
A
C
H
T
R
E
U
N
G
II
2
I
5+
5+
[
Cu
Cu
A
C
H
T
R
E
U
N
G
(BBB)
A
C
H
T
R
E
U
N
G
(TBT)] =[C80
H
66Cu
3
N
14
O
4
]
:
295.62; found: 295.67;
[5]Double helicates with four-coordinate centers: a) J.-M. Lehn, A.
Marquis-Rigault, J. Siegel, J. M. Harrowfield, B. Chevrier, D. Moras,
I
additional peaks: [Cu
3
A
C
H
T
R
E
U
N
G
2 4 3
(BBB) ][BF ] (1).
A
T
E
N
5640
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 5632 – 5641

Helical Structures
FULL PAPER
Proc. Natl. Acad. Sci. USA 1987, 84, 2565; b) J.-M. Lehn, A. Rigault,
Angew. Chem. 1988, 100, 1121; Angew. Chem. Int. Ed. Engl. 1988,
[14]M. M. Harding, U. Koert, J.-M. Lehn, A. Marquis-Rigault, C.
Piguet, J. Siegel, Helv. Chim. Acta 1991, 74, 594.
2
7, 1095; c) T. M. Garrett, U. Koert, J.-M. Lehn, A. Rigault, D.
[15]a) H. Hofmeier, E. Herdtweck, U. Schubert, Z. Anorg. Allg. Chem.
2004, 630, 683; b) B. Hasenknopf, Ph.D. Thesis, UniversitØ Louis
Meyer, J. Fischer, J. Chem. Soc. Chem. Commun. 1990, 557; d) A.
Pfeil, J.-M. Lehn, J. Chem. Soc. Chem. Commun. 1992, 838.
6]B. Hasenknopf, J.-M. Lehn, G. Baum, D. Fenske, Proc. Natl. Acad.
Sci. USA 1996, 93, 1397.
7]V. C. M. Smith, J.-M. Lehn, Chem. Commun. 1996, 2733.
8]a) S. Floquet, M. Borkovec, G. Bernardinelli, A. Pinto, L.-A. Leut-
hold, G. Hopfgartner, D. Imbert, J.-C. G. Bünzli, C. Piguet, Chem.
Eur. J. 2004, 10, 1091; b) J. Mathieu, N. Ghermani, N. Bouhmaida,
B. Fenet, A. Marsura, Eur. J. Inorg. Chem. 2004, 3172.
II
Pasteur, Strasbourg, 1996, p. 94; c) the coordinative plasticity of Cu
[
centers is also reflected in the medium-induced adaptive self-assem-
bly of metallomacrocycles, see: P. N. W. Baxter, R. G. Khoury, J.-M.
Lehn, G. Baum, D. Fenske, Chem. Eur. J. 2000, 6, 4140.
[
[
II
[16]The formation of (B, T, Cu ) units has also been used for perform-
ing metal-directed complementary self-assembly on solid support,
see: E. C. Constable, C. E. Housecroft, J. N. Lambert, D. A. Malar-
ek, Chem. Commun. 2005, 3739.
[
9]a) R. Krꢂmer, J.-M. Lehn, A. Marquis-Rigault, Proc. Natl. Acad.
Sci. USA 1993, 90, 5394; for recent reports on self-recognition see
for instance: b) D. L. Caulder, K. N. Raymond, Angew. Chem. 1997,
[17]D. P. Funeriu, J.-M. Lehn, K. M. Fromm, D. Fenske, Chem. Eur. J.
2000, 6, 2103.
[18]D. P. Funeriu, K. Rissanen, J.-M. Lehn, Proc. Natl. Acad. Sci. USA
2001, 98, 10546.
1
09, 1508; Angew. Chem. Int. Ed. Engl. 1997, 36, 1440; c) E. J. Ene-
mark, T. D. P. Stack, Angew. Chem. 1998, 110, 977; Angew. Chem.
Int. Ed. 1998, 37, 932; d) R. Stiller, J.-M. Lehn, Eur. J. Inorg. Chem.
[19]For a case of assembly controlled by steric effects see: M. Tominaga,
T. Kusukawa, S. Sakamoto, K. Yamaguchi, M. Fujita, Chem. Lett.
2004, 33, 794.
[20]A. Livoreil, C. O. Dietrich-Buchecker, J.-P. Sauvage, J. Am. Chem.
Soc. 1994, 116, 9399.
1
1
998, 977; e) M. Albrecht, M. Schneider, H. Rçttele, Angew. Chem.
999, 111, 512; Angew. Chem. Int. Ed. 1999, 38, 557; f) A. Wu, L.
Isaacs, J. Am. Chem. Soc. 2003, 125, 4831; g) M. Schmittel, V. Kalsa-
ni, R. S. K. Kishore, H. Cçlfen, J. W. Bats, J. Am. Chem. Soc. 2005,
[21]M. Griep, S. Whitney, M. Nelson, H. Viljoen, AIChE J. 2006, 52,
384.
1
27, 11544, and references therein.
[
[
10]a) J.-M. Lehn, Chem. Eur. J. 1999, 5, 2455; b) S. Otto, R. Le Furlan,
J. K. M. Sanders, Curr. Opin. Chem. Biol. 2002, 6, 321.
11]a) J.-M. Lehn, Proc. Natl. Acad. Sci. USA 2002, 99, 4763; b) J.-M.
Lehn, Science 2002, 295, 2400; c) J.-M. Lehn, Prog. Polym. Sci. 2005,
[22]For a recent review on self-replicating systems see: A. Robertson,
A. J. Sinclair, D. Philp, Chem. Soc. Rev. 2000, 29, 141.
[23]H. E. Gottlieb, V. Kotlyar, A. Nudelman, J. Org. Chem. 1997, 62,
7512.
3
0, 814.
[
[
12]B. Hasenknopf, J.-M. Lehn, Helv. Chim. Acta 1996, 79, 1643.
13]A. Marquis-Rigault, A. Dupont-Gervais, A. Van Dorsselaer, J.-M.
Lehn, Chem. Eur. J. 1996, 2, 1395.
Received: February 2, 2006
Published online: May 8, 2006
Chem. Eur. J. 2006, 12, 5632 – 5641
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5641