Efficient energy sensitization of C60 and application to organic photovoltaics

Article abstract of DOI:10.1021/ja4043356

Fullerenes are currently the most popular electron-acceptor material used in organic photovoltaics (OPVs) due to their superior properties, such as good electron conductivity and efficient charge separation at the donor/acceptor interface. However, low absorptivity in the visible spectral region is a significant drawback of fullerenes. In this study, we have designed a zinc chlorodipyrrin derivative (ZCl) that absorbs strongly in the visible region (450-600 nm) with an optical density 7-fold higher than a C₆₀ film. ZCl efficiently transfers absorbed photoenergy to C₆₀ in mixed films. Application of ZCl as an energy sensitizer in OPV devices leads to an increase in the photocurrent from the acceptor layer, without changing the other device characteristics, i.e., open circuit voltage and fill factor. For example, C ₆₀-based OPVs with and without the sensitizer give 4.03 and 3.05 mA/cm², respectively, while both have V_OC = 0.88 V and FF = 0.44. Our ZCl sensitization approach improves the absorbance of the electron-acceptor layer while still utilizing the beneficial characteristics of C₆₀ in OPVs.

Full text of DOI:10.1021/ja4043356

Article
pubs.acs.org/JACS
Efficient Energy Sensitization of C60 and Application to Organic
Photovoltaics
Cong Trinh,† Kent O. Kirlikovali,† Andrew N. Bartynski,‡ Christopher J. Tassone,§ Michael F. Toney,§
George F. Burkhard,∥ Michael D. McGehee,⊥ Peter I. Djurovich,† and Mark E. Thompson*,†,‡
‡
†Department of Chemistry and Department of Chemical Engineering, University of Southern California, Los Angeles, California
90089, United States
§Materials Science Department, Stanford Synchrotron Radiation Lightsource, Menlo Park, California 94025, United States
⊥
∥Geballe Laboratory for Advanced Materials and Department of Materials Science and Engineering, Stanford University, Stanford,
California 94305, United States
S
* Supporting Information
ABSTRACT: Fullerenes are currently the most popular
electron-acceptor material used in organic photovoltaics
(OPVs) due to their superior properties, such as good
electron conductivity and efficient charge separation at the
donor/acceptor interface. However, low absorptivity in the
visible spectral region is a significant drawback of fullerenes. In
this study, we have designed a zinc chlorodipyrrin derivative
(ZCl) that absorbs strongly in the visible region (450−600
nm) with an optical density 7-fold higher than a C60 film. ZCl
efficiently transfers absorbed photoenergy to C60 in mixed
films. Application of ZCl as an energy sensitizer in OPV
devices leads to an increase in the photocurrent from the acceptor layer, without changing the other device characteristics, i.e.,
open circuit voltage and fill factor. For example, C60-based OPVs with and without the sensitizer give 4.03 and 3.05 mA/cm2,
respectively, while both have VOC = 0.88 V and FF = 0.44. Our ZCl sensitization approach improves the absorbance of the
electron-acceptor layer while still utilizing the beneficial characteristics of C60 in OPVs.
INTRODUCTION
■
Organic solar cells, or organic photovoltaics (OPV), are a
promising energy source as they offer potential advantages over
the conventional inorganic photovoltaics, including lightweight,
flexibility, and low cost roll-to-roll production. A typical OPV
device consists of separate electron-donor and -acceptor
materials, which are responsible for light absorption and charge
generation, as excitons dissociate at the donor/acceptor (D/A)
interface.1−3 While research has largely focused on the
development of new donor materials, fullerenes remain the
most popular acceptor materials because of their good electron
conductivity4−6 and ability to promote efficient D/A charge
Figure 1. Absorption spectra of C60 in neat films (black circles),
toluene solution (blue squares), and AM1.5G solar photon flux (red).
separation.7,8 Fullerenes have been shown to perform well with
a variety of donor materials,8 and all of the reported high
efficiency OPV devices utilize fullerene acceptors.9 A key
near-infrared (NIR), the green-to-orange part of the spectrum
is left unabsorbed due to the relatively narrow absorption bands
of organic dyes.11−19
In order to efficiently harvest solar energy, it would thus be
desirable to have new acceptor materials that strongly absorb
visible light, while also retaining the favorable electron mobility
drawback of the most commonly used fullerene C60 is its low
absorption in the visible part of the spectrum, due to the
symmetry forbidden nature of the lowest energy electronic
transition, λ = 670 nm. A neat C60 film shows a high optical
density between 400−500 nm assigned to an intermolecular
charge-transfer (CT) absorption.10 The poor overlap of this C60
absorption with the AM1.5G solar flux (Figure 1) requires that
the donor material(s) collect the majority of the incident
photons. As absorption of the donor material is extended to the
Received: May 10, 2013
Published: July 17, 2013
© 2013 American Chemical Society
11920
dx.doi.org/10.1021/ja4043356 | J. Am. Chem. Soc. 2013, 135, 11920−11928

Journal of the American Chemical Society
Article
and charge separation efficiencies found in fullerenes. In this
way, light absorption could be more equitably distributed
between the donor and acceptor materials, potentially allowing
the device’s photoresponse to be fully extended into the NIR.
Unfortunately, the performance of nonfullerene acceptors is
generally inferior to that of fullerenes.7,8,20,21 Energy sensitiza-
tion of fullerenes is an alternative method to improve both the
breadth and the efficiency of visible light absorption in
fullerene-based acceptors.
The sensitization of phenyl-C61-butyric acid methyl ester
(PCBM) using a perylene diimide (PDI) derivative has been
recently reported by Hesse et al.22 It was shown that the
PCBM-PDI blend worked well with the UV-absorbing
hexabenzocoronene donor but not with a red-NIR-absorbing
polymer donor.22 This dependence on donor material might be
the result of dissimilarities between the molecular shapes of
PDI (flat) and PCBM (ball-like), leading to unfavorable phase
separation between the two acceptors. Moreover, while the
singlet energy of PDI is higher than that of PCBM, the triplet
energy of PDI (1.2 eV)23,24 is lower (PCBM triplet = 1.55 eV).
Thus, it is possible that singlet energy passed from PDI to
PCBM can, after intersystem crossing (ISC), undergo triplet
transfer back to PDI as was previously observed in a PDI-C60
dyad.25−27 From the energetic standpoint, this energy ping-
pong effect (illustrated in Figure 2a) is an undesirable process
excitons and electrons and a sensitizer that serves as a light
absorbing guest. Upon excitation of the sensitizer, energy is
transferred to the C60 host and subsequently transported to the
D/A interface where, after charge separation, electrons are
conducted by the C60 host to the electrode. In our sensitization
approach, metal dipyrrin complexes are employed to enhance
the ability of a C60 based acceptor layer to generate
photocurrent from absorption in the green-to-orange part of
the solar spectrum. Energy sensitization to fullerenes has been
previously explored in solution by studying covalently linked
dyads containing fullerene and dipyrrins.30,31 However, these
compounds have not been used as sensitizers in solid state
devices, such as OPVs, where the CT absorption band of C60
dominates the photoresponse.
RESULTS AND DISCUSSION
■
Molecular Design for Sensitizers. Three possible path-
ways to sensitize C60 via photoexcitation of a sensitizer are
shown in Figure 2b: (A) singlet energy transfer, (B) triplet
energy transfer, or (C) electron transfer from sensitizer to C60
followed by charge recombination to the triplet state of C60. It
is important that all three pathways be viable for the
sensitization of C60, because the nonviable pathway can act as
an exciton or charge trap in the mixed C60 film. The three
pathways impose several requirements for an efficient
sensitizer: (1) Both singlet and triplet energies of the sensitizer
need to be greater than C60 to guarantee efficient energy
transfer to C60. (2) the oxidation potential has to be sufficiently
high to ensure that the energy of the CT state (i.e., Sen−C60+),
if formed, is greater than that of the triplet state of C60, E
C60
.
triplet
sen
To a first approximation, the CT energy is given by −q(E
−
ox
ECre6d0) − ΔEC, where q is the elementary charge, E and E
sen
ox
C60
red
are the oxidation potential of sensitizer and reduction potential
of C60, respectively, and ΔEC (estimated ∼0.3 eV)32 is the
Columbic interaction between the sensitizer cation and C60
Figure 2. Energy diagram showing sensitization pathways in two cases:
(a) a sensitizer (e.g., PDI or DCV3T) having higher singlet energy but
lower triplet energy than C60 and (b) a sensitizer having both singlet
and triplet energies higher than that of C60. There are three possible
pathways for sensitization: A, singlet transfer; B, triplet transfer; and C,
electron transfer to form a CT state, which decay to triplet state of C60
if the CT state energy is higher than C60 triplet. If the energy of the
CT state is lower than C60 triplet, it will act as a trap state (CT trap).
anion. Given a C60 reduction potential of −1.06 V (vs Fc/
Fc+),33
E
must be >0.74 V to satisfy this criterion. If the
ox
sen
energy of the CT state Sen+C60− is lower than C60 triplet state,
it will serve as a net electron donor to C60,34 forming a charge
trap state in our host−guest acceptor layer (Figure 2b). This
type of charge trapping is a potential problem in previous works
that involved blending an additional donor material into C60
layer.35,36 (3) In order to maintain good electron conductivity
by the C60 host, the reduction potential of the sensitizer has to
be lower (more negative) than C60 to ensure that electrons are
conducted by C60 without being trapped by the sensitizer. (4)
In addition, it is desirable that the molecular size and shape of
the sensitizer should be similar to C60 so as to not severely
disrupt the molecular packing and thus maintain good electron
conductivity in the C60-based acceptor layer. It should be
emphasized that our design strategy is different from the
previous works with PDI or DCV3T,22,29 as all energy absorbed
in the acceptor layer will ultimately be located on the C60 host.
Singlet and triplet energies as well as reduction and oxidation
potentials of C60 are summarized in Table 1. While there are
many organic dyes that have strong absorption in the visible
spectrum and higher singlet energy than C60, there are only a
few classes of organic compounds exhibiting higher or similar
triplet energy to C60 as well. Dipyrrins (ET = 1.82 eV)37 and
porphyrins (1.60−1.65 eV)38,39 meet these criteria, however
the flat molecular shape of porphyrins is very different from
C60. Zinc dipyrrin complexes40 have a quasi-spherical shape and
size similar to C60. Both singlet40 and triplet37 energies of
because the absorbed light energy is not confined to the
fullerene, causing energy loss due to the low energy of
sensitizer’s triplet. In addition, even though a photoresponse
from PDI was observed in a device using a blended PCBM-PDI
acceptor layer, it remains unclear whether the observed
photocurrent comes from energy sensitization of PCBM or
from direct charge separation between PDI and the donor at
the D/A interface. A similar energy ping-pong effect in a
dicyanovinyl-terthiophene (DCV3T)-C60 acceptor blend has
been intentionally utilized in OPV devices28,29 where it was
postulated that the triplet of DCV3T induced charge separation
at the D/A interface.29 Even though, as stated above, triplet
transfer back to the sensitizer is undesired, in this case the
conductivity and fill factor (FF) of OPV devices using these
blended acceptors were improved compared to the devices
using a neat DCV3T acceptor.
To take full advantage of the superior electron transport
properties of C60 while minimizing possible energy loss by back
triplet energy transfer from C60 to the sensitizer, the sensitized
energy is confined to C60. We have designed a blended acceptor
layer where C60 is used as the host material that conducts both
11921
dx.doi.org/10.1021/ja4043356 | J. Am. Chem. Soc. 2013, 135, 11920−11928

Journal of the American Chemical Society
Article
Table 1. Photophysical and Electrochemical Properties of
ZH, ZCl, and C60 in Solution and Thin Films
solution
thin film
1/2
E
ET
(eV)
Eox
b
E
QE
τ
E
ET
(eV)
red
S a
b
f
f
S a
(%) (ns) (eV)
(eV)
(V)
(V)
c
ZH
ZCl
C60
2.54
2.37
1.86
1.82
+0.71
+1.22
−1.94
−1.30
−1.06
41
18
−
3.9
2.2
−
2.33
2.22
1.84
−
1.65
c
1.75
d
e
1.55
+1.26
1.50,
1.44
Figure 3. (a) ORTEP diagram of ZCl and (b) space filling models of
ZCl (upper) and C60 (lower) with the proportional sizes as
determined from X-ray single crystal data. The atom marked in red
can be either H or Cl with an occupancy ratio H:Cl = 1:3.
d
a
Determined by the midpoint between the normalized absorption and
b
c
emission spectra recorded in cyclohexane. vs Fc+/Fc. Measured in 2-
d
e
f
MeTHF at 77K. Ref 45. Ref 33. In cyclohexane.
molecular volume of ZCl (1000 Å3) is similar to that of C60
(725 Å3) as estimated from the crystal structure.44
Electrochemical measurements of ZCl carried out in
dichloromethane with Fc+/Fc as the internal standard are
presented in Figure 4. ZCl undergoes irreversible oxidation at
dipyrrin complexes are higher than those of C60. Another
benefit of the dipyrrin complexes over porphyrins is that the
energy levels of zinc dipyrrins complexes are readily tunable via
chemical modification of pyrrole rings or substituents at the
meso position and prepared by straightforward synthetic
procedures.41 Zinc dipyrrins have been used as linkers and/or
light absorbers in multichromorphoric light absorbing40,42 or
charge separating systems43 in solution, but have not been used
in OPV devices.
Synthesis and Characterization of the Sensitizer. The
unsubstituted bis(5-mesityldipyrrinato) zinc (ZH) was synthe-
sized and characterized following a published procedure.42
Unfortunately, the oxidation potential of ZH (+0.71 V vs Fc+/
Fc) is not sufficient to satisfy requirement (2) above, and a low-
energy CT state is expected, i.e., ZH+C60−. Indeed, ZH can be
used as a donor material in OPV devices using C60 as the
acceptor (see SI). In order to raise the oxidation potential of
zinc dipyrrin, high electron affinity substituents, i.e., Cl atoms,
were added to the pyrrole rings.
The synthesis of a chlorinated zinc dipyrrin is summarized in
Scheme 1. The final product was obtained as an inseparable
mixture of two complexes, having either 11 (ZCl11) or 12
(ZCl12) chlorine atoms, in a respective molar ratio of 1:3 (17%
yield). The unique site of the residual pyrrolic hydrogen of the
ZCl11 derivative (indicated in red in Scheme 1) was confirmed
by standard and 1D NOESY NMR techniques (SI). Attempts
to separate these derivatives using column chromatography,
recrystallization, or thermal gradient sublimation were un-
successful. The ZCl12:ZCl11 ratio could be increased using
either longer reaction times or elevated temperatures; however,
the yields obtained under these conditions were so
unacceptably low (1−5%) that the 1:3 mixture of ZCl11:ZCl12
(abbreviated ZCl) was used for all subsequent experiments. An
X-ray analysis was performed on a co-crystal of ZCl11 and
ZCl12, and the molecular structure is shown in Figure 3 (the
atom marked in red can be either H or Cl with an occupancy
ratio H:Cl = 1:3). The two dipyrrin ligands are held nearly
perpendicular to each other through Zn center with a dihedral
angle of 87.6°, forming a quasi-spherical molecular shape. The
Figure 4. CV and DPV diagrams of ZCl in dichloromethane (vs Fc+/
Fc). Scan rate = 100 mV/s.
+1.22 V and two reversible reductions at −1.30 and −1.54 V
(vs Fc+/Fc). Since no discernible difference was observed in the
redox potentials of ZCl11 and ZCl12, the electronic properties of
the two compounds were evaluated using DFT calculations. In
agreement with the experimental data, ZCl11 and ZCl12 have
similar energies for their respective highest occupied molecular
orbitals (HOMO = −6.13 and −6.25 eV), lowest unoccupied
molecular orbitals (LUMO = −3.23 and −3.29 eV), and triplet
states (ET = 1.70 eV for both).
The absorption and emission spectra of ZCl in cyclohexane
solution and neat film are shown in Figure 5; the photophysical
properties are summarized in Table 1. ZCl displays an intense
absorption band between 450−600 nm and is strongly
fluorescent in cyclohexane at room temperature (quantum
yield = 18%, τ = 2.2 ns). Phosphorescence (λmax = 710 nm,
lifetime of ∼0.5 ms) of ZCl was observed in 2-methyltetrahy-
drofuran (2-MeTHF) at 77 K (Figure 4a). Chlorination slightly
lowers both singlet and triplet states of ZCl compared to ZH by
170 and 70 meV, respectively. However, the singlet and triplet
energies of ZCl are still significantly higher than C60, satisfying
requirement (1) for the sensitizer. Chlorination significantly
Scheme 1
11922
dx.doi.org/10.1021/ja4043356 | J. Am. Chem. Soc. 2013, 135, 11920−11928

Journal of the American Chemical Society
Article
Figure 6. Absorption (solid symbols) and emission under excitation at
514 nm (open symbols) of 50 nm C60 film (black circle) and 59 nm
C60:ZCl film (blue triangle, 15% ZCl by volume) and 75 nm C60:ZCl
film (red square, 35% ZCl by volume). Also shown is an energy level
diagram (eV) for energy transfer from ZCl to C60.
(λmax = 737 nm) match well with published fluorescence
spectra of C6046,47 and differ markedly in peak position from the
phosphorescence of ZCl (Figure 5). The fluorescence intensity
increased with higher ZCl content and consequent rise in
absorbance from the blended films. Thus, the PL measurements
indicate that photoexcited ZCl transfers energy to the C60 host
in the thin films, resulting in increased fluorescence from C60.
The luminescent quenching of ZCl could occur via either
Figure 5. Absorption (blue) and emission (red) spectra of (a) solution
of ZCl in cyclohexane at room temperature, inset is the
phosphorescence of ZCl measured in 2-MeTHF at 77K, excitation =
460 nm and (b) ZCl film at room temperature, excitation = 460 nm.
The thin film absorption spectrum of C60 is also shown for
comparison.
Forster energy transfer or through an electron transfer
̈
mechanism to form a ZCl+C60 CT intermediate. The energy
−
of this CT state is estimated to be 2.0 eV, so it would be
expected to transfer energy to C60, promoting it to its singlet
excited state.31,34
increases the oxidation potential from +0.71 (ZH) to +1.22 V
(ZCl). In addition, the reduction potential of ZCl is 240 mV
more negative than that of C60. Thus, the oxidation and
reduction potentials of ZCl satisfy the sensitizer requirements
(2) and (3), respectively.
The optical density of neat ZCl thin films is nearly 7-fold
greater than that of C60 in the range between 470 and 570 nm
(Figure 5b). Interestingly, both fluorescence and phosphor-
escence were observed in a neat film at room temperature
(Figure 5b). Despite a 60 nm red shift of the phosphorescence
peak of the thin film compared to that in 2-MeTHF solution at
77K, the triplet energy of ZCl is still 0.15 eV higher than that of
C60. Thus, the energy levels of ZCl in the neat films also fulfill
the requirement (1) given above for the sensitizer. Moreover,
significant overlap exists between ZCl emission and C60
absorption spectra, assisting singlet energy transfer from the
Another sensitization pathway in the C60:ZCl blend is triplet
energy transfer from ZCl to C60 (pathway B in Figure 2b).
Luminescent spectra of ZCl in both solution and neat solid
clearly show competitive ISC to the triplet state. However, the
rate of ISC is at best 3.7 × 108 s−1, which is orders of magnitude
smaller than the quenching rate constant by C60 determined
from Stern−Volmer quenching experiments. Regardless, any
triplets formed on ZCl can be subsequently transferred to the
lower triplet state of C60. Thus, in all possible pathways of
either singlet, triplet, or electron transfers from the ZCl to the
C60, the triplet energy of C60 is the lowest state, assuring that
the sensitized energy is cascaded to the C60 host according to
the design presented in Figure 2b.
Morphology of C60:ZCl Films. Since the blended C60:ZCl
films will be used in OPV devices, it is important to understand
any effect that the presence of ZCl has on molecular packing of
C60 in the solid state or phase separation of the mixture. Atomic
force microscopy (AFM) and grazing incident X-ray diffraction
(GIXD) measurements were carried out on thin films of C60
(50 nm), ZCl (50 nm), and mixed C60:ZCl (1:1, 100 nm), and
the results are presented in Figure 7. GIXDs of neat ZCl films
and mixed C60:ZCl films at different concentrations are
presented in SI. The neat C60 film shows six diffraction peaks
at 0.77, 1.27, 1.48, 2.01, 2.19, and 2.32 Å−1, which are indexed
to the fcc crystal phase as the 111, 220, 311, 420, 422, and 333
peaks, respectively,48 whereas the neat films of ZCl are
amorphous. Upon addition of ZCl, the C60 phase becomes
substantially more disordered as indicated by the dramatic
broadening of the diffraction rings, reduced peak intensity, and
the inability to resolve the 220 and 311 peaks. Thus, the GIXD
measurements show that the mixed films become nanocrystal-
line or amorphous upon addition of the ZCl sensitizer.
sensitizer to the C60 host by the Forster mechanism.
̈
Photosensitization Study. The photosensitization effi-
ciency was quantified by Stern−Volmer quenching measure-
ments of ZCl:C60 mixtures in toluene solution (see SI). ZCl
emission is efficiently quenched by C60, with a rate constant for
luminescent quenching at the diffusion controlled limit. To
examine whether energy transfer processes from ZCl to C60 will
lead to sensitization in the solid state, absorption, and
photoluminescence (PL) spectra were collected from neat
C60 and mixed C60:ZCl films with varied blending ratios (Figure
6). Note that the amount of C60 in these films is kept constant.
In all blended samples, the fluorescence of ZCl was completely
quenched, and emission bands with spectral profiles identical to
neat C60 were observed between 700−900 nm. The
luminescence from the thin films is assigned to the excited
singlet state of C60 since the band shape and peak maximum
11923
dx.doi.org/10.1021/ja4043356 | J. Am. Chem. Soc. 2013, 135, 11920−11928

Journal of the American Chemical Society
Article
Figure 7. GIXD data of (a) neat C60 and (b) mixed C60:ZCl 1:1 films
on Si substrates. AFM images (5 × 5 μm) of (c) neat C60, (d) mixed
C60:ZCl 1:1 by volume and (e) neat ZCl films on Si substrates.
Analysis by AFM techniques also supports a decrease in
molecular order in the blended ZCl:C60 films. AFM images of
neat films of C60 were found to be rough (rms = 1.85 nm),
whereas neat films of ZCl were smooth (rms = 0.59 nm).
Interestingly, the AFM image of the 1:1 mixed film shows no
sign of phase separation and is quite smooth (rms = 0.39 nm).
The phase sensitive AFM image showed no features, consistent
with homogeneously mixed film. Thus, the AFM and GIXD
data indicate that ZCl guest molecules are homogeneously
dispersed in the C60 host, which is likely due to similar the sizes
and shapes of both compounds. Homogeneous dispersion of
ZCl into C60 is also beneficial for efficient photosensitization as
the ZCl molecules will be located in close proximity to the C60
molecules.
Application in OPV Devices. OPVs were fabricated with a
common multilayer structure of ITO/MoO3/donor/acceptor/
buffer/Al. Since the purpose of this work is to explore the
sensitization process in the visible spectrum, the UV absorbing
N,N′-di-[(1-naphthyl)-N,N′-diphenyl]-1,1′-biphenyl)-4,4′-dia-
mine (NPD) was chosen as the donor material. NPD shows
negligible absorption at wavelengths longer than 400 nm, which
allows the photoresponse from the ZCl sensitizer and C60 to be
unambiguously assigned. MoO3 is used to improve hole
extraction efficiency from the donor layer. The acceptor layer
can be neat C60 or mixed C60:ZCl with various architectures,
and the buffer layer is bathocuproine (BCP). In the reference
device, the optimal thickness of the neat C60 acceptor layer is
40 nm, increasing C60 thickness results in the decrease in
photocurrent and subsequently lower device efficiency (see
Table S1). Thus, the thickness of C60 used in reference devices
is kept constant at 40 nm.
There are several ways to construct the sensitized devices;
one straightforward method is to deposit the mixed
C60:sensitizer film with varied sensitizer concentrations directly
on the donor layer, as was done in previous works.22,29
Unfortunately, this simple architecture creates ambiguity as to
whether any photoresponse from the sensitizer originates from
photosensitization of C60 or from direct charge separation
between donor and sensitizer at the D/A interface. Thus,
sensitized devices with a structure ITO/MoO3/NPD/C60/
C60:ZCl/BCP/Al were constructed to prevent any direct
contact of ZCl with the donor layer. Current−voltage (J−V)
and external quantum efficiency (EQE) curves of OPV devices
utilizing a C60:ZCl layer with 15% (D2) and 50% (D3) of ZCl
by volume on top of a thin C60 layer (5 nm) are presented in
Figure 8. The device characteristics are summarized in Table 2.
The values for the open circuit voltage (VOC) and fill factor
(FF) in both sensitized devices remain unchanged from those
for the reference device D1. Seeing as all three devices have the
Figure 8. Structure and characteristics of OPV devices using an NPD
donor layer with and without ZCl. (a) Plot of EQE, inset is the device
structure, and (b) J−V curves of devices under 1 sun AM1.5G
illumination.
same NPD/C60 interface, the thermodynamics and kinetics of
the charge separation/recombination processes at this DA
boundary are identical, and consequently, the VOC is unaffected.
The unaltered FF value in devices D2−3 indicates that the
charge collection efficiency is well maintained despite the lower
crystallinity of the mixed C60:ZCl film. The high conductivity of
C60, along with the similar size and shape of ZCl and C60,
allows a homogeneous C60:ZCl layer to form with sufficient
percolation pathways in C60 to enable good electron
conduction. In a related study, we have investigated the effect
of mixing an inert material, BCP, into C60 and found that the
conductivity of C60 was not affected even in films that had up to
50% BCP by volume, i.e., the same ZCl loading used here.49
The photocurrents of the devices drop as the ZCl
concentration increases, D1 > D2 > D3. The reduction in
photocurrent can be understood from EQE measurements
(Figure 8a). EQE clearly indicates the contribution of the
sensitizer ZCl (peak maximum at 550 nm) to the photocurrents
in devices D2 and D3, with the ZCl photoresponse
proportional to its concentration. As there is no direct contact
between ZCl and the donor layer, the increased photoresponse
between 500 and 600 nm in D2 and D3 must come from
photoenergy absorbed by ZCl and subsequently transferred to
C60. In contrast, the photoresponse from C60 decreases as ZCl
concentration increases in devices D2 and D3. The loss of C60
response is due to the decrease in absorption of C60 as blending
concentration increases.49 This decrease is slightly larger than
the increase due to ZCl sensitization, resulting in a net decrease
in the photocurrent of D2 and D3.
It is desirable to maintain the photoresponse from C60 while
maximizing the photoresponse from the sensitizer ZCl, thus
11924
dx.doi.org/10.1021/ja4043356 | J. Am. Chem. Soc. 2013, 135, 11920−11928

Journal of the American Chemical Society
Article
a
Table 2. Characteristics of OPV Devices under Spectral Mismatch Factor Corrected 1 sun AM1.5G Illumination
device
structure of the acceptor layers
JSC ( 0.05 mA/cm2)
VOC ( 0.01 V)
FF ( 0.01)
η ( 0.04%)
Donor = NPD, 11 nm
D1
D2
D3
D4
D5
D6
D7
D8
C60, 40 nm
3.04
0.88
0.89
0.88
0.89
0.89
0.88
0.88
0.90
0.44
0.46
0.46
0.43
0.44
0.44
0.44
0.44
1.19
1.17
1.06
1.48
1.35
1.25
1.56
1.48
C60, 5 nm/C60:ZCl (5.6:1), 40 nm
C60, 5 nm/C60:ZCl (1:1), 40 nm
C60, 15 nm/C60:ZCl (1:1), 30 nm
C60, 25 nm/C60:ZCl (1:1), 20 nm
C60, 35 nm/C60:ZCl (1:1), 10 nm
C60, 15 nm/C60:ZCl (1:1), 50 nm
C60, 25 nm/C60:ZCl (1:1), 50 nm
2.92
2.51
3.87
3.53
3.23
4.03
3.90
Donor = SQ, 11 nm
D9
C60, 40 nm
5.88
7.07
0.76
0.78
0.53
0.51
2.39
2.78
D10
C60, 15 nm/C60:ZCl (1:1), 50 nm
a
Common device structure: ITO/MoO3 (10 nm)/D/A/BCP (7 nm)/Al.
sensitized devices with a high ZCl concentration (50%) were
chosen for further optimization. In order to determine the
thickness of neat C60 needed to recover the its photoresponse, a
series of devices consisting of a constant thickness of the
acceptor layer (45 nm) and varied thicknesses of the neat C60
layer at D/A interface, ITO/MoO3/NPD/C60 (x nm)/C60:ZCl
(1:1, 45 − x nm)/BCP/Al, has been fabricated. EQE
measurements of these devices are presented in Figure 9, and
Figure 9. EQE measurements of sensitized devices with varied C60
layer thicknesses at the D/A interface.
the device characteristics are summarized in Table 2. As the
thickness of the neat layer of C60 increases from 5 nm (D3) to
15 nm (D4), the photoresponse of C60 is fully recovered, while
a significant contribution from ZCl is still observed. A further
increase of C60 thickness to 25 nm (D5) markedly decreases
the ZCl photoresponse along with the same response
photoresponse from C60. When 35 nm of C60 is used (D6),
no photoresponse from ZCl is observed despite much higher
optical density of ZCl compared to C60 (Figure 5b), likely due
to limited diffusion length of the sensitized excitons.
Sensitized devices with a 15 nm of C60 at the D/A interface,
the minimal thickness necessary for C60 photoresponse
recovery, were further optimized by varying the thicknesses
of the mixed C60:ZCl (1:1) layer. The results are presented in
Figure 10a. Similar to other sensitized devices, the FF and VOC
remains largely unchanged, and only photocurrent JSC varies as
the thickness of mixed C60:ZCl layer changes. The JSC value
maximizes at a mixed layer thickness of 50 nm. Compared to
the control device D1 using only neat C60 as acceptor layer, D7
exhibits a markedly higher photocurrent (ΔJSC = 1.0 mA/cm2).
Thus, the improved JSC, in combination with unchanged VOC
and FF, results in an increase of the power conversion efficiency
Figure 10. (a) Characteristics of sensitized devices with a neat layer of
C60 (x = 15 nm) and varied thicknesses (y nm) of the blended C60:ZCl
layers; y = 0 represents the reference device with 40 nm of neat C60 as
the acceptor layer; and (b) plot of EQE of D1 and optimized device
using sensitizer (D7), inset is the device structure.
from 1.16% (D1) to 1.56% (D7). Further increasing the
thickness of neat C60 at the D/A interface from 15 nm (D7) to
25 nm (D8) results in slight decrease in photocurrent and
subsequent efficiency (Table 2). Thus, the acceptor layer used
in D7 is the optimal structure for the sensitized devices.
As the absorption of donor materials is extended to the red-
NIR part of spectrum, the green-to-orange part is left
unabsorbed; 2,4-bis[4-(N,N-diisobutylamino)-2,6-dihydroxy-
phenyl] squaraine (SQ)50 and zinc phthalocyanine (ZnPc)
donor-based OPVs illustrate this problem (Figure 11 and SI).
There are minima in the EQE curves at roughly 550 nm in both
SQ/C60 and ZnPc/C60 devices where neither donor (ZnPc or
SQ) nor C60 absorb effectively. The absorption of ZCl matches
well with the EQE minima, thus ZCl can potentially be used to
fill up the absorption gap in SQ and ZnPc OPV devices.
11925
dx.doi.org/10.1021/ja4043356 | J. Am. Chem. Soc. 2013, 135, 11920−11928

Journal of the American Chemical Society
Article
energy transfer from the photoexcited state of ZCl to C60 can
increase the photoresponse of the acceptor layer up to 33%
without changing other device characteristics. The sensitization
approach presented here could potentially be used for multiple
sensitizers to extend the absorption range of the electron-
acceptor layer. While we have focused on the C60 acceptor layer
in the present study, a similar approach could be used with
other fullerenes as well as sensitized donor layers to broaden
the photoresponse of OPVs.
EXPERIMENTAL SECTION
■
Instrumentation. GIXD measurements were performed at the
Stanford Synchrotron Radiation Lightsource (SSRL) on beamline 11-3
with a photon wavelength of 0.0974 nm. The diffraction intensity was
detected on a 2D image plate (MAR 345) with a pixel size of 150 μm
(2300 × 2300 pixels). GIXD images were analyzed using the software
package WxDiff, provided by Dr. Stefan Mannsfeld. The samples were
20 mm long in the direction of the beam path, and the detector was
located 398.8 mm from the sample center. The beam incidence angle
was 0.12°, and the beam size was 50 × 150 μm. The data presented
here were corrected for the grazing incidence geometry. NMR
measurements were performed on a Varian Mercury 500 MHz
spectrometer. AFM was performed using a Digital Instruments
Nanoscope Dimension 3100 atomic force microscope. UV−vis spectra
were recorded on a Hewlett-Packard 4853 diode array spectrometer.
Fluorescence lifetime measurements were performed by a time-
correlated single-photon counting method using an IBH fluorocube
lifetime instrument by equipped with a 405 nm LED excitation source.
Quantum efficiency measurements were carried out using a
Hamamatsu C9920 system equipped with a xenon lamp, calibrated
integrating sphere and model C10027 photonic multichannel analyzer.
CV and DPV measurements were performed using an EG&G
Potentiostat/Galvanostat model 283. Dry dichloromethane from
VWR was used as solvent under a N2 atmosphere with 0.1 M
tetra(n-butyl)ammonium hexafluorophosphate (Aldrich) as the
supporting electrolyte. A glassy carbon working electrode and a
platinum counter electrode were used, with a silver wire as the
pseudoreference electrodes. The oxidation potential was measured
relative to a ferrocenium/ferrocene (Fc+/Fc) redox couple as an
internal standard.
PL of ZCl in Solution and Thin Film. Steady-state emission
measurements of ZCl in the thin film and solutions at room
temperature and 77 K were performed using a Photon Technology
International QuantaMaster Model C-60SE spectrofluorimeter.
PL measurements of C60 and C60:ZCl Films. No emission was
detected from neat C60 and mixed C60:ZCl films using the Photon
Technology International QuantaMaster Model C-60SE spectro-
fluorimeter. The samples under nitrogen atmosphere were excited
by 514 nm emission lines from a Spectra-Physics Stability 2017 argon
ion laser. PL was measured by imaging the excited area of the sample
onto the entrance slit of an Acton Research Spectrapro 500i
spectrograph coupled to a cooled Hamamatsu CCD detector. The
acquired data were corrected for the responses of the CCD array and
the diffraction grating.
Figure 11. Structure and characteristics of OPV devices using a SQ
donor layer with and without a sensitizer. (a) J−V curves of devices
under 1 sun AM1.5G illumination, inset is the SQ structure (b) plot of
EQE.
Sensitized devices with the optimal acceptor structure have
been fabricated with ZnPc and SQ. As indicated by EQE
measurements (Figure 11), the photoresponses from SQ and
C60 remain unchanged, while the ZCl response nicely fills up
the dip in the EQE curve, leading to a photocurrent
improvement of 1.0−1.2 mA/cm2 without any other change
in the device characteristics. This increase of photocurrent is
comparable to what was observed in the NPD/C60 device
(Figure 10). Similar improvement is observed when ZCl is used
in ZnPc/C60 OPV devices (see SI). Thus, the obtained results
demonstrate that the ZCl sensitizer can be used with a variety
of donor materials.
CONCLUSION
■
In summary, we have shown a set of energetic requirements
and molecular design for a C60 sensitizer. An ideal sensitizer
should have higher singlet and triplet energies than C60 and
sufficiently high oxidation energy to give a Sen+C60− CT state a
higher energy than the C60 triplet exciton. A chlorinated zinc
dipyrrin ZCl, which satisfies all requirements for the sensitizer,
was synthesized and fully characterized. Stern−Volmer
quenching experiment and PL measurements of mixed
C60:ZCl films show that ZCl efficiently transfers absorbed
photoenergy to C60.
X-ray Data and Collection for ZCl. A metallic intense pink-green
rhombic-like specimen of C36H22.25Cl11.75N4Zn, approximate dimen-
sions 0.26 × 0.29 × 0.38 mm, was used for the X-ray crystallographic
analysis. The X-ray intensity data were measured on a Bruker APEX II
CCD system equipped with a TRIUMPH curved-crystal mono-
chromator and a MoKα fine-focus tube (λ = 0.71073 Å).
A total of 2520 frames were collected. The total exposure time was
3.50 h. The frames were integrated with the Bruker SAINT software
package using a SAINT V7.68A algorithm. The integration of the data
using a monoclinic unit cell yielded a total of 96 239 reflections to a
maximum θ angle of 30.56° (0.70 Å resolution), of which 12 079 were
The OPV devices employing the C60:ZCl host−guest
acceptor layer have been constructed and optimized to
demonstrate the sensitization approach. The optimal acceptor
layer consisting of a neat C60 (15 nm) and a mixed C60:ZCl
(1:1, 50 nm) films has been used to maintain the beneficial
properties of C60 while taking advantages of ZCl absorption. In
the sensitized devices with various donor materials, efficient
independent (average redundancy 7.967, completeness = 98.3%, Rint
=
3.01%, Rsig = 1.81%) and 10 440 (86.43%) were >2σ (F2). The final
cell constants of a = 8.4166(4) Å, b = 23.9579(10) Å, c = 20.3078(9)
11926
dx.doi.org/10.1021/ja4043356 | J. Am. Chem. Soc. 2013, 135, 11920−11928

Journal of the American Chemical Society
Article
Å, β = 101.4330(10)°, volume = 4013.7(3) Å3, are based upon the
refinement of the XYZ-centroids of 9012 reflections above 20 σ(I)
with 4.431° < 2θ < 60.92°. Data were corrected for absorption effects
using the multiscan method (SADABS). The ratio of minimum to
maximum apparent transmission was 0.901. The calculated minimum
and maximum transmission coefficients (based on crystal size) are
0.6150 and 0.7086.
The structure was solved and refined using the Bruker SHELXTL
software package, using the space group P121/c1, with Z = 4 for the
formula unit, C36H22.25Cl11.75N4Zn. The final anisotropic full-matrix
least-squares refinement on F2 with 485 variables converged at R1 =
2.78%, for the observed data and wR2 = 6.65% for all data. The
goodness-of-fit was 1.031. The largest peak in the final difference
electron density synthesis was 0.945 e−/Å3, and the largest hole was
−0.619 e−/Å3 with an rmsd of 0.060 e−/Å3. On the basis of the final
model, the calculated density was 1.643 g/cm3 and F(000), 1984 e−.
Computational Methods. All calculations were performed using
Titan software package (Wave function Inc.). The gas-phase geometry
optimization was calculated using B3LYP functional with the
LACVP** basis set as implemented in Titan. The energy levels and
orbital diagrams of the HOMO and LUMO were obtained from the
optimized geometry of the singlet state. The energy of the triplet state
was calculated from the optimized triplet geometry.
Synthesis of ZCl. Synthesis of 5-mesityldipyrromethane (DPM)
was performed following a published procedure.51 A solution of DPM
(0.62 g, 2.34 mmol) in 60 mL freshly distilled tetrahydrofuran (THF)
was prepared; the solution was cooled down using dry ice/acetone
bath and was bubbled with nitrogen for 5 min. A solution of N-
chlorosuccinimide (3.1 g, 23.4 mmol) in 70 mL THF was prepared
and slowly added to the DPM solution under nitrogen. The reaction
mixture was stirred and allowed to warm up to room temperature
during 2 h. The reaction mixture turned a dark-red color. After stirring
for additional 2 h at room temperature, the reaction was stopped. The
solvent was removed under reduced pressure, and the crude products
were dissolved in 300 mL dichloromethane. The crude products were
washed with NaHCO3 solution twice and brine twice. The dark-red
solution of the products in dichloromethane was used without further
purification.
A solution of Zn(OAc)2·2H2O (2.5 g) in 30 mL CH3OH was
prepared and added to the solution of the dark-red products. The
reaction turned into dark-green color and emitted a green-yellow color
under UV lamp illumination. The reaction mixture was stirred
overnight. The solvents were removed under reduced pressure. The
solid reaction mixture was dissolved in dichloromethane, and inorganic
salts were filtered off. The solution was washed with Na2CO3 solution
twice and brine once. The solvent was removed. The crude product
was passed through short neutral Al2O3 flash column using
dichloromethane/hexanes (1/9) as eluent, and the orange-red fraction
was collected. After removing the solvents, a dark-red product was
collected. The product was dissolved in DCM and recrystallized by
layering MeOH on top. Red-green crystals (ZCl) were collected (0.2
g, 17% total yield). ZCl was further purified by gradient sublimation
under vacuum (10−5 Torr) at 280, 200, and 140 °C gradient
temperature zones. The sublimed crystals were qualified for X-ray
single crystal structure determination.
were cleaned with soap and boiled in tetrachloroethylene, acetone, and
propanol (5 min each). ITO substrates were exposed to ozone
atmosphere (UVOCS T10 × 10/OES) for 10 min immediately before
loading into the high-vacuum chamber. Deposition rates for layers of
neat materials: MoOx (0.05 nm/s), NPD (0.15 nm/s), SQ (0.1 nm/s),
ZnPc (0.2 nm/s), C60 (0.2 nm/s), BCP (0.15 nm/s), and Al (0.2 nm/
s). Deposition rates for mixed films (% ZCl content by volume):
C60:ZCl (15% ZCl) − codeposition C60 (0.2 nm/s):ZCl (0.035 nm/
s); C60:ZCl (35% ZCl) − codeposition C60 (0.1 nm/s):ZCl (0.05 nm/
s); and C60:ZCl (50% ZCl) − codeposition C60 (0.05 nm/s):ZCl (0.05
nm/s). After organic depositions, masks with 2 mm stripe width were
placed on substrates under N2, and 100 nm of Al electrode was
deposited. Area of devices is 4 mm2. Current−density dependence on
applied test voltage J(V) measurements were performed in air at 25 °C
using a Keithley 2420 Sourcemeter (sensitivity = 100 pA) in the dark
and under ASTM G173-03 spectral mismatch corrected 1000 W/m2
white light illumination from an AM1.5G filtered 300 W xenon arc
lamp (Newport Oriel). Routine spectral mismatch correction was
performed using a silicon photodiode (Hamamatsu S1787−04,8RA
filter) calibrated at the National Renewable Energy Laboratory
(NREL). Chopped and filtered monochromatic light (250 Hz, 10
nm fwhm) from a Cornerstone 260 1/4 M double grating
monochromator (Newport 74125) was used in conjunction with an
EG&G 7220 lock-in amplifier to perform all spectral responsivity and
spectral mismatch correction measurements.
ASSOCIATED CONTENT
■
S
* Supporting Information
A detailed attribution for contributions from funding agencies;
electrochemical measurements of ZH; 1H, 13C and NOESY 1H
NMR of ZCl; emission of ZCl in 2-MeTHF at 77 K; calculated
energy levels and MOs of ZCl11 and ZCl12; details of the
Stern−Volmer quenching experiment; GIXD of mixed C60:ZCl
films at various concentrations; characteristics of sensitized
devices with various C60:ZCl (1:1) layer thicknesses; character-
istics of devices using ZnPc and ZH donor layers; crystal data of
ZCl. This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
met@usc.edu
Notes
The authors declare the following competing financial
interest(s): One of the authors (M.E.T.) has a financial interest
in the Global Photonic Energy Corporation.
ACKNOWLEDGMENTS
■
Financial support from Global Photonic Energy Corporation
(GPEC), Department of Energy (DOE), Center for Energy
Nanoscience (CEN) at USC, National Science Foundation
(NSF) and King Abdullah University of Science and
Technology (KAUST), through the Center for Molecular
Photovoltaics (CAMP) is gratefully acknowledged. Details of
how each funding agency supported this work are given in the
Supporting Information. We thank Dr. Ralf Haiges for the help
in refining single crystal structure, Dr. Travis Williams for the
The product ZCl is a mixture of two compounds C36H22Cl12N4Zn
and C36H23Cl11N4Zn with a molar ratio of 3:1 (general formula:
1
C144H89Cl47N16Zn4). H NMR (400 MHz, CDCl3): δ ppm 6.95 (s,
4H, mesityl), 6.49 (s, 0.23 H, pyrrole), 2.40−2.35 (m, 6H, mesityl),
2.09−2.04 (m, 12H, mesityl). 13C NMR (500 MHz, CDCl3): δ (ppm)
147.00, 143.29, 139.29, 136.04, 135.55, 132.46, 132.33, 130.30, 130.26,
129.74, 129.13, 128.43, 119.36, 118.70, 21.34, 21.23, 19.49, 19.42.
HRMS: C36H22Cl12N4Zn: calcd, 999.7304; found, 999.7333.
C36H22Cl11N4Zn: calcd, 965.7694; found, 965.7705. Elemental
analysis: theoretical C, 43.55%; H, 2.26%; N, 5.64%. Found: C,
43.54%; H, 1.96%; N, 5.50%.
1
help in NOESY H NMR experiment, and Drs. Sarah Conron
and Zhiwei Liu for helpful discussions.
REFERENCES
Device Fabrication and Characterization. Glass substrates
■
coated with patterned ITO (width of patterned stripes is 2 mm,
(1) Clarke, T. M.; Durrant, J. R. Chem. Rev. 2010, 110, 6736−6767.
(2) Hains, A. W.; Liang, Z.; Woodhouse, M. A.; Gregg, B. A. Chem.
Rev. 2010, 110, 6689−6735.
thickness = 150
10 nm; sheet resistance = 20
5 Ω cm−2;
transmission 84% at 550 nm; courtesy of Thin Film Devices, Inc.)
11927
dx.doi.org/10.1021/ja4043356 | J. Am. Chem. Soc. 2013, 135, 11920−11928

Journal of the American Chemical Society
Article
(3) Schlenker, C. W.; Thompson, M. E. Chem. Commun. 2011, 47,
3702−3716.
(36) Kubo, Y.; Watanabe, K.; Nishiyabu, R.; Hata, R.; Murakami, A.;
Shoda, T.; Ota, H. Org. Lett. 2011, 13, 4574−4577.
(37) Hanson, K.; Tamayo, A.; Diev, V. V.; Whited, M. T.; Djurovich,
P. I.; Thompson, M. E. Inorg. Chem. 2010, 49, 6077−6084.
(38) Azenha, E. G.; Serra, A. C.; Pineiro, M.; Pereira, M. M.; de Melo,
J. S.; Arnaut, L. G.; Formosinho, S. J.; Gonsalves, A. Chem. Phys. 2002,
280, 177−190.
(4) Anthopoulos, T. D.; Singh, B.; Marjanovic, N.; Sariciftci, N. S.;
Ramil, A. M.; Sitter, H.; Colle, M.; de Leeuw, D. M. Appl. Phys. Lett.
2006, 89, 213504.
(5) Zhang, X. H.; Kippelen, B. Appl. Phys. Lett. 2008, 93, 133305.
(6) Zhang, X.-H.; Kippelen, B. J. Appl. Phys. 2008, 104, 104504.
(7) Lin, Y.; Li, Y.; Zhan, X. Chem. Soc. Rev. 2012, 41, 4245−4272.
(8) Mishra, A.; Baeuerle, P. Angew. Chem., Int. Ed. 2012, 51, 2020−
2067.
(9) You, J.; Dou, L.; Yoshimura, K.; Kato, T.; Ohya, K.; Moriarty, T.;
Emery, K.; Chen, C.-C.; Gao, J.; Li, G.; Yang, Y. Nat. Commun. 2013,
4, 1446.
(39) Pineiro, M.; Carvalho, A. L.; Pereira, M. M.; Gonsalves, A.;
Arnaut, L. G.; Formosinho, S. J. Chem.−Eur. J. 1998, 4, 2299−2307.
(40) Sazanovich, I. V.; Kirmaier, C.; Hindin, E.; Yu, L. H.; Bocian, D.
F.; Lindsey, J. S.; Holten, D. J. Am. Chem. Soc. 2004, 126, 2664−2665.
(41) Ulrich, G.; Ziessel, R.; Harriman, A. Angew. Chem., Int. Ed. 2008,
47, 1184−1201.
(42) Yu, L. H.; Muthukumaran, K.; Sazanovich, I. V.; Kirmaier, C.;
Hindin, E.; Diers, J. R.; Boyle, P. D.; Bocian, D. F.; Holten, D.;
Lindsey, J. S. Inorg. Chem. 2003, 42, 6629−6647.
(10) Kazaoui, S.; Minami, N.; Tanabe, Y.; Byrne, H. J.; Eilmes, A.;
Petelenz, P. Phys. Rev. B 1998, 58, 7689−7700.
(11) Schlenker, C. W.; Thompson, M. E. Chem. Commun. 2011, 47,
3702−3716.
(43) Rio, Y.; Sanchez-Garcia, D.; Seitz, W.; Torres, T.; Sessler, J. L.;
Guldi, D. M. Chem.−Eur. J. 2009, 15, 3956−3959.
(12) Peumans, P.; Uchida, S.; Forrest, S. R. Nature 2003, 425, 158−
162.
(44) Dorset, D. L.; McCourt, M. P. Acta Crystallogr., Sect. A 1994, 50,
344−351.
(13) Peumans, P.; Yakimov, A.; Forrest, S. R. J. Appl. Phys. 2003, 93,
(45) Vandenheuvel, D. J.; Chan, I. Y.; Groenen, E. J. J.; Schmidt, J.;
Meijer, G. Chem. Phys. Lett. 1994, 231, 111−118.
3693−3723.
(14) Lassiter, B. E.; Wei, G.; Wang, S.; Zimmerman, J. D.; Diev, V.
V.; Thompson, M. E.; Forrest, S. R. Appl. Phys. Lett. 2011, 98,
243307−3.
(46) Capozzi, V.; Celentano, G.; Perna, G.; Lorusso, G. F.; Minafra,
A. J. Lumin. 2000, 86, 129−135.
(47) Ichida, M.; Sakai, M.; Yajima, T.; Nakamura, A. Prog. Cryst.
Growth Charact. Mater. 1996, 33, 125−128.
(15) Wei, G.; Lunt, R. R.; Sun, K.; Wang, S.; Thompson, M. E.;
Forrest, S. R. Nano Lett. 2010, 10, 3555−3559.
(48) Li, J.; Komiya, S.; Tamura, T.; Nagasaki, C.; Kihara, J.; Kishio,
K.; Kitazawa, K. Physica C 1992, 195, 205−208.
(16) Wei, G.; Xiao, X.; Wang, S.; Sun, K.; Bergemann, K. J.;
Thompson, M. E.; Forrest, S. R. ACS Nano 2012, 6, 972−978.
(17) Zimmerman, J. D.; Xiao, X.; Renshaw, C. K.; Wang, S.; Diev, V.
V.; Thompson, M. E.; Forrest, S. R. Nano Lett. 2012, 12, 4366−4371.
(18) Wei, G.; Wang, S.; Renshaw, K.; Thompson, M. E.; Forrest, S.
R. ACS Nano 2010, 4, 1927−1934.
(49) Bartynski, A. N.; Trinh, C.; Panda, A.; Bergemann, K.; Lassiter,
B. E.; Zimmerman, J. D.; Forrest, S. R.; Thompson, M. E. Nano Lett.
2013, 13, 3315−3320.
(50) Wang, S.; Mayo, E. I.; Perez, M. D.; Griffe, L.; Wei, G.;
Djurovich, P. I.; Forrest, S. R.; Thompson, M. E. Appl. Phys. Lett. 2009,
94, 233304.
(51) Laha, J. K.; Dhanalekshmi, S.; Taniguchi, M.; Ambroise, A.;
Lindsey, J. S. Org. Process Res. Dev. 2003, 7, 799−812.
(19) Wei, G.; Xiao, X.; Wang, S.; Zimmerman, J. D.; Sun, K.; Diev, V.
V.; Thompson, M. E.; Forrest, S. R. Nano Lett. 2011, 11, 4261−4264.
(20) Anthony, J. E. Chem. Mater. 2011, 23, 583−590.
(21) Sonar, P.; Lim, J. P. F.; Chan, K. L. Energy Environ. Sci. 2011, 4,
1558−1574.
(22) Hesse, H. C.; Weickert, J.; Hundschell, C.; Feng, X.; Muellen,
K.; Nickel, B.; Mozer, A. J.; Schmidt-Mende, L. Adv. Energy Mater.
2011, 1, 861−869.
(23) Ford, W. E.; Kamat, P. V. J. Phys. Chem. 1987, 91, 6373−6380.
(24) Kircher, T.; Lohmannsroben, H. G. Phys. Chem. Chem. Phys.
1999, 1, 3987−3992.
(25) Baffreau, J.; Leroy-Lhez, S.; Van Anh, N.; Williams, R. M.;
Hudhomme, P. Chem.−Eur. J. 2008, 14, 4974−4992.
(26) Baffreau, J.; Leroy-Lhez, S.; Hudhomme, P.; Groeneveld, M. M.;
van Stokkum, I. H. M.; Williams, R. M. J. Phys. Chem. A 2006, 110,
13123−13125.
(27) Liu, Y.; Zhao, J. Chem. Commun. 2012, 48, 3751−3753.
(28) Schueppel, R.; Uhrich, C.; Pfeiffer, M.; Leo, K.; Brier, E.;
Reinold, E.; Baeuerle, P. ChemPhysChem 2007, 8, 1497−1503.
(29) Uhrich, C.; Schueppel, R.; Petrich, A.; Pfeiffer, M.; Leo, K.;
Brier, E.; Kilickiran, P.; Baeuerle, P. Adv. Funct. Mater. 2007, 17,
2991−2999.
(30) Wu, W.; Zhao, J.; Sun, J.; Guo, S. J. Org. Chem. 2012, 77, 5305−
5312.
(31) Ziessel, R.; Allen, B. D.; Rewinska, D. B.; Harriman, A. Chem.−
Eur. J. 2009, 15, 7382−7393.
(32) Vauthey, E. ChemPhysChem 2012, 13, 2001−2011.
(33) Yang, Y. F.; Arias, F.; Echegoyen, L.; Chibante, L. P. F.;
Flanagan, S.; Robertson, A.; Wilson, L. J. J. Am. Chem. Soc. 1995, 117,
7801−7804.
(34) Veldman, D.; Meskers, S. C. J.; Janssen, R. A. J. Adv. Funct.
Mater. 2009, 19, 1939−1948.
(35) Dastoor, P. C.; McNeill, C. R.; Frohne, H.; Foster, C. J.; Dean,
B.; Fell, C. J.; Belcher, W. J.; Canipbell, W. M.; Officer, D. L.; Blake, I.
M.; Thordarson, P.; Crossley, M. J.; Hush, N. S.; Reimers, J. R. J. Phys.
Chem. C 2007, 111, 15415−15426.
11928
dx.doi.org/10.1021/ja4043356 | J. Am. Chem. Soc. 2013, 135, 11920−11928