"Helter-skelter-like" perylene polyisocyanopeptides

Article abstract of DOI:10.1002/chem.200801746

We report on a combined experimental and computational investigation on the synthesis and thorough characterization of the structure of perylene-functionalized polyisocyanides. Spectroscopic analyses and extensive molecular dynamics studies revealed a well defined 4) helix in which the perylene molecules form four "helter skelter-like" overlapping pathways along which excitons and electrons can rapidly migrate. The well-defined polymer scaffold stabilized by hydrogen bonding, to which the chromophores are attached, accounts for the precise architectural definition, and molecular stiffness observed for these molecules. Molecular-dynamics studies showed that the chirality present in these polymers is expressed in the formation of stable right-handed helices. The formation of chiral supramolecular structures is further supported by the measured and calculated bisignated Cotton effect. The structural definition of the chromophores aligned in one direction along the backbone is highlighted by the extremely efficient exciton migration rates and charge densities measured with Transient Absorption Spectroscopy.

Full text of DOI:10.1002/chem.200801746

DOI: 10.1002/chem.200801746
“Helter-Skelter-Like” Perylene PolyACTHNUTRGNEiNUG socyanopeptides
Erik Schwartz,[a] Vincenzo Palermo,[b] Chris E. Finlayson,[c] Ya-Shih Huang,[c]
Matthijs B. J. Otten,[a] Andrea Liscio,[b] Sara Trapani,[d] Irene Gonzꢀlez-Valls,[a]
Patrick Brocorens,[d] Jeroen J. L. M. Cornelissen,[a] Kalina Peneva,[e] Klaus Mꢁllen,[e]
Frank C. Spano,[f] Arkady Yartsev,[g] Sebastian Westenhoff,[c] Richard H. Friend,*[c]
David Beljonne,*[d] Roeland J. M. Nolte,[a] Paolo Samorꢂ,*[b, h] and Alan E. Rowan*[a]
Abstract: We report on a combined ex-
perimental and computational investi-
gation on the synthesis and thorough
characterization of the structure of per-
ylene-functionalized polyisocyanides.
Spectroscopic analyses and extensive
molecular dynamics studies revealed a
well defined 41 helix in which the pery-
lene molecules form four “helter skel-
ter-like” overlapping pathways along
which excitons and electrons can rapid-
ly migrate. The well-defined polymer
scaffold stabilized by hydrogen bond-
ing, to which the chromophores are at-
tached, accounts for the precise archi-
tectural definition, and molecular stiff-
ness observed for these molecules. Mo-
lecular-dynamics studies showed that
the chirality present in these polymers
is expressed in the formation of stable
right-handed helices. The formation of
chiral supramolecular structures is fur-
ther supported by the measured and
calculated bisignated Cotton effect.
The structural definition of the chro-
mophores aligned in one direction
along the backbone is highlighted by
the extremely efficient exciton migra-
tion rates and charge densities mea-
sured with Transient Absorption Spec-
troscopy.
Keywords: electron transport
·
perylene diimides · polyisocyanides ·
polymers
[a] E. Schwartz, M. B. J. Otten, I. Gonzꢀlez-Valls,
Dr. J. J. L. M. Cornelissen, Prof. Dr. R. J. M. Nolte,
Prof. Dr. A. E. Rowan
[e] Dr. K. Peneva, Prof. Dr. K. Mꢄllen
Max-Planck Institute for Polymer Research,
Ackermann 10, 55124 Mainz (Germany)
Institute for Molecules and Materials
Radboud University Nijmegen
Toernooiveld 1, 6525 Ed Nijmegen (The Netherlands)
Fax : (+31)-24-3652929
[f] Prof. Dr. F. C. Spano
Department of Chemistry, Temple University
Beury Hall 201, Philadelphia, PA 19122 (USA)
[g] Prof. A. Yartsev
E-mail: a.rowan@science.ru.nl
Lund University
[b] Dr. V. Palermo, Dr. A. Liscio, Prof. Dr. P. Samorꢁ
Instituto per la Sintesi Organica e la Fotoreattivitꢂ, (Institution)
Consiglio Nazional dlle Ricerche, Via Gobetti 101
40129 Bologna (Italy)
P.O. Box 124, 22100 Lund (Sweden)
[h] Prof. Dr. P. Samorꢁ
Nanochemistry Laboratory, ISIS, Universitꢃ Louis Pasteur and CNRS
(UMR 7006), 8 allꢃe Gaspard Monge
67000 Strasbourg (France)
E-mail: samori@isis-ulp.org
[c] Dr. C. E. Finlayson, Y.-S. Huang, Dr. S. Westenhoff,
Prof. Dr. R. H. Friend
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200801746.
Cavendish Laboratory, University of Cambridge
J. J. Thomson Avenue Cambridge CB3 0HE (UK)
E-mail: rhf10@cam.ac.uk
[d] S. Trapani, Dr. P. Brocorens, Dr. D. Beljonne
Universitꢃ de Mons-Hainaut, Place du Parc 20
7000 Mons (Belgium)
E-mail: David@averell.umh.ac.be
2536
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 2536 – 2547

FULL PAPER
Introduction
terials with potentially interesting properties. Such a high
order is reflected in the extraordinary high stiffness (as
proven by a persistence length of 76 nm) observed by AFM
measurements.[37]
The synthesis of complex molecular systems ordering chro-
mophores in well-defined positions in order to achieve ex-
tremely high energy and electron transfer pathways repre-
sents a great challenge in chemistry and materials science.
This goal lies at the foundation of various optoelectronic ap-
plications, such as photovoltaic devices and field effect tran-
sistors (FETs). Within the frame of fundamental research
focused on finding new systems for implementing these ap-
plications, the endeavour has been addressed towards the
development of new structures featuring a cofacial arrange-
ment of chromophores such as polycyclic aromatic systems
to allow high degree of p-orbital overlap and significant ex-
citon coupling. High electron mobilities[1,2] and also fast ex-
citon transport over large distances[3] have been measured in
such cofacial systems. Several approaches to obtain well or-
ganized perylene-bis(dicarboximides) (PDIs) structures for
application in solar cells and FETs are described in the liter-
ature including studies on perylene crystals,[4–6] aggre-
gates,[7–18] and liquid crystals.[19–21] Alongside the aggregation
of monomeric PDIs units, also several examples have been
reported in which face-to-face aggregation of PDIs is pro-
moted by an oligomer or polymer architecture.[22–28] These
investigations have been mainly concentrated on flexible
polymers architectures. We therefore decided to investigate
the incorporation of PDI units in the backbone that can
adopt a very stable well-defined 41 helical conformation
when the side chains exhibit a bulky group or bears addi-
tional stabilizing interactions, such as p–p stacking and/or
hydrogen bonding between the repeat units in the n and n+
4 positions.[29–32] The introduction of peptide substituents
orders the side groups into a precise architecture, resulting
in an attractive scaffold onto which functional groups can be
arranged into well-defined arrays, such as thiophenes,[33] por-
phyrins[34] and perylenes diimides,[35,36] thus creating new ma-
These polymers could therefore be ideal candidates for
photovoltaic cells, although unfortunately they suffer from a
very poor solubility in organic solvents.[35,36] Moreover, the
organization of the perylenes makes the material interesting
for both nanoscale or macroscopic wires and field-effect
transistors (FETs). In this paper we focus on the relation-
ship between the inter-chromophore interactions in two dif-
ferent PDI functionalized polyisocyanopeptides and the
(photophysical) properties of the material.
We report on the synthesis and characterization of two
new PDI-polyisocyanide systems both having an improved
solubility in organic solvents. In particular we have devel-
oped polymer P2 exposing sterically demanding phenoxy
substituents in the bay area of the PDI, and polymer P3
bearing longer alkyl tails in the side chains of “ultra-rigid”
polyisocyanides. Polyisocyanides possess a unique polymer
periphery (Figure 1). The incentive to prepare and study P2
was that the bulky phenoxy groups in the PDI bay area may
hinder the well-known tendency of PDIs to stack through
p–p interactions. Furthermore, the bulky character of the
non-planar PDIs in P2 may inhibit the formation of exci-
mers, which might be beneficial for excitons transport, since
excimer sites act as energy traps in which energy is lost as
fluorescence. In order to achieve a thorough understanding
on these three polymer systems, we combined the (photo-
physical) characterization in solution with an extensive mod-
elling of the structural properties of the perylene-functional-
ized polymer P3. For the sake of comparison, polymer P1,
which consists of a central PIC backbone exposing an ala-
nine–alanine segment and terminal methoxy groups, was
studied.
Figure 1. Molecular structure of the polymers P1–P3. The amide unit that is responsible for the stabilization of the helical structure by a hydrogen bond-
ing network is encircled in green. A schematic structure of P3 is depicted, with the backbone in blue and the PDI groups in red.
Chem. Eur. J. 2009, 15, 2536 – 2547
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2537

R. H. Friend, D. Beljonne, P. Samori, A. E. Rowan et al.
Results and Discussion
agent. Dehydration of formamide 8 with diphosgene and N-
methylmorpholine (NMM) resulted in the formation of iso-
cyanide monomer 9. In an analogous manner of the synthet-
ic route from amine 5 to isocyanide 9, the isocyanide mono-
mer 13, containing bulky phenoxy substituents, was pre-
pared following the same procedure starting from amine 10
(Scheme, Supporting Information).
Isocyanide 9 was polymerized in 60 minutes using nickel
complex 14 (1/1000 equiv) (Figure S2). The azide in the ini-
tiator complex 14 can be used as a possible further reaction
site for modification of the polymer terminus with for exam-
ple, a second polyisocyanide block or other functional
groups because it ends up in the end group of the polyiso-
cyanide, however, work in this area is not presented in this
paper.
Upon polymerization of isocyanide 9, the reaction mix-
ture turned from yellow to red, which is a typical for the ag-
gregation of PDIs and in this particular case indicative of
the polymerization leading to intermolecular aggregation of
the PDIs. The polymer was purified by repetitive precipita-
tion; first in methanol/water (1:1 v/v) and then in methanol,
until no monomer fluorescence was visible. The solubility of
P3 was found to be much better than that of the perylene
functionalized polyisocyanide described in previous
work,[35,36] and therefore polymer P3 is much better process-
Synthesis of monomers and polymers
Polymers P1–P3 were prepared via a nickel-catalyzed poly-
merization reaction of isocyanide monomers. The synthesis
and characterization of polymer P1 has been described in
detail in the literature and will not be discussed here.[38] The
perylene-bis(dicarboximide) functionalized polymer P3 was
synthesized following a slightly modified strategy to that de-
scribed for the synthesis of PDI functionalized polyisocya-
nides.[35,36] First, hexylheptyl amine was treated with pery-
lene dianhydride 1 to give the symmetrically substituted per-
ylene diimide 2, which after partial hydrolysis with potassi-
um hydroxide in tert-butanol yielded perylene mono-anhy-
dride 3 (Scheme 1). Subsequently 3 was functionalized with
a tert-butyloxycarbonyl (Boc)-protected amine by reaction
with mono-Boc-1,3-diaminopropane. The protected amine 4
was converted into the free amine 5 by removal of the Boc
protection with an excess of trifluoroacetic acid (TFA). In
the following step, amine 5 was coupled to Boc-l-alanine to
give 6, using 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide
hydrochloride (EDC) and hydroxybenzotriazole (HOBt) as
coupling reagents. After removal of the Boc protection,
amine 7 was formylated using 4-nitrophenyl formate as re-
Scheme 1. Synthesis of PDI functionalized polyisocyanides P3 (R=CH2CH2CH2N3). a) 1-hexylheptylamine, DMF/Imidazole, 1108C; b) KOH, tert-buta-
nol; c) Boc-1,3-diaminopropane, DMF/imidazole, 958C; d) TFA, CH2Cl2, quant.; e) Boc-l-alanine, EDC, HOBt, DIPEA, CH2Cl2, 67%; f) TFA, CH2Cl2,
quant.; g) 4-nitrophenyl formate, CH2Cl2, 92%; h) diphosgene, N-methylmorpholine, CH2Cl2, ꢀ308C, 88%; i) 14, CH2Cl2, 98%.
2538
www.chemeurj.org
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 2536 – 2547

Perylene Polyisocyanopeptides
FULL PAPER
able. Polymer P3 was soluble (to at least up to 10 mgmLꢀ1)
in chlorinated solvents (CH2Cl2, CHCl3, dichloroethane, tet-
rachloroethane) and solvents such as benzene, toluene and
tetrahydrofuran.
a
shift in absorption from 3376 cmꢀ1 for PDI 13 to
3308 cmꢀ1 for P2 characteristic of hydrogen bonding be-
tween the amides (Figure S3). From this IR data it can be
concluded that the attachment of the perylene to the central
polymer chain does not affect the hydrogen bonding net-
work of the polyisocyanide. (Table S1).
In contrast to the fast polymerisation of 9, it took seven
days before monomer 13 (having phenoxy substituents) was
completely consumed as was concluded from TLC analysis.
This slow polymerization can be attributed to the bulkiness
of the PDIs and occurs at a similar rate to that observed for
the polymerization of tert-butyl isocyanide.[39] During the
polymerization of 13 to P2 the reaction mixture turned
slightly darker, from purple to deep purple, but a clear
colour change as visible in the preparation of P3 was not ob-
served. Polymer P2 was purified by size-exclusion chroma-
tography and was seen to be soluble in most organic sol-
vents, toluene and THF. Determination of the absolute mo-
lecular weight of P2 and P3 using gel-permeation chroma-
tography (GPC) is difficult because the behavior of a rod-
like polymer in a gel is not straightforward. Therefore, we
estimated the MWs by AFM using dilute solutions of the pol-
yisocyanide (10ꢀ6 in CHCl3) spincoated on freshly cleaved
mica. This approach reveals single fibers and due to the ri-
gidity the polymers are extended on the surface and their
contour lengths can be accurately measured. This approach
has been used by Prokhorova et al.[40] to determine the abso-
lute molecular weight of polymethacrylates and polystyrenes
having bulky substituents, and could be applied to our polyi-
socyanopeptides as well.[41] According to AFM images of P3
on muscovite mica, the average length of P3 is 180 nm.
From the analysis of AFM images of P2 on muscovite mica,
an average polymer chain length of 110 nm was determined.
Assuming a 41 type helix with a helical pitch of 0.46 nm
(measured by Powder X-ray diffraction of P1,[41,42], 180 nm
corresponds to 1.2ꢆ103 repeat units and an Mn value of 9ꢆ
105 gmolꢀ1. For P2 this leads to an Mn value of 1.1ꢆ
106 gmolꢀ1.
UV/Vis spectroscopy: The absorption spectrum of P3 in
chloroform shows vibronic bands of the PDI S0–S1 optical
transition between 400 and 600 nm (Figure 2). Compared
with the spectrum of monomer 9, the absorption spectrum
of the polymer revealed a blue shifted maximum and a rela-
tively increased oscillator strength for the 0–1 and 0–2 vibra-
tional transitions over the 0–0 transition; this can be quanti-
fied by the ratio of the absorptions A0!0/A0!1 which has
value of 1.66 for the monomer and 0.66 for the poly-
mer.[24,26,27] In addition, the maxima of all vibronic bands
were red shifted by 2–6 nm and a broad absorption tail was
found to be present at lower energy than the 0–0 transition
of monomer 9. These spectral changes observed upon poly-
merization are reminiscent of weak exciton coupling interac-
tions (strong exciton–phonon coupling) between the PDIs in
H-type aggregates.[44,45] Polyisocyanides are known to adopt
a 41 helical structure, in which dipole–dipole interactions
(C1–5) between the nth and the (n+4)th units dominate. In
such weakly coupled aggregates, excited states are grouped
into vibronic bands that correspond to the single molecule
vibronic lines (0–0, 0–1, etc.) in the case of vanishingly small
Sepctroscopical methods
Infrared spectroscopy: To obtain greater insight into the
structures of the polymers, the polymerization of 9 with cat-
alyst 14 (1/740 equiv) was followed by IR spectroscopy
(Figure S2). Most characteristic for the polymerization is the
disappearance of the isocyanide peak at n˜ 2140 cmꢀ1 and
from this signal the reaction was determined to be com-
pletely finished after ꢁ1000 s. Upon polymerization the ab-
sorptions at 3438 and 3391 cmꢀ1, attributed to the trans and
[43]
ꢀ
cis amide N H stretch vibrations, respectively, shifted to
3294 cmꢀ1 indicative for hydrogen-bond formation of the
amides, exclusively in the trans configuration. An intensity
decrease at about 1686 cmꢀ1 and an intensity increase at
about 1649 cmꢀ1 as observed in Figure S2 showing the differ-
ence spectrum, are ascribed to a shift of the amide I vibra-
tion (C=O stretch) due to hydrogen bonding. The slow poly-
merization of monomer 13 was not followed by IR spectros-
copy, however, the solid-state IR spectrum of polymer P2
compared to that of the monomer in the solid also revealed
Figure 2. a) Absorption spectrum of P3 (grey) and monomer 9 (black)
(CHCl3; 10ꢀ5 m). Inset: illustration of the different interactions between
the PDIs in polymer P3. b) Absorption spectrum of P2 (grey) and mono-
mer 12 (black). Inset: illustration of the S0–S1 and S0–S2 transition dipoles
(CHCl3; 10ꢀ5 m).
Chem. Eur. J. 2009, 15, 2536 – 2547
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2539

R. H. Friend, D. Beljonne, P. Samori, A. E. Rowan et al.
electronic interactions. Of particular interest is the relative
intensity of the A0!0 band as the intensity of this band di-
minishes with increasing exciton bandwidth and is thus a
useful measure of the electronic coupling.[44–46]
The formation of exciton states resulting from intermolec-
ular interactions leads to broader spectral features in P3
compared to 9. The loss of oscillator strength in the A0!0
band (and simultaneous increase in higher energy bands) is
primarily due to the interband coupling between zero-order
vibronic excitons in the A0!0 band with zero-order vibronic
excitons in the higher energy bands (A0!1, A0!2, etc.), medi-
ated by excitonic interactions.[47] Such interactions lead to a
decrease in the A0!0/A0!1 ratio as well as a slightly in-
creased spectral separation between the 0–0 and 0–1 absorp-
tion peaks. The demise of the A0!0/A0!1 ratio is a precursor
to the more conventional blue-shifted main absorption peak
characteristic of H-aggregates with much stronger excitonic
coupling. The rigid red shift of the whole spectrum upon
going from monomer 9 to polymer P3 is likely due to a gas-
to-crystal shift induced by aggregation (though it is possible
that the weak C1–2 interactions also contribute to a small red
shift). A detailed analysis of the absorption spectrum of P3
is provided in the modelling section.
Figure 3. CD spectrum of polymer P3 (top) and polymer P2 (bottom)
(CHCl3; 10ꢀ5 m).
same stack (C1–5) and with units in adjacent stacks (C1–2
)
(see Figure 5), while the helicity is close to 41. While the
signal arising from interactions between two nearest PDIs in
different stacks reflects the helicity of the polymer back-
bone, the one between two perylenes within the same stack
originates from the helicity of the S0–S1 transitions within
the PDI stacks.[51] The positive signal at higher wavelength,
most likely originating from a positive bisignate signal,
would correspond to a right handed polymer helix.[52,53] This
is in agreement with the smaller positive cotton effect at
317 nm originating from the helicity in the polyimine back-
bone. The latter signal is similar to that observed for right
handed helices of P1 and also reflects the robustness of the
isocyano-l-alanyl backbone, which is not influenced by the
outer PDI substituents.[29,49,54] This robustness is also reflect-
ed by the identical CD spectra of polymer P3 in different
solvents such as, chloroform, tetrachloroethane and toluene.
The CD spectrum of P2 displayed a signal for the pery-
lene absorptions with a lower intensity (30 times) than that
observed for P3 (Figure 3). This suggests that the PDIs in
the C1–4 geometry (Figure 2) have either a very small twist
angle with respect to each other or that this angle fluctuates
between positive and negative values along the stack. It is
possible that the weak CD effect originates not from exciton
coupling, but from induced chirality in the perylene plane;
for these phenoxy substituted PDIs an excess of an M or P
twist is likely to be induced by the incorporation of the
PDIs in the helical polymer backbone. Such an effect has
previously been observed for aggregates of similar PDIs for
which a similar CD effect was observed.[50] A relatively
strong signal for P2 was visible at l=284 nm, preceded by a
broad signal around l=338 nm. The signal located at
280 nm is most likely caused by chiral interactions between
the phenoxy substituents, but also the imine backbone signal
is expected to be present in this region. It is therefore diffi-
cult to conclude from the CD signal in this region what is
the helix conformation of the polymers.
In the absorption spectra of P2 and 12, besides the S0–S1
transition, the S0–S2 transition was also visible at l=448 nm
(Figure 5). The S0–S1 transitions of P2, compared to mono-
mer 12, revealed relatively increased intensities of the 0–1
and 0–2 oscillations with respect to the 0–0 oscillation; A0!0
/
A0!1 equals 1.48 for the monomer and 1.07 for the polymer,
suggesting (weak) electronic interactions between the chro-
mophores.[24,26,27] In addition, a blue shift of 4 nm of the ab-
sorption maxima was detected and in contrast to P3, no red
shifted component was visible for the S0–S1 transitions. For
S0–S2 transition a tiny red shift of 2 nm was observed. The
relatively increased oscillator strength for the S0–S1 transi-
tion in the blue is similar to observations for PDI dimers
and might be indicative for H-aggregation of the PDIs, how-
ever, the absence of a red shifted component is atypical and
has not been observed in the dimer model systems.[48] Be-
sides exciton coupling, changes in the spectrum might also
be caused by conformational changes induced by steric in-
teractions between the bulky PDIs, which makes the inter-
pretation of the absorption spectra more complex.
CD spectroscopy: Since the polymers adopt a helical ar-
rangement induced by the l-alanine unit in the side arms of
the polymer, circular dichroism (CD) is a powerful tool to
analyze the arrangement of the chromophores in the polyi-
socyanopeptides.[34,49] The CD spectrum of P3 revealed
strong positive and weaker negative signals for the exciton
coupled S0–S1 vibronic transitions between 420 and 600 nm
(Figure 3). No CD signal was detected for the monomer.
The CD effect is unlike any other signals observed for chiral
perylene bisimides stacks and can be rationalized by consid-
ering the multiple chiral interactions in which every pery-
lene bisimide unit is involved.[14,50] In P3 every chromophor-
ic unit has chiral interactions with adjacent units within the
The thermal stability of the perylene-functionalized poly-
mers P2 and P3 was investigated by performing tempera-
ture-dependent CD measurements. The CD signal of P3 at
l=504 nm in tetrachloroethane (TCE) gradually decreased
in changing from ꢀ10 to 1108C, most likely due to increased
thermal motion of the perylene bisimide. The initial signal
2540
www.chemeurj.org
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 2536 – 2547

Perylene Polyisocyanopeptides
FULL PAPER
was almost completely restored cooling the solution back to
ꢀ108C, and remained the same even after five heating-cool-
ing cycles (Figure S4).[55]
identified to be a right-handed helix (for an l-alanine amino
acid) with an average angle of 21.38 between two successive
units along the helix axis, 3.76 units per turn and a pitch
length of 67.8 ꢇ. Such a conformation was very robust
against long molecular dynamics simulations at room tem-
perature.
The predicted geometric parameters are in excellent
agreement with the corresponding data extracted from a de-
tailed spectroscopic studyof polyisocyanides substituted with
porphyrin derivatives (Table 1).[34] Thus, the positioning in
In contrast, upon heating a solution of P2 in TCE from 20
to 1108C all absorption, emission and CD signals decreased
and were not restored upon cooling. The signal also slowly
decreased in time without any heating, while this was not
observed for the solution in CH2Cl2. This indicates that the
conformation in TCE is thermodynamically not stable. In
conclusion, these CD and DSC measurements indicate that
heating of P3 up to 1108C does not result in irreversible
changes in the polymer, reflecting the stable and thermody-
namic nature of these systems.
Table 1. Comparison of the calculated (Dreiding force field) structural
parameters of unsubstituted polyisocyanides with the corresponding
values extracted from optical studies of polyisocyanides bearing porphy-
rin derivatives.[34]
Modeling studies
Angle between units Distance between units Units
Pitch
length
[ꢇ]
n and n+4 (b) [8]
n and n+4 (C1–5) [ꢇ]
per
Morphology investigations: As a first step in the modeling
study of polyisocyanides with pendant perylene diimides, a
complete morphological investigation was performed. The
conformation of the polyisocyanide template in the absence
of chromophores was first explored using the Dreiding force
field as it is particularly suited to reproduce hydrogen
bonds, which play a crucial role in the stability of the polyi-
socyanide backbone.[56] The torsion potential around the po-
lymer backbone, however, is not properly accounted for by
the original Dreiding force field and has been corrected
against ab initio MP2 6-1G** quantum-chemical calcula-
tions. The polymer helices were built according to a “step-
by-step” approach where oligomers of increasing size were
grown. At each step, the lowest energy structures were re-
tained for the generation of the larger size oligomers. From
this conformational study, the most stable conformer was
turn
Dreiding
experiment
21.3
22
4.34
4.2
3.76
3.75
67.8
68.7
space of the chromophores in these helices is fixed to a
large degree by the template effect enforced by the polyiso-
cyanide core, which in turn arises mostly from the formation
of an H-bonded network (Figure 4). In addition, dynamic
modelling revealed that all the H-bonds point in the same
direction, explaining the extremely high dipole moment of
these molecules (in the range 2.8–3.6 D per repeating
unit).[57] The next step of the modelling involved the grafting
of PDI chromophores onto the helical polymer. The PDI
molecules were linked to the backbone via a short saturated
Figure 4. Most stable conformation of the unsubstituted polyisocyanide helix as obtained from Dreiding force field calculations (top). The H-bonds are
indicated by dashed lines. Top view of h1, h2, and h3 (bottom).
Chem. Eur. J. 2009, 15, 2536 – 2547
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2541

R. H. Friend, D. Beljonne, P. Samori, A. E. Rowan et al.
spacer that was submitted to a conformational search. Based
on the local minima found for the relevant torsions, all pos-
sible helical stacks were constructed, among which the most
realistic conformations based on simple steric arguments
were selected for further investigations. The values of the
torsion angles defining the conformation of the spacer and
their combination had a considerable impact on the relative
arrangement of the chromophores in the resulting 3D struc-
tures. Geometric optimizations were performed on systems
of increasing complexity and converged to the identification
of three local minima on the potential energy surface. To
avoid finite size effects along the polymer direction, periodic
boxes out of these three structures were subjected to further
structural refinements. These helices, hereafter referred to
as h1, h2 and h3, differ by the relative orientation of the
PDI chromophores with respect to the main axis of the pol-
yisocyanide backbone (Figure 4). In h1, the PDIs are orient-
ed with their long axis forming an angle of ꢁ608 with the
helical direction. This is an open structure that shows the
lowest density among the three helical arrangements.
tural changes as far as the relative arrangement of the PDI
units is concerned (while the stiff helical backbone remained
largely unaffected). The spacers connecting the PDI units to
the carbon backbone thus allows some flexibility in the spa-
tial positioning of the chromophores that organize into or-
dered domains separated by defects. This is particularly pro-
nounced in the case of the sparse helix h1, which collapses
into a structure similar to h2 along the MD run. The mea-
sured angles between the PDI main axis and the helix direc-
tion indeed switched from an average value of 608 at time
zero to 748 after 300 ps. The denser structures h2 and h3
maintained their overall organization during the full MD
simulations (the corresponding rotation angles vary from 94
to 888 for h2 and from 26 to 278 for h3). Thus, the morpho-
logical analysis has allowed pinpointing three structures that
differ by the conformation of the arms connecting the PDI
to the template and yield different relative spatial orienta-
tions of the chromophores. From the three modeled helical
conformations, h2 and h3 appear to be the most stable con-
formers.[58]
The helix h3, in contrast to h2, presents a much more
compact, pinetree-like structure where the PDI chromo-
phores are lying almost down on the helical core (the aver-
age angle between the PDI main axis and the helix direction
is 268). A similarly dense structure is achieved in h2, al-
though in contrast to h3, the chromophores are maintained
in an orthogonal orientation (average angle of 948) with re-
spect to the helical growth direction. The interactions be-
tween the PDIs are different for all three helices. In h1 the
PDIs interact mostly via face-on interactions, while in h2 the
chromophores are lying more in an edge-on organization. In
h3 the PDI interactions also involve a cofacial arrangement,
yet with translational displacement along the PDI long axes.
In each of the three cases, the polymer self-organizes into a
right-handed helix with 3.7–3.9 molecules per helical turn
(resulting in an average rotation angle between two succes-
sive PDI along the stacking direction of ꢁ218 for h1, ꢁ248
for h2 and ꢁ118 for h3). With the exception of h3, the
angles are close to the value measured for the porphyrin-
substituted polyisocyanide (228), thus confirming that graft-
ing of the PDI units does not perturb the structural organi-
zation of the polyisocyanide template.[34] From an energetic
point of view, the potential energies calculated for these
local minima indicate that h3 is the most stable(mainly due
to attractive Van der Waals interactions), followed by h2
and h1.
The three local minima, identified from the geometric op-
timizations at 0 K then were then subjected to molecular dy-
namics simulations at room temperature. The aim of this
study was twofold: i) it allows gauging the stability of the
conformers with respect to thermal fluctuations and, in addi-
tion, ii) the snapshots extracted along the molecular dynam-
ics (MD) run can subsequently be used to assess the influ-
ence of conformational motion on the electronic and optical
properties of the polymer. A comparison between the mor-
phologies of the various structures obtained along the MD
trajectory revealed that these follow quite substantial struc-
Spectroscopic investigations
The geometric structures generated from the morphology
studies described above were used as input for quantum-
chemical calculations of the electronic excited states (per-
formed at the INDO/CCSD level).[59,60] The results of these
calculations were in turn injected into a phenomenological
Holstein model to predict the spectroscopic properties of
the helical supramolecular structures, namely the optical ab-
sorption and circular dichroism spectra. A single effective
high-frequency (0.17 eV) vibrational mode was included
whose exciton-phonon coupling (Huang–Rhys factor of 0.6)
was fitted against the absorption spectrum of an isolated
PDI chromophore in solution. The model and its applica-
tions to spectroscopy have been largely described in previ-
ous works.[61,62] The polyisocyanide helices possess less than
four monomer units per turn, and therefore the PDIs graft-
ed onto the template yield four quasi-one-dimensional
stacks. The absorption and CD spectra were simulated for
20-mer chains (i.e., chains formed by four 1D stacks of five
molecules each); test calculations on longer oligomers yield
very similar results.[63]
The comparison of the measured and simulated (from
snapshots at the beginning and end of the MD run) absorp-
tion spectra for the three helical structures is shown in
Figure 5. A reasonable agreement between theory and ex-
periment is observed in all cases, with a slightly better
match for helix h1. The spectra show the characteristic fea-
tures of a weakly coupled aggregate (see above).[64] At this
stage it is very difficult to rule out any structure from the
comparison between measured and simulated absorption
spectra, since linear absorption is not very sensitive to long-
range interactions.[46]
As the nearest neighbour couplings are very similar from
one conformer to another, the computed spectra are only
weakly dependent on the helical conformation and on its
2542
www.chemeurj.org
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 2536 – 2547

Perylene Polyisocyanopeptides
FULL PAPER
in fact, all present in solution and contribute to the overall
optical properties of the polymer.
CD spectroscopy provides a more critical test for the spa-
tial organization of the chromophores in the helical polymer
as it is far more sensitive to long-range interactions than
linear absorption.[65,66] When two chromophores are elec-
tronically coupled in a chiral structure, the two electronic
excited states resulting from the interactions feature rota-
tional strengths with opposite signs. Therefore, the intermo-
lecular interactions between conjugated segments in helical
conformations usually result in a double-peak signal (bisig-
nated Cotton effect) in the CD spectrum; right-handed heli-
ces are characterized by a positive peak followed by a nega-
tive peak with increasing energy, while the reverse is true
for left-handed helices. All conformers generated by the
force field calculations were right-handed, at least from the
mere inspection of the twisting angles between closely
spaced PDI units along the four individual stacks. The ex-
perimental CD spectrum complies with this and shows the
“+/-” sequence expected for a right-handed helix.
The simulated CD spectra for the three conformers,
before and after the MD run are shown in Figure 6. Strik-
ingly, h3 shows a CD signal that is inverted with respect to
h2 and to experimental results. This is also the case for h1 at
t=0 ps, but the bisignated effect switches sign during the
MD simulation. All individual stacks extracted from the
three conformations are, however, right-handed, as proven
by calculations retaining only molecules n, n+4, n+8, etc.
Calculation of the rotational strength for all possible PDI
pairs within the full structures (not shown) revealed the
presence of both positive and negative interactions. In fact,
on top of the right handedness conferred by the relative po-
sitions of the chromophores within each stack, molecules be-
longing to different stacks also arrange according to helical
structures with opposite (left) handedness. The balance be-
tween the “right” and “left” interactions depends on the
conformation of the polymer chains, with only h2 providing
a qualitative match with experiment. It is interesting to see
that the change in structure of h1 evolving toward that of h2
along the MD trajectory is accompanied by a reversal of the
CD spectrum. It can be concluded from the modelling stud-
ies that the most probable conformation for these helical
arrays in solution and the solid state is the architecture h2.
This modelled architecture accounts not only for all the
physical observations (UV/Vis and CD), but predicts that
the polyisocyano-perylenes would be ideal polymers for ex-
citon and electron migration. To study this we examined the
photophysical behaviour of P2 and P3.
Figure 5. Theoretical absorption spectra of structures a) h1, b) h2 and c)
h3, before (t=0 ps, a) and after the MD run (t=300 ps, g), overlaid
to the experimental spectrum of P3 (c) measured in a chloroform so-
lution. Note that the theoretical spectra have been rigidly shifted to re-
produce the position of the measured absorption maxima,
Fluorescence spectroscopy: The fluorescence spectrum of
P3, compared to that of monomer 9, showed a broad struc-
tureless red shifted emission at l=625 nm with a fluores-
structural fluctuations during the MD run. This is particular-
ly striking for h1, which shows very similar absorption line
shapes before and after the MD simulation, despite pro-
nounced structural reorganizations. In addition, it is also
conceivable that the limiting structures considered here are,
cence quantum yield of 3.7% (in solution (CHCl3) in air[67]
)
upon excitation at the absorption maximum at l=492 nm
(Figure 7). The low quantum yield and long fluorescence
lifetime (see photophysical studies) of the emission are rem-
iniscent of a parallel orientation of the transition dipoles
Chem. Eur. J. 2009, 15, 2536 – 2547
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2543

R. H. Friend, D. Beljonne, P. Samori, A. E. Rowan et al.
Figure 7. a) Solution fluorescence spectra of polymer P3 (intensity ꢆ5,
grey) and monomer 9 (black) (CHCl3; 10ꢀ6 m). b) Solution fluorescence
spectra of polymer P2 (grey) and PDI 12 (black) (CHCl3; 10ꢀ6 m).
formed only after excitation of monomeric species and
therefore the emission, but not the absorption, is red-shifted.
The chromophores in an excimer are typically 0.3–0.4 nm
apart and can involve more than two chromophoric units in
molecular aggregates.[35,36,68,69] The formation of excimers of
aromatic hydrocarbons is restricted to a parallel cofacial
configuration and confirms the conclusion that the PDIs
stack in an H-aggregated fashion.
The fluorescence spectrum of P2 has a slightly redshifted
emission at l=607 nm compared to l=605 nm observed for
the monomer. This value is comparable to the shifts in the
absorption spectrum (Figure 7). The fluorescence quantum
yield of P2 amounted to 3.0% with respect to the monomer.
This low quantum yield is in line with H-aggregated PDIs,
however, it can also be attributed to electron transfer be-
tween the PDI and neighbouring phenoxy substituents. The
emission spectra gave no evidence for excimer emission, as
was observed for P3, which is consistent with the conclusion
that the steric bulk of the phenoxy groups of the PDIs in P2
seem to be able to prevent excimer formation.
To investigate the photophysical properties and excited
state dynamics of P3, time-correlated single photon counting
(TCSPC) and transient absorption spectroscopy were per-
formed. The TCSPC technique (Figure 8a) revealed an ex-
traordinarily long radiative lifetime of ꢁ24 ns in a CHCl3
solution, as compared to the usual value of 3–4 ns in the per-
ylene monomer;[70,71] this is strongly indicative of excimer-
like excited states in the polymer.[26,35,36] When the polymer
was spun onto a film, the system decayed rapidly and multi-
exponentially due to the three dimensional energy migration
in the film.
Figure 6. Theoretical CD spectra before (t=0 ps, a) and after the MD
run (t=300 ps, g) of structures a) h1, b) h2 and c) h3 overlaid to the
experimental spectrum of P3 (c) measured in a chloroform solution;
spectra simulated for individual 1D stack a.
and are in agreement with the absorption characteristics.
The featureless emission of the polymer, which is red-shifted
by more than 70 nm, together with the fact that the shift in
the emission spectrum is much larger than the shift in the
absorption spectrum, is indicative of the formation of an in-
tramolecular excimer-like species, although the effect of ag-
gregate states can not be ruled out. Excimer species are
2544
www.chemeurj.org
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 2536 – 2547

Perylene Polyisocyanopeptides
FULL PAPER
Polarization-sensitive transient absorption measurements
of P3 (Figure 8b) showed that the anisotropy in photo-in-
duced absorption decays on timescales of the order of pico-
seconds, which is faster than in other known p-conjugated
polymers, such as polythiophene.[72]
polymers are promising compounds for light-harvesting ap-
plications, such as photovoltaics[75] and thin-film transistor
architectures.[76]
We consider that the faster energy transfer and depolari-
zation process which take place in P3 are due to its compact
helical structure,[73] which promotes rapid exciton migration
along the chiral arrangement of perylene stacks. The depola-
rization process is especially fast in thin films, reflecting the
higher dimensionality of migration, relative to solution
measurements.
The transient absorption spectra (Figure 8c–d) show
ground-state absorption bleaching around l=525 nm, as
well as a broad (from l=550 nm to at least l=750 nm)
photo-induced absorption (PA) band; the latter of these
being indicative of strong interactions between the chromo-
phores.[74] The spectra obtained from both solution and film
showed these features, although the shapes of the PA bands
were rather different; this might be due to charge genera-
tion in the film. Furthermore, the non-linear dependence of
the PA intensity on pump-fluence (Figure S5) suggests that
bimolecular annihilation processes takes place in the system.
Early time decay kinetics are very much faster than the ns
decay rates for excitons at longer times.
Conclusion
A careful study on two novel soluble perylene-based poly-
isocyanopeptides performed using various experimental and
computational methodologies provided unambiguous evi-
dence that these ultrastiff polymers are ideal scaffolds for
precisely organizing chromophoric arrays into functional 2D
wires. The precise organization of the polymer and its side
chain has been, for the first time, fully analyzed for two dif-
ferent polymers exposing perylene dyes. Extensive molecu-
lar modelling dynamics refines previous models, revealing a
41 helix in which the chromophores overlap along the poly-
mer backbone. The calculated and spectroscopically ob-
served architectures predict that these unique polymers are
highly favorable systems for electron transport. This was
confirmed by transient absorption spectroscopy studies,
which indicates extremely high exciton migration rates and
charge densities. The application of these materials in thin-
film transistors (TFTs) and photovoltaics is currently under
study. The scaffolding approach employed here is very ver-
satile and can be employed to dictate the spatial localization
of functional groups (possessing for example, optical, electri-
cal or catalytic activity) in pre-determined positions, paving
High fractional changes in optical density (above ca. 1%
in solution) are observed at short times showing an extreme-
ly high excitation density on chains. This suggests that the
Figure 8. a) Time-resolved photoluminescence decay dynamics, measured at l=625 nm. b) Photo-induced absorption anisotropy measurements on P3,
detected at 690 nm, in both film and in chloroform solution. The excitation was at 500 nm with a fluence of 424 mJcmꢀ2. Transient absorption (fractional
change in optical density {OD}) measurements on P3, being excited at 500 nm, are shown in solution (c) and in film (d). The fluences of the transient ab-
sorption measurements for the solution and film were 353 and 389 mJcmꢀ2, respectively. The spectral features of bleaching and photo-induced absorption
(PA), as a function of delay time, are indicated for clarity.
Chem. Eur. J. 2009, 15, 2536 – 2547
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2545

R. H. Friend, D. Beljonne, P. Samori, A. E. Rowan et al.
[21] F. Wꢄrthner, C. Thalacker, S. Diele, C. Tschierske, Chem. Eur. J.
2001, 7, 2245–2253.
[22] J. J. Han, W. Wang, A. D. Q. Li, J. Am. Chem. Soc. 2006, 128, 672–
673.
the way towards applications in various fields including op-
toelectronics, catalysis and medicine.
[23] C. Hippius, F. Schlosser, M. O. Vysotsky, V. Bohmer, F. Wꢄrthner, J.
Am. Chem. Soc. 2006, 128, 3870–3871.
[24] A. D. Q. Li, W. Wang, L. Q. Wang, Chem. Eur. J. 2003, 9, 4594–
4601.
Acknowledgement
We would like to thank B. Simons for help with AFM measurements.
This work was supported by the Netherlands Organization for scientific
research chemistry section (top grant for R.J.M.N and V.I.C.I, A.E.R),
Nanoned STW A.E.R., the Royal Netherlands Academy for Arts and
Sciences (KNAW), ESF-SONS2-SUPRAMATES and ESF-SONS-BION-
ICS projects, the EU through the projects Marie Curie EST-SUPER
(MEST-CT-2004-008128), Marie Curie RTNs PRAIRIES (MRTN-CT-
2006-035810), ForceTool (NMP4-CT-2004-013684), the ERA-Chemistry
project SurConFold and the Regione Emilia-Romagna PRIITT Nanofab-
er Net-Lab. C.E.F. thanks the Leverhulme Trust (UK) for an “Early
Career Fellowship”.The work at Temple is supported by the National Sci-
ence Foundation, grant #DMR-0606028. The work in Mons is partly sup-
ported by the Interuniversity Attraction Pole program of the Belgian
Federal Science Policy Office (PAI 6/27) and by FNRS-FRFC. D.B. is re-
search director of FNRS. S.T. acknowledges a grant from Fonds pour la
Formation ꢂ la Recherche dans lꢈIndustrie et dans lꢈAgriculture (FRIA).
[25] E. E. Neuteboom, R. A. J. Janssen, E. W. Meijer, Synth. Met. 2001,
121, 1283–1284.
[26] E. E. Neuteboom, S. C. J. Meskers, E. W. Meijer, R. A. J. Janssen,
Macromol. Chem. Phys. 2004, 205, 217–222.
[27] W. Wang, L. S. Li, G. Helms, H. H. Zhou, A. D. Q. Li, J. Am. Chem.
Soc. 2003, 125, 1120–1121.
[28] W. Wang, W. Wan, H. H. Zhou, S. Q. Niu, A. D. Q. Li, J. Am. Chem.
Soc. 2003, 125, 5248–5249.
[29] J. J. L. M. Cornelissen, A. E. Rowan, R. J. M. Nolte, N. A. J. M. Som-
merdijk, Chem. Rev. 2001, 101, 4039–4070.
[30] R. J. M. Nolte, Chem. Soc. Rev. 1994, 23, 11–19.
[31] Y. Okamoto, T. Nakano, Chem. Rev. 1994, 94, 349–372.
[32] M. Suginome, Y. Ito, Adv. Polym. Sci. 2004, 171, 77–136.
[33] D. M. Vriezema, J. Hoogboom, K. Velonia, K. Takazawa, P. C. M.
Christianen, J. C. Maan, A. E. Rowan, R. J. M. Nolte, Angew. Chem.
2003, 115, 796–800; Angew. Chem. Int. Ed. 2003, 42, 772–776.
[34] P. A. J. de Witte, M. Castriciano, J. J. L. M. Cornelissen, L. M. Sco-
laro, R. J. M. Nolte, A. E. Rowan, Chem. Eur. J. 2003, 9, 1775–1781.
[35] P. A. J. de Witte, J. Hernando, E. E. Neuteboom, E. M. H. P. van
Dijk, S. C. J. Meskers, R. A. J. Janssen, N. F. van Hulst, R. J. M.
Nolte, M. F. Garcia-Parajo, A. E. Rowan, J. Phys. Chem. B 2006,
110, 7803–7812.
[36] J. Hernando, P. A. J. de Witte, E. M. H. P. van Dijk, J. Korterik,
R. J. M. Nolte, A. E. Rowan, M. F. Garcia-Parajo, N. F. van Hulst,
Angew. Chem. 2004, 116, 4137–4141; Angew. Chem. Int. Ed. 2004,
43, 4045–4049.
[37] P. Samori, C. Ecker, I. Gossl, P. A. J. de Witte, J. J. L. M. Cornelis-
sen, G. A. Metselaar, M. B. J. Otten, A. E. Rowan, R. J. M. Nolte,
J. P. Rabe, Macromolecules 2002, 35, 5290–5294.
[38] J. J. L. M. Cornelissen, W. S. Graswinckel, P. J. H. M. Adams, G. H.
Nachtegaal, A. P. M. Kentgens, N. A. J. M. Sommerdijk, R. J. M.
Nolte, J. Polym. Sci. Part A-1 Polym. Chem. 2001, 39, 4255–4264.
[39] P. C. J. Kamer, R. J. M. Nolte, W. Drenth, J. Am. Chem. Soc. 1988,
110, 6818–6825.
[40] S. A. Prokhorova, S. S. Sheiko, M. Moller, C. H. Ahn, V. Percec,
Macromol. Rapid Commun. 1998, 19, 359–366.
[41] J. J. Cornelissen, J. J. Donners, R. de Gelder, W. S. Graswinckel,
G. A. Metselaar, A. E. Rowan, N. A. Sommerdijk, R. J. Nolte, Sci-
ence 2001, 293, 676–680.
[1] C. D. Dimitrakopoulos and P. R. L. Malenfant, Adv. Mater. 2002, 14,
99–117.
[2] C. W. Struijk, A. B. Sieval, J. E. J. Dakhorst, M. van Dijk, P. Kimkes,
R. B. M. Koehorst, H. Donker, T. J. Schaafsma, S. J. Picken, A. M.
van de Craats, J. M. Warman, H. Zuilhof, E. J. R. Sudholter, J. Am.
Chem. Soc. 2000, 122, 11057–11066.
[3] B. A. Gregg, J. Sprague, M. W. Peterson, J. Phys. Chem. B 1997, 101,
5362–5369.
[4] Z. J. Chen, M. G. Debije, T. Debaerdemaeker, P. Osswald, F. Wꢄrth-
ner, ChemPhysChem 2004, 5, 137–140.
[5] R. J. Chesterfield, J. C. McKeen, C. R. Newman, P. C. Ewbank,
D. A. da Silva, J. L. Bredas, L. L. Miller, K. R. Mann, C. D. Frisbie,
J. Phys. Chem. B 2004, 108, 19281–19292.
[6] P. R. L. Malenfant, C. D. Dimitrakopoulos, J. D. Gelorme, L. L.
Kosbar, T. O. Graham, A. Curioni and W. Andreoni, Appl. Phys.
Lett. 2002, 80, 2517–2519.
[7] M. J. Ahrens, L. E. Sinks, B. Rybtchinski, W. H. Liu, B. A. Jones,
J. M. Giaimo, A. V. Gusev, A. J. Goshe, D. M. Tiede, M. R. Wasie-
lewski, J. Am. Chem. Soc. 2004, 126, 8284–8294.
[8] K. Balakrishnan, A. Datar, T. Naddo, J. L. Huang, R. Oitker, M.
Yen, J. C. Zhao, L. Zang, J. Am. Chem. Soc. 2006, 128, 7390–7398.
[9] D. Franke, M. Vos, M. Antonietti, N. A. J. M. Sommerdijk, C. F. J.
Faul, Chem. Mater. 2006, 18, 1839–1847.
[42] J. Cornelissen, W. S. Graswinckel, A. E. Rowan, N. Sommerdijk,
R. J. M. Nolte, J. Polym. Sci. Part A-1 Polym. Chem. 2003, 41, 1725–
1736.
[43] L. Bellamy, The Infra Red Spectra of Complex Molecules, Springer,
Heidelberg, 1975.
[10] J. M. Giaimo, A. V. Gusev, M. R. Wasielewski, J. Am. Chem. Soc.
2002, 124, 8530–8531.
[11] Y. Guan, Y. Zakrevskyy, J. Stumpe, M. Antonietti, C. F. J. Faul,
Chem. Commun. 2003, 894–895.
[12] K. Sugiyasu, N. Fujita, S. Shinkai, Angew. Chem. 2004, 116, 1249–
1253; Angew. Chem. Int. Ed. 2004, 43, 1229–1233.
[44] F. C. Spano, J. Chem. Phys. 2005, 122, 234701.
[13] M. J. Tauber, R. F. Kelley, J. M. Giaimo, B. Rybtchinski, M. R. Wa-
sielewski, J. Am. Chem. Soc. 2006, 128, 1782–1783.
[14] C. Thalacker, F. Wꢄrthner, Adv. Funct. Mater. 2002, 12, 209–218.
[15] Y. Zakrevskyy, C. F. J. Faul, Y. Guan, J. Stumpe, Adv. Funct. Mater.
2004, 14, 835–841.
[16] V. Palermo, A. Liscio, D. Gentilini, F. Nolde, K. Mꢄllen, P. Samori,
Small 2007, 3, 161–167.
[17] G. De Luca, A. Liscio, P. Maccagnani, F. Nolde, V. Palermo, K.
Mꢄllen, P. Samori, Adv. Funct. Mater. 2007, 17, 3791–3798.
[18] A. Liscio, G. De Luca, F. Nolde, V. Palermo, K. Muellen, P. Samori,
J. Am. Chem. Soc. 2008, 130, 780–781.
[45] F. C. Spano, J. Chem. Phys. 2007, 126, 159901.
[46] F. C. Spano, S. C. J. Meskers, E. Hennebicq, D. Beljonne, J. Am.
Chem. Soc. 2007, 129, 7044–7054.
[47] F. C. Spano, Chem. Phys. 2006, 325, 22–35.
[48] T. van der Boom, R. T. Hayes, Y. Y. Zhao, P. J. Bushard, E. A.
Weiss, M. R. Wasielewski, J. Am. Chem. Soc. 2002, 124, 9582–9590.
[49] J. J. L. M. Cornelissen, N. A. J. M. Sommerdijk, R. J. M. Nolte, Mac-
romol. Chem. Phys. 2002, 203, 1625–1630.
[50] F. Wꢄrthner, Z. J. Chen, F. J. M. Hoeben, P. Osswald, C. C. You, P.
Jonkheijm, J. von Herrikhuyzen, A. P. H. J. Schenning, P. P. A. M.
van der Schoot, E. W. Meijer, E. H. A. Beckers, S. C. J. Meskers,
R. A. J. Janssen, J. Am. Chem. Soc. 2004, 126, 10611–10618.
[51] M. Kasha, H. Rawls, M. El-Bayoumi, Pure Appl. Chem. 1965, 11,
371.
[19] S. G. Liu, G. D. Sui, R. A. Cormier, R. M. Leblanc, B. A. Gregg, J.
Phys. Chem. B 2002, 106, 1307–1315.
[20] F. Wꢄrthner, Z. J. Chen, V. Dehm, V. Stepanenko, Chem. Commun.
2006, 1188–1190.
2546
www.chemeurj.org
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 2536 – 2547

Perylene Polyisocyanopeptides
FULL PAPER
[52] N. Harada, K. Nakanishi, Circular Dichroism Spectroscopy. Exciton
Coupling in Organic Stereochemistry, University Science Books, Mill
Valley, 1983.
couplings (computed within the transition density approach) were
then used to solve the Holstein Hamiltonian within the two-particle
approximation.(see refs. [61,62])
[53] N. Harada, K. Nakanish, Acc. Chem. Res. 1972, 5, 257–263.
[54] G. A. Metselaar, J. J. L. M. Cornelissen, A. E. Rowan, R. J. M.
Nolte, Angew. Chem. 2005, 117, 2026–2029; Angew. Chem. Int. Ed.
2005, 44, 1990–1993.
[64] The low-energy shoulder corresponds to the purely electronic 0–0
transition of the individual molecule and its relative intensity, in
comparison to the higher-energy phonon sidebands, is sensitive to
the magnitude of the excitonic couplings. The couplings are all in
the range of 400–500 cmꢀ1 for the nearest neighbor interactions and
significantly smaller than the geometric relaxation energy (0.6ꢆ
0.17 eV=823 cmꢀ1).
[65] F. C. Spano, S. C. J. Meskers, E. Hennebicq, D. Beljonne, J. Chem.
Phys. 2008, 129, 024704.
[66] F. C. Spano, Z. Zhao and S. C. J. Meskers, J. Chem. Phys. 2004, 120,
10594–10604.
[67] A Guide to Recording Fluorescence Quantum Yields, Vol. A Guide to
Recording Fluorescence Quantum Yields, http://www.jobinyvon.com/
usadivisions/fluorescence/applications/quantumyieldstrad.pdf.
[68] R. Katoh, S. Sinha, S. Murata, M. Tachiya, J. Photochem. Photobiol
A 2001, 145, 23–34.
[69] M. Masuko, H. Ohtani, K. Ebata, A. Shimadzu, Nucleic Acids Res.
1998, 26, 5409–5416.
[70] P. B. Bisht, K. Fukuda, S. Hirayama, Chem. Phys. Lett. 1996, 258,
71–79.
[55] After one week in tetrachloroethane solution no changes in the CD
spectrum were observed, however, the fluorescence spectrum re-
vealed a small monomer like signal (<1.5% monomer), which
slowly became stronger in time. The monomer-like emission might
be the result of the unwinding of the helix at its terminals or slow
decomposition of the polymers at its terminals, because of reaction
of the polyimide backbone with carbene- or radical-like species in
solution. In the solid state the polymer is very stable over periods of
up to 2 years. Differential scanning calorimetry (DSC) showed an
endothermic peak at 858C in the heating trace with a corresponding
exothermic peak at 588C in the cooling trace, which can be attribut-
ed to the melting and crystallization of the alkyl tails in the periph-
ery of the polymer, respectively. (Figure S4).
[56] S. L. Mayo, B. D. Olafson, W. A. Goddard, J. Phys. Chem. 1990, 94,
8897–8909.
[57] M. Kauranen, T. Verbiest, C. Boutton, M. N. Teerenstra, K. Clays,
A. J. Schouten, R. J. M. Nolte, A. Persoons, Science 1995, 270, 966–
969.
[58] The calculations have been done in gas phase, hence the more col-
lapsed conformation is also the most stable. Because solvent effects
are not taken into account, we have not ruled out stable structures
on the basis of the sole energetic argument. Absorption and CD
spectra have been done for a number of conformers and the com-
parison to experiment allowed us to identity the most likely confor-
mations in solution.
[59] J. Ridley, M. Zerner, Theor. Chim. Acta 1973, 32, 111–134.
[60] Z. Shuai, J. L. Bredas, Phys. Rev. B 2000, 62, 15452–15460.
[61] F. C. Spano, Annu. Rev. Phys. Chem. 2006, 57, 217–243.
[62] H. Sun, Z. Zhao, F. C. Spano, D. Beljonne, J. Cornil, Z. Shuai, J. L.
Bredas, Adv. Mater. 2003, 15, 818–822.
[71] W. E. Ford, P. V. Kamat, J. Phys. Chem. 1987, 91, 6373–6380.
[72] S. Westenhoff, C. Daniel, R. H. Friend, C. Silva, V. Sundstrom, A.
Yartsev, J. Chem. Phys. 2005, 122, 094903.
[73] C. Daniel, S. Westenhoff, F. Makereel, R. H. Friend, D. Beljonne,
L. M. Herz, C. Silva, J. Phys. Chem. C 2007, 111, 19111–19119.
[74] C. Silva, D. M. Russell, M. A. Stevens, J. D. Mackenzie, S. Setayesh,
K. Mꢄllen, R. H. Friend, Chem. Phys. Lett. 2000, 319, 494–500.
[75] V. Palermo, M. Otten, A. Liscio, E. Schwartz, P. de Witte, M. A.
Castriciano, M. M. Wienk, F. Nolde, G. De Luca, J. J. L. M. Cornelis-
sen, R. A. J. Janssen, K. Mꢄllen, A. E. Rowan, R. J. M. Nolte, P.
Samorꢁ, J. Am. Chem. Soc. 2008, 130, 14605–14614.
[76] C. E. Finlayson, R. H. Friend, M. B. J. Otten, E. Schwartz, J. J. L. M.
Cornelissen, R. J. M. Nolte, A. E. Rowan, P. Samorꢁ, V. Palermo, A.
Liscio, K. Peneva, K. Mꢄllen, S. Trapani, D. Beljonne, Adv. Funct.
Mater. 2008, 18, 3947–3955.
[63] For each chain, the ground-state geometries of the individual chro-
mophores were extracted from the full helical structures and single
point INDO/CCSD calculations were performed on the H-adjusted
molecules. The obtained vertical excitation energies and excitonic
Received: August 22, 2008
Published online: January 28, 2009
Chem. Eur. J. 2009, 15, 2536 – 2547
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2547