Bulk heterojunction solar cells based on a low band gap soluble bisazopyrrole and the corresponding BF2-azopyrrole complex

Article abstract of DOI:10.1039/c0jm00122h

The diazonium salt derived from 4-(hexyloxy)benzamine reacted with half an equivalent of pyrrole to afford 2,5-bis(4-hexyloxyphenylazo)-1H-pyrrole (A). The reaction of the latter with BF₃·Et₂O gave the corresponding BF₂-azopyrrole complex (B). The thin film absorption spectra of A and B were broad and extended into the near infrared region with optical band gaps of 1.54 and 1.49 eV, respectively. The photovoltaic properties have been investigated using blends of A or B with PCBM, and it was found that the device based on B:PCBM showed higher power conversion efficiency (PCE) in comparison to that for the device with A:PCBM blend. This has been attributed to the higher hole mobility and lower band gap of B relative to A. The effect of different thermal annealing on the photovoltaic response of the devices has been investigated and it was found that the contact annealed device displayed PCE of approximately 2.7% and 3.15% for A:PCBM and B:PCBM blends, respectively. The increase in the PCE for the contact annealed device has been interpreted in terms of more balanced charge transport.

Full text of DOI:10.1039/c0jm00122h

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Bulk heterojunction solar cells based on a low band gap soluble bisazopyrrole
and the corresponding BF2-azopyrrole complex
a
bc
J. A. Mikroyannidis, A. N. Kabanakis,a D. V. Tsagkournos,a P. Balrajubd and G. D. Sharma
*
*
Received 22nd January 2010, Accepted 9th April 2010
DOI: 10.1039/c0jm00122h
The diazonium salt derived from 4-(hexyloxy)benzamine reacted with half an equivalent of pyrrole to
afford 2,5-bis(4-hexyloxyphenylazo)-1H-pyrrole (A). The reaction of the latter with BF3$Et2O gave the
corresponding BF2-azopyrrole complex (B). The thin film absorption spectra of A and B were broad
and extended into the near infrared region with optical band gaps of 1.54 and 1.49 eV, respectively.
The photovoltaic properties have been investigated using blends of A or B with PCBM, and it was
found that the device based on B:PCBM showed higher power conversion efficiency (PCE) in
comparison to that for the device with A:PCBM blend. This has been attributed to the higher hole
mobility and lower band gap of B relative to A. The effect of different thermal annealing on the
photovoltaic response of the devices has been investigated and it was found that the contact annealed
device displayed PCE of approximately 2.7% and 3.15% for A:PCBM and B:PCBM blends,
respectively. The increase in the PCE for the contact annealed device has been interpreted in terms of
more balanced charge transport.
ability of the material, the charge-separation efficiency, and the
high and balanced carrier mobilities. On the other hand, Voc is
limited by the difference in the highest occupied molecular
orbital (HOMO) of the donor and the lowest unoccupied
molecular orbital (LUMO) of the acceptor, where a small Voc
represents a smaller driving force for the PV process. An effective
method to improve the Voc of polymer solar cells is to manipulate
the HOMO level of the donor and/or LUMO level of the
acceptor.16
Introduction
Polymer solar cells have evolved as a promising cost-effective
alternative to inorganic-based solar cells1,2 due to their potential
to be low-cost, light-weight and flexible. Since the discovery of
ultrafast photoinduced charge transfer from a conjugated poly-
mer to fullerene molecules, followed by the introduction of the
bulk heterojunction (BHJ) concept,3 intensive research with
potential materials has been carried out for future photovoltaic
(PV) technology.4–8 Two organic materials with distinct donor
and acceptor properties are required to form a heterojunction in
the bulk film, which is often achieved by solution processing. In
such a case, the BHJ not only provides abundant donor/acceptor
interfaces for charge separation, but also forms an inter-
penetrating network for charge transport.8,9 Highly efficient
polymer solar cells based on poly(3-hexylthiophene) (P3HT) and
[6,6]-phenyl C61 butyric acid methyl ester (PC61BM) have been
reported with PCEs of 4–5%6,10–12 So far, besides the P3HT
system, there are very few polymer solar cell systems reported
which exceed 5% in PCE.13 Recent developments concerning
processes for the production of flexible large area polymer and
generally organic solar cells have been reported.14 Finally, some
recent reviews on organic solar cells have been reported.15
The two most decisive parameters regarding polymer solar cell
efficiencies are the open circuit voltage (Voc) and the short circuit
current (Jsc). Jsc is mostly determined by the light absorption
The photoconductive properties of azo compounds were
reported as early as 1968 by Rau who observed photocurrents
from thin layers of 1-(phenylazo)naphthol.17 Several years later,
Champ and Shattuck reported the use of a bisazo pigment
derived from 3,30-dichlorobenzidine as
a photogenerating
pigment in xerographic devices.18 Then, azo pigments were
widely studied as charge generation materials because of their
technological advantages such as high sensitivity, wide spectral
range (450–800 nm) and excellent stability.19
Recently, much effort has been spent in the synthesis of novel
fluorine–boron complexes (FBC) and their application in PV
devices; it is found that the modification of the FBC can not only
adjust their optical characteristics but also improve their capa-
bilities of electron-accepting and transportation.20–22 Although
several FBCs were reported to be used in organic light emitting
diodes and field effect transistors, the exploration of FBCs as
charge transport materials in organic solar cells is still inade-
quate. Finally, a strategy has been developed for functionalizing
and solubilizing boron dipyrromethene (Bodipy) dyes at the
central boron atom and changing the color by increasing delo-
calization on the central core. This approach leads to the
formation of stable B–C and pyrrole–C linkages suitable for use
in TiO2-sensitized devices.23
aChemical Technology Laboratory, Department of Chemistry, University
of Patras, GR-26500 Patras, Greece. E-mail: mikroyan@chemistry.
upatras.gr; Fax: +30 2610 997118; Tel: +30 2610 997115
bPhysics Department, Molecular Electronic and Optoelectronic Device
Laboratory, JNV University, Jodhpur (Raj.), 342005, India. E-mail:
sharmagd_in@yahoo.com; Fax: +91-0291-2720856; Tel: +91-0291-
2720857
A series of symmetrical 2,5-bisazopyrroles and a BF2–azo-
pyrrole complex which showed near infrared absorption has been
synthesized.24 Very recently, we have synthesized an alternating
cJaipur Engineering College, Kukas, Jaipur (Raj.), India
dMolecular Electronics Laboratory, JNCASR, Bangalore, India
6464 | J. Mater. Chem., 2010, 20, 6464–6471
This journal is ª The Royal Society of Chemistry 2010

View Article Online
phenylenevinylene copolymer with 2,5-bisazopyrrole segments
along the backbone and the corresponding BF2–azopyrrole
complex. The latter has been used for BHJ solar cells.25
In addition, we have synthesized a series of low band gap dyes
containing the BF2–azopyrrole complex and carboxy or hydroxy
anchoring groups which has been used for dye-sensitized solar
cells.26 In continuation of this research line, herein we synthesized
a novel symmetrical bisazopyrrole (A) bearing hexyloxyphenyl
terminal moieties and the corresponding BF2–azopyrrole
complex (B). Their simple and convenient synthesis started from
inexpensive starting materials and gave products in relatively
high yields. The presence of the hexyloxy chains enhanced the
solubility of A and B. The latter showed a broader absorption
band and a lower optical band gap relative to A, a feature which
is attractive for PV applications.
(m, 12H, O(CH2)2(CH2)3CH3); 0.91 (t, J ¼ 6.8 Hz, 6H,
O(CH2)5CH3).
Anal. Calcd. for C28H37N5O2: C, 70.71; H, 7.84; N, 14.72.
Found: C, 70.24; H, 7.92; N, 14.58.
2,5-Bis(4-hexyloxyphenylazo)-1-boron difluoride (B). To
a solution of A (0.30 g) and triethylamine (2 mL) in dry chlo-
roform (20 mL) was slowly added boron trifluoride diethyl
etherate (1 mL) under N2. The resulting solution was refluxed for
2 h and then evaporated to dryness. The residue was triturated
with ether, filtered and dried to afford B as a dark purple product
(0.27 g, 83%).
FT-IR (KBr, cmꢁ1): 3198, 1068 (BF2-azopyrrole complex),
2930, 2859 (C–H stretching of hexyloxy groups); 1248, 1160
(ether bond); 1600, 1508, 1458 (aromatic).
We have also investigated the PV response of the A:PCBM and
B:PCBM blend sandwiched between PEDOT:PSS-coated ITO
and Al electrode. The effect of different annealing conditions on
the overall PCE of the device has been studied and it was found
that the contact annealed devices had relatively higher PCE in
comparison to other devices. This has been attributed to the
more balanced charge transport in the contact annealed device,
resulting in the high value of incident photon to current efficiency
(IPCE) and Jsc.
1H NMR (CDCl3) d/ppm: 7.90 (m, 4H, aromatic ortho to azo);
6.93 (s, 2H, pyrrole); 6.83 (m, 4H, aromatic ortho to oxygen);
3.98 (t, J ¼ 6.2 Hz, 4H, OCH2(CH2)4CH3); 1.80 (m, 4H,
OCH2CH2(CH2)3CH3); 1.37 (m, 12H, O(CH2)2(CH2)3CH3);
0.90 (t, J ¼ 6.7 Hz, 6H, O(CH2)5CH3).
Anal. Calcd. for C28H36BF2N5O2: C, 64.25; H, 6.93; N, 13.38.
Found: C, 63.86; H, 6.85; N, 13.12.
Characterization methods
IR spectra were recorded on a Perkin-Elmer 16PC FT-IR spec-
1
trometer with KBr pellets. H NMR (400 MHz) spectra were
Experimental
obtained using a Bruker spectrometer. Chemical shifts (d values)
are given in parts per million with tetramethylsilane as an
internal standard. UV-vis spectra were recorded on a Beckman
DU-640 spectrometer with spectrograde THF. Elemental anal-
yses were carried out with a Carlo Erba model EA1108 analyzer.
Electrochemical properties of both polymers were examined
using cyclic voltammetry (CV) (HCH instrument). In particular,
the compounds were coated on a platinum disk and immersed in
0.1 M Bu4NPF6 acetonitrile solution. CV was recorded using
a platinum disk as working electrode and Ag/Ag+ as the reference
Reagents and solvents
1-(Hexyloxy)nitrobenzene (1)27 and 4-(hexyloxy)benzamine (2)28
were synthesized according to reported methods. Tetrahydro-
furan (THF) was dried by distillation over CaH2. Triethylamine
was purified by distillation over KOH. All other reagents and
solvents were commercially purchased and were used as supplied.
Synthesis of compounds A and B
2,5-Bis(4-hexyloxyphenylazo)-1H-pyrrole (A). A flask was
charged with a suspension of 4-(hexyloxy)benzamine (2) (1.00 g,
5.18 mmol) in water (10 mL). Hydrochloric acid (2 mL) was
added to the suspension. The mixture was cooled and kept at
0–5 ꢀC in an ice bath and diazotized by adding a solution of
NaNO2 (0.37 g, ꢀ5.18 mmol) in water (5 ml) followed by stirring
for 0.5 h at 0–5 C. The solution of the diazonium salt (3) thus
prepared was immediately used for the next coupling reaction.
The solution of 3 was slowly added to a solution of pyrrole
(0.17 g, 2.59 mmol) and pyridine (2 mL) in methanol (30 mL) at
0–5 ꢀC. The resulting mixture was stirred for 10 h and then
concentrated under reduced pressure. The precipitate was
filtered, washed with water and dried to afford A. It was purified
electrode at a scan rate of 20 mV sꢁ1
.
Fabrication and characterisation of organic solar cells
Organic solar cells based on compound A or B were fabricated in
the traditional sandwich structure through several steps. Indium
tin oxide (ITO) coated glass substrates were cleaned by ultra-
sonication sequentially in detergent, de-ionized water, acetone
and 2-propanol. After drying the substrate, PEDOT:PSS (Bay-
tron) was spin coated (3000 rpm for 30 s) onto the ITO substrate.
The film was dried at 100 ꢀC for 10 min. After cooling the
substrate, a THF solution of the compound A or B and PCBM
mixture was spin coated onto the top of the PEDOT:PSS. After
drying the solvent, the substrate was transferred into a thermal
evaporator chamber. An Al electrode (ꢂ70 nm) was evaporated
onto the top of the compound A or B:PCBM layer under high
vacuum (10ꢁ5 Pa). For the blend annealedꢀdevice, the blend layer
was thermally annealed in vacuum (120 C) for 10 min, before
the top Al electrode deposition. For the contact annealed device,
the as cast device (i.e. Al toꢀp contact on the un-annealed blend
layer) was annealed at 120 C for 10 min. The current–voltage
(J–V) characteristics of the devices in dark and under
by column chromatography, eluting with
a mixture of
dichloromethane and hexane (1 : 1). Yield 78% (1.92 g).
FT-IR (KBr, cmꢁ1): 3454 (NH stretching); 2934, 2858 (C–H
stretching of hexyloxy groups); 1246, 1146 (ether bond); 1598,
1496 (aromatic).
1H NMR (CDCl3) d/ppm: 9.75 (s, 1H, NH); 7.78 (m, 4H,
aromatic ortho to azo); 6.88 (s, 2H, pyrrole); 6.80 (m, 4H,
aromatic ortho to oxygen); 3.97 (t,
J
¼
6.3 Hz, 4H,
OCH2(CH2)4CH3); 1.81 (m, 4H, OCH2CH2(CH2)3CH3); 1.35
This journal is ª The Royal Society of Chemistry 2010
J. Mater. Chem., 2010, 20, 6464–6471 | 6465

View Article Online
illumination were measured by semiconductor parameter
analyzer (Keithley 4200-SCS). A xenon light source (Oriel, USA)
was used to give an irradiance of 100 mW cmꢁ2 (the equivalent of
one sun at AM1.5) at the surface of the device. The photoaction
spectrum of the devices was measured using a monochromator
(Spex 500 M, USA) and the resulted photocurrent was measured
with a Keithley electrometer (model 6514), which is interfaced to
the computer by LABVIEW software. The device fabrication
procedure for the analysis carrier transport is identical to that of
PV device fabrication, except a Au electrode was used as the top
electrode instead of an Al electrode.
IR spectrum of B, as compared to A, was broader and showed
new intense absorption bands at 3198 and 1068 cmꢁ1 associated
1
with the BF2–azopyrrole complex. The H NMR spectra of A
and B were similar, but A displayed an additional signal at d 9.75
assigned to the NH proton of pyrrole.
A significant change in the colour of the reaction solution was
observed during the complexation. In particular, the colour of
the chloroform solution of A in the presence of Et3N changed
from red to dark purple when BF3$Et2O was added to this
solution. This indicates a bathochromic shift of the absorption
spectrum during the complexation reaction. It has been reported
that the BF2–azopyrrole complexes show a near infrared
absorption which presumably results from considerable p-elec-
tron delocalization by extending the conjugation around the
N]N bond.24
Results and discussion
Synthesis and characterization
Scheme 1 outlines the general synthetic strategy for the inter-
mediate compounds 1–3, as well as the target compounds A and
B. In particular, 4-nitrophenol reacted with 1-bromohexane in
the presence of K2CO3 to afford the ether derivative 1.27 The
catalytic hydrogenation of the latter in a Parr apparatus afforded
the diamine 2.28 Compound 2 reacted subsequently with NaNO2/
aqueous HCl at temperature 0–5 ꢀC to yield the diazonium salt 3.
Then, it reacted with half an equivalent of pyrrole in methanol
neutralized with pyridine to give the symmetrical bisazopyrrole
A. Finally, A reacted with boron trifluoride etherate in the
presence of triethylamine to afford the corresponding BF2–azo-
pyrrole complex B.24 It has been reported that such complexes
are stable in nonprotic solvents like THF, chloroform, acetone,
acetonitrile etc., but unstable in protic solvents like ethanol.24
Consequently, they are also humidity sensitive, a drawback for
these complexes. Both A and B were readily soluble in common
organic solvents owing to the hexyloxy chains.
Photophysical properties
Fig. 1 shows the UV-vis absorption spectra of A and B in THF
solution (10ꢁ5 M) and thin film. Their photophysical character-
istics are summarized in Table 1. A and B displayed in solution
an absorption maximum (la,max) at ꢂ420 nm and a shoulder at
550 nm. This shoulder was more pronounced in B. The absorp-
tion spectra in thin films were broader relative to solution and
covered the visible and near infrared region. This feature was
attributed to intramolecular interactions in solid state. The thin
films showed la,max at 494 nm for A and 515 nm for B. In
addition, B showed a shoulder at 615 nm, which widened its
absorption curve. Apparently, the complexation reaction
broadened the absorption curve and extended it into to the near
infrared region. This is attributable to the formation of rigid azo
configuration which extended the p-conjugation.24 The thin film
absorption onset of A and B is located at 808 and 829 nm cor-
responding to an optical band gap (Eogpt) of 1.54 and 1.49 eV,
respectively. These values of Eogpt are comparable to those of
other related materials which have been synthesized in our
laboratory.25,26
The chemical structures of A and B were confirmed by FT-IR
1
and H NMR spectroscopy. Both these compounds displayed
common absorption bands at approximately 2934, 2858 (C–H
stretching of aliphatic moieties) and 1246, 1146 (ether bond).
Moreover, A showed a sharp band at 3454 cmꢁ1, assigned to the
NH stretching of pyrrole, while this absorption was absent in B
due to complexation of the pyrrole nitrogen. It has been well
established that the N]N frequency of aromatic azo compounds
is difficult to identify by IR. The N]N frequencies are not only
weak but are overlaid by the other aromatic bands as well.29 The
Electrochemical properties
We have carried out the electrochemical cyclic voltammetry for
both A and B films on glassy carbon electrodes to get informa-
tion about the positions of the highest occupied molecular orbital
Scheme 1 Synthesis of compounds A and B.
This journal is ª The Royal Society of Chemistry 2010
6466 | J. Mater. Chem., 2010, 20, 6464–6471

View Article Online
The reduction potential, oxidation potential, HOMO and
LUMO levels, and electrochemical band gap of compounds A
and B are collected in Table 1. The band gaps of A and B, which
were estimated from the electrochemical method, are similar to
the optical band gaps from the UV-Vis absorption onset. The
values of the HOMO and LUMO levels of PCBM were also
estimated from CV and were found to be ꢁ6.3 and ꢁ4.0 eV,
respectively. It can be seen from Table 1 that the difference
between the LUMO of PCBM and compound A or B is greater
than the binding energy of excitons in the organic semiconductor
(0.35 eV).31 Consequently, the blend of A or B as donor with
PCBM as acceptor can be used as a photoactive layer for BHJ PV
devices.
Electrical and photovoltaic properties
We have measured the J–V characteristics of the devices based
on pure A or B sandwiched between the PEDOT:PSS and Al
electrodes, under dark and under illumination intensity of
100 mW cmꢁ2 at room temperature, and they are shown in
Fig. 2(a) and 2(b). The J–V characteristics of both devices in the
dark show a rectification effect, when positive potential is applied
to the PEDOT:PSS-coated ITO electrode with respect to the Al
electrode. The work function of PEDOT:PSS-coated ITO
(5.2 eV) is very close to the HOMO level of A (5.2 eV) or B
(5.1 eV) and forms a nearly Ohmic contact for hole injection
from PEDOT:PSS to the HOMO of the organic materials.
However, the LUMO level of both A and B compounds is very
far from the work function of Al (4.2 eV) and forms the Schottky
barrier for the electron injection from Al into the LUMO level of
A or B. Therefore, the rectification effect is due to the formation
of the Schottky barrier at the Al–A or Al–B interfaces for the
respective devices.
Fig. 1 Normalized absorption spectra of compounds A and B in THF
solution (top) and thin film (bottom).
Table 1 Optical and electrochemical properties of A and B
Charge carrier mobility is an important parameter for mate-
rials used in organic solar cells. Hole mobility in pure A or B was
determined by the space charge limited current (SCLC)
method,32,33 with a device structure ITO/PEDOT:PSS/A or B/Au.
As shown in Fig. 3, the results were plotted as
Compound
A
B
la,max in solution/nma
418, 550d
494
808
420, 550d
515, 615d
829
la,max in thin film/nma
Thin film absorption onset/nm
Eogpt/eVb
1.54
ꢁ1.10
1.49
ꢁ1.25
"
#
ox
onset
red
onset
E
E
/V
/V
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Jd3
=
2
0.50
0.40
ln
vs. ðVappVbiÞ. The analysis of the SCLC
ðVapp ꢁ VbiÞ
HOMO/eV
LUMO/eV
Eegl/eVc
ꢁ5.20
ꢁ3.60
1.60
ꢁ5.10
ꢁ3.55
1.55
results gives hole mobility of 2.15 ꢃ 10ꢁ5 cm2 Vꢁ1 sꢁ1 and
7.18 ꢃ 10ꢁ5 cm2 Vꢁ1 sꢁ1 for A and B, respectively. The PV
parameters were estimated from the J–V characteristics under
illumination (Fig. 2b) and are summarized in Table 2. It can be
seen that the Jsc and PCE of the devices using B is higher than
that for A, under the same conditions. The lower band gap and
higher hole mobility of B as compared to A leads to the relatively
higher PV performance of the device based on B.
a
la,max: The absorption maxima from the UV-vis spectra in THF
b
solution or in thin film. Eogpt: Optical band gap determined from the
c
absorption onset in thin film.
Eegl: Electrochemical band gap
determined from cyclic voltammetry. d Shoulder of the absorption curve.
(HOMO) and lowest unoccupied molecular orbital (LUMO) of
onset
Fig. 4(a) and 4(b) show the J–V characteristics of the devices
using A:PCBM and B:PCBM blends under different conditions
and illumination of AM 1.5 (100 mW cmꢁ2). The corresponding
open circuit voltage (Voc), Jsc, fill factor (FF) and PCE of the
devices are summarized in Tables 3 and 4. It can be seen from
Table 3 that the PCE of the contact annealed device is higher
than that for both the as cast and blend annealed devices. Two
important changes in J–V characteristics are responsible for the
observed improvement in the PCE of the contact annealed
these compounds. The onset reduction potential (E ) and
onset
red
oxidation potential (E ) have been measured with respect to
ox
a Ag/Ag+ electrode. HOMO and LUMO levels were estimated
from the following equations:30
ꢀ
ꢁ
onset
ox
EHOMO ¼ ꢁe E
ELUMO ¼ ꢁe E
ꢁ 4:7
ꢁ 4:7
(1)
ꢀ
ꢁ
onset
red
where the unit of potential is V vs. Ag/Ag.+
This journal is ª The Royal Society of Chemistry 2010
J. Mater. Chem., 2010, 20, 6464–6471 | 6467

View Article Online
Fig. 3 Current–voltage curves for ITO/PEDOT:PSS/A or B/Au devices
in the form of ln[Jd3/(Vappl ꢁ Vbi)2] vs. [(Vappl ꢁ Vbi)/d]1/2
.
Table 2 Photovoltaic parameters for devices based on pristine A and B
using ITO/PEDOT:PSS and Al electrodes
Power
conversion
Short Open
Compound (Jsc)/mA cmꢁ2 (Voc)/V
circuit current circuit voltage Fill factor efficiency
(h) (%)
(FF)
A
B
0.12
0.24
0.74
0.78
0.38
0.40
0.034
0.074
devices is very similar to the absorption spectra of the respective
blends, indicating that all the absorption of compound A or B
contributed to the PV conversion. The IPCE spectrum of the
device based on the B:PCBM blend has been extended up to
820 nm. The improved IPCE for the device based on the
B:PCBM blend as compared to the device based on the A:PCBM
may be due to the higher hole mobility and lower band gap of B
relative to A. The driving force for the dissociation of excitons
into free carriers in the BHJ device is the difference between the
LUMO levels of the donor and acceptor used in the blend. This
difference in the B:PCBM blend is higher than that for the
A:PCBM blend, and may also be responsible for the higher value
of Jsc and IPCE in the PV devices based on the B:PCBM blend.
We can understand the effect of processing the blend active
layer by examining the details of hole and electron conduction
through the blend network. We have measured the electronic
transport through two independent networks in the blend layer
using the appropriate choice of metallic top and bottom elec-
trodes.34,35 We have fabricated the BHJ devices using the
B:PCBM blend with PEDOT:PSS (bottom) and Au (top)
contacts (Fig. 7), to investigate the hole and electron transport in
B and PCBM domains. Under reverse bias (Fig. 7b), where Au is
positively biased with respect to PEDOT:PSS, the device permits
only hole transport via the HOMO of B, because hole injection
from Au to the HOMO of B occurs with nominal energy barrier.
Fig. 2 Current–voltage (J–V) characteristics of the devices based on
compounds (a) in dark (b) under illumination intensity of 100 mW cmꢁ2
.
device. The enhancement in the Jsc originates from the changes in
the internal material properties other than the blend
morphology, because we observed only a slight increase in the Jsc
in the devices similarly annealed prior to contact deposition
(i.e. blend annealed). The increase in both Voc and FF can also be
correlated with the decrease in current leakage for the contact
annealed device. Both the as cast and blend annealed devices
show more dark current in comparison with the contact annealed
device, as shown in the J–V characteristics in the dark for the
device using B:PCBM blend (Fig. 5).
The incident photon to current efficiency (IPCE) spectra of the
BHJ PV devices using the as cast blends under monochromatic
light are shown in Fig. 6. The shape of the IPCE spectra of the
6468 | J. Mater. Chem., 2010, 20, 6464–6471
This journal is ª The Royal Society of Chemistry 2010

View Article Online
Table 4 Photovoltaic parameters for the as cast, blend annealed and
contact annealed devices based on the B:PCBM (1 : 1) blend using
ITO/PEDOT:PSS and Al electrodes
Short circuit Open circuit
current
voltage
(Jsc)/mA cmꢁ2 (Voc)/V
Fill factor Power conversion
(FF)
efficiency (h) (%)
As cast
Blend
6.30
7.00
0.72
0.69
0.47
0.47
2.22
2.27
annealed
Contact
annealed
8.20
0.74
0.52
3.15
Fig. 5 Current–voltage characteristics in the dark for three devices with
B:PCBM blend BHJ devices with different annealed conditions.
Fig. 4 Current–voltage characteristics under illumination of intensity
100 mW cmꢁ2 for three devices with compound (a) A:PCBM (1 : 1) and
(b) B:PCBM blend BHJ photovoltaic devices with different annealed
conditions.
Table 3 Photovoltaic parameters for the as cast, blend annealed
and contact annealed devices based on A:PCBM (1 : 1) blend using
ITO/PEDOT:PSS and Al electrodes
Short circuit Open circuit
current voltage
(Jsc)/mA cmꢁ2 (Voc)/V
Power
Fill factor conversion
(FF)
efficiency (h) (%)
As cast
Blend
5.20
6.22
0.75
0.72
0.45
0.46
1.76
2.00
annealed
Contact
annealed
6.90
0.80
0.49
2.70
Fig. 6 IPCE spectra of BHJ photovoltaic devices with compound A or
B:PCBM blends.
This journal is ª The Royal Society of Chemistry 2010
J. Mater. Chem., 2010, 20, 6464–6471 | 6469

View Article Online
Fig. 7 (a) Schematic energy diagram for a PEDOT:PSS/B:PCBM/Au
device in an open circuit condition. Energy values are relative to the
vacuum level. The difference in work function values of PEDOT:PSS
(5.2 eV) and Au (5.1 eV)31 results in Vbi of 0.1 eV. (b) Energy diagram
under reverse voltage bias showing hole injection from Au. (c) Energy
diagram under forward voltage bias, showing hole injection from
PEDOT:PSS and electron injection from Au.
Electron injection from PEDOT:PSS to the LUMO of PCBM is
suppressed by the 1.2 eV Schottky barrier, as evidenced by
negligible reverse bias current in the device having PEDOT:PSS
as the bottom contact. In our device with a Au top electrode, the
electrons can also be injected from the Au into the LUMO level
of the PCBM due to the reduction in Schottky barrier height by
interface dipoles (in forward bias condition). Consequently, the
total current in forward bias (Fig. 7c) consists of contributions
from both electron and hole currents.35,36,37
The hole transport through the B network in both the as cast
and contact blend annealed devices at low reverse applied voltage
Va < 1.5 V is well described using the SCLC model in trap filled
limit, as the square root of reverse bias current (JR1/2) increases
nearly with bias across the blend over the entire voltage range
(Fig. 8a). In the trap filled SCLC model, the J–V characteristics
can be described as:
9
V2
d3
J ¼ 3o3rm
(2)
8
where V ¼ Va ꢁ Vbi, Va is the applied voltage, Vbi is the built in
voltage due to the difference in the electrode work functions, 30 is
vacuum permittivity, 3r is the relative dielectric constant of the
transport medium, m is the mobility of charge carriers and d is the
thickness of the blend layer in the device. From the slope of
the curves shown in Fig. 8a, we have estimated the hole mobility
and found it to be 1.15 ꢃ 10ꢁ5 and 6.56 ꢃ 10ꢁ5 cm2 Vꢁ1 sꢁ1 for the
as cast and contact annealed devices, respectively.
Fig.
8
(a) Low voltage J–V characteristics of reverse biased
PEDOT:PSS/B:PCBM/Au devices for the as-cast and contact annealed
conditions, displaying trap-filled SCLC behavior. (b) J–V characteristics,
i.e. Jf-r vs. (Va ꢁ Vbi) for electron transport in the B:PCBM blend, the plot
indicating SCLC conduction at Va ꢁ Vbi > 1 V.
We have estimated the electron mobility through the PCBM
domain in the blend layers through analysing the forward bias
J–V characteristics of the devices. In the forward bias condition,
the Au electrode is biased negatively with respect to PEDOT:PSS
(Fig. 7c). The total device current consists of both an electron
current, Je (electron injection from Au electrode into the LUMO
level of PCBM), and hole current, Jh (hole injection from
PEDOT:PSS into the HOMO level of compound B). The total
forwards bias current, JF ¼ (Je + Jh), which is analogous to
double injection of electrons and holes into an insulator in the
space charge current, limit with high recombination proba-
bility.38 Since the work functions of Au and PEDOT:PSS are
similar, we can assume that the hole conduction through the
HOMO of compound B is symmetric with respect to the bias
voltage direction. Therefore, we can separate the forward bias
6470 | J. Mater. Chem., 2010, 20, 6464–6471
This journal is ª The Royal Society of Chemistry 2010

View Article Online
7 P. W. M. Blom, V. D. Mihailetchi, L. J. A. Koster and D. E. Markov,
Adv. Mater., 2007, 19, 1551.
electron current through the LUMO level of PCBM by sub-
tracting the total reverse current from the forward bias current,
i.e. Je ¼ Jf ꢁ Jh ¼ Jf ꢁ Jr ¼ Jf-r. The variation of Jf-r with Va ꢁ Vbi
is shown in Fig. 8b for both the as cast and contact annealed
devices. It can be seen from these curves that at sufficiently high
forward bias voltages, both devices show J–V behaviour
consistent with space charge limited electron conduction through
PCBM in trap filled limit i.e. Jf-r ꢂ (Va ꢁ Vbi)2. The J–V char-
acteristics can be described by a SCLC model including a field
dependent mobility m(E).39
8 G.Dennler,M.C.ScharberandC.J.Brabec,Adv.Mater.,2009,21,1323.
9 S. E. Shaheen, C. J. Brabec, N. S. Sariciftci, F. Padinger, T. Fromherz
and J. C. Hummelen, Appl. Phys. Lett., 2001, 78, 841.
10 J. Peet, J. Y. Kim, N. E. Coates, W. L. Ma, D. Moses, A. J. Heeger
and G. C. Bazan, Nat. Mater., 2007, 6, 497.
11 G. Li, V. Shrotriya, Y. Yao, J. Huang and Y. Yang, J. Mater. Chem.,
2007, 17, 3126.
12 W. Ma, C. Yang, X. Gong, K. Lee and A. J. Heeger, Adv. Funct.
Mater., 2005, 15, 1617.
13 (a) D. Muhlbacher, M. Scharber, M. Morana, Z. G. Zhu, D. Waller,
R. Gaudiana and C. Brabec, Adv. Mater., 2006, 18, 2884; (b) J. Peet,
J. Y. Kim, N. E. Coates, W. L. Ma, D. Moses, A. J. Heeger and
G. C. Bazan, Nat. Mater., 2007, 6, 497; (c) E. G. Wang, L. Wang,
L. F. Lan, C. Luo, W. L. Zhuang, J. B. Peng and Y. Cao, Appl.
Phys. Lett., 2008, 92, 33307; (d) J. H. Hou, H. Y. Chen,
S. Q. Zhang, G. Li and Y. Yang, J. Am. Chem. Soc., 2008, 130,
16144; (e) Y. Liang, D. Feng, Y. Wu, S.-T. Tsai, G. Li, C. Ray and
L. Yu, J. Am. Chem. Soc., 2009, 131, 7792.
14 (a) F. C. Krebs, Energy Environ. Sci., 2010, 3, 512; (b) F. C. Krebs,
M. Jorgensen, K. Norman, O. Hagemann, J. Alstrup,
T. D. Nielsen, J. Fyenbo, K. Larsen and J. Kristensen, Sol. Energy
Mater. Sol. Cells, 2009, 93, 422; (c) F. C. Krebs, S. A. Gevorgyan,
B. Gholamkhass, S. Holdcroft, C. Schlenker, M. E. Thompson,
B. C. Thompson, D. Olson, D. S. Ginley and S. E. Shaheen, et al.,
Sol. Energy Mater. Sol. Cells, 2009, 93, 1968; (d) F. C. Krebs,
S. A. Gevorgyan and J. Alstrup, J. Mater. Chem., 2009, 19, 5442.
15 (a) M. Helgesen, R. Sondergaard and F. C. Krebs, J. Mater. Chem.,
2010, 20, 36; (b) B. Kippelen and J.-L. Bredas, Energy Environ. Sci.,
2009, 2, 251; (c) F. C. Krebs, Sol. Energy Mater. Sol. Cells, 2009,
93, 394.
9
V2
d3
J ¼ 303rmðEÞ
(3a)
8
where m(E) is given by
!
rffiffiffiffi
V
mðEÞ ¼ mð0Þexp 0:891g
(3b)
d
where g is the field activation factor and m(0) is the zero field
mobility. The analysis of J–V characteristics is shown in Fig. 8b
and eqn 3a and 3b. We have estimated the values of electron
mobility, which are 3.2 ꢃ 10ꢁ4 cm2 Vꢁ1 sꢁ1 and 4.3 ꢃ 10ꢁ4 cm2 Vꢁ1
sꢁ1 for the as cast and contact annealed device, respectively.
Comparing the electron and hole mobility for the as cast and
contact blend device, it is found that the difference in the electron
and hole mobility is about 28 and 6.5, respectively. This decrease
in the difference causes the reduction of recombination in the
contact annealed device and leads to an improvement in the PCE
of the PV device.
16 C. J. Brabec, A. Cravino, D. Meissner, N. S. Sariciftci, T. Fromherz,
M. T. Rispens, L. Sanchez and J. C. Hummelen, Adv. Funct. Mater.,
2001, 11, 374.
17 H. Rau and B. Bunsenges, Physikalische Chemie, 1968, 72, 637.
18 R. B. Champ and M. D. Shattuck, US Patent, 3898084, 1975.
19 K. Y. Law, Chem. Rev., 1993, 93, 449.
20 D. Liu, M. Mudadu, R. Thummel, Y. Tao and S. Wang, Adv. Funct.
Mater., 2005, 15, 143.
Conclusions
21 B. Domercq, C. Grasso, J. L. Maldonado, M. Halik, S. Barlow,
S. R. Marder and B. Kippelen, J. Phys. Chem. B, 2004, 108, 8647.
22 Y. Sun, D. Rohde, Y. Liu, L. Wan, Y. Wang, W. Wu, C. Di, G. Yu
and D. Zhu, J. Mater. Chem., 2006, 16, 4499.
23 D. Kumaresan, R. P. Thummel, T. Bura, G. Ulrich and R. Ziessel,
Chem.–Eur. J., 2009, 15, 6335.
24 Y. Li, O. Brian Patrick and D. Dolphin, J. Org. Chem., 2009, 74, 5237.
25 J. A. Mikroyannidis, G. D. Sharma, S. S. Sharma and Y. K. Vijay,
J. Phys. Chem. C, 2010, 114, 1520.
26 J. A. Mikroyannidis, M. S. Roy and G. D. Sharma, J. Power Sources,
2010, 195, 5391.
A symmetrical bisazopyrrole (A) and the corresponding BF2–
azopyrrole complex (B) were successfully synthesized and used
for BHJ solar cells. Compounds A and B were soluble in
common organic solvents, but B was stable only in nonprotic
solvents. The thin film absorption spectrum of B was broader
and showed lower band gap (1.49 eV) than that (1.54 eV) of A.
The HOMO and LUMO levels of both compounds A and B are
found suitable for the BHJ photoactive layer. We have shown
that
a
suitable post-device-fabrication thermal treatment
27 S. Kim, A. Oehlhof, B. Beile and H. Meier, Helv. Chim. Acta, 2009,
92, 1023.
(contact annealed) results in high performance A:PCBM or
B:PCBM blend BHJ solar cells. The PCE values are 2.70% and
3.15% for the devices based on the A:PCBM and B:PCBM blend,
respectively. We found that the improved PCE for the contact
annealed device results from the improved balanced charge
transport, due to the enhanced hole mobility.
28 M.J.Jeong,J.H.Park,C.LeeandJ.Y.Chang,Org.Lett.,2006,8,2221.
29 L. J. Bellamy, The infra-red spectra of complex molecules, Chapman
and Hall, (London), John Wiley & Sons (New York), 1975, p 304.
30 Q. J. Sun, H. Q. Wang, C. H. Yang and Y. F. Li, J. Mater. Chem.,
2003, 13, 800.
31 G. Li, V. Shrotriya, J. Huang, Y. Yao, T. Moriarity, K. Emery and
Y. Yang, Nat. Mater., 2005, 4, 864.
32 G. G. Malliaras, J. R. Salem, P. J. Brock and C. Scott, Phys. Rev. B:
Condens. Matter Mater. Phys., 1998, 58, R13411.
33 D. Chirvase, Z. Chiguvane, M. Knipper, J. Parisi, V. Dyakonov and
J. C. Hummelen, J. Appl. Phys., 2003, 93, 3376.
References
34 V. Shrotriya, Y. Yao, G. Li and Y. Yang, Appl. Phys. Lett., 2006, 89,
063505.
35 V. D. Mihailetchi, H. X. Xie, B. de Boer, L. J. A. Koster and
P. W. M. Blom, Adv. Funct. Mater., 2006, 16, 699.
36 V. D. Mihailetchi, P. W. M. Blom, J. C. Hummelen and
M. T. Rispens, J. Appl. Phys., 2003, 94, 6849.
37 S. C. Veenstra, A. Heeres, G. Hadziioannou, G. A. Sawatzky and
H. T. Jonkman, Appl. Phys. A: Mater. Sci. Process., 2002, 75, 661.
38 M. A. Lampert and P. Mark, Current Injection in Solids, Academic,
New York, 1970.
1 C. J. Brabec, V. Dyakonov, J. Parisi and N. S. Sariciftci, in Organic
Photovoltaics: Concepts and Realization, Springer, Heidelberg, 2003.
2 C. J. Brabec, Sol. Energy Mater. Sol. Cells, 2004, 83, 273.
3 G. Yu, J. Gao, J. C. Hummelen, F. Wudl and A. J. Heeger, Science,
1995, 270, 1789.
4 F. Padinger, R. S. Rittberger and N. S. Sariciftci, Adv. Funct. Mater.,
2003, 13, 85.
5 J. Xue, B. P. Rand, S. Uchida and S. R. Forrest, Adv. Mater., 2005,
17, 66.
6 G. Li, V. Shrotriya, J. Huang, Y. Yao, T. Moriarty, K. Emery and
Y. Yang, Nat. Mater., 2005, 4, 864.
39 P. N. Murgatroyd, J. Phys. D: Appl. Phys., 1970, 3, 151.
This journal is ª The Royal Society of Chemistry 2010
J. Mater. Chem., 2010, 20, 6464–6471 | 6471