Green emission in ladder-type quarterphenyl: Beyond the fluorenone-defect

Article abstract of DOI:10.1002/adfm.201402638

In polyfluorenes it is generally accepted that (photo)degradation leads to fluorenone type defects that accept the excitation energy and emit green-to-yellow light with rather low efficiency. Although initial spectroscopic studies suggest the same to hold

Full text of DOI:10.1002/adfm.201402638

www.afm-journal.de
www.MaterialsViews.com
Green Emission in Ladder-Type Quarterphenyl: Beyond the
Fluorenone-Defect
Björn Kobin, Francesco Bianchi, Simon Halm, Joachim Leistner, Sylke Blumstengel,
Fritz Henneberger,* and Stefan Hecht*
In polyfluorenes (PF) the emerging green
In polyfluorenes it is generally accepted that (photo)degradation leads to
fluorenone type defects that accept the excitation energy and emit green-to-
yellow light with rather low efficiency. Although initial spectroscopic studies
suggest the same to hold true for ladder-type poly(para-phenylene)s (LPPPs),
kinetic studies of the degradation process are not compatible with the estab-
lished mechanism. In general, the observed green emission can be caused
by the introduction of carbonyl groups; however, only if associated with an
additional disruption of the backbone rigidity and hence planarity of the entire
π-system. This is clearly shown by comparison with synthesized model com-
pounds, which are bearing the fluorenone motif yet possess very different
optical properties as compared to the defects, which are actually formed.
Degradation can be caused by solvent specific, yet substrate nonspecific
aromatic formylation but mainly originates from reaction with in-situ gener-
ated singlet oxygen, both in solution as well as in thin films. Time-dependent
photoluminescence measurements on thin films show that green emission is
enhanced by energy transfer from intact molecules to defect centers.
emission has been attributed to excimers
and fluorenone defects, which are capable
of trapping the excitation energy.[4] Fur-
thermore chain end defects, such as
hydroxyl groups, which originate from
Suzuki cross couling polymerization,
may lead to green emission as well.[5] In
LPPPs a break of the bridging units of the
polymer backbone was taken into consid-
eration, as well as fragments, such as aryl
ketones and fluorenones were identified
by vibrational spectroscopy.[6,7] The optical
properties of these fragments have also
been characterized by means of quantum-
chemical calculations and compared to the
properties of thermally degraded LPPP
films.[7] The most prominent mecha-
nisms of degradation proposed in the
literature involve a 9-fluorenylhydroper-
oxide formed from synthetic defect sites,
such as residual 9-hydrogen-substituted
fluorene.[4a] In addition, free-radical autoxidation leading to
fluorenone type defects has been proposed for thermally[8] as
well as photochemically[9] induced degradation, typically neces-
sitating synthetic defect-sites. There is also at least one report
on highly purified oligo(alkylfluorene)s, which alternatively
claims that gradual degradation of the side chains due to sin-
glet oxygen leads to the fluorenone defect.[10]
1. Introduction
In the past years there has been a lot of research in the field
of organic electronics with the prospect of fabricating low-
cost, light-weight, and flexible devices.[1] However, in organic
photovoltaics (OPV) and in organic light-emitting diodes
(OLED), especially the blue emitting ones, the lifetimes of
the devices remains a critical issue. Polyfluorenes, as well as
ladder-type poly(para-phenylene)s (LPPPs)[2] have already been
exploited in blue emitting OLEDs for some time,[3] but they
suffer from relatively fast degradation. There are several studies
on the degradation of such compounds. The degradation can
be initiated thermally, photochemically or electrochemically.
In contrast to most of these studies dealing with PF, we focus
on small molecules, such as the ladder-type quarterphenyl L4P,
which in combination with ZnO is supposed to be a potent
candidate for creating hybrid inorganic-organic systems for
opto-electronic devices.[11] In order to assemble these systems,
L4P was designed to be compatible with organic molecular
vapor-deposition, and therefore the molecular weight was mini-
mized by limiting the number of repeat units and omitting
solubilizing side chains. In this context it should be empha-
sized that the small-molecule approach—after proper purifica-
tion by gradient-sublimation and crystallization—provides a
pure sample compound of one well-defined molecular entity,
which can be unambiguously structurally characterized. Here,
we present a detailed study of the photodegradation of L4P,
with detailed insights into its kinetics in solution as well as the
chemical structure of formed degradation products. Particular
attention is on the active species of oxygen being responsible
for degradation. The optical properties of the isolated structural
defects, deliberately synthesized fluorenone-type defects, and
crude degraded mixtures are being compared, showing that
B. Kobin, Dr. J. Leistner, Prof. S. Hecht
Department of Chemistry and IRIS Adlershof
Humboldt-Universität zu Berlin
Brook-Taylor-Str. 2 12489, Berlin, Germany
E-mail: sh@chemie.hu-berlin.de
F. Bianchi, Dr. S. Halm, Dr. S. Blumstengel,
Prof. F. Henneberger
Institut für Physik and IRIS Adlershof
Humboldt-Universität zu Berlin
Newtonstr. 15 12489, Berlin, Germany
E-mail: fh@physik.hu-berlin.de
DOI: 10.1002/adfm.201402638
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201402638
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1

www.afm-journal.de
www.MaterialsViews.com
a
)
n
B
u
L
i ,
f
B
C
(
l
O
n
B
H
u
)
3
the observed absorption and emission
characteristics of the degraded samples
cannot be explained by the conventional
fluorenone-type defect in LPPPs. Finally,
O
O
a)
b
)
P
d
(
d
p
p
)
*
C
C
l 2
,
2
2
C
s
C
O
2
3
Br
Br
Br
+
5
5
%
time-dependent
photoluminescence
1
3
2
O
O
O
(PL) measurements show that intact L4P
is efficiently transferring its excitation
energy to defects.
K
M
E
n
O
A
4
T
B
C
9
4
%
O
O
O
C
H
T
(
o
O
s
C
O
H
H
)
3
3
2. Synthesis
The synthesis of the two ketones L4P-O
and L4P-O2[12] involves in both cases
cross-coupling of properly substituted
phenyl residues to a central fluorene
building block (Scheme 1). In the first
case of L4P-O, annulation proceeds by
twofold Friedel-Crafts alkylation of the
protected fluorenone core. On the other
hand, twofold Friedel-Crafts acylation of
a central fluorene building block creates
the two terminal ketone functionalities in
L4P-O2. Recently, we also published a syn-
thesis yielding L4P,[11d,12] which has been
designed to prepare longer oligomers
and therefore involves many reaction
steps. Comparing the different syntheses,
we recommend the route via diester 10
described herein for facile access of L4P.
For the synthesis of L4P-O (Scheme
1a), 2,7-dibromofluorenone 2 was chosen
as the central building block as well as
9
5
%
O
O
O
O
4
5
O
O
O
O
a
b
)
)
C
B
H
L
i
3
F
*
T
H
F
,
3
O
SH
HS
5
1
%
S
S
L4P-O
61%
N
I
S
+
O
6
+ isomers
7
9%
O
a
)
n
B
u
L
i ,
f
B
C
(
l
O
n
B
H
u
)
3
b
)
P
d
(
d
p
p
)
*
C
C
l
,
2
2
2
b)
C
s
C
O
2
3
Br
Br
Br
+
3
9
%
1
8
9
O
O
K
M
n
O
A
2-(methoxymethyl)-bromobenzene
as
4
T
E
B
C
the outer building block carrying the
masked groups for intramolecular cycli-
zation. Ether 1 was synthesized from
2-bromobenzyl bromide and sodium
methoxide. The synthesis of 2,7-dibro-
mofluorenone was adapted from an old,
yet simple protocol by Goldschmiedt and
Schranzhofer.[13] Ether 1 was converted to
a boronic ester via lithiation and coupled
twice to 2 via Suzuki crosscoupling using
conditions, which were adapted from a
paper by Knochel involving Pd(dppf)Cl2 as
catalyst and Cs2CO3 as base.[14] Oxidation
of diether 3 with KMnO4 yielded diester 4
in good yield.[15] Then, the ketone was pro-
tected as its dimethyl ketal using trimethyl
orthoformate and toluenesulfonic acid.
Subsequently, the ester-functionalities of 5 c)
were reacted with methyl lithium to yield
a tertiary alcohol intermediate, which was
not isolated. For the twofold condensation
to the desired ring-closed product, a Lewis
acid needs to be employed, which – if
used alone – was found to deprotect the
ketone and eliminate the tertiary alcohols,
8
0
%
a
C
)
N
a
C
O
H
H
b
)
S
O
l
,
S
O
H
2
3
3
7
3
%
O
O
L4P-O2
1 0
O
O
O
O
a
)
)
u
C
H
F
L
i
3
b
A
B
s
*
u
T
t
H
e
F
3
b
e
!
L4P
O
O
P
d
(
d
p
p
f
)
C
l
*
C
H
C
l 2
,
2
2
C
s
C
O
OH
2
3
B
Br
Br
+
8
3
%
OH
11
2
FLO-Ph2
Scheme 1. Investigated fluorene and fluorenone derivatives and their synthesis: a) Synthesis of
L4P-O, b) synthesis of L4P-O2 and L4P, and c) synthesis of FLO-Ph2.
©
2
wileyonlinelibrary.com
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201402638

www.afm-journal.de
www.MaterialsViews.com
followed by further decompostion. In contrast, addition of
1,2-ethanedithiol (in the presence of the Lewis acid) led to con-
version of the ketone to 1,3-dithiolane 6, which is stable in the
presence of the Lewis acid, as well as a minor amount of iso-
mers (that have not been separated). Deprotection with N-iodo-
succinimide and acetic acid gave the desired L4P-O as the main
product as well as traces of 7, which can easily be separated by
column chromatography or crystallization.
The synthesis of L4P-O2 (Scheme 1b) is rather straightfor-
ward: Ether 1 was coupled to 2,7-dibromofluorene 8 to yield
diether 9, which was subsequently oxidized to the diester 10
and hydrolyzed to the free acid. Condensation to L4P-O2 was
achieved by reaction of the bisacid with thionyl chloride and
methanesulfonic acid as catalyst. Depending on the purity and
morphology, L4P-O2 appears as orange powder or deep orange
to reddish crystals.
order to elucidate the primary photo-degradation mechanism,
the measurements have to be carried out at concentrations suf-
ficiently low to neglect energy transfer. Over the course of irra-
diation, the narrow absorption peaks around 350 and 370 nm
strongly decrease and a broad featureless absorption between
380 and 500 nm rises (Figure 1a). In a similar manner the blue
L4P emission between 370 and 450 nm decreases and loses its
vibronic structure (Figure 1c). Simultaneously, a broad green
PL appears with a maximum at 480 nm (Figure 1b). The time
dependent decrease of both the absorption at the 0–0 transition
as well as the blue PL is very similar (Figure 1d), suggesting
that these two spectroscopic signatures correspond to one and
the same molecular structure, i.e. L4P. Interestingly, after the
L4P is mostly consumed and the maximum concentration of
the primary photoproduct is reached, the green emission even-
tually also starts to decay.
The semi-bridged FLO-Ph2 was obtained by coupling phe-
nylboronic acid 11 and 2,7-dibromofluorenone 2 (Scheme 1c).
During this reaction only few amounts of side-products were
formed and hence FLO-Ph2 could be crystallized from the
crude product in good yields.
These findings agree somehow with the common
photodegradation mechanisms proposed in the literature that
involve formation of carbonyl (C = O) group defects, typically
at the bridging benzylic sites.[8–10] If photodegradation would
therefore follow this kind of mechanism, structures carrying
the fluorenone motif, such as L4P-O and L4P-O2 (Scheme 1)
should be formed. However, the product mixture of degraded
L4P in methylene chloride or ethyl acetate displays very dif-
ferent optical spectra as compared to the synthesized ketones
L4P-O and L4P-O2 (Figure 2). The photodegraded mixtures
(Figure 2b) show a broad absorption down to 2.5 eV, signifying
the presence of transitions with considerable strength below
the absorption edge of L4P (3.3 eV, see Figure 1a). This low-
energy absorption is still substantial in the region of L4P emis-
sion, suggesting that energy transfer might be quite efficient,
if donor (L4P) and acceptor (defect) are in close proximity. The
defect emission is bright and at around 2.5 eV. In contrast, the
near-UV absorption (3.3–3.9 eV) of the ketones L4P-O and L4P-
O2 (Figure 2a) resembles the absorption of L4P (Figure 1a),
although the extinction coefficient ε is significantly smaller and
the maxima are blue-shifted. Additionally there is a low-energy
absorption between 2.5 and 3.3 eV, making these compounds
appear orange. Yet this broad transition has a very low oscil-
lator strength such that energy transfer for this type of defects is
rather unlikely. Additionally the orange emission of these com-
pounds is farther in the low-energy region than observed for any
degradation product or mixture and the emission quantum yield
is extremely low. From these data it becomes apparent that both
quantum yield of these ketones as well as their spectral overlap
with L4P is rather low and hence such fluorenone structures are
most likely not involved in the observed defect emission.
3. Photodegradation in Solution
To investigate the chemical reactions, which give rise to photo-
degradation of samples containing L4P, some initial experi-
ments were carried out in solution. In contrast to thin films
it is relatively simple to keep the samples homogenous and
hence the experimental conditions are defined more precisely
while at the same time chemical analysis is facilitated. In an
initial set of experiments, the influence of different solvents on
the process of photodegradation was analyzed by exposing the
respective solutions to strong UV-irradiation while acquiring
photoluminescence (PL) spectra (see Supporting Information:
Figure S5.1). Photodegradation is strongly dependent on the
solvent, for example in toluene and tetrahydrofuran (THF) it
is relatively slow whereas solutions in methylene chloride and
ethyl acetate degrade much more rapidly (over the course of
few hours). By far the fastest photodegradation (over the course
of few minutes) occurs in chloroform associated with a rapidly
increasing green PL.
3.1. Degradation Kinetics from Absorption and Emission
Spectroscopy
Based on these initial experiments, methylene chloride and
ethyl acetate were chosen for further investigation assuming
that in these solvents a general representation of the degrada-
tion behavior of L4P can be achieved in a reasonable duration
of UV-irradiation. To get a better understanding of the process
of degradation, kinetic studies of three observables were carried
out: The absorbance, the blue PL (excited at 330 nm), which
is related to native defect-free L4P, and the green emission,
which is excited below the absorption edge of L4P (at 400 nm)
for addressing defect-related states directly. As will be detailed
below, energy transfer between L4P and the defects is an addi-
tional process reducing the LOPP's radiative yield. Therefore, in
3.2. Chemical Analysis of Photodegradation Products
Another indication arises from comparison with the non-
bridged p-quarterphenyl 4P, which should behave very different
in case photodamage would occur at the benzylic bridges.
However, when analogous degradation experiments were car-
ried out in methylene chloride 4P behaves similarly to L4P (see
Supporting Information: Figure S5.2) providing another piece
of evidence that the formation of fluorenone defects is rather
unlikely to be the main photodegradation process.
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201402638
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3

www.afm-journal.de
www.MaterialsViews.com
Wavelength [nm]
Wavelength [nm]
517 477
496
413
354
310
276
564
443
413
0.8
0.6
0.4
0.2
0.0
(a)
(b)
2.5
3.0
3.5
Energy [eV]
4.0
4.5
2.2
2.4 2.6
Energy [eV]
2.8
3.0
Wavelength [nm]
517 477 443 413 387 365 344
(c)
(d)
2.4
2.6
2.8
Energy [eV]
3.0
3.2
3.4
3.6
0
50
100
Time [min]
150
200
Figure 1. Spectral changes upon irradiation of a 6.8 × 10−6 mol/L solution of L4P in CH2Cl2 (25 °C) with 300–400 nm light: a) Absorbance spectra,
b) corrected photoluminescence (PL) spectra in green region upon 400 nm excitation, and c) corrected PL spectra in blue region upon 330 nm excita-
tion. Spectra shown in a)-c) were recorded every 30 min. d) Resulting plot of changes in absorbance at 369 nm (black diamonds), blue PL detected at
394 nm (blue squares), and green PL detected at 500 nm (green circles) as a function of irradiation time.
In order to elucidate the origin of the defect emission, product
formation over time was studied once again in methylene chlo-
ride at concentrations compatible with analysis by means of chro-
matography coupled to mass spectrometry (UPLC-MS). Using
UV detection (Figure 3b) the main peak at a retention time of
6.86 min is attributed to the starting material L4P. Upon irradia-
tion, the first product, which is formed relatively rapidly, exhibits
a m/z = 460.2 and a characteristic isotopic pattern, such that it
can be assigned to the monochlorinated product L4P-Cl (peak at
7.44 min in Figure 3b). Within the proceeding degradation also a
relatively large amount of dichlorinated L4P-Cl2 (see Figures S5.5.
and S5.6.) was formed. However, the influence of the chlorina-
tion on the optical properties of L4P is rather negligible as can be
concluded from the transparency at long wavelength. In addition
to the singly and doubly chlorinated products, three low-energy
absorbing defect-species L4P-deg were formed (intense peaks
between 6.65 and 6.76 min in Figure 3a, also seen in 3b). All of
them show the same m/z = 455.2, which corresponds to a gain
of 28 mass units as compared to L4P. It turned out to be very dif-
ficult to correlate the observed gain in mass with any oxidation
process and therefore the chemical nature of the defect cannot
be derived from the above analysis. To rule out fragmentation or
complex formation upon ionization the ketone L4P-O was also
analyzed via UPLC-MS, but only a peak at m/z = 413 was found,
which corresponds to the [L4P-O+H]+-ion. Accordingly, no evi-
dence was found in the photodegraded samples for the presence
of the ketones L4P-O or L4P-O2.
To resolve the chemical structure of the photodegradation
products, two of the three main defects L4P-deg identified via
UPLC-MS monitoring (Figure 3a) were isolated on a scale suf-
ficient for subsequent analysis by NMR spectrocopy (Figure 4).
1
Comparison of the H-NMR spectra of one isomer (L4P-deg-a)
of the photodegradation products (L4P-deg) with L4P shows
the presence of an additional singlet slightly above 10 ppm,
the loss of one proton signal in the aromatic region and the
reduced symmetry of the latter. The aliphatic proton signals
also indicate a reduction of the molecular symmetry. The cor-
responding 13C-NMR spectra may be interpreted the same way
and furthermore show the presence of a carbonyl signal. Based
on this spectral information, we conclude that upon irradiation
of L4P in chloroform or methylene chloride a formylation reac-
tion (Figure 4c) occurs, that somewhat resembles the thermal
Reimer-Tiemann reaction[16] and gives rise to formation of
various monoformylated regioisomers. This would also explain
the much more rapid degradation in chloroform as compared
to methylene chloride. In the latter, L4P primarily undergoes
©
4
wileyonlinelibrary.com
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201402638

www.afm-journal.de
www.MaterialsViews.com
Wavelength [nm]
620
496
413
354
310
276
10
8
(a)
6
4
2
0
0.20
(b)
0.15
0.10
0.05
0.00
2.0
2.5
3.0
3.5
4.0
4.5
Energy [eV]
Figure 2. a) Absorption spectra of L4P-O (solid, black) and L4P-O2 (solid, red) as well as photoluminescence (PL) spectra of L4P (dotted, blue), L4P-O
(dotted, black), and L4P-O2 (dotted, red) upon excitation at 330 nm (25 °C, CH2Cl2). The spectral region between 2.5 and 3.3 eV is not shown because
a cutoff filter was used to remove the 2nd harmonic of the excitation source and PL of residual, highly emissive L4P in the samples. b) Absorption (solid
lines) and PL spectra (dotted lines) of L4P after irradiation with 300–400 nm light in CH2Cl2 (black, initial concentration: 6.8 × 10−6 mol/L) and ethyl
acetate (red, initial concentration: 6.8 × 10−6 mol/L).
chlorination while methylene chloride is being chlorinated
itself to in-situ generate chloroform, which is the more reac-
tive formylation agent. Although we can only speculate about
the mechanism of this photochemical conversion (potentially
involving dichlorocarbenes or precursors of phosgene), it
becomes clear now, why especially chloroform, but also meth-
ylene chloride, are not suited to process polyfluorenes, polyphe-
nylenes and ladder-type polyphenylenes (since residual solvent
is difficult to be removed completely).
3.3. Optical Properties of Isolated Photodegradation Products
Naturally, the spectra of the isolated defects resemble the mixture
of photodegraded products (Figure 2b). The absorption spectra of
the isolated defects overlap extensively with the emission of L4P
in the region between 3.0 and 3.5 eV (Figure 5a) and, as already
mentioned above, energy transfer is expected to occur efficiently.
Furthermore the efficiency of the PL of the isolated defects is
much higher than that of the ketones L4P-O and L4P-O2 and the
emission is situated in the green to yellow region (2.6–2.7 eV).
In this context it is interesting to note that careful comparison
of the absorption and emission spectra can provide some further
insight into the molecular origin of the defects. In the case of
L4P-O there is a very weak low-energy absorption between 2.3
and 3.2 eV (depicted in Figure 5b in tenfold amplification), which
can be attributed to the symmetry-forbidden n→π* transition
of the ketone. In stark contrast, in FLO-Ph2, which has exactly
the same π-system in terms of connectivity, the same transi-
tion is significantly stronger, although the π→π* transition is
much weaker. This can readily be explained by the freely rotat-
able phenyl groups providing more flexibility to the molecule
and hence some mixing of the in-plane π→π* transition and
the out-of-plane n→π* may occur thereby enhancing the low-
energy transition.[17] This influence of backbone rigidity is not
only observed in absorption but also in emission: In contrast to
L4P-O, in FLO-Ph2 some of the PL is recovered and the quantum
yield amounts at least to a few per cent (≈4%) in CH2Cl2. This
implies that the reason for strong low-energy emission of the
(a)
L4P
L4P-deg
(b)
L4P-Cl
6.4
6.6
6.8
7.0
7.2
7.4
7.6
Retention time [min]
Figure 3. Chromatography traces of L4P after irradiation with 350 nm
light in CH2Cl2 solution using UV-detection at a) 390–410 nm and
b) 325–375 nm giving rise to L4P-deg and L4P-Cl.
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201402638
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
5

www.afm-journal.de
www.MaterialsViews.com
1
Figure 4. Formylation of L4P during irradiation in CH2Cl2 or CHCl3. H-NMR spectra of a) pristine L4P (300 MHz, CD2Cl2) and b) defect L4P-deg-a
(500 MHz, CD2Cl2). c) Proposed photochemical formylation reaction to explain the observed photodegradation products.
isolated photodegradation products is not the broken symmetry
due to the reduced planarity of the aromatic system but instead
the ability of the carbonyl group to rotate out-of-plane.
Although in chlorinated solvents such as chloroform and
methylene chloride we have proven formation of a formylation
product, the general validity of these findings and their applica-
bility to explain photodegradation in different solvents or films
needed to be investigated. For this purpose, the aging behavior
of solutions in methylene chloride and ethyl acetate was com-
pared by means of absorption and PL measurements as well
©
6
wileyonlinelibrary.com
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201402638

www.afm-journal.de
www.MaterialsViews.com
oligomer L6P,[11d] it seems most likely that
structures related to the “phenol defect”[7] are
responsible for the green emission under the
conditions employed by us.
Wavelength [nm]
413 354
620
496
310
276
5
4
3
2
1
(a)
Although the products formed in ethyl
acetate behave in a very similar manner they
are not the same than the ones formed in
methylene chloride, as for example proven
by their mass spectra. Because typically
degradation is referred to an oxidation pro-
cess in the literature,[4a,7–10] the influence of
oxygen on photodegradation was evaluated.
Samples of L4P in different solvents, each of
them under Argon as well as under air, were
irradiated at 350 nm at equal conditions. In
several solvents (methylene chloride, ethyl
acetate, tetrahydrofuran) bubbling Argon
through the solution prior to irradiation leads
to a significant slow-down of the degradation.
The most pronounced difference was found
for methylene chloride, where in the Argon-
saturated solution the typical L4P absorption
decreased much more slowly as compared to
the aerated sample (Figure 6a). Clearly, the
absence of oxygen inhibits photodegradation
0
8
(b)
6
4
2
0
x10
2.0
2.5
3.0
Energy [eV]
3.5
4.0
4.5
Figure 5. a) Absorption (solid lines) and photoluminescence (PL, dotted lines) of two different
isolated formyl defects (L4P-deg in Figure 4c, excitation at 380 and 395 nm, respectively) and
PL of L4P (dotted blue, excitation at 330 nm) in CH2Cl2 (25 °C). b) Absorption (solid and
dashed dotted (zoom) lines) and PL spectra (dotted lines) of L4P-O (red, excitation at 330 nm) or at least it slows it down (also knowing that
and FLO-Ph2 (black, excitation at 435 nm) in CH2Cl2 (25 °C).
water was present in all the samples).
To get another hint towards the oxida-
as UPLC-MS analysis. In the case of methylene chloride, three
main products absorbing around 400 nm were found and
are attributed to the three isomers of the formylated product
L4P-deg. In ethyl acetate there is a large variety of products that
absorb above 400 nm, making it—at least by the means that
were available for us—not feasible to isolate and characterize
them. However, the overall absorption spectra of the irradiated
solutions in ethyl acetate (Figure 2b) show qualitatively the
same behavior when compared to methylene chloride: Upon
irradiation the sharp peaks of the L4P absorption are vanishing
and a broad tail below 3.3 eV appears, which again overlaps
with the L4P emission. Even the kinetics of the degradation in
ethyl acetate (see Supporting Information: Figure S5.3) is sim-
ilar to that in methylene chloride.
A very thorough theoretical analysis of the optical properies
of potential degradation products of larger oligomers of LPPP
is given in ref. [7] that discusses three types of defects: a) a
“ketonic defect”, resembling fluorenone-type defect and hence
the structure of L4P-O, b) a “phenol defect”, and c) a “pheno-
late defect”.[7] While for the first, the calculations of the optical
transitions are in good agreement with our findings, the latter
two defects arise from the oxidative cleavage of a methylene
bridge, resulting in a freely rotatable arylketone fragment adja-
cent to the phenol unit. Interestingly, in the excited state, these
types of structure show allowed transitions below the pristine
LPPP and in the case of deprotonated species, i.e. the “pheno-
late defect”,[7] the energy of the transition is even below the flu-
orenone-type defect. In view of our observations and employed
experimental conditions as well as the fact that the π-system of
the “phenol defect” somehow resembles L4P-deg, which indeed
shows low energy emission, even compared to the longer
tion mechanism, different quenchers for active species of
oxygen were added to solutions of L4P in ethyl acetate prior to
Wavelength [nm]
443
2.0
387
344
310
(a)
1.5
1.0
0.5
0.0
0.4
(b)
0.2
0.0
2.8
3.2
3.6
4.0
Energy [eV]
Figure 6. a) Absorption (0.2 mm path length) of L4P in CH2Cl2 (initial
concentration 10−3 mol/L) after irradiation with 350 nm light under Argon
(red) and under air (black). b) Absorption (1 mm path length) of L4P
solutions in ethyl acetate (initial concentration 4.4 × 10−5 mol/L) after
irradiation with 350 nm light in the presence of DTBP (green, 0.02 mol/L)
or DABCO (red, 0.02 mol/L), without an additive (black) as well as the
original solution (blue). For solutions containing additives, proper back-
ground correction has been carried out (see Supporting Information
Figure S5.4 for unprocessed spectra).
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201402638
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
7

www.afm-journal.de
www.MaterialsViews.com
irradiation. Namely di-2,6-tert-butylphenol (DTBP) was added,
which is a known scavenger for radical auto-oxidation pro-
cesses, as well as 1,4-diazabicyclo[2.2.2]octan (DABCO), which
is a quencher for singlet oxygen.[18] Surprisingly, the addi-
tion of DTBP did not lead to a longer lifetime of the L4P in
solution and in contrast the aging even seemed to accelerate
(Figure 6b). However, upon addition of DABCO practically no
photodegradation was observed after the same duration of irra-
diation. Hence, photodegradation of L4P in solution primarily
seems to involve singlet oxygen as the main active oxidant. Sin-
glet oxygen engages in variety of reactions, mainly [2+2]- and
[4+2]-cycloadditions as well as ene-reactions, involving mostly
olefins or dienes, in some cases even aromatic systems, as sub-
strates. Primarily (endo)peroxides and oxetanes are formed in
these reactions, which eventually would lead to the formation
of ketones and aldehydes.[19] L4P does not display one of these
particularly priviledged substrate structures giving rise to rather
low selectivity during oxidation and the large number of dif-
ferent degradation products obtained in ethyl acetate.
hybrid inorganic-organic systems mentioned in the intro-
duction, intermolecular processes, such as crosslinking,
exciton migration and excimer formation may have a strong
influence.[4d,g,h,k,l,m,p,r,8] Thus it is necessary to work with solid
films containing a high concentration of molecules. Therefore,
thin layers of L4P embedded in a polystyrene (PS) matrix were
investigated. Three types of 100 nm thick films (A, B, and C)
with increasing L4P concentrations of 1, 10, and 20 wt% L4P/
PS were prepared by spin-casting from toluene solutions onto
sapphire substrates. Assuming a homogeneous distribution of
the molecules in the matrix, the associated average intermo-
lecular distances decrease from 4 nm over 2 nm to 1.5 nm for
samples A, B, and C, respectively. PL spectra of the pristine,
unexposed films normalized to L4P’s 0–0 transition (Figure 7a)
are rather similar to the PL spectra in solution (see Figure 1c).
However, the high concentration of L4P in all films as com-
pared to dilute solutions results in reabsorption of the PL
above 3.25 eV, hence the 0–1 transition dominates the emis-
sion spectra. Furthermore, a spectrally broad, blue-green emis-
sion becomes prominent at increasing concentration. All the
experimental data presented below consistently demonstrate
that the origin of this emission is the same photochemically
generated defect as observed for solutions, while aggrega-
tion or excimer formation can be excluded. The defect mole-
cules are already present as impurities in the pristine films
and, owing to the much larger molecular concentrations,
4. Degradation in the Solid State and Energy
Transfer
While photodegradation of L4P is more conveniently studied
in dilute solutions, for device applications, such as the
Figure 7. Energy transfer from L4P to defects in L4P/polystyrene (PS) layers on sapphire. a) PL spectra of 100 nm thick pristine films with increasing
amount of L4P: 1 wt% (blue), 10 wt% (orange), and 20 wt% (red, solid). Spectra are normalized to the 0–0 transition. PL spectrum of the 20 wt%
sample after partial photodegradation (red, dash-dotted). b) Absorbance spectra of a pristine (solid, red) and a strongly photo-aged film (dash-dotted,
red; spectrum enlarged by a factor of 3 for better visibility) of the 20 wt% sample as well as PL excitation (PLE) spectrum corresponding to the defect
emission of a partially photo-aged film (solid, purple; Edet = 2.38 eV) of the 20 wt% sample. “Partial” and “strong” aging denotes Ti:Sapphire laser irra-
diation (λ = 350 nm) under ambient conditions for 15 min at 5 W/cm2 and 30 min at 50 W/cm2, respectively. c,d) PL decay of pristine films containing
either 1 wt% (blue, shifted for clarity) or 20 wt% (red) of L4P. c) PL of L4P (EPL = 3.28) and d) defect PL (EPL = 2.38 eV). Black curves are fitting results
using a Förster type energy transfer model (for details, see text) convoluted with the instrument response function (IRF, grey lines).
©
8
wileyonlinelibrary.com
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201402638

www.afm-journal.de
www.MaterialsViews.com
efficient energy transfer occurs here even without extensive
photodegradation.
energy transfer rate where the prefactor β is proportional to the
number of unexcited acceptors of density ρA within a spher-
ical volume V of Förster radius R0. With the initial conditions
ND(0) = ND,0 and NA(0) = NA,0 for a δ-like excitation pulse, the
solutions to Equation (1) are given by
Similarly to solutions (see Figure 1, 6), extended UV irradi-
ation of L4P/PS layers leads to a severe reduction of the UV
emission and an increase of the blue-green PL component as
illustrated exemplarily for a photodegraded layer with 20 wt.%
L4P (red dash-dotted line in Figure 7a). The increase observed
in the low-energy emission upon photodegradation makes it
unlikely to originate from aggregates. Addition of DABCO to
the layer was confirmed to considerably slow down the degra-
dation (see Supporting Information: Figure S4.1). Therefore
we assign the process at least in part to singlet-oxygen based
photo-oxidation.
The above conclusions are further substantiated by studying
the absorption of the layers (Figure 7b). While the absorption
of the pristine layer is that of L4P, the irradiated film exhibits
a reduced and nearly featureless spectrum, which extends to
energies below the L4P absorption onset at 3.25 eV. Again, this
behavior closely resembles that in solution: There, direct excita-
tion of photocreated defects via the low-energy absorption tail is
demonstrated (see Figure 1b), leading to the exclusion of exci-
mers as the main source of the blue-green PL. Here, we verify
by means of PL excitation (PLE) and time-resolved PL spectros-
copy that Förster-type energy transfer from L4P to the defects
is taking place for sufficiently high concentrations. In PLE,
the defect emission at EPL = 2.38 eV of a partially aged layer
of type C is monitored as a function of excitation energy. The
resulting spectrum (Figure 7b, purple curve) is an envelope to
the absorption spectra of the defects and of L4P, demonstrating
that the defects can be excited either directly or indirectly via
the L4P.
t
ND
(
t
)
= ND,0e−kDte−β kD and NA
(
t
)
= NA,0nA
(
t
)
+ k N * n
(
t)
)
(2)
(
T
D
A
At
with nA
(
t
)
= e−k and * being the convolution operator. The
acceptor population NA(t) includes both direct excitation
At
nA
(
t)
= e−k and excitation via energy transfer kT ≠ 0. The
model functions (2) are convoluted by the system's instru-
ment response function (Figures 7c and 7d, grey curves)
before fitting. As indicated by the black lines in Figures 7c
and 7d, both the donor and acceptor decay curves are consist-
ently described by one set of fitting parameters, namely kD
=
(1.52 ns)−1, β = 3.82, kA = (0.56 ns)−1, α = NA,0/ND,0 = 0.12.
Efficient energy transfer is evidenced by β > 2 which implies
that kT(t) > kD up to t > kD−1, that is, energy transfer domi-
nates the total donor decay up to times that are larger than
the intrinsic decay time. Being aware of the model’s limita-
tions, such as the assumed single exponential intrinsic donor
and acceptor decays or the neglect of energy diffusion between
donors,[22] the obtained fitting parameters are not discussed
in further detail. Finally, it is remarked that the observation of
energy transfer from L4P molecules to photochemically cre-
ated defects is not restricted to solid films, but equally applies
to highly concentrated solutions (see Supporting Information:
Figure S4.3).
The dynamics of the ET process was elucidated by time-
resolved PL spectroscopy. Pristine layers of type A and C with
notably different average intermolecular distances are investi-
gated. In order to avoid effects of photodegradation during the
experiment, such as changes in the donor-acceptor ratio,[20] a low
excitation density of 0.35 W/cm2 and short integration time of 20
s were chosen. In addition, the samples were kept under vacuum
(10−4 mbar) inside a cryostat, and each time trace was acquired
on a new, previously unexposed sample position. The excited
L4P and defect populations are monitored at emission energies
of 3.28 eV and 2.38 eV, respectively. The decay of the L4P emis-
sion in layer C (Figure 7c, red line) is markedly non-exponential
and much faster than in layer A (Figure 7c, blue line). Thus, an
additional decay channel with a time-dependent decay rate has
to be present here. Concerning the acceptor dynamics, a rise
time is clearly observed in layer C (Figure 7d, red line) but not
in layer A (Figure 7d, blue line). This proves the existence of a
feeding process to the defect population in the first case.
A Förster-type model[21] that takes into account energy
transfer from L4P donors to statistically distributed defect
acceptors is readily employed to describe the decay dynamics
in layer C. Accordingly, the number of excited donors ND and
acceptors NA follow the rate equations:
5. Conclusion
It is demonstrated that small, ladder-type L4P molecules
dispersed in solution or a solid matrix degrade upon UV
irradiation in the presence of oxygen. This photodegradation
results in green-yellow emitting defects, which for sufficiently
small intermolecular distances are efficiently excited via energy
transfer from the original L4P. The overall emission pro-
cess from the defect traps agrees well with previous extensive
studies on photodegraded polyfluorene-type polymers.[4c,4k,23]
Considering the process of photodegradation, our system at
first appears to be rather similar to what has been proposed for
polyfluorenes in the literature (i.e., greenish-yellow emission
upon extensive irradiation). But already kinetic studies with
L4P suggest that neither the fluorenyl anion mechanism,[4a,4f]
which should follow typical first order photokinetics, nor the
radical chain mechanism[8] may explain our observations. The
inhibition of photodegradation by adding DABCO to the sam-
ples proved, that singlet oxygen is the major source of decom-
position, as has already been proposed by Abbel.[10] Thereby, the
accelerating kinetics can readily be explained: At the beginning
the L4P substrate acts as singlet oxygen sensitizer, yet with a
rather low sensitization quantum yield.[24] But as soon as the
aromatic carbonyl defect has been formed, it acts as a much
more efficient singlet oxygen sensitizer due to the enhanced
intersystem crossing leading to efficient population of the car-
bonyl triplet,[25] and hence the photochemical degradation pro-
cess is auto-catalytic in nature.
ꢀ
ꢀ
)ND and NA = −kANA + kT t)ND
(
ND = −kDND − kT
(
t
(1)
Herein, kD and kA are the intrinsic donor and acceptor decay
rates, respectively, and kT
(
t
)
= β/2 kD/t is the time-dependent
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201402638
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
9

www.afm-journal.de
www.MaterialsViews.com
Phys. 2003, 119, 12017; i) P. Papadopoulos, G. Floudas, C. Chi,
G. Wegner, J. Chem. Phys. 2004, 120, 2368–2374; k) M. Sims,
D. D. C. Bradley, M. Ariu, M. Koeberg, A. Asimakis, M. Grell,
D. G. Lidzey, Adv. Func. Mater. 2004, 14, 765–781; l) C. Gadermaier,
L. Romaner, T. Piok, E. J. W. List, B. Souharce, U. Scherf, G. Cerullo,
G. Lanzani1, Phys. Rev. B 2005, 72, 045208; m) C. Chi, C. Im,
V. Enkelmann, A. Ziegler, G. Lieser, G. Wegner, Chem. Eur. J.
2005, 11, 6833–6845; n) K. Becker, J. M. Lupton, J. Feldmann,
B. S. Nehls, F. Galbrecht, D. Gao, U. Scherf, Adv. Funct. Mater.
2006, 16, 364; o) C. Chi, C. Im, G. Wegner, J. Chem. Phys. 2006, 124,
024907; p) Y.-S. Wu, J. Li, X.-C. Ai, L.-M. Fu, J.-P. Zhang, Y.-Q. Fu,
J.-J. Zhou, L. Li, Z.-S. Bo, J. Phys. Chem. A 2007, 111, 11473;
r) J. Kang, J. Jo, Y. Jo, S. Y. Lee, P. E. Keivanidis, G. Wegner,
D. Y. Yoon, Polymer 2008, 49, 5700–5704.
At this point it should be emphasized that upon photodeg-
radation no evidence of a derivative of L4P carrying the flu-
orenone motif has been found and that the synthesized ketones
L4P-O and L4P-O2 did not exhibit optical properties reminis-
cent of the actually formed defects in photodegraded L4P sam-
ples. Our results—in agreement with previous studies[4]—imply
that upon photodegradation carbonyl defects are being intro-
duced into the LPPP structure; however and more importantly,
they suggest that a rigid LPPP carrying the fluorenone motif
itself cannot explain the typical yellow-to-green defect emission!
It appears that the aromatic backbone loses rigidity (as proven
in other works)[6,7] by potential cleavage of bridging methylene
groups or introduction of rotatable carbonyl groups and there-
fore symmetry-forbidden n→π* transitions become partially
allowed as reflected in the corresponding optical spectra. This
finding is highly relevant for the critical discussion of reaction
mechanisms accounting for the photodegradation in LPPPs as
well as PF, and in organic materials in general.
[5] S. Kappaun, H. Scheiber, R. Trattnig, E. Zojer, E. J. W. List,
C. Slugovc, Chem. Commun. 2008, 5170–5172.
[6] a) W. Graupner, M. Sacher, M. Graupner, C. Zenz, G. Grampp,
A. Hermetter, G. Leising in Electrical, Optical, and Magnetic Properties
of Organic Solid-State Materials IV, (Ed.: J. R. Reynolds, A. K.-Y. Jen,
M. F. Rubner, L. Y. Chiang, L. R. Dalton), Materials Research Society,
Warrendale, PA 1998, 789; b) E. J. W. List, J. Partee, J. Shinar,
U. Scherf, K. Müllen, E. Zojer, K. Petritsch, G. Leising, W. Graupner,
Phys. Rev. B 2000, 61, 10807; c) L. Liu, S. Qiu, B. Wang, W. Zhang,
P. Lu, Z. Xie, M. Hanif, Y. Ma, J. Shen, J. Phys. Chem. B 2005, 109, 23366.
[7] L. Romaner, G. Heimel, H. Wiesenhofer, P. Scandiucci de Freitas,
U. Scherf, J.-L. Brédas, E. Zojer, E. J. W. List, Chem. Mater. 2004, 16,
4667–4674.
[8] a) B. W. Zhao, T. Cao, J. M. White, Adv. Funct. Mater. 2004, 14, 783;
b) R. Grisorio, G. P. Suranna, P. Mastrorilli, C. F. Nobile, Adv. Funct.
Mater. 2007, 17, 538.
[9] a) L. Liu, S. Tang, M. Liu, Z. Xie, W. Zhang, P. Lu, M. Hanif, Y. Ma,
J. Phys. Chem. B 2006, 110, 13734; b) R. Grisorio, G. Allegretta,
P. Mastrorilli, G. P. Suranna, Macromolecules 2011, 44, 7977.
[10] R. Abbel, M. Wolffs, R. A. A. Bovee, J. L. J. van Dongen, X. Lou,
O. Henze, W. J. Feast, E. W. Meijer, A. P. H. J. Schenning, Adv.
Mater. 2009, 21, 597.
Supporting Information
Supporting Information (synthesis and characterization details,
description of degradation experiments and time-resolved luminescence
spectroscopy) is available from the Wiley Online Library or from the
author.
Acknowledgements
Generous support by the German Research Foundation (DFG via SFB
951) and the Helmholtz Association (via Helmholtz Energy Alliance) is
gratefully acknowledged. BASF AG, Bayer Industry Services and Sasol
Germany are thanked for generous donations of chemicals.
Received: August 4, 2014
Revised: August 31, 2014
Published online:
[11] a) V. M. Agranovich, Y. N. Gartstein, M. Litinskaya, Chem. Rev.
2011, 111, 5179; b) S. Blumstengel, S. Sadofev, C. Xu, J. Puls,
R. L. Johnson, H. Glowatzki, N. Koch, F. Henneberger, Pys. Rev.
B 2008, 77, 085323; c) S. Blumstengel, H. Glowatzki, S. Sadofev,
N. Koch, S. Kowarik, J. P. Rabe, F. Henneberger, Phys. Chem. Chem.
Phys. 2010, 12, 11642; d) B. Kobin, L. Grubert, S. Blumstengel,
F. Henneberger, S. Hecht, J. Mater. Chem. 2012, 22, 4383.
[12] L4P,as well as L4P-O2 have been described in M. Kimura (Idemitsu
Kosan Co., Ltd.), EP 1860 097 A1, 2007; however the synthetic route
involves harsh conditions, characterization data for the “intermedi-
ates” was not provided, and some of the few properties given are
obviously different in reality.
[13] G. Goldschmiedt, F. Schranzhofer, Monatshefte für Chemie 1895, 16,
807.
[14] B. A. Haag, C. Sämann, A. Jana, P. Knochel, Angew. Chem. Int. Ed.
2011, 50, 7290; Angew. Chem. 2011, 123, 7428.
[15] The procedure was adapted from: J. H. Markgraf, B. Y. Choi, Synth.
Commun. 1999, 29, 2405.
[16] a) K. Reimer, F. Tiemann, Chem. Ber. 1876, 9, 824; b) L. Kürti,
B. Czakó, Strategic Applications of Named Reactions in Organic Syn-
thesis, Elsevier, Burlington, San Diego, London 2005, 378.
[17] N. J. Turro, V. Ramamurthy, J. C. Scaiano, Modern Molecular Photo-
chemistry of Organic Molecules, University Science Books, Sausalito,
CA 2010, Sec. 4 23–4.26.
[18] A. G. Griesbeck, T. T. El-Idreesy, W. Adam, O. Krebs in CRC Hand-
book of Organic Photochemistry and Photobiology, 2nd Ed, (Eds.
W. Horspool, F. Lenci), CRC Press Boca Raton, FL 2004, Ch. 8.
[19] a) C. S. Foote, Acc. Chem. Res. 1968, 1, 104–110; b) W. Ando,
S. Kohmoto, Chem. Commun. 1978, 120–121; c) CRC
[1] a) Organic Electronics, (Ed.: F. So), CRC Press, Boca Raton, FL
2010; b) Organic Light Emitting Devices, (Ed.: K. Müllen, U. Scherf),
WILEY-VCH, Weinheim 2006.
[2] U. Scherf, K. Müllen, Makromol. Chem. Rapid Commun. 1991, 12, 489.
[3] a) R. H. Friend, R. W. Gymer, A. B. Holmes, J. H. Burroughes,
R. N. Marks, C. Taliani, D. D. C. Bradley, D. A. D. Santos,
J. L. Brédas, M. Logdlund, W. R. Salaneck, Nature 1999, 397, 121;
b) J. Huber, K. Müllen, J. Salbeck, H. Schenk, U. Scherf, T. Stehlin,
R. Stern, Acta Polymer. 1994, 45, 244.
[4] For an overview on degradation of polyfluorenes see: a) U. Scherf,
E. J. W. List, Adv. Mater. 2002, 14, 477; b) S. Gamerith,
C. Gadermaier, U. Scherf, E. J. W. List, in Physics of Organic Semicon-
ductors, (Ed.:, (Ed.: W. Brütting), WILEY-VCH, Weinheim 2005, Ch.
6; c) S. Gamerith, C. Gadermaier, U. Scherf, E. J. List, Phys. Status
Solidi 2004, 201, 1132; d) For original research contributions see:
d) V. N. Bliznyuk, S. A. Carter, J. C. Scott, G. Klärner, R. D. Miller,
D. C. Miller, Macromolecules 1999, 32, 361; e) E. Zojer, A. Pogantsch,
E. Hennebicq, D. Beljonne, J.-L. Brédas, P. S. de Freitas,
U. Scherf, E. J. W. List, J. Chem. Phys. 2002, 117, 6794;
f) E. J. W. List, R. Guentner, P. Scandiucci de Freitas, U. Scherf, Adv.
Mater. 2002, 14, 374; g) J. M. Lupton, M. R. Craig, E. W. Meijer,
Appl. Phys. Lett. 2002, 80, 4489; h) S. I. Hintschich, C. Rothe,
S. Sinha, A. P. Monkman, P. S. de Freitas, U. Scherf, J. Chem.
©
10 wileyonlinelibrary.com
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201402638

www.afm-journal.de
www.MaterialsViews.com
Handbook of Organic Photochemistry and Photobiology, 2nd Ed,
(Eds: W. Horspool, F. Lenci), CRC Press, Boca Raton, FL 2004,
Ch. 4, 8, 25, 45, 108.
[23] H. Marciniak, M. Teicher, U. Scherf, S. Trost, T. Riedl, M. Lehnhardt,
T. Rabe, W. Kowalksy, S. Lochbrunner, Phys. Rev. B 2012, 85, 214204.
[24] S. M. Fonseca, J. Pina, L. G. Arnaut, J. Seixas de Melo,
H. D. Burrows, N. Chattopadhyay, L. Alcácer, A. Charas,
J. Morgado, A. P. Monkman, U. Asawapirom, U. Scherf, R. Edge,
S. Navaratnam, J. Phys. Chem. B 2006, 110, 8278.
[20] That photodegradation is indeed influencing the ET dynamics by
increasing the acceptor-donor ratio is demonstrated in Figure S4.2
of the supporting information.
[21] Th. Förster, Z. Naturforsch. 1949, 4a, 321.
[22] E. Engel, K. Leo, M. Hoffmann, Chem. Phys. 2006, 325,
170.
[25] N. J. Turro, V. Ramamurthy, J. C. Scaiano, Modern Molecular Photo-
chemistry of Organic Molecules, University Science Books, Sausalito,
CA 2010, Sec. 14.6.
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201402638
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
11