From supramolecular porphyrin tweezers to dynamic AnB mClDk multiporphyrin arrangements through orthogonal coordination

Article abstract of DOI:10.1002/chem.200600463

A dynamic, supramolecular, three-component A_nB_mC _l bis(zinc porphyrin) tweezer has been prepared quantitatively using the heteroleptic bisphenanthroline (HETPHEN) concept. Upon addition of nitrogenous spacers of different

Full text of DOI:10.1002/chem.200600463

DOI: 10.1002/chem.200600463
From Supramolecular Porphyrin Tweezersto Dynamic A BmClDk
n
Multiporphyrin ArrangementsThrough Orthogonal Coordination
Ravuri S. K. Kishore, Thomas Paululat, and Michael Schmittel*[a]
Abstract: A dynamic, supramolecular,
three-component AnBmCl bis(zinc por-
phyrin) tweezer has been prepared
quantitatively using the heteroleptic
bisphenanthroline (HETPHEN) con-
cept. Upon addition of nitrogenous
spacers of different length, namely, the
extended bipyridine 3a, 4,4’-bipyridine
Nspacer), three structurally different, still
dynamic, four-component AnBmClDk
assemblies were cleanly formed, as in-
dicated by UV/Vis and NMR titrations
as well as by DOSY investigations. The
structures were identified as a bridged
monotweezer A2BC2D,
bridged double tweezer A4B2C4D2, and
triply bridged double tweezer
A4B2C4D3, the latter resembling a por-
phyrin stack. Notably, the same struc-
tures were equally formed directly
from a mixture of the constituents A,
B, C, and D put together in any se-
quence if the correct stoichiometry was
applied.
a
doubly
a
Keywords: N ligands · nanostruc-
tures · porphyrinoids · self-assem-
bly · supramolecular chemistry
(3b), and 1,4-diazabicyclo
A
(DABCO; 3c), to set up an additional
orthogonal binding motif (ZnPor
–
Introduction
and as a whole unit. An ideal mimic of the natural process
would therefore require the self-assembly of various compo-
nents, each supplying an input through specific molecular in-
formation and orthogonal binding modes, which under equi-
librium conditions would develop into one single aggregate.
Over the past two decades there have been zealous at-
tempts to fabricate intricate supramolecular structures by
utilizing diverse intermolecular interactions and multiple
molecular components.[2] At present, higher-order, nanoscale
supramolecular structures are forged by the use of multiples
of up to three components and a single binding algorithm
(AnBm or AnBmCl),[1,2] while, surprisingly, very few examples
are known with four or more molecular components
(AnBmClDk).[3] Hence, in spite of some fascinating examples
of supramolecular assemblies by Lehn, Fujita, Stang, Rein-
houdt, and many others,[4] the design and preparation of
higher-order aggregates with four or more different compo-
nents[5] (i.e., modular building blocks) using orthogonal
complexation motifs[6] still remain a challenge. The clear so-
The ultimate goal of supramolecular science is to attain a
level of complexity and sophistication in structure and func-
tion similar to that commonly observed in natural supramo-
lecular assemblies.[1] Nature adopts an omnifarious strategy
for the creation of tertiary supramolecular structures from
simple molecules by exploiting a combination of noncova-
lent and weak interactions in a reversible and thermody-
namically controlled process, to finally achieve the required
topology along with an appreciable conformational rigidity
for high-end physiological functions. A major factor respon-
sible for such exceptional organization in natural systems is
the use of modular and hierarchical self-assembly. The mo-
lecular information necessary for self-assembly is designated
to and disseminated over multiple levels of organization,
which results in a structural and functional hierarchy that
delivers a high-fidelity output at each level of organization
A
[a] R. S. K. Kishore, Dr. T. Paululat, Prof. Dr. M. Schmittel
Center of Micro- and Nanochemistry and Engineering
Organische Chemie, Universität Siegen
are co-existent in the final assembly. The present study is an
endeavor in this direction.
Adolf-Reichwein-Strasse, 57068 Siegen (Germany)
Fax : (+49)271-740-3270
E-mail: schmittel@chemie.uni-siegen.de
Supporting information (experimental procedures, 1H NMR spectra,
ESI-MS data, Jobs plot analyses, and DOSY plots) for this article is
available on the WWW under http://www.chemeurj.org/ or from the
author.
Starting from our ongoing work directed toward hetero-
leptic nanoscale architectures,[7] we decided to elaborate the
three-component self-assembly into a four-component one
utilizing noninterfering orthogonal binding algorithms under
thermodynamic equilibration conditions (Figure 1). The CuI
heteroleptic bisphenanthroline (HETPHEN) complexation
8136
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8136 – 8149

FULL PAPER
motif[8] was thus supplemented by the ZnPor–Nspacer (zinc por-
phyrin–DABCO/bipyridyl; DABCO=1,4-diazabicyclo-
[2.2.2]octane) interaction, a motif that has been used earlier
metal salt. The phenanthroline–porphyrin hybrid 1 and bis-
phenanthroline 2 were synthesized by Sonogashira coupling
A
reactions based on earlier reports.[7d,13] Equally, 3a was pre-
pared in 68% yield by a Sonogashira coupling based on an
analogous compound [Eq. (1)].[14]
to assemble cofacial higher-order porphyrin structures[9] and
to study the dynamics therein.[10] Interest in such systems
arises from the fact that successful light harvesting occurs
solely by congruous arrangement of constituent porphyrins
and appropriate cofacial stacking.[11] Moreover, cofacial por-
phyrin arrangements have raised vast interest due to their
suitability for controlled electrochemical processes, such as
the four-electron reduction of dioxygen to water.[12] Al-
though considerable research has gone into the generation
of highly preorganized, cofacially arranged bis-porphyrins,
almost all studies concern a covalently designed frame-
work.[9,10] The present work not only describes for the first
time the preparation of supramolecular tweezers, but also
demonstrates how the tertiary structure of a supramolecular
arrangement can be drastically modified as a function of the
length of an orthogonally operating building block.
Results and Discussion
Preparation of porphyrin tweezer PT by a three-component
assembly: As a result of the HETPHEN algorithm[8] im-
planted into the ligands, quantitative heteroleptic complexa-
The components used in this study include the phenanthro-
line-appended porphyrin 1, the linear bisphenanthroline 2,
the spacers 3a–c [1,4-bis(4’-pyridylethynyl)durene (3a;
tion of ligand 1 in the presence of [Cu(CH3CN)4]PF6 with
A
d
NN =16 Š), 4,4’-bipyridine (3b; dNN =7 Š), and DABCO
1,10-phenanthroline (5) and bisphenanthroline 2 was ob-
served, which afforded 4 and the supramolecular porphyrin
tweezer PT, respectively, as ex-
clusive products (Scheme 1).
The formation of PT is reminis-
cent of the formation of other
supramolecular racks whose
structure (by X-ray diffraction)
and dynamics have been inves-
tigated extensively.[7f]
(3c; dNN =3 Š)], and [Cu(CH3CN)4]PF6 as the coordinating
AHCTREUNG
Complexes 4 and PT were
characterized by ESI-MS, and
gave signals of 100% intensity
at m/z 1580 and 1825 (dication),
respectively. The isotopic split-
ting patterns obtained in ESI-
MS fitted accurately with the si-
mulated splitting patterns (Fig-
ure S4 in Supporting Informa-
tion). Moreover, the 1H NMR
spectra displayed a characteris-
tic upfield shift of the 3’,5’-me-
sitylene protons (3’,5’-MesH) of
1 due to shielding from the
second phenanthroline p system
(5 or 2).
The 1H NMR spectrum of
complex 4 shows a single signal
for the enantiotopic 3’,5’-MesH
due to the plane of symmetry
ꢀ
Figure 1. A four-component self-assembly process that is controlled by the length of the linear ligand 3.
along the phenanthroline– –
Chem. Eur. J. 2006, 12, 8136 – 8149
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8137

M. Schmittel et al.
Scheme 2. Dynamic equilibration of PT by dissociation/association and
bond rotation. The roman numerals represent the assignment to the five
sets of Mes-CH3 protons which appear in the 1H NMR spectrum of PT.
five sets (Figure S6 in Supporting Information) assignable to
a set of rapidly equilibrating configurations (meso and P,P/
M,M) as shown in Scheme 2. Due to the dynamic nature of
its CuI metal ion–ligand interaction, PT cannot be isolated
Scheme 1. Synthesis of 4 and the supramolecular porphyrin tweezer PT.
1
porphyrin axis. In contrast, the porphyrin assembly PT ex-
hibits two sets of diastereotopic protons due to the nonequi-
valence of the 3’,5’-MesH (Figure 2), since the plane of sym-
metry has disappeared. Similarly, Mes-CH3 protons of the
porphyrin (2’’’,4’’’,6’’’-CH3) in 4 at dꢁ1.83–1.86 ppm appear
only as two singlets (Figure S5 in Supporting Information),
while the corresponding Mes-CH3 protons in PT show up as
or differentiated by H NMR spectroscopy. In addition, vari-
ous conformations (cisoid=Pac-Man and transoid) may
arise because of rotation about the bisphenanthroline axis.
Although the porphyrin units should be free to rotate per-
pendicularly to the bisphenanthroline backbone, free rota-
tion is only possible in the transoid conformation. In the
cisoid conformation, the simultaneous rotation of the two
Figure 2. Comparison of the aromatic region in the NMR spectra of PT (top) and 4 (bottom).
8138
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8136 – 8149

Supramolecular Multicomponent Porphyrin Tweezers
FULL PAPER
porphyrins of PT is prohibited due to hindrance caused by
the mesityl groups, thus confining cisoid-PT to a tweezer-
like conformation with cofacial porphyrins as shown in
Scheme 2 (left).
7.5 ppm) at the porphyrin (PorMes-H) were found to shift
upfield increasingly in the series PT<PT·3a<PT·3bꢁ
PT·3c (Figure 3). The signal pattern of the 3’,5’-Mes-H pro-
tons at d=6.2 ppm served as a diagnostic tool for the sym-
metry of the molecule and provided insight into the possible
structure. Thus, while a single signal was observed for 4, two
singlets were noticed in PT due to a breakdown in symme-
try. Similarly, one singlet was noticed with PT·3a, while four
singlets were observed for PT·3b,c. Notably, a distinct simi-
Four-component assemblies: The porphyrin tweezer PT dis-
plays a “Pac-Man”-like arrangement in its cisoid (P,M) con-
figuration exhibiting a distance of 19 Š between the cofacial
porphyrins, as evaluated from the PM3 optimized model.[15]
Most known Pac-Man porphyrins[11] require an optimum dis-
larity of the various shifts was noticed in the H NMR spec-
1
[9a,b]
tance of 7 Š
for a single “bite” complexation to a
trum of PT·3b (1:1) and PT·3c (1:1.5). It could thus be
speculated that a similar structural composition is achieved
in both the complexes.
DABCO molecule. In the case of larger inter-porphyrin dis-
tances, as in PT, clearly a longer bis-nitrogen ligand is
needed to achieve a 1:1 PT·(bidentate base) host–guest ag-
gregate.[16]
After the initial construction of a 1:1 assembly of PT and
3a (dNN =16 Š), it was decided to shorten the N···N distance
of the bis-nitrogen ligand by using 3b (dNN =7 Š) and 3c
(dNN =3 Š) and to investigate the nature of the assemblies
formed by UV and NMR spectroscopy. Upon addition of
3a–c to PT distinct changes were noticed in the spectra, as
summarized in Table 1. A Jobs plot analysis revealed an op-
Further evidence for the formation of the adducts was de-
rived from the significant 1H NMR shifts of the diaza
spacers 3a–c. In the case of PT·3a, the 3,3’ protons of 3a at
d=8.61 ppm shifted upfield to 6.50 and 6.70 ppm (Figure 3),
as is known for related complexes.[9a,b] Additionally, the
signal of the CH3 protons of 3a shifted upfield from d=2.50
to 2.10 ppm. In PT·3b, the signals of the 2,2’ protons of 3b
shifted upfield to d=2.32 ppm, while those of the
3,3’ protons were observed at d=4.96 ppm.
The addition of 1.5 equiva-
lents of DABCO (3c) to PT
(based on the stoichiometry ob-
tained from the Jobs plot) pro-
Table 1. Data from UV/Vis and NMR spectroscopy investigations (for numbering, see Figure 1).
Jobs plot Soret band
analysis[a] abs [nm][a]
d of Hb
[ppm][b]
d of porphyrin
d of 3’,5’-MesH [ppm][b]
A
Mes-CH3 [ppm][b]
duced
substantial
changes
PT
420
8.69–8.78 (m) 1.17–1.92 (m)
6.12 (s), 6.16(s)
which were monitored by NMR
PT·3a 1:1
PT·3b 1:1
PT·3c 1:1.5
430
8.64–8.70 (m) 1.79 (s), 1.81(s), 1.82 6.12 (s)
(s)
1
and UV spectroscopy. H NMR
425 (1:1), 430
(1:2)
425 (1:1.5), 430
(1:2)
8.50–8.63 (m) 1.56–1.60 (m)
6.05 (s), 6.06 (s), 6.16 (s), 6.19
titration of PT with 3c at the
millimolar concentration level
provided particular insight.
Upon addition of one equiva-
lent of 3c, two signals at d=
(s)
8.28–8.35
(m)[c]
1.43–1.51 (m)
6.06 (s), 6.09 (s), 6.21 (s), 6.23
(s)[c]
[a] Measured in CH2Cl2 at 258C. [b] Measured in CD2Cl2. [c] NMR measured at a PT·3c ratio of 1:1.5.
timum 1:1 stoichiometry for
PT·3a and PT·3b, but a repro-
ducible ratio of 1:1.5 was ob-
tained for PT·3c (Figure S17 in
Supporting Information).
NMR analysis: In all adducts at
ideal stoichiometry (based on
the Jobs plot analysis), that is,
PT·3a,b (1:1) and PT·3c (1:1.5),
the porphyrin Hb signals exhib-
ited significant upfield shifts
due to axial coordination of
ligand 3, the degree varying
with the binding affinity.[9,10]
While the upfield shift was
Dd(Hb)=0.05 ppm for PT·3a
versus PT, Dd(Hb) was in-
creased to 0.2 ppm for PT·3b
and to 0.4 ppm for PT·3c. In
Figure 3. Comparison of PT with PT·3a, PT·3b (1:1 composition), and PT·3c (1:1.5 composition). (*) Porphyr-
the same manner, the 3’’’,5’’’
mesitylene protons (d=7.0– 3a.
~
*
in Hb signals in PT·3a–c; ( ) protons corresponding to 3’,5’-MesH; ( ) upfield-shifted 3,3’-pyridyl protons of
Chem. Eur. J. 2006, 12, 8136 – 8149
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8139

M. Schmittel et al.
1
ꢂ4.5 and ꢂ4.7 ppm emerged in the H NMR spectrum. As a
Hb signals from d=8.78 to 8.35 ppm until 1.5 equivalents
had been added. Subsequent addition of 3c led to a small
downfield shift to d=8.53 ppm. The signal corresponding to
the 3’,5’-MesH protons at d=6.12 and 6.16 ppm gradually
disappeared as the titration proceeded, thus giving rise to
four singlets at d=6.0, 6.09, 6.21, and 6.23 ppm upon addi-
tion of 1.5 equivalents of 3c. After subsequent addition of
3c the signals gradually coalesced back to two signals. Also,
a clear change was observed in the signal at d=9.02 ppm
corresponding to the 4-H of the phenanthroline and the aa’
and bb’ signals on PT. Upon addition of 1.5 equivalents of
3c, a new set of signals of equal integrated area correspond-
ing to 4-H and aa’ and bb’ was noticed. The signals, upon
subsequent addition of 3c, coalesced into a single set. The
above-noted changes point to the transformation of PT into
PT·3c (2:3) upon addition of up to 1.5 equivalents of 3c,
while addition of larger amounts of 3c led to complex
PT·3c with a 1:2 composition.
sharp singlet in this region is unequivocal proof for the for-
mation of a porphyrin–3c (2:1) complex,[9,10] the presence of
two signals indicates two distinct environments of the
DABCO methylene protons. The signal at d=ꢂ4.5 ppm ap-
peared as a sharp singlet from the very first few aliquots
added, whereas the latter developed gradually and gained
equal integrated peak area and intensity during addition
until a 1:1 mole ratio of DABCO to PT was reached. As a
COSY spectrum of PT·3c (1:1) revealed no cross-peaks be-
tween the two signals, we concluded that two distinct
DABCO–porphyrin sandwich complexes were present in the
assembly. The formation of the second complex is preceded
by the first complex, as seen from the development in the
NMR titration.
At a ratio of PT/3c=1:1.5, the aromatic region evolved
into sharp assignable signals, while the signals in the nega-
tive region coalesced into a broad singlet at d=ꢂ4.2 ppm
(Figure 4). Subsequent addition of DABCO (3c) gradually
UV/Visintviegsations
:
For
monitoring the binding of 3a–c
to PT, the change in the Soret
band of the zinc porphyrin unit
of PT proved to be of great
help.
Formation of PT·3a: During
the sequential addition of 3a to
PT, the Soret absorption band
exhibited a bathochromic shift
from 420 to 430 nm due to the
axial binding of the pyridyl resi-
dues of 3a to the two zinc por-
phyrin units of PT. The data
obtained from the UV/Vis titra-
tion were subjected to global
Figure 4. Evolution of the upfield-shifted DABCO signals in the NMR spectra (left) and graphically (right).
shifted the signal further down-
field until it was no longer ob-
servable at PT/3c=1:2.5.
A
1H COSY spectrum of PT·3c
(1:2) revealed cross-peak signals
between the broad signal now
at d=ꢂ1.6 ppm with a singlet
at 1.8 ppm. This behavior is
consistent with
a porphyrin–
DABCO 2:1 sandwich complex
gradually breaking down into a
1:1 complex upon addition of
DABCO.[10h]
The titration provided addi-
tional insight, as judged by the
change of the signals between
d=5 and 9 ppm (Figure 5). For
example, sequential addition of
3c to complex PT gave rise to
1
an upfield shift of the porphyrin Figure 5. Evolution of the H NMR spectra of PT upon addition of DABCO (3c).
8140
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8136 – 8149

Supramolecular Multicomponent Porphyrin Tweezers
FULL PAPER
curve fitting by SPECFIT,[17] which provided an association
constant for PT·3a, that is, logb of 6.7 (ꢃ0.6) and K=5”
106 mꢂ1 (see Table 2). The analysis by both SPECFIT and the
Jobs plot supported a 1:1 ratio for PT·3a in the final assem-
bly.
Formation of PT·3c: Addition of one equivalent of
DABCO (3c) to PT revealed a characteristic red shift of the
Soret band from 420 to 425 nm that was paralleled by analo-
gous red shifts in the Q-band region. Based on earlier stud-
ies by Hunter and others,[19] this finding provided tentative
evidence for the binding of 3c within a porphyrin sandwich.
To gain further insight, spectrometric titrations were carried
out at micromolar concentrations and monitored at the
Soret band. The peak at 420 nm corresponding to the un-
bound zinc porphyrin decreased in intensity, with a new
band appearing at 425 nm that is characteristic of a porphy-
Formation of PT·3b: The sequential addition of 3b to PT
monitored by UV/Vis spectroscopy displayed a titration pro-
file markedly different from that of the 1:1 adduct PT·3a.
Noticeably, during the addition of a first equivalent of 3b
the Soret band experienced a bathochromic shift from 420
to 425 nm, while further addition of two equivalents of 3b
shifted it to 430 nm (Figure 6). Jobs plot analysis revealed a
1:1 stoichiometry. Using SPECFIT the best fit was obtained
for a 2:2 model with logb values of 16.4 (ꢃ0.6) for (PT)2·
A
centration of DABCO increased, the peak at 425 nm de-
creased with a new maximum emerging at 430 nm which is
typical of a simple 1:1 porphyrin·DABCO complex.
(3b)2 and 6.4 (ꢃ1.1) for PT·
C
Table 2. Macroscopic binding constants obtained from a three-state bind-
ing model for PT and 3a,b.
logb
K
PT·3a
PT·(3b)2
PT2·(3b)2
6.7 (ꢃ0.6)
6.4 (ꢃ1.1)
16.4 (ꢃ0.6)
5”106 mꢂ1
A
2.5”106 mꢂ2
2.5”1016 mꢂ3
ACHTREUNG
Figure 7. Top: UV/Vis titration of DABCO against micromolar concen-
trations of PT. Bottom: Fitting of UV/Vis titration data to a four-state
binding model.
When the titration results were analyzed by SPECFIT,[17]
the data fitted best to a 2:3 model providing logb values of
9.7ꢃ0.2 for PT·
A
G
0.4 for (PT) ·
AHCTREUNG
Table 3. Macroscopic binding constants obtained from a four-state bind-
ing model for 4, PT, and 3c.
logb
K
4·3c
PT·(3c)2
5.1 (ꢃ0.03)
9.7 (ꢃ0.2)
1.2”105 mꢂ1
5”109 mꢂ2
4”1015 mꢂ3
2”1021 mꢂ4
AHCTREUNG
Figure 6. Top: UV/Vis titration profile of 3b against PT at micromolar
concentrations. Bottom: Fitting of UV/Vis titration data to a three-state
binding model.
(PT)2·
(PT)2·
A
15.6 (ꢃ0.8)
21.3 (ꢃ0.4)
AHCTREUNG
Chem. Eur. J. 2006, 12, 8136 – 8149
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8141

M. Schmittel et al.
supported by the Jobs plot, which gives independently a
1:1.5 mole ratio of PT to DABCO, and from the evolution
of the signals in the NMR titrations.
In contrast, titration of the simple monoporphyrin com-
plex 4 with 3c at micromolar concentrations provided a
monomeric 1:1 model for the binding of DABCO to por-
phyrin. The logb value of 5.12ꢃ0.03 (Km) matches well with
values known in the literature (see Table 3).[21]
Diffusion NMR studies: To verify beyond doubt the dynam-
ic formation of the three multicomponent assemblies in so-
A
N
to diffusion NMR studies. The method to analyze the size/
weight of PT·3a–c in solution followed that adopted by
Reinhoudt, Cohen et al.,[22] which utilized the fact that the
ratio of the diffusion coefficients for two different molecular
species (Di/Dj) is inversely proportional to the square root
or to the cubic root of the ratio of their molecular weights
Mr [Eq. (2)] for rodlike and spherical molecules, respective-
ly.[23] Equation (2) allows determination of the molecular
weights of the assemblies under investigation, once a cali-
bration curve that correlates molecular weights and experi-
mentally determined diffusion coefficients of known com-
pounds containing similar structural units has been ob-
tained.
Figure 8. Graphical analysis of diffusion coefficients D versus molecular
weight M. The solid lines represent the theoretical correlation of diffu-
sion coefficients and molecular weights as a function of the two models
[sphere (black) or rod (gray)] according to Equation (a), while the dots
~
*
represent various compositions of the self-assemblies. 4 ( ); PT ( );
!
^
^
(PT)2·
(PT)2·
G
(
); (PT)2·
U
(
); PT·
U
(
) formed at excess of 3c;
&
*
(3b)2 ( ); PT·
(3b)2 (3) formed at excess of 3b; PT·3a ( ).
~
the diffusion coefficient of 4 ( ) falling right on the black
curve. Importantly, all experimental diffusion coefficients
for the other aggregates fitted the calibration curves only
for those stoichiometries suggested by the Jobs plot and
UV/Vis titration global simulation results. Accordingly,
mixing PT and 3a in a ratio of 1 equiv:1 equiv afforded an
aggregate PT·3a (1:1) with a molecular weight of 4276 a.m.u
*
( ). In contrast, mixing PT and 3b in a ratio of 1 equiv:1
equiv furnished an aggregate with a molecular weight of
&
8192 a.m.u ( ) that is in agreement with (PT)2·
ACHTREUNG
Table 4 shows the experimental diffusion coefficients
measured for structures 4, PT, and PT·3a as well as for
PT·3b and PT·3c in various proportions. By inserting the
diffusion coefficient values for PT into Equation (2), two
theoretical calibration curves for molecular weights between
0 and 10000 a.m.u. were obtained with PT as anchor point
(Figure 8). The calibration was nicely confirmed by having
mixture of 1 equiv:2 equiv led to PT·
ACHTREUNG
molecular weight (3). Similarly, the aggregates formed by
mixing PT and 3c in ratios of 1 equiv:1 equiv or 1 equiv:
1.5 equiv fit the calibration curve only with molecular
weights around 8100 a.m.u, that is, (PT)2·(3c)2 and (PT)2·
A
(3c)3, respectively. Since the difference between the molecu-
lar weights of the 2:3 and 2:2 compositions amounts to only
112 a.m.u, the difference is not well resolved in the DOSY
spectra. When the ratio of PT and 3c was increased to
1 equiv:2 equiv, only the small aggregate PT·(3c)2 with a
molecular weight of 4164 a.m.u emerged.
A
Table 4. Experimental diffusion coefficients measured in CD2Cl2 at
298 K and their respective molecular weights.
Based on the proof from UV spectroscopy, NMR titra-
tions, and DOSY, the composition of a double sandwich as-
sembly was ascertained in the case of PT·3b and PT·3c as
against the 1:1 adduct in the case of PT·3a. Moreover, as is
noticeable from the single set of signals seen in the 2D
DOSY plots (Figures S19–S25 in Supporting Information),
each PT·3 assembly showed up as a clean species without
any side products of lower molecular weight.
Composition of
the mixture
(mol equivalent)
Experimental diffusion
coefficient
Possible aggregates
and their molecular
[10ꢂ10 m2 sꢂ1
]
weight [gmolꢂ1
]
4 (1 equiv)
PT (1 equiv)
PT (1 equiv) and 3a
(1 equiv)
PT (1 equiv) and 3b
(1 equiv)
PT (1 equiv) and 3b
(2 equiv)
PT (1 equiv) and 3c
(1 equiv)
PT (1 equiv) and 3c
(1.5 equiv)
PT (1 equiv) and 3c
(2 equiv)
12.2
9.2
9.4
1725/4
3940/PT
4276/PT·3a
6.2
12.2
5.5
4096/PT·3b; 8192/
(PT)2·(3b)2
4252/PT·(3b)2
AHCTREUNG
ACHTREUNG
Inferences
4052/PT·3c; 8104/
(PT)2·(3c)2
8216/(PT)2·
AHCTREUNG
PT: a dynamic supramolecular tweezer: As against a number
of known examples of covalent porphyrin tweezers,[24] PT
represents the first example of a supramolecular porphyrin
tweezer. The use of CuI-directed heteroleptic phenanthro-
7.3
A
8.9
4164/PT·(3c)2
E
8142
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8136 – 8149

Supramolecular Multicomponent Porphyrin Tweezers
FULL PAPER
diffusion NMR spectroscopy re-
sults bear witness to a remark-
A
of PT as a function of the
added bis-nitrogen spacers 3a–
c. In the presence of 3a, the dy-
namic bis-porphyrin tweezer
PT operates as a supramolecu-
lar bidentate host (dZnZnffi19 Š)
for the bidentate guest mole-
cule 3a (dNN =16 Š) due to a
nearly perfect geometric com-
plementarity of the two constit-
uents, which gives rise to the
formation of the 1:1 aggregate
PT·3a (Figure 10). As a result
of a lack of configurational con-
trol in the formation of PT and
PT·3a, however, we expect the
formation of two diastereomer-
ic assemblies, namely, meso-
and rac-PT·3a. Interestingly,
the NMR analysis of the diag-
nostic 3’,5’-MesH protons in
PT·3a, although impeded by
the nearly isochronous signals
of the two diastereomeric ag-
gregates, shows a clear prefer-
Figure 9. Investigations on the dynamic nature of PT.
ence for one diastereomer (ca.
line complexation enables exclusive formation of PT as a dy-
namic assembly, as tested by ligand exchange experiments
(Figure 9). Thus, when a solution of PT was titrated against
ligand 2a, a bisphenanthroline analogue of 2 with a shorter
alkoxy side chain, a facile exchange of the bisphenanthroline
ligands was observed after five minutes which provided a
mixture of two tweezers PT and PTa. The ligand exchange
was monitored by ESI-MS through the gradual increase in
the relative abundance of the dication of PTa at m/z 1768.5
(Figure S3 in Supporting Information).
4:1 excess), in contrast to the situation in PT where a dia-
stereomeric ratio of 1:1 is registered. The presence of two
diastereomeric aggregates is also registered by the appear-
ance of two closely placed signals for the 4-H and Hb,Hb’
protons of 1. Association constants derived for PT+3a!
PT·3a revealed a logb value of 6.7 (ꢃ0.6) and hence an as-
sociation constant K=5”106 mꢂ1. This value is typical for di-
topic interactions of a bis-porphyrin system with a bis-nitro-
gen ligand.[25] The magnitude of the association constant in
this case is largely independent of any cooperativity effects,
thus amounting to about twice that of the association of the
parent pyridine to ZnTPP (2”logbpy =2”3.8=7.6; TPP=
tetraphenylporphyrin).[26]
The macroscopic binding constant for the formation of
PT was calculated from UV/Vis titrations by addition of ali-
quots of CuI to a solution of 1 and 2 (2:1 ratio). The forma-
tion of the complex was accom-
panied by a distinct decrease in
the intensity of the Soret band
at 420 nm along with the ap-
pearance of a low and broad ab-
sorption at 420–525 nm, which is
a characteristic metal-to-ligand
charge transfer (MLCT) band
for a CuI complex (Figure S14 in
Supporting Information). The
association constant was logb=
20.1ꢃ0.6 (K=1.1”1020 mꢂ4).
Formation of PT·3a (1:1): The
combined UV/Vis, NMR, and Figure 10. Formation of PT·3a (1:1) from the reaction of 3a with PT.
Chem. Eur. J. 2006, 12, 8136 – 8149
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8143

M. Schmittel et al.
Formation of PT·3b (2:2): Although the distance between
the two zinc porphyrins in PT (dZnZnffi19 Š) can be modulat-
ed to some extent by a distortion about the two CuI bis-
A
N
enthalpically less favorable to coordinate a much smaller
bis-nitrogen spacer, such as 3b (dNN =7 Š), in a 1:1 adduct.
As such, it was not astonishing to find exclusively the 2:2 di-
meric assembly (PT)2·(3b)2 from the reaction of one equiva-
G
lent of PT with one equivalent of 3b (Figure 11). This as-
signment is unequivocally supported by combined evidence
from the UV/Vis and NMR titrations and the DOSY experi-
ments.
Figure 12. Coordination and conformational equilibria present in a mix-
ture of PT and 3b.
While the above results confirm the 2:2 composition of
(PT)2·(3b)2, they do not provide a complete picture of its
G
structure in solution. This is due to the fact that the dynamic
nature of the assembly would allow for multiple conforma-
tions with the same composition (see Figure 12).
2:2 aggregate (PT)2·(3c)2. The double-sandwich aggregate,
A
however, does not show any indications of breaking down at
1
the mole ratio of 1:1.5 (PT:3c), as in the H NMR spectrum
Formation of PT·3c (2:3): A further decrease of the size of
the bis-nitrogen spacer down to 3 Š by using 3c (DABCO)
the porphyrin Hb protons maintain the upfield shift position.
Furthermore, the two signals corresponding to the 3’,5’-
MesH protons continue to appear as four separate signals
due to the nonequivalence of these protons. Hence, by
virtue of the appropriately spaced porphyrins, the aggregate
dynamically creates a binding site which anomalously ac-
cepts a third DABCO guest, contrary to known linear bis-
porphyrins which only assemble in a 2:2 sandwich composi-
tion.[10a,h] Thus, it can be concluded that there is a fast equili-
bration of the three DABCO molecules, which keeps the
“double Pac-Man” porphyrin framework intact (Figure 13).
A distinct assignable set of NMR signals for the aromatic
region at 2:3 stoichiometry (Figure S11 in Supporting Infor-
mation) provides evidence for the same.
produced the unexpected 2:3 aggregate (PT) ·
cated by the Jobs plot and NMR analysis. At first an analo-
gous 2:2 assembly similar to (PT)2·(3b)2 had been expected,
A
AHCTREUNG
but the internal cavity produced by a 2:2 complex offers an
optimal-sized bis(zinc porphyrin) complexation site for the
inclusion of a third molecule of 3c. However, the attach-
ment of a sixth ligand at the zinc atom is known to labilize
and destabilize the already bound axial ligand, and conse-
quently association complexes of the type DABCO–(zinc
porphyrin)–DABCO have only been observed in the solid
state[27] and never in solution. Thus, binding of the third
molecule of 3c is expected to weaken and break down the
Figure 11. Formation of the 2:2 dimeric assembly (PT)2·(3b)2 from the reaction of 3b with PT.
G
8144
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8136 – 8149

Supramolecular Multicomponent Porphyrin Tweezers
FULL PAPER
Figure 13. Proposed chemical structure of the 2:3 aggregate (PT)2·(3c)3.
G
On the basis of the experimental binding constants, a
simple model could be proposed for the dynamics observed
in the system PT+3c (Figure 14). The value of Km was cal-
culated from K12 to be logb=4.6, close to the value ob-
tained in the titration of 4 to DABCO. From the value of
K22 it is possible to calculate Em (effective molarity) which is
found to be 0.002m.
Figure 15. Fitting of NMR titration data to concentration profiles of a 2:3
binding model obtained from the UV/Vis titration. The plot depicts the
changes in the mole fraction (c) of the four different species against the
concentration of DABCO: free PT; 2:2 aggregate (PT)2·
A
gate (PT)2·(3c)3; 1:2 aggregate PT·(3c)2. The symbols represent the ex-
A
ACHTREUNG
perimental values as observed from NMR titration.
ular weight of one DABCO molecule), the DOSY experi-
ment does not distinguish the two species and thus provides
a single value for the diffusion coefficient.
As in the case of PT·3b, the experimental studies allow
the composition to be determined unequivocally, but due to
the dynamic nature of the complex configurational issues
have to remain unaddressed.
Figure 14. Equilibria present in a mixture of PT and DABCO.
One-pot approach to the four-component assemblies: The
foregoing discussion demonstrates the successful formation
of four-component assemblies in a stepwise process, that is,
by the initial formation of a three-component assembly PT
that develops into tertiary assemblies by the coordination of
bases 3a–c. Importantly, all assemblies could equally be pre-
pared in one-pot reactions, where all components were put
together as solids and then brought into solution by dissolv-
ing them in dichloromethane (Figure 16). The resulting solu-
tion was then tested by NMR and UV/Vis spectroscopy. The
The veracity of the model in Figure 14 was further inter-
rogated by fitting the NMR titration data to the simulated
concentration profiles for a 2:3 model obtained from the
UV titration (Figure 15). The NMR data fit agreeably to the
concentration profile. The concentration profile also reveals
that the 2:2 complex shows 100% formation, while the 2:3
complex is never completely formed and breaks down on
addition of excess of 3c (DABCO). Since the mass differ-
ence between the 2:3 and 2:2 complexes is only 112 (molec-
Chem. Eur. J. 2006, 12, 8136 – 8149
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8145

M. Schmittel et al.
Figure 16. Hierarchy of self-assembly processes leading to a tertiary structure.
1H NMR spectra were exactly the same as those obtained in
a stepwise approach. The final proof of the formation of the
same structural composition was obtained by performing a
DOSY experiment, which yielded the same diffusion coeffi-
cients. This result led us to conclude that the same supramo-
lecular composition could be achieved by either a one-pot
or a sequential approach owing to the noninterfering nature
of the two binding algorithms.
organized manner to the present generation of supramole-
cules. A logical step in this direction would be to utilize su-
pramolecules themselves as building blocks to build tunable
and meaningful tertiary superstructures. This would call for
the use of more than one kind of noncovalent interactions
(which are noninterfering and orthogonal) and multiple mo-
lecular components (typically ꢄfour) in the self-assembly.
Our future investigations aim at exploring this aspect fur-
ther.
Conclusion
Experimental Section
In summary, we provide a unique example of the translation
of a set of simple covalent components into a tertiary assem-
bly obtained through a binary binding algorithm, with two
noninterfering binding interactions driving the system ther-
modynamically into a unique supramolecular assembly. We
demonstrate that careful tuning of components at a lower
level of a multicomponent assembly (the length of one of
the components in the present case) could translate into
substantial changes in the structure of the assembly.
Porphyrin 1 and bisphenanthroline 2 were synthesized by known proce-
dures from earlier reports.[7d,13] 3a was synthesized based on an analogous
report.[14] DABCO and 4,4’-bipyridine were obtained from ACROS
Chemicals. [Cu(MeCN)4]PF6 was prepared according to known proce-
G
dures.[28] 1H NMR spectra were recorded on a Bruker AC 400 (400 MHz)
or Bruker AC 200 (200 MHz) spectrometer and 13C NMR spectra were
obtained on a Bruker AC 400 (100 MHz) or Bruker AC 200 (60 MHz) in-
strument. 1H NMR spectroscopy was carried at room temperature in
CD2Cl2. ESI-MS was performed on an LCQ Deca ThermoQuest instru-
ment. Typically, 25 scans were accumulated for one spectrum. All com-
In moving toward highly functional multimolecular aggre-
gates, there is a need to introduce further complexity in an
1
plexes were characterized by H NMR, 13C NMR, and elemental analysis.
8146
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8136 – 8149

Supramolecular Multicomponent Porphyrin Tweezers
FULL PAPER
Zinc(ii) meso-5-{4-[3-(2-(4-bromo-2,3,5,6-tetramethyl)-9-(2,4,6-trimethyl-
phenyl))-1,10-phenanthrolin-3-ylethynylethynyl]phenyl}porphyrin (1): In
a three-necked round-bottomed flask fitted with a reflux condenser,
meso-5-(4-iodophenyl)-10,15,20-trimesityl zinc porphyrin (138 mg,
148 mmol) was taken up in dry benzene/triethylamine (20 mL, 15:5).
Then 3-[2-(2,4,6-trimethylphenyl)-9-(4-bromo-2,3,5,6-tetramethyl)-1,10-
phenanthrolinyl]ethyne (79.0 mg, 148 mmol) was added to this solution.
The mixture was degassed for 30 min under a steady flow of nitrogen.
mental analysis (%) calcd for C98H81BrCuF6N8PZn: C 68.25, H 4.73, N
6.50; found: C 68.77, H 4.68, N 6.56.
Complex PT: Bisphenanthroline 2 (0.5 equiv) was added to a solution of
1 and [Cu(CH3CN)4]PF6 (1:1 in dichloromethane). The deep red solution
A
obtained was evaporated and complex PT was isolated in quantitative
yield. 1H NMR (400 MHz, CD2Cl2): d=8.96 (s, 2H; 4-H), 8.75 (d, J=
8.3 Hz, 2H; 7-H), 8.72 (d, J=4.5 Hz, 2H; pyrrol-H1), 8.68 (d, J=4.8 Hz,
2H; pyrrol-H3), 8.67 (d, J=4.8 Hz, 2H; pyrrol-H4), 8.66 (d, 3J=4.6 Hz,
2H; pyrrol-H2), 8.61 (d, 4J=1.5 Hz, 2H; 4’-H), 8.54 (d, 4J=1.5 Hz, 4H;
2’-H), 8.53 (d, 4J=1.5 Hz, 2H; 9’-H), 8.47 (d, J=8.3 Hz, 4H; 7’-H), 8.29
(d, J=9.1 Hz, 1H; 5’-H), 8.25 (d, J=9.1 Hz, 1H; 6’-H), 8.08 (d, J=
7.7 Hz, 4H; Ar-Ha, -Ha’), 7.95 (d, J=8.1, 1H; 8-H), 7.94 (d, J=9.1 Hz,
[Pd2(dba)3] (10.0 mg, 1.48 mmol) and AsPh3 (45.3 mg, 148 mmol) were
G
then added as solids. The reaction was heated at 408C for 4 h, after
which the reaction mixture was evaporated in a vacuum. The residue ob-
tained was dissolved in dichloromethane and washed with a solution of
2% KCN dried over sodium sulfate. The resulting solid was chromato-
graphed on silica gel with dichloromethane as eluent to give a violet
solid containing 1 as a crude product. The solid was then dissolved in tol-
uene (1 mL), the solution was loaded on a size-exclusion gel containing
BioRad Bio-Beads SX-1 swollen in toluene, and a chromatography run
was carried out under gravity flow. The bright red fractions were isolated
to give the product. Yield: 85 mg (43%); 1H NMR (400 MHz, CD2Cl2):
d=8.88 (d, J=4.5 Hz, 2H; pyrrol-H1), 8.81 (d, J=4.5 Hz, 2H; pyrrol-
H2), 8.77 (d, J=4.5 Hz, 4H; pyrrol-H3, -H4), 8.68 (s, 1H; 4-H), 8.35 (d,
J=8.2 Hz, 1H; 7-H), 8.20 (d, J=7.9 Hz, 2H; Ar-Hb, -Hb’), 7.96 (s, 2H;
5-, 6-H), 7.64 (d, J=8.2 Hz, 1H; 8-H), 7.53 (d, J=7.9 Hz, 2H; Ar-Ha, -
Ha’), 7.33 (s, 6H; por-mes-H), 7.01 (s, 2H; 3’’’-, 5’’’-H), 2.68 (s, 9H; por-
mes-Me), 2.59 (s, 6H; 8’’-, 9’’-H), 2.39 (s, 3H; 8’’’-H), 2.23 (s, 6H; 7’’’-,
9’’’-H), 2.20 (s, 6H; 7’’-, 10’’-H), 1.92 (s, 9H; por-mes-Me), 1.90 ppm (s,
9H; por-mes-Me); 13C NMR (100 MHz, CDCl3): d=163.0, 161.1, 150.3,
150.1, 149.9, 146.3, 145.2, 144.3, 143.8, 140.7, 139.7, 138.9, 138.4, 138.0,
137.7, 136.5, 135.1, 134.9, 134.2, 133.8, 132.1, 131.5, 131.1, 130.9, 130.0,
129.7, 129.3, 128.9 ,128.8, 128.7, 128.0, 127.6, 127.4, 127.1, 126.1, 125.7,
122.0, 120.7, 119.2, 119.1, 96.3, 88.1, 22.2, 21.8, 21.5, 21.4, 21.0, 19.2, 18.7,
14.6 ppm; ESI-MS: m/z (%): 1335.9 (100) [M+H]+; elemental analysis
(%) calcd for C86H73BrN6Zn: C 77.32, H 5.51, N 6.29; found: C 77.74, H
5.68, N 6.28.
3
3
1H; 5-H), 7.89 (d, J=9.1 Hz, 1H; 6’-H), 7.78 (dd, J=8.1 Hz, J=4.8 Hz,
2H; 8’-H), 7.45 (d, J=7.7 Hz, 4H; Ar-Hb, -Hb’), 7.26 (s, 6H; por-mes-
Ha), 7.24 (s, 6H; por-mes-Hb), 7.12 (s, 2H; phenyl-H), 6.11 (s, 2H; 3’’’-,
5’’’-H1), 6.07 (s, 2H; 3’’’-, 5’’’-H2), 4.03 (t, J=6.2 Hz, 4H; OCH2), 2.59 (s,
12H; 7’’-, 9’’-H), 2.56 (s, 12H; 7’’’-, 10’’’-H),1.92 (s, 6H; por-CH3), 1.89 (s,
6H; por-CH3), 1.85 (s, 6H; por-CH3), 1.80 ( s, 18H; por-CH3), 1.76 (s,
18H; por-CH3), 1.62 (s, 6H; 8’’’-H), 1.59 (s, 12H; 8’’-, 9’’-H), 1.17 (m,
40H; OCH2-C10H20-CH3), 0.82 ppm (t, J=6.8 Hz, 6H; OC11H22-CH3);
13C NMR (100 MHz, CD2Cl2): d=185.4, 161.3, 159.9, 154.3, 150.3, 150.0,
149.7, 149.5, 148.4, 144.7, 144.1, 142.9, 142.6, 141.6, 140.2, 140.2, 139.9,
139.4, 139.3, 139.2, 139.1, 138.3, 137.9, 137.7, 137.0, 135.1, 135.0, 134.9,
133.8, 132.7, 132.6, 131.8, 131.5, 131.4, 130,9,130.0, 129.7,
129.4,129.2,128.6, 128.4, 127.9, 127.6, 127.2, 126.8, 126.6, 125.2, 123.2,
121.8, 121.0, 119.6, 119.3, 118.9, 117.0, 113.8, 111.3, 110.8, 97.9, 92.4, 90.9,
86.1, 84.9, 69.8, 32.2, 30.0, 29.9, 29.7, 29.6, 26.3, 23.1, 21.8, 21.7, 21.5, 20.5,
20.5, 20.3, 18.7, 18.6, 14.3 ppm; ESI-MS: m/z (%): 1824.9 (100) [M]2+; el-
emental analysis (%) calcd for C230H212Br2Cu2F12N16O2P2Zn2: C 70.12, H
5.42, N 5.69; found: C 70.35, H 5.42, N 6.22.
ComplexesPT-3a–c : PT-3a–c were prepared by adding the respective
equivalents of 3a–c to PT in CH2Cl2. For NMR purposes, the complex
was prepared by adding 3a–c directly to an NMR tube containing PT,
and subsequent measurements were made without any further isolation
or purification. Extensive characterization is described in the Results sec-
tion.
1,4-Bis(4’-pyridylethynyl)durene (3a): In a two-necked round-bottomed
flask fitted with a reflux condenser, 1,4-diiododurene (138 mg, 357 mmol),
Titrations: UV/Vis titrations were performed on a Cary Varian UV in-
strument with a quartz cuvette of path length 1.0 cm at 258C in CH2Cl2.
Aliquots of millimolar concentrations of bases 3a, 3b, and 3c were
added to complex PT at micromolar concentrations with microliter sy-
ringes. UV/Vis titrations were analyzed by fitting the whole series of
spectra at 0.5-nm intervals using the software SPECFIT version 3.0.22
(Spectrum Software Associates, P.O. Box 4494, Chapel Hill, NC 27515-
4494, USA), which uses a global analysis system with expanded factor
analysis and a Marquardt least-squares minimization to obtain globally
optimized parameters.[29]
4-ethynylpyridine hydrochloride (120 mg, 859 mmol), [Pd(PPh3)2Cl2]
G
(7.50 mg, 10.6 mmol), and copper iodide (10 mg, 52 mmol) were mixed
under a nitrogen atmosphere. Benzene (10 mL) and diethylamine (5 mL)
were added to the flask. The reaction mixture was refluxed for 12 h, after
which the solvents were removed and the residue was dissolved in tol-
uene and eluted over a pad of silica. The resulting light yellow solution
1
was evaporated to give a yellow solid 3a. Yield: 180 mg (68%); H NMR
(400 MHz, CD2Cl2): d=8.63 (dd, 3J=4.5 Hz, 4J=1.6 Hz, 4H), 7.41 (dd,
3J=4.5 Hz, 4J=1.6 Hz, 4H), 2.51 ppm (12H; CH3); 13C NMR (100 MHz,
CDCl3): d=149.7, 136.3, 131.7, 125.3, 123.1, 95.5, 92.9, 18.3 ppm; ESI-
MS: m/z (%): 337.4 (100) [M+H]+; elemental analysis (%) calcd for
C24H20N2: C 85.68, H 5.99, N 8.33; found: C 85.38, H 5.87, N, 8.70.
1H NMR titrations were performed in CD2Cl2 in a 5-mm NMR tube at
298 K on a Bruker AC 400 (400 MHz) instrument by sequential addition
of the bases into the NMR tube with a microliter syringe.
Complex 4: Anhydrous 1,10-phenanthroline (1 equiv) was added to equi-
DOSY: Diffusion experiments were performed on the Bruker Avance
400-MHz NMR spectrometer, with a 5-mm BBI probe head, equipped
with a pulsed field gradient unit capable of producing magnetic field gra-
dients in the z direction of about 5.35 Gcmꢂ1. All experiments were car-
ried out at 298 K in a 5-mm NMR tube at 2 mm concentration. The bipo-
lar magnetic field pulse gradients (d) were of 2.5–4.5 ms duration, and
the diffusion time (D) was 50 ms. The pulse gradients were increased
from 0.10 to 5.08 Gcmꢂ1 in 32 steps. Signals were averaged over 30–45
scans. In each experiment the peaks were analyzed using an inbuilt inten-
sity fit function “simfit” which utilizes Equation (3)], where g is the gyro-
magnetic radius (radsꢂ1 Gꢂ1), d is the length of the diffusion gradients
(Gcmꢂ1), D is the time of separation between the gradients, G is the
pulsed gradient strength, and D is the diffusion coefficient.
molar amounts of 1 and [Cu(CH3CN)4]PF6 in dichloromethane. The re-
A
sulting solution showed an instantaneous change in color to deep red.
The complex was isolated without any further purification and was found
to be 4, obtained in quantitative yield. 1H NMR (400 MHz, CD2Cl2): d=
8.95 (s, 1H; 4-H), 8.72 (d, J=7.8 Hz, 1H; 7-H), 8.72 (d, J=4.6 Hz, 2H;
pyrrol-H1), 8.68 (d, J=4.8 Hz, 2H; pyrrol-H3), 8.67 (d, J=4.8 Hz, 2H;
pyrrol-H4), 8.66 (d, J=4.6 Hz, 2H; pyrrol-H2), 8.52 (dd, 3J=4.6 Hz, 4J=
1.3 Hz, 2H; 2’-, 9’ -H), 8.44 (dd, 3J=8.1 Hz, 4J=1.5 Hz, 2H; 4’-, 7’-H),
8.27 (d, J=9.1 Hz, 1H; 5-H), 8.24 (d, J=9.1 Hz, 1H; 6-H), 8.08 (d, J=
8.2 Hz, 2H; Ar-Ha, -Ha’), 7.92 (s, 2H; 5’-, 6’-H), 7.91 (d, J=7.8 Hz, 1H;
8-H), 8.27 (dd, 3J=8.0 Hz, 3J=4.8 Hz, 2H; 3’-, 8’-H), 7.44 (d, J=8.2 Hz,
2H; Ar-Hb, -Hb’), 7.26 (s, 6H; por-mes), 6.02 (s, 2H; 3’’’-, 5’’’-H), 2.59 (s,
12H; 7’’-, 10’’-, 7’’’-, 9’’’-H), 1.81 (s, 18H; por-mes-Me), 1.78 (s, 9H; por-
mes-Me), 1.64 (s, 6H; 8’’-, 9’’-H), 1.54 ppm (s, 3H; 8’’’-H); 13C NMR
(100 MHz, CD2Cl2): d=160.5, 159.0, 149.4, 149.2, 149.2, 148.9, 147.2,
144.4, 143.4, 142.4, 141.8, 139.0, 138.9, 138.5, 137.3, 136.9, 135.9, 134.2,
132.8, 131.8, 131.0, 130.4, 130.1, 129.1, 128.3, 127.8, 127.3, 127.1, 126.6,
126.4, 125.8, 124.2, 122.3, 120.0, 118.1, 117.9, 97.0, 85.3, 21.1, 20.7, 19.6,
19.5 (2), 19.4 (2), 17.7 ppm; ESI-MS: m/z (%): 1579.6 (100) [M]+; ele-
I ¼ Ið0Þ e½Dð-g2G2d2ÞðDꢂd=3Þꢅ
ð3Þ
Chem. Eur. J. 2006, 12, 8136 – 8149
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8147

M. Schmittel et al.
343; g) C. A. Hunter, R. Tregonning, Tetrahedron 2002, 58, 691–697;
h) L. Baldini, P. Ballester, A. Casnati, R. M. Gomila, C. A. Hunter,
F. Sansone, R. Ungaro, J. Am. Chem. Soc. 2003, 125, 14181–14189;
i) M. C. Lensen, S. J. T. van Dingenen, J. A. A. W. Elemans, H. P.
Dijkstra, G. P. M. van Klink, G. van Koten, J. W. Gerritsen, S. Spell-
er, R. J. M. Nolte, A. E. Rowan, Chem. Commun. 2004, 762–763.
[11] a) J. Barber, B. Andersson, Nature 1994, 370, 31–34; b) W. Kühl-
brandt, Nature 1995, 374, 497–498; c) G. McDermott, S. M. Prince,
A. A. Freer, A. M. Hawthornthwaite-Lawless, M. Z. Papiz, R. J.
Cogdell, N. W. Isaacs, Nature 1995, 374, 517–521; d) T. Pullerits, V.
Sundstrçm, Acc. Chem. Res. 1996, 29, 381–389.
[12] a) C. K. Chang, I. Abdalmuhdi, J. Org. Chem. 1983, 48, 5388–5390;
b) J. P. Collman, P. S. Wagenknecht, J. E. Hutchinson, Angew. Chem.
1994, 106, 1620–1639; Angew. Chem. Int. Ed. Engl. 1994, 33, 1537–
1556; c) Y. Deng, C. J. Chang, D. J. Nocera, J. Am. Chem. Soc. 2000,
122, 410–411.
Acknowledgements
We are very much indebted to the DFG and the Fonds der Chemischen
Industrie for continued financial support.
[1] J.-M. Lehn in Nobel Lectures, Chemistry(1981–1990) (Ed.: B. G.
Malmstrçm), World Scientific Publishing Co., Singapore, 1992.
[2] a) S. Leininger, B. Olenyuk, P. J. Stang, Chem. Rev. 2000, 100, 853–
907; b) J.-M. Lehn, Supramolecular Chemistry: Concepts and Per-
spectives, VCH, Weinheim, 1995; c) J. W. Steed, J. L. Atwood, Su-
pramolecular Chemistry, Wiley, New York, 2001; d) A. V. Davis,
R. M. Yeh, K. N. Raymond, Proc. Natl. Acad. Sci. USA 2002, 99,
4793–4796; e) D. L. Caulder, K. N. Raymond, Acc. Chem. Res. 1999,
32, 975–982; f) J. A. R. Navarro, B. Lippert, Coord. Chem. Rev.
1999, 185–186, 653–667; g) R.-D. Schnebeck, E. Freisinger, B. Lip-
pert, Angew. Chem. 1999, 111, 235–238; Angew. Chem. Int. Ed.
1999, 38, 168–171; h) M. Fujita, Acc. Chem. Res. 1999, 32, 53–61.
[3] a) Y. L. Cho, H. Uh, S.-Y. Chang, H.-Y. Chang, M.-G. Choi, I. Shin,
K.-S. Jeong, J. Am. Chem. Soc. 2001, 123, 1258–1259. (Host–guest
supramolecular systems which are inclusion complexes of the type
DꢆAnBmCl have not been considered, since the included guest does
not produce a structural enhancement to the supramolecule.)
[4] a) M. Schmittel, V. Kalsani, Top. Curr. Chem. 2005, 245, 1–53, and
references therein; b) M. Yoshizawa, J. Nakagawa, K. Kumazawa,
M. Nagao, M. Kawano, T. Ozeki, M. Fujita, Angew. Chem. 2005,
117, 1844–1847; Angew. Chem. Int. Ed. 2005, 44, 1810–1813; c) M.
Fujita, N. Fujita, K. Ogura, K. Yamaguchill, Nature 1999, 400, 52–
55; d) C. J. Kuehl, Y. K. Kryschenko, U. Radhakrishnan, S. R.
Seidel, S. D. Huang, P. J. Stang, Proc. Natl. Acad. Sci. USA 2002, 99,
4932–4936; e) B. Olenyuk, J. A. Whiteford, A. Fechtenkr, P. J.
Stang, Nature 1999, 398, 796–799.
[5] By multiple molecular components we mean different molecular
components AnBmClDk…Zx and not a large number of the same set
of components, that is, AnBm. Numerous examples of the latter kind
are known.
[6] a) M. C. T. Fyfe, J. F. Stoddart, Coord. Chem. Rev. 1999, 183, 139–
155; b) H. Hofmeier, U. S. Schubert, Chem. Commun. 2005, 19,
2423–2432.
[7] a) M. Schmittel, A. Ganz, D. Fenske, Org. Lett. 2002, 4, 2289–2292;
b) M. Schmittel, H. Ammon, V. Kalsani, A. Wiegrefe, C. Michel,
Chem. Commun. 2002, 2566–2567; c) M. Schmittel, V. Kalsani, A.
Wiegrefe, Chem. Commun. 2004, 490–491; d) M. Schmittel, R. S. K.
Kishore, Org. Lett. 2004, 6, 1923–1926; e) V. Kalsani, H. Ammon, F.
Jäckel, J. P. Rabe, M. Schmittel, Chem. Eur. J. 2004, 10, 5481–5492;
f) V. Kalsani, H. Bodenstedt, D. Fenske, M. Schmittel, Eur. J. Inorg.
Chem. 2005, 1841–1849; g) M. Schmittel, V. Kalsani, J. W. Bats,
Inorg. Chem. 2005, 44, 4115–4117; h) M. Schmittel, V. Kalsani,
R. S. K. Kishore, H. Cçlfen, J. W. Bats, J. Am. Chem. Soc. 2005, 127,
11544–11545.
[13] M. Schmittel, C. Michel, A. Wiegrefe, V. Kalsani, Synthesis 2001,
1561–1567.
[14] T. M. Fasina, J. C. Collings, D. P. Lydon, D. A. Jove, A. S. Batsanov,
J. A. K. Howard, P. Nguyen, M. Bruce, A. J. Scott, W. Clegg, S. W.
Watt, C. Viney, T. B. Marder, J. Mater. Chem. 2004, 14, 2395–2404.
[15] Carried out on PC Spartan Pro 1.0.7, Wavefunction Inc., Irvine, CA
92612, USA.
[16] a) L. Flamigni, M. R. Johnston, New J. Chem. 2001, 25, 1368–1370;
b) K. Yamada, H. Imahori, E. Yoshizawa, D. Gosztola, M. R. Wasie-
lewski, Y. Sakata, Chem. Lett. 1999, 235–236.
[17] a) H. Gampp, M. Maeder, C. J. Meyer, A. D. Zuberbühler, Talanta
1985, 32, 95–101; b) F. J. C. Rossoti, H. S. Rossoti, R. J. Whewell, J.
Inorg. Nucl. Chem. 1971, 33, 2051–2065; c) H. Gampp, M. Maeder,
C. J. Meyer, A. D. Zuberbühler, Talanta 1985, 32, 257–264; d) H.
Gampp, M. Maeder, C. J. Meyer, A. D. Zuberbühler, Talanta 1986,
33, 943–951.
[18] SPECFIT generates multiple fit models. Thus, a 1:1 model was also
generated with logb=5.9ꢃ0.1 for PT·3b and logb=6.4ꢃ0.9 for
PT·(3b)2.
E
[19] a) C. A. Hunter, L. D. Sarson, Angew. Chem. 1994, 106, 2424–2426;
Angew. Chem. Int. Ed. Engl. 1994, 33, 2313–2316; b) X. Chi, A. J.
Guerin, R. A. Haycock, C. A. Hunter, L. D. Sarson, J. Chem. Soc.
Chem. Commun. 1995, 2567–2569; c) C. A. Hunter, R. K. Hyde,
Angew. Chem. 1996, 108, 2064–2067; Angew. Chem. Int. Ed. Engl.
1996, 35, 1936–1939; d) C. A. Hunter, R. J. Shannon, Chem.
Commun. 1996, 1361–1362; e) C. C. Mak, N. Bampos, J. M. K. Sand-
ers, Angew. Chem. 1998, 110, 3169–3172; Angew. Chem. Int. Ed.
1998, 37, 3020–3023; f) N. Armaroli, F. Diederich, L. Echegoyen, T.
Habicher, L. Flamigni, G. Marconi, J.-F. Nierengarten, New J.
Chem. 1999, 23, 77–83.
[20] SPECFIT also generates a 1:1 model with logb=5.7ꢃ0.1 for PT·3c
and logb=9.5ꢃ0.1 for PT·(3c)2, and a 2:2 model with logb=16.8ꢃ
U
0.2 for (PT)2·
A
ACHTREUNG
[21] M. Gardner, A. J. Guerin, C. A. Hunter, U. Michelsen, C. Rotger,
New J. Chem. 1999, 23, 309–316.
[8] M. Schmittel, A. Ganz, Chem. Commun. 1997, 999–1000.
[9] a) C. A. Hunter, M. N. Meah, J. K. M. Sanders, J. Am. Chem. Soc.
1990, 112, 5773–5780; b) H. L. Anderson, C. A. Hunter, M. N.
Meah, J. K. M. Sanders, J. Am. Chem. Soc. 1990, 112, 5780–5789;
c) S. Yagi, I. Yonekura, M. Awakura, M. Ezoe, T. Takagishi, Chem.
Commun. 2001, 557–558; d) S. Yagi, M. Ezoe, I. Yonekura, T. Taka-
gishi, H. Nakazumi, J. Am. Chem. Soc. 2003, 125, 4068–4069;
e) D. M. Guldi, C. Luo, A. Swartz, M. Scheloske, A. Hirsch, Chem.
Commun. 2001, 1066–1067; f) D. Jokic, Z. Asfari, J. Weiss, Org.
Lett. 2002, 4, 2129–2132; g) M. Dudic, P. Lhotak, H. Petrickova, I.
Stibor, K. Lang, J. Sykora, Tetrahedron 2003, 59, 2409–2415.
[10] a) H. L. Anderson, Inorg. Chem. 1994, 33, 972–981; b) P. N. Taylor,
H. L. Anderson, J. Am. Chem. Soc. 1999, 121, 11538–11545;
c) T. E. O. Screen, J. R. G. Throne, R. G. Denning, D. G. Bucknail,
H. L. Anderson, J. Am. Chem. Soc. 2002, 124, 9712–9713; d) C. C.
Mak, D. Pomeranc, M. Montalti, L. Prodi J. K. M. Sanders, Chem.
Commun. 1999, 1083–1084; e) A. L. Kieran, A. D. Bond, A. M. Be-
lenguer, J. K. M. Sanders, Chem. Commun. 2003, 2674–2675; f) N.
Bampos, V. Marvaud, J. K. M. Sanders, Chem. Eur. J. 1998, 4, 335–
[22] T. Timmerman, J.-L. Weidmann, K. A. Jolliffe, L. J. Prins, D. N.
Reinhoudt, S. Shinkai, L. Frish, Y Cohen, J. Chem. Soc. Perkin
Trans. 2 2000, 2077–2089.
[23] a) M. Holtz, X. Mao, D. Seiferling, A. Sacco, J. Chem. Phys. 1996,
104, 669–679; b) A. R. Waldeck, P. W. Kuchel, A. J. Lennon, B. E.
Chapman, Prog. Nucl. Magn. Reson. Spectrosc. 1997, 30, 39–68.
[24] a) G. Proni, G. Pescitelli, X. Huang, K. Nakanishi, N. Berova, J. Am.
Chem. Soc. 2003, 125, 12914–12927; b) Z.-Q. Wu, X.-B. Shao, C. Li,
J.-L. Hou, K. Wang, X.-K. Jiang, Z.-T. Li, J. Am. Chem. Soc. 2005,
ˇ
ˇ ˇ
127, 17460–17468; c) M. Dudic, P. Lhotꢁk, I. Stibor, H. Petrꢂckovꢁ,
K. Lang, New J. Chem. 2004, 28, 85–90; d) V. V. Borovkov, J. M.
Lintuluoto, M. Sugiura, Y. Inoue, R. Kuroda, J. Am. Chem. Soc.
2002, 124, 11282–11283; e) V. V. Borovkov, J. M. Lintuluoto, G. A.
Hembury, M. Sugiura, R. Arakawa, Y. Inoue, J. Org. Chem. 2003,
68, 7176–7192; f) J. Brettar, J.-P. Gisselbrecht, M. Gross, N. Solladiꢃ,
Chem. Commun. 2001, 733–734; g) R. Rein, M. Gross, N. Solladiꢃ,
Chem. Commun. 2004, 1992–1993.
[25] A. Satake, Y. Kobuke, Tetrahedron 2005, 61, 13–41.
8148
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8136 – 8149

Supramolecular Multicomponent Porphyrin Tweezers
FULL PAPER
[26] M. Nappa, J. S. Valentine, J. Am. Chem. Soc. 1978, 100, 5075–5080.
[27] a) I. Goldberg, Chem. Eur. J. 2000, 6, 3863– 3870; b) R. K. Kumar,
S. Balasubramanian, I. Goldberg, Inorg. Chem. 1998, 37, 541–552.
[28] G. J. Kubas, Inorg. Synth. 1979, 19, 90–92.
[29] a) M. Maeder, A. D. Zuberbühler, Anal. Chem. 1990, 62, 2220–
2224; b) H. Gampp, M. C. J. Maeder, A. D. Zuberbühler, Talanta
1986, 33, 943–951.
Received: April 3, 2006
Published online: July 24, 2006
Chem. Eur. J. 2006, 12, 8136 – 8149
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8149