High efficiency solution-processed two-
dimensional small molecule organic solar cells
obtained via low-temperature thermal annealing†
Zhengkun Du,
ab
Weichao Chen,
a
Yanhua Chen,
ac
Shanlin Qiao,
a
Xichang Bao,
a
Shuguang Wen,
a
Mingliang Sun,
c
Liangliang Han
a
and Renqiang Yang
*
a
A new two-dimensional (2D) organic small molecule, DCA3T(T-BDT), was designed and synthesized for
solution-processed organic solar cells (OSCs). DCA3T(T-BDT) exhibited a deep HOMO energy level
(5.37 eV) and good thermal stability. The morphologies of the DCA3T(T-BDT):[6,6]-phenyl-C61-butyric
acid methyl ester (PC
61
BM) blends were investigated by atomic force microscopy and the crystallinity
was explored by X-ray diffraction (XRD) and 2D grazing incidence wide-angle X-ray scattering (GIWAXS),
respectively. The morphologies of the blends were strongly influenced by the blend ratio of DCA3T(T-
BDT):PC
61
BM and annealing temperature. The effect of thermal annealing on the photovoltaic
performance of DCA3T(T-BDT)-based small molecule organic solar cells (SMOSCs) was studied in detail.
When DCA3T(T-BDT) was used as a donor with PC
61
BM as an acceptor, high efficiency SMOSCs with a
power conversion efficiency of 7.93%, a high V
oc
of 0.95 V, J
sc
of 11.86 mA cm
2
and FF of 0.70 were
obtained by a thermal annealing process at only 60
C, which offers obvious advantages for large scale
production compared with solvent additive or interfacial modification treatment.
1. Introduction
Over the past two decades, organic solar cells (OSCs) as a
promising renewable energy source have attracted considerable
attention of chemical, material and physical scientists because
of their low cost, light weight, and large area fabrication on
exible substrates.
1–9
At present, the power conversion efficiency
(PCE) of bulk heterojunction (BHJ) OSCs based on low band gap
conjugated polymer is over 9% using an inverted device archi-
tecture.
10
As an alternative candidate for polymer solar cells,
small molecule organic solar cells (SMOSCs) have recently
achieved great development, and PCE of over 8% have been
realized through carefully modifying the molecular structure
and using additives.
11,12
Although PCE is still relatively lower
than that of polymer solar cells, SMOSCs have a better prospect
of industrialization due to their unique advantages of well-
dened molecular structure, synthetic reproducibility, mono-
dispersity, and convenient energy level control.
13–22
Thus,
further work on designing new small molecules and gaining a
better understanding of the molecular structure-function rela-
tionships are needed for their future commercial applications.
Compared to their polymeric counterparts,
23–25
in most
cases, SMOSCs have relatively low ll factor (FF) and short
circuit current density (J
sc
) due to their poorer lm formation
ability and this has become the main hindrance for developing
high performance devices.
26
To date, conjugated organic small
molecules with wide absorption, high hole mobility, as well as
appropriate miscibility with the fullerene acceptor to form
uniform interpenetrating networks were designed and synthe-
sized for high efficiency SMOSCs.
17,27
Among the vast variety of
the developed electron-donating materials for OSCs, two-
dimensional (2D) alkylthiophene substituted benzo[1,2-b:4,5-b
0
]-
dithiophene (BDT) has been proved to be a promising material
for photovoltaic applications.
28–30
The incorporation of thio-
phene side chains could extend the vertical p–p conjugation
and promote p–p stacking in the solid state.
31
This extended
conjugated structure is benecial in improving charge transport
and correspondingly achieve a high J
sc
.
31,32
Compared with a
one-dimensional (1D) BDT-based conjugated polymer,
6
the 2D
BDT-based polymer generally represents a higher open-circuit
voltage (V
oc
) in the organic solar cell due to its lower highest
occupied molecular orbital (HOMO) levels.
33,34
Hou et al.
reported a series of copolymers based on 2D BDT derivatives,
which showed excellent photovoltaic performance with PCEs of
over 7%.
31,35
Considering the great progress that has been ach-
ieved in polymer OSCs using 2D BDT derivatives, the 2D BDT
unit is a promising donor material for SMOSCs.
12,15,36
Therefore,
a
CAS Key Laboratory of Bio-based Materials, Qingdao Institute of Bioenergy and
Bioprocess Technology, Chinese Academy of Sciences, Qingdao 266101, China.
E-mail: yangrq@qibebt.ac.cn
b
University of Chinese Academy of Sciences, Beijing 100049, China
c
Institute of Materials Science and Engineering, Ocean University of China, Qingdao
266100, China
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c4ta03314k
Cite this: J. Mater. Chem. A,2014,2,
15904
Received 29th June 2014
Accepted 28th July 2014
DOI: 10.1039/c4ta03314k
www.rsc.org/MaterialsA
15904 | J. Mater. Chem. A,2014,2,15904–15911 This journal is © The Royal Society of Chemistry 2014
Journal of
Materials Chemistry A
PAPER
Published on 19 August 2014. Downloaded on 27/10/2014 01:25:25.
View Article Online
View Journal
| View Issue
novel small molecules with high efficiency would be designed
and synthesized if the HOMO and LUMO levels of the 2D BDT-
based small molecules could be effectively tuned by employing
appropriate electron-withdrawing groups. On the other hand, a
suitable method for fabricating the organic solar cells is also
crucial for achieving high efficiency. Thermal annealing of BHJ
lms is one of the most widely used methods for improving the
J
sc
and FF of BHJ OSCs.
37–40
However, as the thermal annealing
temperature increases, the V
oc
of SMOSCs sometimes signi-
cantly decreases.
38
Given the effect of thermal annealing on the
morphology and crystallinity of the lm, a delicate balance
between V
oc
, J
sc
and FF could be achieved by annealing at a low
temperature, and a maximum PCE of solution-processed
SMOSCs will be achieved.
In view of the aforementioned concerns, we have success-
fully synthesized a new solution-processed 2D-conjugated
organic small molecule, DCA3T(T-BDT) (Scheme 1), with 2-
ethylhexyl cyanoacetate units as the end-capped blocks and 4,8-
bis[5-(2-ethylhexyl)thiophen-2-yl]benzo[1,2-b:4,5-b
0
]-dithio-
phene as the central core. BHJ OSCs were fabricated using a
blend of DCA3T(T-BDT) as the donor and PC
61
BM as the
acceptor, with a single layer conventional device architecture of
ITO/PEDOT:PSS/DCA3T(T-BDT):PC
61
BM/Ca/Al. The best device
exhibited a high PCE of 7.93% aer thermal annealing at a low
temperature of 60
C for 10 min, with a high V
oc
of 0.95 V, J
sc
of
11.86 mA cm
2
and FF of 0.70. To the best of our knowledge,
this is the highest PCE for solution-processed SMOSCs without
any solvent additive or interfacial modication treatment found
in the conventional devices.
2. Results and discussion
2.1 Synthesis and characterization
The detailed synthesis and characterization of DCA3T(T-BDT)
are given in the experimental section (Scheme 2). The key
precursor 7 was synthesized using a Pd(PPh
3
)
4
-catalyzed Stille
coupling reaction between 4
0
,4
00
-dioctyl-2,2
0
:5
0
,2
00
-trithiophene-5-
carbaldehyde (5) and 2,6-bis(trimethyltin)-4,8-bis(5-(2-ethyl-
hexyl)thiophen-2yl-)benzo[1,2-b:4,5-b
0
]dithiophene (6)inreux-
ing toluene under argon for 2 days. The target small molecule
DCA3T(T-BDT) was obtained through the Knoevenagel conden-
sation of intermediate 7 with 2-ethylhexyl cyanoacetate using
triethylamine as the catalyst in anhydrous chloroform (CHCl
3
)
for 40 h. The DCA3T(T-BDT) was fully characterized by NMR
spectroscopy, mass spectrometry, electrochemistry, high
performance liquid chromatography with UV-detection (HPLC-
UV), thermogravimetric analysis (TGA) and optical spectroscopy.
DCA3T(T-BDT) possesses good solubility at room temperature in
common organic solvents, such as toluene, chlorobenzene,
o-dichlorobenzene, dichloromethane (CH
2
Cl
2
) and CHCl
3
,due
to eight solubilizing alkyl side chains.
2.2 Thermal stability
TGA and differential scanning calorim etry (DSC) were used to
investigate the thermal properties of the small molecule. As
showninFig.1,theTGAcurveofDCA3T( T-BDT) indicates a
good thermal stability with onset decomposition t empera-
ture with a 5% weight loss (T
d
) o ccurring at 389
Cunderan
nitrogen atmosphere. The thermal stability of DCA3T(T-BDT)
is high enough to ensure the photovoltaic device fabrication.
AsshowninFig.S2,† the main melting endotherm o ccurs
at 163.7
C, and upon cooling DCA3T(T-BDT) exhibits a
major crystallization exotherm at 133.5
C, indicating
that this small molecule has an obvious tendency to
crystallize.
Scheme 1 Chemical structure of DCA3T(T-BDT).
Scheme 2 Synthetic routes towardsDCA3T(T-BDT).
Fig. 1 TGA plot of DCA3T(T-BDT) with a heating rate of 10
Cmin
1
under nitrogen atmosphere.
This journal is © The Royal Society of Chemistry 2014 J. Mater. Chem. A,2014,2,15904–15911 | 15905
Paper Journal of Materials Chemistry A
Published on 19 August 2014. Downloaded on 27/10/2014 01:25:25.
View Article Online
2.3 Optical and electrochemical properties
UV-vis absorption spectra of DCA3T(T-BDT) were investigated in
CHCl
3
solution and as a thin lm, as shown in Fig. 2a. DCA3T-
(T-BDT) exhibited a broad absorption range covering 300 – 700
nm wavelengths in the solid state. In solution, DCA3T(T-BDT)
has a maximal absorption (l
max
) peak at 499 nm corresponding
to the intermolecular charge transfer (ICT) interaction between
D and A moieties.
41,42
In comparison with its absorption spec-
trum in solution, the absorption spectrum of DCA3T(T-BDT) as
a thin lm shows a remarkable red-shied l
max
at 573 nm with a
shoulder peak at 620 nm, indicating that a strong intermolec-
ular p–p interaction exists in the solid state due to the improved
planar construction of the 2D side chains. The optical band gap
(E
opt
g
) value of DCA3T(T-BDT) is about 1.80 eV, which is deduced
from the onset (690 nm) of absorption in the thin lm.
Density functional theory (DFT) calculations were employed
to investigate the electronic structure and geometry of
DCA3T(T-BDT), and the corresponding results are shown in
Fig. 3. The electron density of HOMO was mainly delocalized
over the central BDT core, and the electron density of the LUMO
was concentrated on the 2-ethylhexyl cyanoacetate groups,
demonstrating that an electron can transfer from the central
BDT group to the end-capped blocks during an electronic
excitation process.
43
From the DFT calculations, the HOMO and
LUMO energies of DCA3T(T-BDT) were found to be 4.98 eV
and 2.82 eV, respectively. The electrochemical behavior of
DCA3T(T-BDT) was investigated by cyclic voltammetry (CV). As
shown in Fig. 2b, the onset oxidation potential for the small
molecule was 0.95 V versus a saturated calomel electrode (SCE).
The corresponding HOMO energy level of DCA3T(T-BDT) was
determined to be 5.37 eV. The LUMO energy levels of
DCA3T(T-BDT) was estimated to be around 3.42 eV from the
onset reduction potential. This fairly suitable HOMO energy
level conveys that a high V
oc
of SMOSCs based on DCA3T-
(T-BDT) blended with PC
61
BM could be expected.
31
The LUMO
energy level difference (>0.3 eV) between the donor and acceptor
(PC
61
BM) is large enough for the separation of the excitons,
44–47
which is very useful for increasing the short-circuit current in
DCA3T(T-BDT)-based OSCs.
2.4 Hole mobility
The high carrier charge mobility is important for effective
photovoltaic active layer materials, which would facilitate
charge exciton separation from the donor–acceptor interface,
carrier transport to the electrodes and reduce recombination.
15
The carrier charge mobility of DCA3T(T-BDT) was investigated
by employing organic eld-effect transistors (OFETs) and
vertical diodes. DCA3T(T-BDT) exhibited a typical p-type semi-
conductor behavior (Fig. 4), and the hole mobility was around
0.03 cm
2
V
1
s
1
, calculated according to the transfer charac-
teristic curve. In the vertical direction, the hole mobility was
measured by the space-charge limit current (SCLC)
48–50
model
with a typical device structure of ITO/PEDOT:PSS/DCA3T-
(T-BDT)/Au. SCLC is described by the equation J
SCLC
¼ (9/8)
3
0
3
r
m
h
((V
2
)/(L
3
)), where J stands for current density, 3
0
is the
permittivity of free space, 3
r
is the relative dielectric constant of
the transport medium, m
h
is the hole mobility, V is the internal
potential in the device and L is the thickness of the active layer.
The hole mobility of DCA3T(T-BDT) was about 1.2 10
4
cm
2
V
1
s
1
, calculated using the SCLC model (Fig. S4 †).
2.5 Photovoltaic properties
The BHJ solar cells with a conventional architecture of ITO/
PEDOT:PSS/DCA3T(T-BDT):PC
61
BM/Ca/Al using PC
61
BM as the
acceptor were fabricated to explore the photovoltaic perfor-
mance of DCA3T(T-BDT). The detailed device fabrication and
measurements are described in the experimental section. First,
aer thickness optimization, the relative ideal thickness of
active layer was around 90 nm for DCA3T(T-BDT)-based OSCs.
The effect of different D/A blend ratios on the performance of
SMOSCs based on DCA3T(T-BDT) and PC
61
BM was investigated,
and the corresponding results are collected in Table 1. With an
increase of the D/A blend ratio from 1 : 1 to 4 : 1, V
oc
slightly
Fig. 2 (a) UV-vis absorption spectra of DCA3T(T-BDT) in CHCl
3
solution and as a solid thin film. (b) Cyclic voltammogram of DCA3T-
(T-BDT) in 0.1 M Bu
4
NPF
6
-acetonitrile solution at a scan rate of
100 mV s
1
.
Fig. 3 Optimized molecular geometries and frontier molecular
orbitals (isovalue surface 0.02 au) using DFT evaluated at the B3LYP/6-
31G(d) level of theory.
Fig. 4 (a) Transfer and (b) output characteristics (V
DS
¼80 V) for a
typical DCA3T(T-BDT)-based OFET device.
15906
| J. Mater. Chem. A,2014,2,15904–15911 This journal is © The Royal Society of Chemistry 2014
Journal of Materials Chemistry A Paper
Published on 19 August 2014. Downloaded on 27/10/2014 01:25:25.
View Article Online
increased, but J
sc
and FF rst increased and then gradually
decreased. The optimized device made from the DCA3T(T-BDT)
blended with PC
61
BM (3 : 1, w/w) shows the highest PCE of
7.19%, with a V
oc
of 0.97 V, J
sc
of 11.22 mA cm
2
and FF of 0.66.
The high J
sc
could be partly ascribed to the better absorption of
DCA3T(T-BDT) and the suitable energy matching and larger V
oc
agrees with its deeper HOMO energy level (5.37 eV). The device
performance is consistent with the morphologies of the lms
with different ratio, which are shown in the Fig. 6a–d. The
domain in the lms is gradually increasing with an increase in
the D : A ratio, and the domain boundary is reduced. A good
interpenetrating network could be formed at the ratio of 3 : 1
(D : A) with relatively low boundary density, which is in favor of
the improvement of J
sc
and FF.
To further improve the photovoltaic performance of SMOSCs
based on DCA3T(T-BDT) and PC
61
BM, thermal annealing was
employed. The device parameters with different annealing
temperatures are summarized in Table 1. The performance of
the devices were improved by thermal annealing with higher J
sc
and FF compared to the devices as casted. The optimized device
based on the DCA3T(T-BDT)/PC
61
BM (3 : 1, w/w) blend shows a
PCE of 7.93%, V
oc
of 0.95 V, J
sc
of 11.86 mA cm
2
and FF of 0.70
aer thermal annealing at only 60
C (Fig. 5a). The performance
of the devices with different ratios could also be enhanced aer
annealing, as shown in the Table 1. Therefore, thermal
annealing at relatively low temperature is a good method to
improve the PCE of the devices and has obvious advantages for
large scale production. It is worth noting that all the devices
exhibited high V
oc
(0.94–0.98 V), which were almost insensitive
to the annealed temperature and the D/A weight ratios. The J–V
curves of the devices with different D/A blend ratios under the
illumination of AM 1.5G (100 mW cm
2
) are shown in Fig. S3.†
As a result, DCA3T(T-BDT) would be a promising donor material
for high efficiency SMOSCs.
The EQE curve of the optimized device covered a broad
wavelength range of 310–700 nm with a maximum value of 70%
at 569 nm as shown in Fig. 5b. Aer thermal annealing at 60
C,
the photoresponse is enhanced at the long wavelength region,
which may be the contribution of improvement of the crystal-
lization. J
sc
calculated from the EQE curve is 11.34 mA cm
2
,
which is consistent with the J
sc
value obtained from the J–V
curve.
2.6 Morphology and structural (X-ray) characterization
The enhanced device performance by the thermal annealing
process at low temperature could be partially attributed to a
better interpenetrating network.
27
The morphologies of the
lms with the DCA3T(T-BDT):PC
61
BM (3 : 1, w/w) blend
annealing at different temperatures are given in Fig. S5.† The
lms aer annealing show a lower root mean square compared
to the pristine ones. Upon increasing the annealing tempera-
ture from 40 to 80
C, the domain boundary intensity gradually
reduced, which would decrease the trap state density existing in
the boundary and therefore reduce the recombination current
and improve the transport and collection of the carrier charge.
51
This may also be an important reason for the improvement of J
sc
and FF in this work. The morphologies of lms with different
D : A ratio aer annealing at 60
C are given in Fig. 6e–h. All
lms show a lower boundary intensity compared to their lms
as casted. Therefore, it is believed that thermal annealing at
relatively low temperature may be an effective method to
enhance the performance of devices by reducing the boundary
intensity.
To further understand the performance difference of
DCA3T(T-BDT)-based SMOSCs before and aer thermal
annealing at 60
C, the crystallinity of the DCA3T(T-BDT)
Table 1 The photovoltaic performance of SMOSCs based on DCA3T(T-BDT):PC
61
BM before and after thermal annealing with different blend
ratio
D : A (w/w) Thermal annealing V
oc
(V) J
sc
(mA cm
2
) FF PCE
max
(PCE
ave
a
)(%)
1 : 1 No 0.95 7.64 0.42 3.04 (2.90)
60
C 0.94 9.76 0.52 4.78 (4.61)
2 : 1 No 0.96 11.21 0.60 6.55 (6.27)
60
C 0.95 11.33 0.67 7.25 (7.04)
3 : 1 No 0.97 11.22 0.66 7.19 (6.98)
40
C 0.96 11.27 0.68 7.36 (7.16)
60
C 0.95 11.86 0.70 7.93 (7.79)
80
C 0.95 11.16 0.68 7.21 (7.03)
4 : 1 No 0.98 10.26 0.57 5.98 (5.79)
60
C 0.95 9.16 0.70 11.11 (5.97)
a
The average PCE was obtained from over 10 devices.
Fig. 5 (a) J–V characteristics and (b) EQE curves of the devices based
on DCA3T(T-BDT):PC
61
BM before and after thermal annealing at
60
C with a blend ratio of 3 : 1.
This journal is © The Royal Society of Chemistry 2014 J. Mater. Chem. A,2014,2,15904–15911 | 15907
Paper Journal of Materials Chemistry A
Published on 19 August 2014. Downloaded on 27/10/2014 01:25:25.
View Article Online
blend lms was investigated by X-r ay diffraction (XRD) and
two-dimensional (2D) grazing incidence wide-angle X-ray
scattering (GIWAXS), respectively. As shown in Fig. 7, the
diffraction peak of the annealing lm r emained at the same
position, a nd the intensity was obviously enhanced. B oth of
the lms show an evident diffraction peak (100) at 2q ¼ 4.2
,
corresponding to a d
100
-spacing value of 21.0
˚
A. The second-
order d iffraction peak (200) at 2q ¼ 8. 4
, corresponding to a
d
200
-spacing value of 10.5
˚
A was also clearly observed, indi-
cating the high crystallinity of DC A3T(T-BDT) in the solid
thin lm. The stronger diffract ion peak of the annealing
lm compared to the pristine one reveals that the low
temperature annealing would improve the ordering in the
donor phase.
52,53
The2DGIWAXSpatternsalsoindicatea
well-ordered crystal s tructure throughout the corresponding
thin lm. The high crystallinity could increase the
carrier mobility, which would be b enet for the transport
and collection of the carrier,
54
thus leading to a higher J
sc
and FF.
3. Conclusion
In conclusion, a new low band gap and 2D-conjugated organic
small molecule, DCA3T(T-BDT), was designed and synthesized
for solution-processed organic solar cells. DCA3T(T-BDT)
exhibited good solubility, thermal stability, crystallinity and a
deeper HOMO energy level of 5.37 eV. High efficiency small
molecule organic solar cell with a power conversion efficiency of
7.93%, a high V
oc
of 0.95 V, J
sc
of 11.86 mA cm
2
and FF of 0.70
was obtained by thermal annealing process at only 60
C.
Hence, this encouraging result reveals a prospective strategy for
a new 2D heterocyclic core in constructing donor–acceptor type
small molecules for high performance organic solar cells.
4. Experimental section
4.1 Materials, instruments and characterization
All starting materials and reagents were purchased from
commercial sources and used without further purication,
unless otherwise mentioned. Toluene was distilled over sodium
in the presence of benzophenone to remove water. CHCl
3
and
CH
2
Cl
2
were distilled over calcium hydride (CaH
2
).
1
H and
13
C NMR spectra were obtained on a Bruker Advance
III 600 spectrometer with tetramethylsilane (TMS, d ¼ 0 ppm) as
an internal standard. UV-vis absorption spectra were obtained
on a Hitachi U-4100 spectrophotometer. High resolution mass
spectra were recorded under APCI mode on a Bruker Maxis UHR
TOF spectrometer. TGA and DSC were carried on SDT Q600
simultaneous DSC-TGA instrument and Perkin Elmer diamond
differential scanning calorimeter, respectively. Cyclic voltam-
metry (CV) measurements were taken on a CHI660D electro-
chemical workstation. Moreover, CV experiments were carried
out at room temperature with a conventional three-electrode
system using a platinum wire as the counter electrode, a satu-
rated calomel electrode (SCE) as the reference electrode, and a
glassy carbon electrode as the working electrode. A solution of
tetrabutylammonium phosphorus hexauoride (Bu
4
NPF
6
,
0.1 M) in acetonitrile was used as the supporting electrolyte, at a
scan rate of 100 mV s
1
. Ferrocene/ferrocenium (Fc/Fc
+
) was
used as the internal standard (the energy level of Fc/Fc
+
is
4.8 eV under vacuum) and the formal potential of Fc/Fc
+
was
measured as 0.38 V vs. SCE. HPLC-UV was carried on a Waters
1525 high performance liquid chromatography with UV
detection.
Density functional theory (DFT) calculations were performed
using the Gaussian 09 program suite at the B3LYP/6-31G(d)
level to investigate the electronic structure of the molecule in
the gas phase.
55–58
All the alkyl side chains were replaced with
methyl groups to reduce the computational cost. The optimized
molecular structure was conrmed to be in a stable local
minimum of the ground state potential energy surface by
computing the vibrational frequencies at the same level of
theory.
OFET was fabricated using n-type heavily doped Si as the
gate with a 300 nm thermally oxidized SiO
2
layer as the gate
dielectric (the capacitance was 10 nF cm
2
). The lm was
prepared by spin-coating the solution (6 mg mL
1
in CHCl
3
)on
Fig. 6 Tapping mode AFM height images (4 mm 4 mm) of DCA3T(T-
BDT):PC
61
BM blend films with different D/A ratios before (top: a–d)
and after (bottom: e–h) thermal annealing at 60
C.
Fig. 7 XRD patterns of DCA3T(T-BDT) film spin-coated from CHCl
3
solution before and after thermal annealing at 60
C. The insets are the
corresponding 2D GIWAXS images.
15908
| J. Mater. Chem. A,2014,2,15904–15911 This journal is © The Royal Society of Chemistry 2014
Journal of Materials Chemistry A Paper
Published on 19 August 2014. Downloaded on 27/10/2014 01:25:25.
View Article Online