Controlling the π-stacking behavior of pyrene derivatives: Influence of H-bonding and steric effects in different states of aggregation

Article abstract of DOI:10.1002/cphc.201300242

The performance of opto-electronic devices built from low-molecular-weight dye molecules depends crucially on the stacking properties and the resulting coupling of the chromophoric systems. Herein we investigate the influence of H-bonding amide and bulky

Full text of DOI:10.1002/cphc.201300242

CHEMPHYSCHEM
ARTICLES
DOI: 10.1002/cphc.201300242
Controlling the p-Stacking Behavior of Pyrene Derivatives:
Influence of H-Bonding and Steric Effects in Different
States of Aggregation
Andreas T. Haedler,[a] Holger Misslitz,[a] Christian Buehlmeyer,[b] Rodrigo Q. Albuquerque,[c]
Anna Kçhler,[b] and Hans-Werner Schmidt*[a]
The performance of opto-electronic devices built from low-mo-
lecular-weight dye molecules depends crucially on the stacking
properties and the resulting coupling of the chromophoric sys-
tems. Herein we investigate the influence of H-bonding amide
and bulky substituents on the p-stacking of pyrene-containing
small molecules in dilute solution, as supramolecular aggre-
gates, and in the solid state. A set of four pyrene derivatives
was synthesized in which benzene or 4-tert-butyl benzene was
linked to the pyrene unit either through an ester or an amide.
All four molecules form supramolecular H-aggregates in THF
solution at concentrations above 1ꢀ10ꢀ4 molLꢀ1. These aggre-
gates were transferred on a solid support and crystallized. We
investigate: the excimer formation rates within supramolecular
aggregates; the formation of H-bonds as well as the optical
changes during the transition from the amorphous to the crys-
talline state; and the excimer to monomer fluorescence ratio
in crystalline films at low temperatures. We reveal that in solu-
tion supramolecular aggregation depends predominantly on
the pyrene chromophores. In the crystalline state, however,
the pyrene stacking can be controlled gradually by H-bonding
and steric effects. These results are further confirmed by mo-
lecular modeling. This work bears fundamental information for
tailoring the solid state of functional optoelectronic materials.
1. Introduction
Supramolecular chromophoric or multichromophoric systems
are envisioned to serve as active materials in opto-electronic
devices, in particular for organic photovoltaics (OPVs).[1] For
those applications p-conjugated aromatic systems are promis-
ing candidates. They can transfer charge and energy, and their
absorption and photoluminescence wavelength can be tuned
comparatively easily.[2] Transport as well as the optical proper-
ties and thus the device performance depend largely on the
electronic coupling of the p-conjugated systems.[3] Therefore,
the distance between and the relative orientation of the chro-
mophores are crucial and need to be adjusted thoroughly by
fine-tuning the molecular structure.[4] To solve this difficult
task, tools and concepts of supramolecular chemistry are uti-
lized. However, a good understanding of the interplay of non-
covalent intermolecular interactions like ionic, dipole or Van
der Waals interactions and combinations thereof is needed.[5]
Especially H-bonding, hydrophilic–hydrophobic interactions,
and steric repulsion are widely used in this context.[6] Further-
more, the influence of those intermolecular interactions on the
p–p stacking of the active chromophoric system depends also
on the state of matter and aggregation.
[a] A. T. Haedler,++ H. Misslitz,++ Prof. Dr. H.-W. Schmidt
Macromolecular Chemistry I and
Bayreuther Institut fꢀr Makromolekꢀlforschung (BIMF) and
Bayreuth Zentrum fꢀr Kolloide und Grenzflꢁchen (BZKG)
University of Bayreuth
95447 Bayreuth (Germany)
A well-established chromophore is needed to investigate
the complex relationship between the molecular structure, the
relative orientation of the p-conjugated systems and the re-
sulting optic and electronic properties. Pyrene is a suitable
candidate as its optical behavior is well understood in solution,
the aggregated and the solid state.[7] Fçrster assigned the
quenching of fluorescence in pyrenes to the formation of exci-
mers already in 1955, and Birks later investigated this excimer
formation in detail.[8] An excimer may be considered as a pair
of molecules that, in the ground state, are bound together
only weakly (e.g. in the solid) or not at all (e.g. in solution), and
that absorb light as monomers, but that reorient in the excited
state and then fluoresce as dimers.[9,10] The excited dimer (=ex-
cimer) fluorescence is broad, unstructured and bathochromi-
cally shifted from the monomer fluorescence, and it leads to
the two monomers in their ground state.[11–13] The strong ten-
Fax: (+49)921-55-3206
E-mail: hans-werner.schmidt@uni-bayreuth.de
[b] C. Buehlmeyer, Prof. Dr. A. Kçhler
Experimental Physics II and
Bayreuther Institut fꢀr Makromolekꢀlforschung (BIMF)
University of Bayreuth
95447 Bayreuth (Germany)
[c] Prof. Dr. R. Q. Albuquerque+
Experimental Physics II
University of Bayreuth
95447 Bayreuth (Germany)
[+] Current address:
Institute of Chemistry of S¼o Carlos
University of S¼o Paulo (USP)
13560-970 S¼o Carlos-SP (Brazil)
[
++] Both authors contributed equally
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cphc.201300242.
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 0000, 00, 1 – 13
&
1
&
ÞÞ
These are not the final page numbers!

CHEMPHYSCHEM
ARTICLES
www.chemphyschem.org
dency of pyrene to form excimers prevails both in solution and
in condensed matter such as amorphous films, liquid crystals
and crystals.[9,14] Note that in the condensed phase, the pyrene
chromophores are usually already in close proximity and often,
for example, in the crystal, arranged in sandwich-type pairs.
However, for the excimer formation, a displacement leading to
closer and/or differently oriented arrangement is still needed.
This excimer emission can also be used indirectly to detect
the aggregation process of pyrene derivatives.[10,15,16] These
characteristic properties have been employed in the investiga-
tion of pyrene-based tweezer molecules[17] and multimolecular
aggregates of micelles and membranes,[18] the gelation detec-
tion of organic low-molecular-weight compounds,[19] the in-
spection of the active sites of enzymes,[20] the molecular recog-
nition process of artificial receptors[21] and DNA sequences[22]
and, in the detection of nitroaromatic explosives.[23]
Scheme 1. Synthesis of pyrene derivatives with ester linker 1a,b and amide
linker 2a,b. i) chloroform, triethylamine, DMAP, 608C; ii) NMP, pyridine, 808C.
Herein, we use pyrene derivatives to study the influence of
H-bonding and steric effects on the coupling behaviour of the
chromophores in different phases. For that purpose a set of
four compounds was synthesized consisting of a benzene or
a 4-tert-butyl benzene group connected to the pyrene through
a methylene-ester or -amide linker (Figure 1). The methylene
soluble in apolar solvents and show moderate solubility in THF,
chloroform, dioxane and DMSO.
The thermal properties of compounds 1a,b and 2a,b were
determined by differential scanning calorimetry (DSC) and are
shown in Table 1. Upon first heating, the ester compounds 1a
Table 1. Thermal properties of amide and ester compounds determined
from DSC measurements (Heating and cooling rate: 208Cminꢀ1 under N2)
Tm [8C]
Tcryst [8C]
Trecryst [8C]
Tg [8C]
(first/second heating) (cooling) (second heating) (on heating)
1a 133/130
1b 140/–
2a 199/196
2b 248/245
–
–
123
178
70
–
114
–
4[a]
23[a]
54[a]
77[b]
[a] Determined from second heating. [b] Determined upon first heating
of a previously quenched sample.
Figure 1. Pyrene derivatives consisting of different linker and substituents
separated by a methylene group from the chromophore.
and 1b exhibit melting points (Tm) at 1338C and 1408C, re-
spectively. Both compounds do not crystallize upon cooling
forming a vitrified supercooled liquid. Consequently, upon
second heating 1a and 1b feature glass transitions with tem-
peratures (Tg) at 48C and 238C. On further heating, only 1a re-
crystallizes at 708C and subsequently melts at 1308C. As ex-
pected, the glass transition temperature is shifted to higher
temperatures for compound 1b due to the bulky tert-butyl-
group. In contrast, the amide compounds 2a and 2b possess
more pronounced crystalline behavior. Upon first heating the
melting points were detected at 1998C and 2488C, respective-
ly, while upon cooling both compounds crystallize at 1238C
and 1788C. Upon second heating 2a features a glass transition
at 548C and recrystallizes at 1148C indicating incomplete crys-
tallization upon cooling. In contrast, compound 2b exhibits
complete crystallization upon cooling. Hence, to determine its
glass transition temperature, an amorphous sample was pre-
pared by rapid quenching from the molten state. The first
heating curve reveals the Tg at 778C. Similar to the ester com-
pounds, the Tg and the Tm increase with the introduction of
the tert-butyl group. A significant difference between the ester
and the amide compounds are higher Tm (708C to 1008C) and
group breaks the conjugation between the dye and the rest of
the molecule, hence, the optical properties of the pyrene is
not directly influenced by the linker and the benzene substitu-
ent. However, indirect influence is possible due to the poten-
tial formation of H-bonds in the case of the amide compounds
and due to steric effects arising from the additional tert-butyl
group.
2. Results and Discussion
2.1. Synthesis and Thermal Properties
The synthetic routes to pyrene-based ester derivatives 1a,b
and their corresponding amide derivatives 2a,b are shown in
Scheme 1. (1-Pyrenyl) methanol was reacted with benzoic acid
chloride or p-(tert-butyl) benzoic acid chloride in chloroform
and triethylamine as base to obtain the two ester compounds
1a,b. For the amide compounds 2a,b the same acid chlorides
were reacted with (1-pyrenyl) methylamine in dry N-methyl-2-
pyrrolidone (NMP) and pyridine as base. All products are in-
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 0000, 00, 1 – 13
&
2
&
ÞÞ
These are not the final page numbers!

CHEMPHYSCHEM
ARTICLES
www.chemphyschem.org
Tg (508C) values for the latter. This result is a consequence of
the formation of intermolecular H-bonds by the amide units. A
similar shift of Tg and Tm to higher temperatures by substitut-
ing ester with amide linkages was recently also reported for tri-
sazobenzene derivatives.[24]
Figure 2 are normalized to the peak maximum of the (0–1)
transition of S2 at 326 nm, which enables a comparison of the
absorption intensities originating from the different vibrational
transitions A(0–0), A(0–1), A(0–2), and so forth. With increasing con-
centration, A(0–0)/A(0–1) decreases while the ratio of the (0–1)
transition to higher transitions is almost identical. Similar be-
havior is also seen for the vibrational structure of the third
electronic excitation. This is unusual. A relative reduction of
the 0–0 peak intensity compared to the 0–1 has been reported
for the S1 state of perylene bisimide dyes and of the polymer
P3HT. In both cases, it has been taken as indication for the for-
mation of weakly interacting H-aggregates.[26,27] Higher elec-
tronic excitations have not been considered explicitly so far in
this framework. To the best of our knowledge, this is the first
experimental report explicitly describing such behavior in the
excitation of two higher electronic states. We note, though,
that Winnik, in a review article, also noticed a reduction in rela-
tive 0–0 peak height with increasing degree of pyrene associa-
tion.[10] Winnik assessed this by considering the 0–0 peak
height and comparing it to the intensity of the minimum be-
tween the 0–0 and the 0–1 vibrational peak. In view of today’s
knowledge of electronic interactions between molecules,[26,28]
we consider that the reduction of the 0–0 peak in S2 and S3 in
the pyrene derivatives may be taken to indicate the formation
of such a weakly interacting H-aggregate, that is, associations
of pyrene molecules that electronically interact already in the
ground state.
2.2. Structural Influences on the Pyrene Interaction in
Diluted to Concentrated THF Solution
In this section we first focus on discussing the photophysical
properties inherent to all four pyrene-derivatives before then
considering the differences resulting from the substitutents.
The UV/Vis absorption spectra were recorded at different con-
centrations in THF solution and are shown in Figure 2 for 1a
In the same concentration range (1ꢀ10ꢀ4 molLꢀ1–1ꢀ
10ꢀ3 molLꢀ1), dynamic light scattering (DLS) measurements
were conducted to obtain an indication for the size of the ag-
gregates. We used a laser beam of 632.8 nm wavelength to
avoid the excitation of the pyrene units during the measure-
ments. For all compounds (1a,b; 2a,b) the obtained scattering
signal implies that at a concentration of 1ꢀ10ꢀ4 molLꢀ1 in THF,
the size of any supramolecular aggregates is below the detec-
tion threshold, whereas particles of at least a few tens of nm
Figure 2. UV/Vis absorption spectra of compound 1a at different concentra-
tions from 1ꢀ10ꢀ4 to 1ꢀ10ꢀ3 molLꢀ1 in THF normalized to the (0–1) transi-
tion at 326 nm; the enlargement shows the 10 times magnified transition to
the first electronically excited state for the concentration of 1ꢀ10ꢀ4 molLꢀ1
.
as an example and for all compounds in the Supporting Infor-
mation, Figure S1. For all absorption measurements 1 mm cuv-
ettes were used to account for the high concentrations at
which the measurements were performed keeping the maxi-
mum OD values below 3. All four compounds show almost
identical absorption characteristics (1a,b; 2a,b), which proves
the electronic decoupling of the chromophore from the sub-
stituent by the methylene spacer. The absorption closely re-
sembles that of pyrene. It shows a very weak feature at about
375 nm and two intense bands, one in the 350–300 nm range
and one from 280–240 nm. By comparison with the pyrene
spectrum these features can be readily assigned. The weak
375 nm feature is the symmetry-forbidden S0!S1 transition in
pyrene, the first intense band with a (0–0) peak at 344 nm and
vibronic replicas at 326 nm, 313 nm and 301 nm corresponds
to the pyrene S0!S2 transition and the peak at 278 nm is due
to the pyrene S0!S3 (0–0) transition, with vibronic replicas at
shorter wavelength.[25] Those characteristics are likewise report-
ed in literature for similar pyrene derivatives.[15]
in size are detected at concentrations above 1ꢀ10ꢀ3 molLꢀ1
.
From the DLS measurements we cannot give a reliable size dis-
tribution of the particles as the supramolecular aggregates are
non-spherical in all cases. The morphology of the supramolec-
ular aggregates at concentrations of 1ꢀ10ꢀ2 molLꢀ1 was vi-
sualized by (SEM) using freeze-dried samples from dioxane
(Figures S2.1 and S2.2).
The steady-state photoluminescence (PL) spectra of the
compounds in THF were investigated at different concentra-
tions, ranging from 6ꢀ10ꢀ6 to 6ꢀ10ꢀ2 molLꢀ1. The spectra are
displayed in Figure 3. The evolution of the monomer fluores-
cence intensity and the relative A(0–0)/A(0–1) absorption peak
height with concentration is shown in Figure 4. All compounds
(1a,b; 2a,b) show the same concentration-dependent PL prop-
erties. At low concentrations, the PL spectrum shows a vibra-
tional structure that is characteristic for the S1!S0 fluorescence
of molecularly dissolved pyrenes.[29] With increasing concentra-
tion, a broad unstructured emission appears, centered at
around 470 nm, that is typical for the excimer fluorescence of
pyrene. All four compounds (1a,b; 2a,b) show an identical ex-
cimer fluorescence at a concentration of 6ꢀ10ꢀ2 molLꢀ1, which
An interesting detail concerns the relative intensities of the
vibrational peaks at different concentrations. The curves in
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 0000, 00, 1 – 13
&
3
&
ÞÞ
These are not the final page numbers!

CHEMPHYSCHEM
ARTICLES
www.chemphyschem.org
cies. This species cannot be the sandwich-type pyrene excimer
identified by Birks,[11] because the quantum yield of emission
of the pyrene monomer and the pyrene excimer observed by
Birks are comparable. For example, in cyclohexane solution it is
at 0.65 for the monomer and 0.75 for the excimer and in tolu-
ene solution the values are 0.52 and 0.55, respectively.[25] If the
excited monomer were to result in an excimer, according to
the reaction M+M*!E* (E* denotes the excimer (M+M)*),
the reduction of monomer fluorescence should be accompa-
nied by a concomitant growth of excimer fluorescence. In the
experiment, this is however not the case. We therefore come
to the conclusion that the quenching of monomer fluores-
cence in the intermediate concentration regime (1ꢀ
10ꢀ5 molLꢀ1–3ꢀ10ꢀ4 molLꢀ1) indicates the formation of a non-
emissive precursor-type species. At this stage, we can only
speculate on the nature of this species. It is conceivable that
the molecular approach and orientation required for the for-
mation of an emissive excimer is impeded in compounds
1 and 2 by the substituents, so that only a non-emissive pre-
cursor can be formed. 3) We now focus our attention on the
experimental observation that excimer fluorescence is only ob-
servable at concentrations in excess of 6ꢀ10ꢀ4 molLꢀ1, yet not
at lower concentration. From the A(0–0)/A(0–1) ratio we know that
ground-state associates prevail in this regime, and the DSL
measurements tell us these evolve into supramolecular aggre-
gates of a few tens of nm for concentrations above 1ꢀ
10ꢀ4 molLꢀ1. This implies that for the pyrene-derivatives 1 and
2, the formation of emissive excimers requires a kind of
ground-state stabilization that is only available in the environ-
ment provided by supramolecular aggregates.
Figure 3. Photoluminescence spectra of compound 2b at different concen-
trations from 6ꢀ10ꢀ6 to 6ꢀ10ꢀ2 molLꢀ1 normalized at 434 nm.
While the decay of the monomer fluorescence with increas-
ing concentration is almost identical for all compounds,
a small variation in the ratio of monomer-to-excimer fluores-
cence intensity can be observed for the different substituents.
In order to evaluate how the excimer formation is affected by
chemical modification, we have determined the excimer forma-
tion rates using time-resolved photoluminescence measure-
ments and analyzing the associated rate equations analogous
to the approach taken by Birks et al. in 1963. Birks and cowork-
ers described excimer formation as a process of the type M+
M*ÐE* that takes place with a concentration-dependent rate
Figure 4. Photoluminescence intensities of the (0–0) transition at 375 nm
(left) and the ratio of A(0–0)/A(1–0) (right) of compounds 1a and 2a at different
concentrations. The shaded area depicts the concentration range in which
the formation of aggregates starts to occur.
confirms the formation of the same excimer species in all cases
(Figure S3). This excimer fluorescence is first barely discernible
at a concentration of 6ꢀ10ꢀ4 molLꢀ1 (and more clearly visible
at 1ꢀ10ꢀ3 molLꢀ1) and then increases in intensity with concen-
tration. In contrast, when considering the intensity of the mo-
nomer fluorescence, the intensity increases with concentration
until, at 2ꢀ10ꢀ5 molLꢀ1, it reaches its maximum and then re-
duces drastically. The ratio of A(0–0)/A(0–1) in absorption remains
constant up to a concentration of 3ꢀ10ꢀ4 molLꢀ1 and then
drops steeply.
kFM[M] for the forward reaction (excimer formation) and a con-
centration-independent rate kME for the reverse reaction (exci-
mer dissociation).[12] Here, [M] is the concentration of mono-
mers. In addition, the deactivation processes M*!M and E*!
M+M are considered with rates kM and kE, respectively. After
pulsed excitation, the decay of M* and E* is then described by
the rate equations d[M*]/dt=ꢀ(kM+kEM[M])[M*]+kME[E*] and
d[E*]/dt=ꢀ(kE+kME)[E*]+kEM[M*]. These two coupled differen-
tial equations can be solved to give two concentration-depen-
dent decay constants l1 and l2 from which the excimer forma-
tion rate kEM can be extracted by a linear fit to the concentra-
tion dependence since l1 +l2 =kM +kE +kME +kEM[M].[12]
For the interpretation of this data it is important to recall
that the three spectroscopic signatures, that is, the monomer
fluorescence, the excimer fluorescence and the A(0–0)/A(0–1) ab-
sorption peak height indicate distinct photophysical processes.
1) The sudden decrease of the A(0–0)/A(0–1) ratio in absorption
gives the concentration at which ground-state association
begins to occur. In Figure 4, this is shaded gray. In contrast,
2) the fluorescence quenching at a much lower concentration
marks the regime where some interaction takes place in the
excited state of the molecule, leading to a non-emissive spe-
For the pyrene derivatives investigated here, we need to
consider that the excimer is evidently not formed by the asso-
ciation of two monomers but that it is preceded by the forma-
tion of ground-state associates. As we have no experimental
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 0000, 00, 1 – 13
&
4
&
ÞÞ
These are not the final page numbers!

CHEMPHYSCHEM
ARTICLES
www.chemphyschem.org
information on the number of electronically interacting mole-
cules in these associates, we take the heuristic approach of
considering dimers. The excimer formation is then described
as D+D*ÐE*+D with kED[D] and kDE being the forward and
back reaction rate. The deactivation processes are D*!D and
(E*+D)!(D+D) with rates kD and kE, respectively. The decay
rates can be formulated and analyzed analogously to the Birks-
1963 model, d[D*]/dt=ꢀ(kD+kED[D])[D*]+kE[(E*+D)] and
d[(E*+D)]/dt=ꢀ(kE+kDE)[(E*+D)]+kE[D*]. The signals to moni-
tor as function of concentration are then the fluorescence of
the (dimer stabilized) excimer (E*+D) at 470 nm and the fluo-
rescence of the excited dimer D*. What is the spectral signa-
ture of the dimer fluorescence? Since the dimer is of a weakly
interacting H-aggregate type, as suggested by the absorption
spectra, the emission is at the same spectral position than the
monomer emission, albeit of a different, that is, weaker oscilla-
tor strength. The signal to monitor is therefore the emission at
400 nm.
tion is shown in Figure 5 for compound 2a, along with the
concentration dependence obtained for the decay constants l1
and l2. The decay curves show a buildup of the excimer emis-
sion (gray curve) at short times, for example, below 50 ns at
5.7ꢀ10ꢀ4 molLꢀ1 and below 10 ns at 6ꢀ10ꢀ2 molLꢀ1, followed
by mono-exponential decay. The dimer emission at 400 nm
(black curve) also decays mono-exponentially. The decay of the
dimer fluorescence (black curve) becomes faster with increas-
ing concentration as the formation of excimers gets more
likely. The excimer fluorescence (gray curve) increases in the
beginning due to the time-delayed formation of the excimer
species after the initial excitation of the monomer. The excimer
fluorescence reaches, as expected, its maximum more quickly
with increasing concentration. From those curves, the decay
constants l1 and l2 can be determined and can be plotted
against the concentration of monomers (=twice the dimer
concentration) (Figure 5). The dependence of their sum on the
concentration agrees well with the linear relationship suggest-
ed by the kinetic model and gives, a posteriori, further support
to the heuristic approach taken. Fitting the data with the equa-
tions from Birks et al. yields the relevant values of kED (Table 2).
The entire approach is portrayed in more detail in the support-
ing information (Figures S4,S5).[12]
We have thus measured the photoluminescence decay of
the compounds in THF at different concentrations ranging
from 2ꢀ10ꢀ4 molLꢀ1 to 6ꢀ10ꢀ2 molLꢀ1 using a time-correlated
single-photon counting (TCSPC) setup. The chromophores
were excited with a 375 nm laser and the time-dependent
spectra were recorded at 400 nm and at 485 nm. The typical
change in the fluorescence decays with increasing concentra-
The excimer formation rates (kED) found in this way for all
four compounds (1a,b; 2a,b) for the formation of excimers
Figure 5. Time-resolved photoluminescence transients of compound 2a at 5.7ꢀ10ꢀ4 molLꢀ1 (left) and 6ꢀ10ꢀ2 molLꢀ1 (middle) showing the fall-off of the
&
&
emission at 400 nm (black) and at 485 nm (gray). Right top: Decay rates l1 and l2 of the emission at 485 nm ( ) and at 400 nm ( ) at different concentrations
for compound 1a fitted according to Birks.[12] Right bottom: Sum of l1 and l2 plotted against the concentration yielding the expected linear dependency.
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 0000, 00, 1 – 13
&
5
&
ÞÞ
These are not the final page numbers!

CHEMPHYSCHEM
ARTICLES
www.chemphyschem.org
one, by using polarized optical microscopy (POM), absorption
and fluorescence spectroscopy, and in the case of the amide
compounds also by FT-IR spectroscopy. The POM images of
the crystalline films are presented in the Supporting Informa-
tion (Figure S7).
Table 2. Excimer formation rates for pyrene and the four compounds
1a,b and 2a,b determined from the TCSPC measurements.
Pyrene[a]
6.7
1a
2a
1b
kED [109 Lsꢀ1 molꢀ1
]
6.88ꢁ0.7
5.34ꢁ0.5
3.96ꢁ0.4
[a] Value was determined by Birks et al.[12]
Optically Isotropic Films
All compounds (1a,b; 2a,b) were spin-coated from THF solu-
tion at a concentration of 5ꢀ10ꢀ2 molLꢀ1. At this concentra-
tion supramolecular aggregates are present in solution as pre-
viously demonstrated in DLS experiments. The fresh films were
optically transparent and no crystalline structure could be ob-
served by POM. The absorption spectra are similar for all com-
pounds (Figure S6) and closely resemble that observed in con-
centrated solution, except for a slight red-shift. Interestingly,
the A(0–0)/A(0–1) ratio is slightly lower in concentrated solution
(about 1.0 for concentrations of 1ꢀ10ꢀ3 molLꢀ1 and above)
than in the thin film (about 1.1). This implies that electronic
coupling within aggregates is slightly stronger in solution than
in the film, presumably because the local geometry can be op-
timized more in solution as compared to film. The room-tem-
perature photoluminescence spectra of the fresh films of all in-
vestigated compounds do not show any monomer fluores-
cence. Instead, they show only the same excimer fluorescence
that was already observed in concentrated solution, except
that the center wavelength shifts from 470 nm in solution to
500 nm in the film due to the different dielectric constant (Fig-
ure S8).
from pre-existing dimers are between 6.9ꢀ109 and 3.9ꢀ
10 Lsꢀ1 molꢀ1. This agrees well with the value of 6.7ꢀ
109 Lsꢀ1 molꢀ1 determined by Birks et al. for the formation of
excimers from individual monomers in the unaltered pyrene.[12]
Within the four compounds the ones with the bulky tert-butyl
group, 1b and 2b, show the lowest excimer formation rates of
approximately 3.9ꢀ109 Lsꢀ1 molꢀ1. We attribute this to sterical
hindrance. For both the ꢀH- and the ꢀtBu-substituted com-
pounds the ones with the ester linker (1a and 1b) show
higher excimer formation rates than the corresponding amide
compounds 2a or 2b. Hence, the amide linker does not favor
the formation of an excimer, but rather hinders it.
We now summarize the insight gained from the spectro-
scopic investigations in solution. At low concentrations, that is,
below 1ꢀ10ꢀ5 molLꢀ1, the four compounds (1a,b; 2a,b) are
molecularly dissolved in THF. In an intermediate concentration
range (about 1ꢀ10ꢀ5 molLꢀ1–3ꢀ10ꢀ4 molLꢀ1), non-emissive in-
termediate species are formed. They quench the monomer
fluorescence yet do not impinge on the absorption spectra,
implying that they may be viewed as a non-emissive precursor
to an excimer state. At increased concentrations, that is, above
3ꢀ10ꢀ4 molLꢀ1 weakly interacting ground state H-aggregates
are formed. In contrast to unaltered pyrene, in the compounds
(1a,b; 2a,b) with the comparatively large side chains, the for-
mation of emissive excimers requires stabilization provided by
the aggregate. A significant influence of the structural varia-
tions, namely the linker moiety (ester or amide) and the sterical
hindrance of the substituted benzene unit on the optical and
aggregation properties in THF cannot be observed. Therefore,
we conclude that the pyrene chromophores, which are coher-
ently present in all compounds, are the crucial and determin-
ing part of the compounds in this concentration regime. Only
a closer look at the excimer fluorescence and the excimer for-
mation rates reveal the influence of the variable substitutions
on the pyrene chromophore. The ester compound 1a is most
feasible to form excimers. The introduction of the tert-butyl
group strongly hinders the excimer formation. The same effect
is observed much weaker when the ester group is replaced
with an amide linker. The influence of the linker and the bulky
tert-butyl group on the pyrene chromophore is just secondary
and only observable at concentrations where supramolecular
aggregates are present.
From the solution measurements we know that, in contrast
to unsubsitituted pyrene, the substituted derivatives (1a,b;
2a,b) only show excimer emission when the excimer geometry
is stabilized within a ground state associate. Thus we can inter-
pret the thin film PL data such as to indicate the presence of
aggregates in the film. This is not surprising. For example, for
the polymers poly(9,9’-dioctyl-fluorene) (PFO) and poly(3-hex-
ylthiophene) (P3HT), it is well known that aggregate formation
is enhanced in thin films upon spin-coating when aggregates
already existed in the solution used.[30] The absence of any mo-
nomer fluorescence suggests that either no monomers are
present, or, any excited monomer state is quenched by energy
transfer either to non-emissive states such as the states ob-
served in solution at intermediate concentration or to emissive
excimers.
Crystalline Films
The PL spectra are different when the films are processed such
as to obtain a crystalline structure. To induce crystallization,
the fresh films of amide 2a and amide 2b were annealed
above Tg at 808C and 1308C. The observation of light transmis-
sion under crossed polarizers in an optical microscope con-
firms the presence of crystalline structures in the film (Fig-
ure S7). To obtain further insight into the structural changes as-
sociated with the crystallization process, FT-IR studies were
performed on the films of the amide compounds (2a–b) in the
isotropic and crystalline state. For both compounds, a shift in
2.3. Structural Influences on the Pyrene Interaction in Thin
Films
Thin films were prepared as detailed below and investigated in
two morphologies, an optically isotropic one and a crystalline
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 0000, 00, 1 – 13
&
6
&
ÞÞ
These are not the final page numbers!

CHEMPHYSCHEM
ARTICLES
www.chemphyschem.org
the NꢀH and the carbonyl stretching vibration to lower wave
numbers as well as an increase in the IR absorption intensity is
observed during the crystallization process. Both changes in
the FT-IR spectra strongly indicate the formation of strong hy-
drogen bonds.[31] The shift of the NꢀH and carbonyl stretching
vibration from 3312 cmꢀ1 to 3281 cmꢀ1 and from 1637 cmꢀ1 to
1629 cmꢀ1, respectively, is shown for compound 2a in
Figure 6. Amide 2b exhibits the same trend but much less pro-
Figure 7. Absorption (right) of an amorphous thin film of compound 1a di-
rectly after spin coating and photoluminescence spectra (left) of a thin film
of compound 1a during the aging process resembling different degrees of
crystallization.
excimer fluorescence even at 5 K. In general, the relative frac-
tion of excimer emission compared to the higher energy emis-
sion is higher in compounds 1b and 2b containing the steri-
cally demanding tert-butyl group compared to compounds 1a
and 2a, that only contain a ꢀH atom instead. The relative
amount of excimer emission is also higher in the ester-contain-
ing compounds (1a and 1b) compared to the amide-contain-
ing compounds (2a and 2b).
Figure 6. Enlargements of the relevant parts of the FT-IR spectra of a thin
film of compound 2a during the aging process resembling different degrees
of crystallization; the N-H stretching vibration (left: around 3300 cmꢀ1) and
the carbonyl stretching vibration (right: around 1630 cmꢀ1) are represented.
To understand the changes that take place in the electronic
structure upon crystallization, let us first recall that compound
2a allows for the most ordered structure due to the effect of
the H-bonding amide moieties. On the other hand, the sterical-
ly demanding tert-butyl group and the absence of the stabiliz-
ing amide group implies that compound 1b is likely to have
the least ordered structure. This correlates with the fact that
compound 2a shows the lowest relative amount of excimer
emission while 1b exhibits purely excimer emission. Evidently,
excimer formation is prevented upon crystallization, while it is
promoted by aggregates. This is different to the situation in
unsubstituted pyrene, where excimers form in the crystalline
state. In fact, in pyrene crystals, the basis of the unit cell is
formed by sandwich-type dimer pairs of pyrene molecules that
slip horizontally upon photoexcitation to form an excimer.[9,25]
How can we understand the absence of excimer emission in
crystallites of the substituted pyrene-derivatives? The aggre-
gates prevailing in the optically isotropic film and the crystalli-
tes in the films evidenced by the POM patterns differ in the
degree of structural order. We propose that the aggregates
that are formed in solution or in the non-equilibrium structure
of a spin-cast film will be subject to a certain amount of struc-
tural variation with regard to molecular orientation or distance.
Further, intermolecular distance between two pyrenes may
vary at the interface between the ends of two aggregates.
Some of these local geometries may be suitable for excimer
formation after photoexcitation. As mentioned above, the fact
that in freshly spun films only excimer fluorescence is observed
nounced. This indicates weaker hydrogen bonds due to the
steric hindrance of the tert-butyl group (Figure S10).
For the ester compounds, a different approach was needed
to obtain crystalline films. Unlike the amide compounds, the
fresh films of the ester compounds exhibited dewetting at ele-
vated temperatures. Therefore, they were aged at room tem-
perature in the dark for several months to induce crystalliza-
tion, as confirmed by POM. Hence, in all cases we were able to
convert the optically isotropic fresh film into one containing
crystallites. The aging and annealing processes were conduct-
ed until a stable state was achieved and no further changes
could be observed in the POM, FT-IR and photoluminescence
spectra at room temperature.
For all compounds except 1b, the photoluminescence spec-
tra change upon crystallization. During this process, the exci-
mer emission is reduced and simultaneously a structured
higher-energy emission rises. This is shown in Figure 7 for
ester 1a.
For a clearer understanding of the molecular behavior in the
crystalline state, the thin film photoluminescence was investi-
gated as a function of temperature upon heating from 5 K to
295 K. The spectra at selected temperatures are shown in
Figure 8. For all compounds except 1b, the excimer fluores-
cence vanishes gradually upon decreasing the temperature,
while simultaneously the structured higher energy fluores-
cence intensity rises. At 5 K, only the structured higher energy
emission prevails. In contrast, compound 1b shows exclusively
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 0000, 00, 1 – 13
&
7
&
ÞÞ
These are not the final page numbers!

CHEMPHYSCHEM
ARTICLES
www.chemphyschem.org
Figure 8. Photoluminescence spectra at different temperatures from room temperature to 5 K of the four compounds 1a (A), 1b, (B), 2a (C), and 2b (D) in
crystallized thin films. For a comprehensive representation, graph A was normalized at 2.73 eV, graph B at the maximum intensity, graph C at 2.78 eV and,
graph D at 2.93 eV. For compound 1a the decrease of the excimer fluorescence and the increase in monomer fluorescence with reduced temperature are in-
dicated with arrows. This is also commonly observed for compounds 2a and 2b.
implies that there is an excimer-site within the exciton diffu-
sion range of any photoexcited chromophore. In contrast, the
structure within the crystallites is, by definition, more regular
and, as experiment tells us, of a kind that does not allow for
excimer formation. It seems that the horizontal slip movement
that results in excimers in unsubstituted crystalline pyrene is
impeded by the large side chain in substituted crystallized
pyrene. Thus, upon crystallization, the number of sites that
allow for excimer formation therefore reduces, as manifested
in Figure 7.
cules already in the excimer geometries must be pre-existing.
In passing we point out that the increasing blue-shift of the
high-energy emission tail with increasing temperature is an ex-
ceptionally nice example for the increase in thermal equilibri-
um energy in a Gaussian density of states.[32] We now turn to
compounds 1a, 2a and 2b. At 5 K, these compounds show
a structured emission with a 0–0 peak at 400 nm and a 0–
1 peak at about 425 nm. The relative intensity of the vibration-
al peaks varies between the compounds. This emission can be
attributed to the pyrene monomer unit, possibly slightly modi-
fied by weak H-type interaction. (Since the S1!S0 transition in
pyrene is symmetry-forbidden, one cannot distinguish whether
the 0–0 peak height is modified due to symmetry-selection
rules or due to weak H-type interaction). At 5 K, spectral diffu-
sion of singlet excitons is reduced. Evidently, structural order in
the crystalline compounds 1a, 2a and 2b is increased such
that there are no sites with excimer geometry within reach at
5 K. At elevated temperatures, the exciton diffusion range in-
creases so that some excimer sites can be populated and
In this framework, we can interpret the results of Figure 8.
For compound 1b, which is structurally most disordered, exci-
mer emission is observed at all temperatures even when the
light transmission in the POM indicates that crystallites have
been formed. This leads to two conclusions. First, in this com-
pound, the remaining structural inhomogeneity is sufficient to
result in sites for excimer formation and/or monomer quench-
ing within the exciton diffusion range of each chromophore.
Second, since excimer emission occurs at 5 K, it does not re-
quire any activation energy. This confirms that sites with mole-
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 0000, 00, 1 – 13
&
8
&
ÞÞ
These are not the final page numbers!

CHEMPHYSCHEM
ARTICLES
www.chemphyschem.org
a spectrum showing both, monomer and excimer fluorescence
results.
can be attributed to the additional bulky tert-butyl group. This
is in perfect agreement with the previous results from the FT-
IR measurements. Furthermore, the sterically demanding tert-
butyl group of compound 2b enforces a slight twist in the mo-
lecular structure within the calculated hexamer aggregate.
Although in general p–p stacking of the pyrene moieties is
present in all calculated hexamer aggregates, slight differences
can be observed in the relative orientation of the chromo-
phores. While ester compound 1a features an almost perfect
p–p stacking of the pyrene units an increasing parallel offset is
observed for compound 2a and 2b. Additionally, in the hexa-
mer aggregate of compound 1b a rotational twist of adjacent
pyrene units is present (orange lines, Figure 9). In compound
1a the p–p stacking of the pyrene chromophores is
the only intermolecular interaction. The columnar
stacking is increasingly disturbed by the introduction
of the amide and the tert-butyl moiety. The results
from the molecular modeling calculations are in
good agreement with the previous discussion on
thin film properties, whereas the solution properties
differ. In the film the steric demand and disorder in-
troduced by the tert-butyl group (Figure 9B,D) corre-
lates well with the enhanced propensity to form exci-
mers which was already observed in the low temper-
ature photoluminescence measurements (Fig-
ure 8B,D). Compound 1b, which has, as already as-
sumed previously, the least ordered structure is
unique in two ways—it exhibits a strong rotational
twist of adjacent pyrene chromophores and shows
exclusively excimer fluorescence even after crystalli-
zation and at very low temperatures. On the other
hand, H-bonding amide groups as well as the lack of
the sterically demanding tert-butyl group promote
a more ordered structure reducing the propensity to
form excimers in the crystalline state. Note that in
the crystalline state, unlike the unchanged pyrene,
the investigated molecules (1a,b; 2a,b) do not ar-
2.4. Structural Influences on the Pyrene Interaction:
Molecular Modeling
The conclusions drawn so far on the basis of spectroscopic
data are well supported by molecular modeling of the crystal-
lite structure. In order to assess, how the molecules pack, the
geometry of the monomers of the esters 1a, 1b and the
amides 2a, 2b were optimized using the B3LYP functional and
basis set 6-31G. The optimized structures, together with the
predicted partial charges, were used to build and optimize the
Figure 9. Molecular mechanics geometry optimization of hexamer aggregates of the
esters 1a (A) and 1b (B) and the amides 2a (C) and 2b (D) with the corresponding top
views of two stacked molecules. The two orange lines represent the twist of two adja-
cent pyrene chromophores. Additionally, the distance between two pyrene chromo-
phores and in the case of the amide compounds 2a-b the OꢀH distance of the H-bond-
ing amide moieties are stated. The geometry of the monomers was optimized with DFT.
range in a sandwich-type structure but seem to ex-
hibit equidistance.
3. Conclusions
We were able to understand the influence of H-bond-
structure of the aggregates by means of molecular mechanics
(MM), from which qualitative results could be obtained
(Figure 9).
ing amide linkers and sterically demanding tert-butyl groups
on the p-stacking of pyrene chromophores in different states
of matter. This was achieved by carefully studying a set of four
pyrene derivatives exhibiting either an ester or amide linker
and with or without a bulky tert-butyl substituent. The four
compounds were investigated from molecularly dissolved solu-
tions up to concentrations where supramolecular aggregates
are formed, as well as in spin-cast films in a virgin and in a crys-
tallized state.
As expected, the p–p stacking of the pyrene moieties is the
dominant driving force towards self-aggregation. This can be
seen from the almost parallel orientation of the chromophores
in all cases. The distance between the pyrene units is approxi-
mately 3.50 ꢂ in the calculated hexamer aggregates, which is
in good agreement with crystal structure data of stacked pyr-
enes featuring a distance of 3.53 ꢂ.[29] In the case of the amide
compounds 2a and 2b additional H-bonds are observed in
the MM calculations with OꢀH distances of 1.92 ꢂ and 2.17 ꢂ,
respectively. The longer OꢀH distance of compound 2b indi-
cates weaker H-bonds than in the case of compound 2a which
The influence of the linker and the steric hindrance on the
pyrene stacking gets more significant the closer the molecules
are forced together. In dilute solution up to the threshold con-
centration for the formation of aggregates the molecular be-
havior is independent of the variable groups, as all four com-
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 0000, 00, 1 – 13
&
9
&
ÞÞ
These are not the final page numbers!

CHEMPHYSCHEM
ARTICLES
www.chemphyschem.org
(5%) and water. The solution was dried over Na2SO4 and the sol-
vent was evaporated in vacuum.
pounds show identical spectra and behavior. The four com-
pounds differ, however, from unsubsituted pyrene. Whereas
unsubstituted pyrene readily forms excimers in solution from
monomers, in the substituted derivatives excimer formation re-
quires additional stabilization within aggregates. The rate of
excimer formation depends on the nature of the sidechain.
While the excimer formation rate in the ester compound 1a is
comparable to that of unsubstituted pyrene, excimer forma-
tion proceeds slower with the amide linker of compound 2a,
and is reduced even more by the bulky substituents of 1b and
2b.
In thin films we find that excimers are readily formed within
the supramolecular aggregates present in freshly spin-cast
films, yet that excimer formation is suppressed by crystalliza-
tion. Such crystallization can be induced by heating the amide-
containing compounds, which develop intermolecular H-
bonds, as well as by aging at room temperature in the case of
the ester compounds. The amount of structural order in the
crystalline state is strongly dependent on the variable groups.
H-bonding amide groups in contrast to ester groups promote
a more ordered structure, while the sterically demanding tert-
butyl substituent hinders a highly ordered packing of the mol-
ecules. In the case of compound 1b this leads to a strong rota-
tional twist of adjacent pyrene moieties and an enhanced pro-
pensity to form excimers in the crystalline structure. Com-
pound 2a on the other hand favors a highly ordered packing
resulting in the least amount of excimer fluorescence of all
compounds.
1a: Yield 2.9 g of a yellow powder (92%, 8.9 mmol); Rf =0.76
(Hexane/THF 2:1); m.p. 1338C; 1H NMR (300 MHz, [D6]DMSO, 258C,
TMS): d=6.11 (s, 2H), 7.50 (m, 1H), 7ꢀ.614 (m, 1H), 7.98 (m, 1H),
~
8.08–8.49 ppm (m, 9H); IR: n=1709 cm (C=O); UV/Vis (THF): lmax
(e)=344 nm (31924 molꢀ1 m3 cmꢀ1); elemental analysis calcd (%)
for C24H16O2 C 85.69, H 4.79, O 9.51; found: C 85.55, H 5.07, O 9.17.
1b: Compound 1b was recrystallized from toluene. Yield 1.8 g of
a yellowish powder (43%, 4.6 mmol); Rf =0.83 (Hexane/THF 2:1);
m.p. 1408C; 1H NMR (300 MHz, [D6]DMSO, 258C, TMS): d=1.52 (s,
9H), 6.09 (s, 2H), 7.50 (d, J=8.4 Hz, 1H), 7.90 (d, J=8.4 Hz, 1H),
ꢀ1
~
8.08–8.46 ppm (m, 9H); IR: n=1717 cm (C=O); UV/Vis (THF): lmax
(e)=344 nm (32819 molꢀ1 m3 cmꢀ1); elemental analysis calc (%) for
C28H24O2: C 85.68, H 6.16, O 8.15; found: C 85.66, H 5.95, O 8.39.
General Synthetic Procedure to Pyrenyl-Substituted Amides
2a,b
1-Pyrenyl-methyl-amine-hydrochloride (2.5 g, 9.4 mmol) was sus-
pended in N-methyl-2-pyrrolidone (NMP) in a Schlenk tube under
inert gas. Dry pyridine (20 mL) and LiCl (0.05 g) was added and the
mixture was stirred for 30 min. The solution was cooled to 08C and
benzoic acid chloride (9.4 mmol) was added dropwise and the re-
action mixture was stirred for 2 h at 708C to yield a yellowish solu-
tion. After cooling to room temperature, the mixture was precipi-
tated in ice water (600 mL). The mixtures were filtered to retrieve
the solid, which was washed with water and dried under vacuum
at 808C.
2a: Compound 2a was recrystallized from ethyl acetate/hexane
mixture (1:1). Yield 3.1 g of a white powder (98%, 9.2 mmol); Rf =
0.22 (THF/Toluene 1:20); m.p. 1998C; 1H NMR (300 MHz, [D6]DMSO,
258C, TMS): d=5.24 (d, J=5.6 Hz, 2H), 7.45–7.54 (m, 3H), 7.93–
With this variety of experiments we were able to understand
the molecular behavior and the influence of the variable
groups on the pyrene stacking in different states of matter in
great detail. This study clearly points out how crucially the mo-
lecular design can affect the performance of chromophoric sys-
tems.
~
7.96 (m, 2H), 8.04–8.54 (m, 9H), 9.25 ppm (t, J=5.6 Hz, 1H); IR: n=
1625 (C=O), 3266 cmꢀ1 (N-H); UV/Vis (THF): lmax (e)=344 nm
(29098 molꢀ1 m3 cmꢀ1); elemental analysis calcd (%) for C24H17NO: C
85.95, H 5.18, N 4.18, O 4.77; found: C 85.91, H 5.07, N 3.79, O 5.04.
2b: Compound 2b was recrystallized from toluene. Yield 2.6 g of
a white powder (71%, 6.6 mmol); Rf =0.31 (THF/Toluene 1:20); m.p.
2488C; 1H NMR (300 MHz, [D6]DMSO, 258C, TMS): d=1.29 (s, 9H),
5.24 (d, J=5.7 Hz, 2H), 7.49 (d, J=8.2 Hz, 2H), 7.89 (d, J=8.2 Hz,
Experimental Section
Materials and Methods
~
2H), 8.06–8.53 (m, 9H), 9.18 ppm (t, J=5.7 Hz, 1H); IR: n=1633
(C=O), 3311 cmꢀ1 (N-H); UV/Vis (THF): lmax (e)=344 nm
(31324 molꢀ1 m3 cmꢀ1); elemental analysis calcd (%) for C28H25NO: C
85.90, H 6.44, N 3.58, O 4.09; found: C 85.51, H 6.07, N 3.70, O 4.76.
Solvents were distilled and when necessary dried according to
standard procedures. All starting materials were obtained from Al-
drich, Alfa Aesar, Fluka or Riedel-de Haꢃn and used without further
purification. 1H NMR spectra were recorded on a Bruker Avance
300 spectrometer. Mass spectra were recorded on a Finnigan MAT
8500 apparatus (EI, 70 eV) using direct injection mode. Elemental
analysis (C, H, N) was carried out with an EA 3000 instrument (HE-
KAtech).
Differential Scanning Calorimetry (DSC)
The thermal properties were investigated by DSC measurements
with a PERKIN-ELMER DSC7 (standard heating rate: 20 Kminꢀ1) uti-
lizing 10 mg of the compounds.
General Synthetic Route to Pyrenyl-Substituted Esters 1a,b
1-Pyrenyl-methanol (2.5 g, 10.8 mmol) were dissolved in chloro-
form in a Schlenk tube under inert gas. Triethylamine (1.8 mL) and
4-dimethylaminopyridine (DMAP) (50 mg) were added. The solu-
tion was cooled to 08C and benzoic acid chloride (12.0 mmol) was
added dropwise. The reaction mixture was boiled for 15 h. After
cooling to room temperature, the yellow mixture was filtered
(Alox N) and washed with chloroform. The solvent was evaporated
and the raw product was dissolved in dichloromethane and ex-
tracted subsequently with aqueous HCl (2m), aqueous NaHCO3
Dynamic Light Scattering (DLS)
DLS was performed on an ALV DLS/SLS-SP 5022F compact goniom-
eter system with an ALV 5000/E cross-correlator and a He-Ne laser
(632.8 nm). All measurements were performed at concentrations of
1ꢀ10ꢀ4 molLꢀ1, 1ꢀ10ꢀ3 molLꢀ1, and 1ꢀ10ꢀ2 molLꢀ1 of the four
compounds 1a–b and 2a–b in THF.
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 0000, 00, 1 – 13
&10
&
ÞÞ
These are not the final page numbers!

CHEMPHYSCHEM
ARTICLES
www.chemphyschem.org
Thin Film Preparation
FT-IR Spectroscopy
THF solutions (concentration: 5ꢀ10ꢀ2 molLꢀ1) of the four com-
pounds were spin coated on spectrosil or silicon wafer substrates
(30 s, 1020 rpm).
The infrared measurements were performed on a PerkinElmer FT-IR
Spectrum 100 spectrometer. The solution measurements were con-
ducted in THF solutions at 1ꢀ10ꢀ2 molLꢀ1 using a NaCl cell of
0.5 mm diameter. For the measurements on thin films, the films
had to be spin coated on silica wavers (30 s, 1020 rpm). The result-
ing films were measured at the ATR unit of the spectrometer.
UV/Vis Spectroscopy
UV/Vis spectra on solution and thin film samples were recorded on
a JASCO V-670 spectrophotometer. For the measurements in THF
the path length of the cuvettes was 1 mm. The extinction coeffi-
cient of the 0–0 transition to the S2 state was for all compounds
about 5ꢀ104 molLꢀ1 in the molecularly dissolved state at 1ꢀ
10ꢀ4 molLꢀ1 which results in an OD of 0.5. The further increase of
the concentration aggregation occurs and the extinction coeffi-
cient is substantially reduced so that the OD values are still below
3. Note also that only the extinction coefficients of the two 0–0
transitions to the S2 and S3 states are reduced due to aggregation
at higher concentrations, while the others remain unchanged.
Modeling
Initially a conformational search was carried out to know the most
probable conformation of the molecules 1a, 1b, 2a and 2b. DFT
calculations were carried out for those compounds using the func-
tional B3LYP and basis set 6-31G for all atoms and the program
Gaussian 03 was used. No constraints in the geometry optimiza-
tions were applied. The optimized geometries were checked by
calculation of the vibrational spectra, where no negative frequen-
cies were found. The optimized geometries and the partial charges
obtained from the DFT calculations were used in order to build ag-
gregates containing six monomers each. The geometries of the ag-
gregates were then optimized using the Molecular Mechanics and
considering explicitly the atomic charges (MM+force field, Hyper-
chem 7.5 program).
Photoluminescence Spectroscopy in Solution
For the photoluminescence study THF solutions of the compounds
were prepared and investigated with a FluoroMax 3 spectrometer
at an excitation wavelength of 344 nm.
Acknowledgements
Photoluminescence Spectroscopy on Thin Films
We gratefully acknowledge financial support by the Deutsche
Forschungsgemeinschaft within the DFG Research Training Group
(GRK 1640) “Photophysics of Synthetic and Biological Multichro-
mophoric Systems”. A.T.H. and H.M. would also like to thank the
elite study program “Macromolecular Science” at the University
of Bayreuth. “Elite Netzwerk Bayern” is acknowledged for a fellow-
ship for A.T.H. We are indebted to Doris Hanft and Irene Bauer
for synthetic support and Dr. Klaus Kreger for many fruitful dis-
cussions.
The photoluminescence measurements of the thin films were per-
formed on a home-built setup. An argon laser (Coherent Innova
300C) with wavelengths of 354 nm and 361 nm was used for the
excitation and a CCD-camera (Andor iDus DU420) was used as de-
tector.
Time-Correlated Single-Photon-Counting (TCSPC)
Measurements
The TCSPC measurements were conducted on a FluoTime 200
setup from PicoQuant. The samples were excited with 375 nm
laser pulses and the resulting fluorescence was detected with
a MCP-PMT detector from Hamamatsu.
Keywords: fluorescence · kinetics · molecular modeling ·
photophysics · self-assembly
[1] a) T. Aida, E. W. Meijer, S. I. Stupp, Science 2012, 335, 813–817; b) V.
Percec, M. Glodde, T. K. Bera, Y. Miura, I. Shiyanovskaya, K. D. Singer,
V. S. K. Balagurusamy, P. A. Heiney, I. Schnell, A. Rapp, H.-W. Spiess, S. D.
Hudson, H. Duan, Nature 2002, 417, 384–387; c) F. Wꢄrthner, K. Meer-
holz, Chem. Eur. J. 2010, 16, 9366–9373; d) B. Bodenant, F. Fages, M.-H.
Delville, J. Am. Chem. Soc. 1998, 120, 7511–7519; e) Y. Sun, G. C. Welch,
W. L. Leong, C. J. Takacs, G. C. Bazan, A. J. Heeger, Nat. Mater. 2012, 11,
44–48.
Optical Microscopy
The spin coated thin films were investigated between crossed po-
larizers on an optical microscope (Leica DMRX) equipped with
a hot-stage (Mettler, model FP82TM).
[2] a) K. R. Leight, B. E. Esarey, A. E. Murray, J. J. Reczek, Chem. Mater. 2012,
24, 3318–3328; b) L. F. Dçssel, V. Kamm, I. A. Howard, F. Laquai, W.
Pisula, X. Feng, C. Li, M. Takase, T. Kudernac, S. de Feyter et al., J. Am.
Chem. Soc. 2012, 134, 5876–5886; c) Z. Chen, A. Lohr, C. R. Saha-Mçller,
F. Wꢄrthner, Chem. Soc. Rev. 2009, 38, 564–584.
[3] a) L. van Dijk, P. A. Bobbert, F. C. Spano, J. Phys. Chem. B 2009, 113,
9708–9717; b) L. van Dijk, F. C. Spano, P. A. Bobbert, Chem. Phys. Lett.
2012, 529, 69–73; c) A. A. Voityuk, Phys. Chem. Chem. Phys. 2012, 14,
13789–13793.
Preparation of Freeze-Dried Samples
Supramolecular aggregates of the four compounds were prepared
in dioxane (concentration: 1ꢀ10ꢀ2 molLꢀ1) by heating to dissolu-
tion and subsequent cooling to room temperature. One drop of
the mixture was given on a DSC pan, frozen in liquid nitrogen/pen-
tane mixture and subsequently dried at high vacuum (0.007 mbar).
[4] a) M. R. Molla, S. Ghosh, Chem. Mater. 2011, 23, 95–105; b) F. Wꢄrthner,
Chem. Commun. 2004, 1564–1579.
Scanning Electron Microscopy (SEM)
[5] a) A. Ajayaghosh, S. J. George, A. P. Schenning in Topics in Current
Chemistry (Ed.: F. Wꢄrthner), Springer, Heidelberg, 2005; b) T. Metzroth,
A. Hoffmann, R. Martꢅn-Rapffln, M. M. J. Smulders, K. Pieterse, A. R. A. Pal-
mans, J. A. J. M. Vekemans, E. W. Meijer, H. W. Spiess, J. Gauss, Chem. Sci.
2011, 2, 69–76.
The samples were coated using platinum (about 1.4 nm) in a Cres-
sington sputter coater 208HR. The SEM micrographs were recorded
on a LEO 1530 FE-SEM (Zeiss, Jena) with a Schottky-field-emission
cathode and an in-lens detector.
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 0000, 00, 1 – 13
&
11
&
ÞÞ
These are not the final page numbers!

CHEMPHYSCHEM
ARTICLES
www.chemphyschem.org
[6] a) D. Gonzµlez-Rodrꢅguez, A. P. H. J. Schenning, Chem. Mater. 2011, 23,
310–325; b) T. Shu, J. Wu, M. Lu, L. Chen, T. Yi, F. Li, C. Huang, J. Mater.
Chem. 2008, 18, 886–893; c) M. Blomenhofer, S. Ganzleben, D. Hanft,
H.-W. Schmidt, M. Kristiansen, P. Smith, K. Stoll, D. Mäder, K. Hoffmann,
Macromolecules 2005, 38, 3688–3695; d) J. B. Bodapati, H. Icil, Photo-
chem. Photobiol. Sci. 2011, 10, 1283–1293.
[7] a) S. Sankararaman, G. Venkataramana, B. Varghese, J. Org. Chem. 2008,
73, 2404–2407; b) X. Zhang, S. Rehm, M. M. Safont-Sempere, F. Wꢄrth-
ner, Nat. Chem. 2009, 1, 623–629.
[8] T. Fçrster, K. Kasper, Z. Elektrochem. 1955, 59, 976–980, http://onlineli-
brary.wiley.com/doi/10.1002/bbpc.v59:10/issuetoc.
[9] M. Schwoerer, H. C. Wolf, Organic Molecular Solids, Wiley-VCH, Wein-
heim, 2007.
[20] S. T. Caldwell, G. Cooke, S. G. Hewage, S. Mabruk, G. Rabani, V. Rotello,
B. O. Smith, C. Subramani, P. Woisel, Chem. Commun. 2008, 4126–4128.
[21] H. M. Colquhoun, Z. Zhu, D. J. Williams, M. G. B. Drew, C. J. Cardin, Y.
Gan, A. G. Crawford, T. B. Marder, Chem. Eur. J. 2010, 16, 907–918.
´
´
[22] a) L. Hernandez-Folgado, D. Baretic, I. Piantanida, M. Marjanovic, M.
Kralj, T. Rehm, C. Schmuck, Chem. Eur. J. 2010, 16, 3036–3056; b) R.
Häner, F. Samain, V. L. Malinovskii, Chem. Eur. J. 2009, 15, 5701–5708;
c) R. Varghese, H.-A. Wagenknecht, Chem. Commun. 2009, 2615–2624.
[23] a) Y. H. Lee, H. Liu, J. Y. Lee, S. H. Kim, S. K. Kim, J. L. Sessler, Y. Kim, J. S.
Kim, Chem. Eur. J. 2010, 16, 5895–5901; b) A. D. Hughes, I. C. Glenn,
A. D. Patrick, A. Ellington, E. V. Anslyn, Chem. Eur. J. 2008, 14, 1822–
1827.
[24] K. Kreger, P. Wolfer, H. Audorff, L. Kador, N. Stingelin-Stutzmann, P.
Smith, H.-W. Schmidt, J. Am. Chem. Soc. 2010, 132, 509–516.
[25] J. B. Birks, Photophysics of Aromatic Molecules, Wiley-Interscience, New
York, 1970.
[26] D. Gçrl, X. Zhang, F. Wꢄrthner, Angew. Chem. 2012, 124, 6434–6455;
Angew. Chem. Int. Ed. 2012, 51, 6328–6348.
[27] J. Clark, J.-F. Chang, F. C. Spano, R. H. Friend, C. Silva, Appl. Phys. Lett.
2009, 94, 163306.
[28] a) F. C. Spano, Chem. Phys. 2006, 325, 22–35; b) S. Ghosh, X.-Q. Li, V.
Stepanenko, F. Wꢄrthner, Chem. Eur. J. 2008, 14, 11343–11357.
[29] J. B. Birks, Rep. Prog. Phys. 1975, 38, 903–974.
[30] a) A. L. T. Khan, P. Sreearunothai, L. M. Herz, M. J. Banach, A. Kçhler, Phys.
Rev. B 2004, 69, 085201; b) C. Scharsich, R. H. Lohwasser, M. Sommer, U.
Asawapirom, U. Scherf, M. Thelakkat, D. Neher, A. Kçhler, J. Polym. Sci.
Part B 2012, 50, 442–453.
[31] a) D. J. Skrovanek, S. E. Howe, P. C. Painter, M. M. Coleman, Macromole-
cules 1985, 18, 1676–1683; b) T. Mes, M. M. J. Smulders, A. R. A. Pal-
mans, E. W. Meijer, Macromolecules 2010, 43, 1981–1991.
[32] H. Bässler, Phys. Status Solidi B 1993, 175, 15–56.
[10] F. M. Winnik, Chem. Rev. 1993, 93, 587–614.
[11] J. B. Birks, L. G. Christophorou, Spectrochim. Acta 1963, 19, 401–410.
[12] J. B. Birks, D. J. Dyson, I. H. Munro, Proc. R. Soc. London Ser. A 1963, 275,
575–588.
[13] a) J. B. Birks, M. D. Lumb, I. H. Munro, Proc. R. Soc. London Ser. A 1964,
280, 289–297; b) J. B. Birks, A. A. Kazzaz, T. A. King, Proc. R. Soc. London
Ser. A 1966, 291, 556–569; c) J. B. Birks, A. A. Kazzaz, Proc. R. Soc.
London Ser. A 1968, 304, 291–301.
[14] A. Hayer, V. de Halleux, A. Kçhler, A. El-Garoughy, E. W. Meijer, J.
Barberµ, J. Tant, J. Levin, M. Lehmann, J. Gierschner et al., J. Phys. Chem.
B 2006, 110, 7653–7659.
[15] J. Xiao, Y. Li, Y. Song, L. Jiang, Y. Li, S. Wang, H. Liu, W. Xu, D. Zhu, Tetra-
hedron Lett. 2007, 48, 7599–7604.
[16] a) L. You, G. W. Gokel, Chem. Eur. J. 2008, 14, 5861–5870; b) F. D. Lewis,
J.-S. Yang, C. L. Stern, J. Am. Chem. Soc. 1996, 118, 12029–12037; c) M.
Kimura, N. Miki, D. Suzuki, N. Adachi, Y. Tatewaki, H. Shirai, Langmuir
2009, 25, 776–780.
[17] a) H. M. Colquhoun, Z. Zhu, Angew. Chem. 2004, 116, 5150–5155;
Angew. Chem. Int. Ed. 2004, 43, 5040–5045; b) B. W. Greenland, S. Burat-
tini, W. Hayes, H. M. Colquhoun, Tetrahedron 2008, 64, 8346–8354.
[18] M. S. Becherer, B. Schade, C. Bçttcher, A. Hirsch, Chem. Eur. J. 2009, 15,
1637–1648.
Received: March 8, 2013
[19] B. Adhikari, J. Nanda, A. Banerjee, Chem. Eur. J. 2011, 17, 11488–11496.
Published online on && &&, 2013
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 0000, 00, 1 – 13
&12
&
ÞÞ
These are not the final page numbers!

ARTICLES
A. T. Haedler, H. Misslitz, C. Buehlmeyer,
R. Q. Albuquerque, A. Kçhler,
H.-W. Schmidt*
&& – &&
Controlling the p-Stacking Behavior of
Pyrene Derivatives: Influence of H-
Bonding and Steric Effects in Different
States of Aggregation
The Influence of H-bonding and steric
hindrance on the optical properties and
the p-stacking of pyrene derivatives is
investigated and described in dilute so-
lution, as supramolecular aggregates
and in the crystalline state.
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 0000, 00, 1 – 13
&13
&
ÞÞ
These are not the final page numbers!