Polymerizable C70 derivatives with acrylate functionality for efficient and stable solar cells

Article abstract of DOI:10.1016/j.tet.2019.07.013

Fullerene-based organic solar cells are generally suffering from severe microstructure evolution occurring in their bulk heterojunction active layers and thus are extremely stable. To address it, four polymerizable C₇₀ fullerene derivatives, [6

Full text of DOI:10.1016/j.tet.2019.07.013

Accepted Manuscript
Polymerizable C derivatives with acrylate functionality for efficient and stable solar
70
cells
Yang Bai, Xiang Yao, Jiandong Wang, Jin-Long Wang, Si-Cheng Wu, Shi-Ping Yang,
Wei-Shi Li
PII:
S0040-4020(19)30755-0
https://doi.org/10.1016/j.tet.2019.07.013
TET 30455
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 26 April 2019
Revised Date: 15 June 2019
Accepted Date: 8 July 2019
Please cite this article as: Bai Y, Yao X, Wang J, Wang J-L, Wu S-C, Yang S-P, Li W-S, Polymerizable
C
derivatives with acrylate functionality for efficient and stable solar cells, Tetrahedron (2019), doi:
70
https://doi.org/10.1016/j.tet.2019.07.013.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.
Leave this area blank for abstract info.
Polymerizable C70 derivatives with acrylate
functionality for efficient and stable solar
cells
Yang Bai, Xiang Yao, Jiandong Wang, Jin-Long Wang, Si-Cheng Wu, Shi-Ping Yang,* Wei-Shi Li*

1
ACCEPTED MANUSCRIPT
Tetrahedron
journal homepage: www.elsevier.com
Polymerizable C70 derivatives with acrylate functionality for efficient and stable solar
cells
Yang Bai,a,b Xiang Yao,b Jiandong Wang,b Jin-Long Wang,b Si-Cheng Wu,b Shi-Ping Yang,a,* Wei-Shi
Lib,c*
a Department of Inorganic Chemistry, College of Chemistry and Materials Science, Shanghai Normal University, Shanghai 200234, China. E-mail:
shipingy@shnu.edu.cn
bKey Laboratory of Synthetic and Self-Assembly Chemistry for Organic Functional Molecules, Center for Excellence in Molecular Synthesis, Shanghai Institute
of Organic Chemistry, Chinese Academy of Sciences, 345 Lingling Road, Shanghai 200032, PR China. E-mail: liws@mail.sioc.ac.cn
cEngineering Research Centre of Zhengzhou for High Performance Organic Functional Materials, Zhengzhou Institute of Technology , 6 Yingcai Street, Huiji
District, Zhengzhou 450044, China.
ARTICLE INFO
ABSTRACT
Article history:
Received
Received in revised form
Accepted
Fullerene-based organic solar cells are generally suffering from severe microstructure
evolution occurring in their bulk heterojunction active layers and thus are extremely instable.
To address it, four polymerizable C70 fullerene derivatives, [6,6]-phenyl-C71-ethyl acrylate
(PC71EA), [6,6]-phenyl-C71-propyl acrylate (PC71PrA), [6,6]-phenyl-C71-butyl acrylate
(PC71BA), and [6,6]-phenyl-C71-pentyl acrylate (PC71PeA), have been designed, synthesized,
and investigated. These fullerene compounds have a molecular structure, shape and size very
like the conventional C70 fullerene acceptor, [6,6]-phenyl-C71-butyric acid methyl ester
(PC71BM), and have been found no different in their light absorption, redox potentials, and
frontier orbital energy levels. Using these fullerene acrylates individually as acceptor and
poly(3-hexylthiophene) as donor, organic solar cells have been fabricated and gave optimal
efficiencies ranging from 3.22% to 4.16%, comparable to PC71BM-based reference cells
(4.06%). Owing to their acrylate functionality, these fullerene derivatives can turn into
insoluble upon heating, and thus endow their solar cell devices much better thermostability
than PC71BM-based reference cells. The best one, coming from PC71PeA devices, reported an
optimal efficiency of 4.16%, and maintained 91.7% efficiency after heat treatment at 150 °C
for 35 h. As a sharp contrast, the PC71BM reference cell dropped its optimal efficiency from
4.06% to 0.48% only after 5 h heat treatment. X-ray diffraction, optical and atomic force
microscopy, and space-charge-limited current method have been carried out to understand
active layer structure, morphology, and charge mobility change during heat treatment.
Available online
Keywords:
Organic solar cells
Organic photovoltaic materials
Polymerizable fullerenes
Microstructure evolution
Stability
.
1. Introduction
However, it also complicates the active layer microstructure and
brings a structural evolution problem, an inherent account for
device instability. According to light-to-electricity conversion
mechanism, the idea microstructure of BHJ active layer is a kind
of D/A bicontinuous interpenetrating network with a phase
separation size of tens nanometers.[8] In order to achieve fact, the
reality device fabrications, numerous conditions, including D/A
mixing ratio,[9-10] solvent,[11-13] thermal[14-15] and solvent vapor
annealing,[16-19] and additive engineering,[20-23] should be
optimized thoroughly in the device fabrication. With such
treatments, a number of works achieved a good microstructure
and rewarded with high device performance.[4-6] However, since
there are no specific forces to fix D and A molecules in place
even in optimized structure, they tend to diffuse further during
daily use and evolve into a large phase-separation structure. In
order to tackle it, a strategy has been proposed to use
crosslinkable photovoltaic material as one of active layer
components. After desired microstructure is developed, the
Organic photovoltaics (OPVs) have received considerable
attention for their potential in producing lightweight, low cost
and flexible solar cells.[1-3] In recent years, a big progress has
been achieved in device power conversion efficiency (PCE), in
which the champion record of single junction devices is over
15%,[4-5] while that of tandem cells exceeds 17%.[6] Although
device efficiency needs to be further improved, the OPV
practical application requires more effort on addressing device
stability issues.
Because of low dielectric constants for most organic
materials, the most efficient OPV device structure developed so
far comprises of a bulk-heterojunction (BHJ) active layer formed
by physically blending at least a photovoltaic donor (D) and a
photovoltaic acceptor (A) material.[7] Compared with planar
heterojunction, BHJ can provide a large D/A interface and thus
improve exciton migration and dissociation efficiencies.

2
Tetrahedron
crosslinking reaction is implemented to produce three-
efficiency of 4.12% and 3.78% after heating at 150 °C for 35 h.
However, P3HT:PC71BM devices gave an initial efficiency of
4.06% and dramatically dropped to 0.48% upon the same
heating treatment.
ACCEPTED MANUSCRIPT
dimensionally “fixed” structure, and then to retard or prevent
structure evolution forever. In the past, this strategy has been
mainly applied on photovoltaic donor materials and developed a
variety of crosslinkable donor polymers.[24-26] These polymers
are functionalized with vinyl,[27] azide,[28] bromine,[29] oxetane,[30]
and so on units on their part of side chains and can be
crosslinked upon thermal or light activation. It has been widely
proved that solar cells based on these crosslinkable donor
polymers do have improved their thermostability after their
active layers are cured.
O
O
O
O
n
O
O
O
O
PC61BM
PC71BM
PC61BA
n=1, PC71EA
2, PC71PrA
3, PC71BA
Meanwhile, there is also an interest in developing
crosslinkable/polymerizable acceptor materials, especially for
fullerene-based solar cells. In the field, fullerene derivatives,
particularly [6,6]-phenyl-C61/71-butyric acid methyl ester
(PC61/71BM, Figure 1), are a kind of representative acceptor
materials over the past two decades because of their high
electron-accepting capability, small reorganization energy,
isotropic charge transport with good mobility, and spherical
molecular geometry.[31-32] With continuous effort, the record
power conversion efficiency for fullerene-based single junction
solar cells has exceeded 10%.[33] However, owing to their ball-
like shape, fullerene molecules generally have a high potential to
diffuse and phase segregate from donor component, and
microstructure evolution phenomenon is much severe in
fullerene-based solar cells.[34-35] Therefore, it is highly desirable
to develop fullerene acceptors with crosslinkable or
polymerizable function. In 2005, Drees et al. synthesized an
epoxy-functionalized fullerene derivative and used it to combine
with poly(3-hexylthiophene) (P3HT) in fabrication of solar
cells.[36] The optimized device displayed an initial efficiency of
~2%, but deteriorated within several minutes upon heating at
140 °C. In 2010, PCBSD, a crosslinkable fullerene derivative
bearing two styrene units was developed in Hsu’s group,[37] and
used firstly for fabrication of crosslinked buffer layer between
ZnO and BHJ active layers and later as an acceptor additive to
P3HT:PC61BM BHJ active layer.[37-38] After crosslinking was
performed, both cases demonstrated improved device stability.
However, PCBSD has not been used as the main acceptor
component in solar cell fabrication. After that, several
polymerizable fullerenes functionalized with vinyl,[39] azido,[40]
or benzocyclobutene units[41] have been reported. But as a sole
acceptor component for BHJ active layer, none of them can
compete with PC61BM. Looking at molecular design strategies
for the above crosslinkable/polymerizable fullerene molecules,
one may find that all these compounds were synthesized by
hydrolyzing PC61BM into carboxylic acid intermediate firstly
and then reacting with corresponding polymerizable moiety-
containing reagents. More recently, our research group
developed a new polymerizable fullerene derivative, PC61BA
(Figure 1), bearing an acrylate functionality.[42] Since it has
similar molecular shape and size like PC61BM, the optimized
P3HT:PC61BA device exhibited an efficiency comparable to that
of P3HT:PC61BM (3.54% vs. 3.70%), but displayed much
enhanced device thermostability. Considering C70 derivatives
generally have stronger light absorption properties and higher
photovoltaic performance than C60 compounds, we used C70 in
place of C60 in this work and synthesized a series of new
polymerizable fullerene derivatives, [6,6]-phenyl-C71-ethyl
acrylate (PC71EA), [6,6]-phenyl-C71-propyl acrylate (PC71PrA),
[6,6]-phenyl-C71-butyl acrylate (PC71BA), and [6,6]-phenyl-C71-
pentyl acrylate (PC71PeA) (Figure 1) with different methylene
lengths in their side chains. Compared with PC71BM, these C70
acrylates displayed similar performance in term of device
efficiency, but with substantially enhanced device stability. The
best one, P3HT:PC71PeA-based devices, reported an as-prepared
4, PC71PeA
Figure 1. Chemical structures of PC61BM, PC71BM, PC61BA,
PC71EA, PC71PrA, PC71BA, and PC71PeA.
2. Results and Discussion
2.1. Synthesis and thermal properties
The synthetic route for PC71EA, PC71PrA, PC71BA, and
PC71PeA is outlined in Scheme 1. Compound 1-n (n = 1, 2, 3,
and 4) were prepared according to the literature-reported
methods using 4-phenyl-1,3-dioxane, γ-butyrolactone, δ-
valerolactone, and 6-hexanolactone as starting material,
respectively.[42-43] Then, compound 1-n were refluxed with p-
toluenesulfonyl hydrazide in methanol, affording compound 2-n
(n = 1, 2, 3, and 4).[42] After treating with MeONa in pyridine
and reacting with C70 in o-dichlorobenzene (o-DCB), compound
2-n were successfully converted into fullerene hydroxyl
intermediates, 3-n. Finally, target PC71EA, PC71PrA, PC71BA,
and PC71PeA products were obtained by esterification of
compound 3-n with acryloyl chloride in the presence of
triethylamine. After characterized by 1H and 13C NMR and
MALDI-TOF mass spectroscopies as shown in Experimental
and Figure S1-S8, the molecular structures of these fullerene
acrylates were unambiguously identified. It was found that these
C70 acrylates have good solubility in chlorinated organic
solvents, such as chloroform (CF), chlorobenzene (CB), and o-
DCB.
O
H2O2
O
S
O
O
NH
N
O
HBr,H2O,DCM
r.t.
OH
TsNHNH2
OH
n
O
ether
-78°C
Li
n
methanol
reflux
O
n= 1, 1-1
2, 1-2
n= 1, 2-1
2, 2-2
3, 2-3
CH2
m
3, 1-3
4, 1-4
4, 2-4
m=3,4,5
O
OH
n
O
n
O
Cl
C70
TEA
NaOCH3
pyridine, o-DCB
chloroform
r.t.
reflux
n= 1, 3-1
n=1, PC71EA
2, PC71PrA
3, PC71BA
2, 3-2
3, 3-3
4, 3-4
4, PC71PeA
Scheme 1. The synthetic route for PC71EA, PC71PrA, PC71BA,
and PC71PeA.
The thermal properties of PC71EA, PC71PrA, PC71BA, and
PC71PeA were investigated by thermogravimetry (TGA) and
differential scanning calorimetry (DSC) (Figure 2). From their
TGA curves, we found PC71EA, PC71BA and PC71PeA are very
thermally stable, having only 1% weight loss before 400 °C and

3
indicating the occurrence of the polymerization. Expectedly, the
a Td,5% of 440 °C. However, the TGA curve oAf PCCC71PErAPhTaEd aDn MANUSCRIPT
additional weight loss step during the temperature range of
160~250 °C. But its assignment and reason are unkown at
present. In the first DSC heating traces, all these compounds
exhibited a very broad exothermic peak in the temperature range
of 200-300 °C with an onset of 230, 227, 210 and 220 °C for
PC71EA, PC71PrA, PC71BA and PC71PeA, respectively.
However, such exothermic peaks did not appear in the second
heating traces for all these acrylated fullerenes, illustrating the
acrylate functional groups underwent polymerization reaction in
the first heating procedure. In order to get insight into
functionality change upon heat treatment, FT-IR spectroscopy
was performed on PC71PeA samples before and after heat
treatment at 250 °C for 1 h under N2 atmosphere. As shown in
Figure S9, the pristine PC71PeA sample displayed two IR peaks
at 1638 and 1595 cm-1, which originate from stretching vibration
of the double bond in acrylate. After heat treatment, the sample
became insoluble in CB, CF and o-DCB and afforded a IR
spectrum without the above two peaks. These observation clearly
prove heat treatment afforded the polymerization of acrylate
fullerene into insoluble polymers or oligomers.
insoluble fraction increased along with the raising of heating
temperature and time. When heating at 150 °C, the remained
fraction increased from 7~15% at 0.5 h, to 33~42% at 1 h, and to
53~60% at 3 h.
(a)
60
PC71EA
PC71PrA
50
PC71BA
PC71PeA
40
30
20
10
0
80
100
120
140
160
180
Temperature (oC)
(b)
60
50
40
30
20
10
0
(a)
100
95
90
PC71EA
PC71PrA
PC71BA
PC71PeA
85
PC71EA
PC71PrA
80
PC71BA
0.5
1.0
1.5
2.0
2.5
3.0
PC71PeA
75
70
Time (h)
Figure 3. The remained fractions on glass substrates of PC71EA,
PC71PrA, PC71BA, and PC71PeA films upon heating (a) at
differernt temperature for 1 h, or (b) at 150 oC for different times
and followed by chloroform washing.
200
400
600
800
1000
Temperature (oC)
(b)
PC71EA
Solid line : 1st cycle
Dash line : 2nd cycle
Cooling
PC71PrA
PC71BA
PC71PeA
2.2. Optical and electrochemical properties
2
1
Figure 4a displays UV-vis absorption spectra of the
synthesized C70 acrylates in CB solution with a concentration of
10-4 M and in film state together with those of PC71BM. From
these spectra, it can be seen that the synthesized fullerene
acrylates showed similar UV-vis absorption spectra as PC71BM
in both solution and film state. All of them possess a couple of
absorption bands with peaks centered at 358, 375, 465, and 540
nm in solution, and 380, 476, and 560 nm in film state.
Compared with solution ones, the film absorption peaks had
around 20 nm red shift, suggesting the intense π−π interaction in
film state. Moreover, the absorption onsets in film state for these
acrylated fullerenes appeared at 735 nm, illustrating they have an
optical bandgap of 1.69 eV.
0
-1
-2
Heating
0
50
100
150
200
250
300
Temperature (oC)
Figure 2. (a) TGA profiles and (b) DSC traces of PC71EA,
PC71PrA, PC71BA, and PC71PeA with a temperature ramping rate
of 10 °C min-1 under a N2 atmosphere.
Electrochemical cyclic voltammetry (CV) was used to study
molecular frontier orbital energy levels of the materials. As
shown in Figure 4b, the onset reduction and oxidation potentials
(vs. Ag/Ag+) were measured to be -0.90 and 0.74 V for PC71EA,
-0.87 and 0.72 V for PC71PrA, -0.86 and 0.71 V for PC71BA, and
-0.85 and 0.67 V for PC71PeA, respectively. In comparison,
PC71BM displayed an onset reduction and oxidation potentials at
-0.85 and 0.70 V under the same conditions. Based on these
data, the energy levels of the highest occupied molecular orbitals
(HOMOs) for these fullerene compounds were calculated to be -
5.37~-5.30 eV, while those of lowest unoccupied molecular
orbitals (LUMOs) to be -3.78~-3.73 eV (Figure 4c). It is clear
that all these fullerene acrylates have similar electrochemical
properties and molecular energy levels as PC71BM.
Although DSC indicated that the polymerization starts above
200 °C, it may take place at lower temperature because of the
high heating rate (10 °C min-1) in DSC experiments. For
checking it, the film samples of these fullerene acrylates were
prepared by spin-coating from their chloroform solutions onto
glass substrates, heated at desired temperature for certain time,
and then thoroughly washed with chloroform. The remained
fraction on the substrate, which must be insoluble polymers and
oligomers resulted from the polymerization of the test fullerene
acrylate for a certain extent, was determined by means of light-
absorption spectroscopy to calculate A/A0, where A and A0 are
film absorbances after and before chloroform washing. It was
found that even heated at 80 °C for 1 h, all these fullerene
acrylate films had 5~10% turning into insoluble fraction,

4
Tetrahedron
ACCEPTED MANUSCRIPT
PC71BM
PC71EA
PC71BA
PC71BM
PC71PrA
PC71EA
solution
PC71PeA
PC71PrA
PC71BA
PC71PeA
-2
-1
0
1
2
Potential vs Ag/Ag+ (V)
neat film
-3.5
-4
-3.73
-3.78
-3.76
-3.77
-3.78
-4.5
-5
-5.37
-5.35
-5.34
-5.30
300
400
500
600
700
800
-5.33
-5.5
Wavelength (nm)
Figure 4. (a) UV-vis absorption spectra in solution and film state, (b) CV profiles measured in o-DCB/acetonitrile (4/1, v/v) solution,
and (c) energy level diagram of of PC71EA, PC71PrA, PC71BA, and PC71PeA together with those of PC71BM.
2.3. Solar cell fabrication, performance, and stability
currents (JSC) were calculated to be 8.13, 8.23, 8.7, 8.17, and
8.48 mA cm-2 for the devices based on PC71BM, PC71EA,
PC71PrA, PC71BM and PC71PeA. All these data are within 10%
error from those obtained from J-V curves in Figure 5a.
In order to evaluate photovoltaic performance of the
synthesized C70 fullerene acrylates, P3HT was chosen as donor
material and organic solar cells with
a
structure of
ITO/PEDOT:PSS/BHJ active layer/Ca/Al were fabricated. After
checking the influence of each fabrication condition individually
(Table S1-S5), we established the optimal device fabrication
conditions for all these C70 acrylates as the followings: a
donor/acceptor ratio of 1.5/1, solution concentration of 30 mg
mL-1 in o-DCB, using 1000 rpm spin-coating rate, annealing at
150 °C for 30 min. The detail performances for the devices
fabricated under different conditions were summarized in Table
S1-S5 in the Supporting Information, while those of the best
were displayed their current density-voltage (J-V) curves in
Figure 5a with device parameters listed in Table 1. It can be seen
that the performance of the optimal solar cells improved along
with the elongation of molecular side chain in C70 fullerene
acrylates. The optimized PC71EA devices reported an average
efficiency of 3.28% with the best value of 3.32%, while those of
PC71PeA gave an average PCE of 4.02% and the best value of
4.12%. We also optimized P3HT:PC71BM devices and obtained
an average efficiency of 3.93% with the largest value of 4.06%.
Clearly, only PC71PeA devices outperformed PC71BM reference
cells, whereas those of the other C70 fullerene acrylates were
inferior. But even so, they all displayed the efficiencies over
3%, indicating the right design strategy for polymerizable C70
acceptors.
2
0
P3HT : PC71BM
-2
P3HT : PC71EA
P3HT : PC71PrA
-4
P3HT : PC71BA
P3HT : PC71PeA
-6
-8
-10
0.0
0.2
0.4
0.6
Voltage (V)
P3HT : PC71BM
P3HT : PC71PA
P3HT : PC71PrA
P3HT : PC71BA
P3HT : PC71PeA
60
50
40
30
20
10
0
1.0
0.5
0.0
300
400
500
600
700
800
Figure 5b displays the external quantum efficiency (EQE)
spectra of the best devices showed in Figure 5a. All these cells
were found to have photo-response from 300 to 700 nm, which
are consistent with UV-vis absorption spectra of their
corresponding P3HT:C70 fullerene acrylate BHJ films. By means
of integration of EQE value vs. wavelength, the short-circuit
Wavelength (nm)
Figure 5. (a) J-V curves of the optimized devices under the
illumination of AM 1.5 G with a light intensity of 100 mW cm-2,
and (b) their EQE spectra together with UV-vis absorption
spectra
of
corresponding
BHJ
active
layer.

5
Table 1. Device parameters of the optimized solar cells upon heat treatment at 150 °C for 0.5 and 35 h.
ACCEPTED MANUSCRIPT
a
Annealing time
(h)
VOC
JSC
FF
PCE b
(%)
ꢀe
ꢀh
Device
(V)
(mA cm-2)
(%)
(cm2 V-1 s-1)
(cm2 V-1 s-1)
0.5
35
0.63
0.64
0.57
0.59
0.62
0.62
0.62
0.63
0.63
0.64
8.98 (8.13)
1.72
71.26
42.83
63.61
54.95
63.82
62.30
67.08
63.22
68.73
66.21
4.06 (3.93)
0.48 (0.42)
3.32 (3.28)
1.26 (1.23)
3.36 (3.27)
2.65 (2.52)
3.89 (3.69)
2.92 (2.83)
4.12 (4.06)
3.78 (3.73)
3.11 × 10-4
1.23 × 10-5
6.70 × 10-5
1.12 × 10-5
7.58 × 10-5
5.14 × 10-5
2.42 × 10-4
1.72 × 10-4
4.24 × 10-4
3.77 × 10-4
3.80 × 10-3
1.46 × 10-3
5.00 × 10-3
3.31 × 10-3
4.93 × 10-3
2.19 × 10-3
4.58 × 10-3
2.83 × 10-3
5.10 × 10-3
3.27 × 10-3
P3HT:PC71BM
0.5
35
9.10 (8.23)
4.15
P3HT:PC71EA
P3HT:PC71PrA
P3HT:PC71BA
P3HT:PC71PeA
0.5
35
8.50 (8.70)
6.84
0.5
35
9.07 (8.17)
7.31
0.5
35
9.37 (8.48)
8.88
a Data in parenthesis are the calculated JSC values from their EQE spectra. b Data in parenthesis are the average values based on 5 devices.
To evaluate the capability of our designed acceptors in
maintaining device performance under heat stress, we made a
number of half devices without the deposition of top Ca/Al
electrode. These half cells had the same layer structure and
active layer composition and were fabricated under the same
conditions. Then, the half cells were subjected to thermal
treatment at 150 °C for various times, and finally the top Ca/Al
electrodes were deposited to finish the whole device.
Meanwhile, the P3HT:PC71BM reference cells were treated with
the same manner for thermostability investigation. After
characterization, the J-V curves of all the obtained devices with
various heat-treating times are displayed in Figure S10, while the
changes of their device parameters as a function of heating time
are plotted in Figure 6. It can be seen that the P3HT:P71BM
reference cells exhibited an extremely quick deterioration
behavior upon 150 °C heating. Their PCEs dropped from 4.06%
(the best value, which appeared after 0.5 h heat treatment) to
0.48% only after 5 h heat treatment, i.e. maintained only 11.3%
of the best value. In comparison, the devices based on the newly
prepared C70 fullerene acrylates showed more or less
improvement in device stability. Although the PC71EA devices
seemed still unable to endure the heat treatment, their
efficiencies kept about 30% of initial value after 5 h treatment,
larger than the above PC71BM reference cells. In contrast, the
devices based on PC71PrA, PC71BA, and PC71PeA maintained
64.3%, 70.9% and 91.7% efficiencies of their best values,
respectively, even after 35 h heat treatment, indicating greatly
improved device thermostability. It is also implied the existence
of a trend that the device thermostability increases along with the
side chain elongation in C70 fullerene acrylates. In association
with the detail device parameter changes, we found that the VOC
was stable for the systems during the heat treatment, but the JSC
had significant drop for the devices based on PC71BM or
PC71EA. This should be the main account for their device
performance deterioration. Meanwhile, the PC71BM and PC71EA
cells also showed 20~30% decrease in device fill factor (FF)
after heat treatment, which should be another factor contributing
for their device instability.
(a)
(b)
100
80
60
40
20
0
P3HT:PC71PeA
100
80
60
40
20
0
P3HT:PC71BA
P3HT:PC71PrA
P3HT:PC71PeA
P3HT:PC71BA
P3HT:PC71PrA
P3HT:PC71EA
P3HT:PC71BM
P3HT:PC71EA
P3HT:PC71BM
0
5
10
15
20
25
30
35
40
0
5
10
15
20
Time (h)
25
30
35
40
Time (h)
(c)
(d)
P3HT:PC71PeA
100
80
60
40
20
0
P3HT:PC71PeA
100
P3HT:PC71BA
P3HT:PC71PrA
P3HT:PC71BA
P3HT:PC71EA
80
60
40
20
0
P3HT:PC71PrA
P3HT:PC71EA
P3HT:PC71BM
P3HT:PC71BM
0
5
10
15
20
25
30
35
40
0
5
10
15
20
25
30
35
40
Time (h)
Time (h)
Figure 6. The changes of device parameters, including (a) PCE, (b) VOC, (c) JSC, and (d) FF during the heat treatment at 150 oC for the
devices based on PC71BM, PC71EA, PC71PrA, PC71BA and PC71PeA.

6
Tetrahedron
2.4. Film structure, morphology and charge mobility of BHJ
films
broad in the as-prepared films, but turned to intense and sharp
after the films were heated for 0.5 h (Figure 7f). This is not
ACCEPTED MANUSCRIPT
surprising since thermal annealing can generally promote P3HT
crystallization and result in device performance improvement. In
this work, device optimization found all solar cells
unexceptionally showed their best performance just after thermal
annealing at 150 °C for 0.5 h. However, when the active layers
were heated at 150 °C for 35 h, the intensity of this peak
decreased a lot for the BHJ films based on PC71EA, PC71PrA,
and PC71BA, indicating long-term heating leaded to the destroy
of P3HT crystalline domains and thus the deterioration in their
device performance. In comparison, the PC71PeA-based BHJ
film almost maintained its intensity after 35 h heating,
coinciding with its excellent device stability. But in the case of
PC71BM reference film, heat treatment for 35 h resulted in
further increase in the intensity of 5.2 ° diffraction peak,
suggesting a higher crystallization degree for P3HT component.
Since this is conflict with its device performance change, the
instability of the reference cell should originate from other
accounts.
In order to get insight into the above device performance
change upon heat treatment, the film structure, morphology, and
charge mobility of their corresponding BHJ active layers were
studied in detail. Firstly, the neat films of C70 fullerene acrylates
and PC71BM were analyzed by X-ray diffraction (XRD), but
found no detectable peaks (Figure 7a-7e, bottom lines). This
indicates that these fullerene derivatives tend to form an
amorphous film. Then, the film structures of the studied BHJ
active layers were characterized in their as-prepared stage, and
after heating at 150 °C for 0.5 h and for 35 h, and the obtained
XRD profiles are shown in Figure 7a-7e. It can be seen that a
diffraction peak around 5.2 ° appeared in all profiles. Clearly,
the peak originates from the (100) diffraction of a layered chain-
packing structure of P3HT component. The observation indicates
that P3HT polymer adopted a semi-crystalline structure in all
BHJ active layers and different fullerene derivatives did not
hamper its crystallization. Moreover, the peak was weak and
P3HT:PC71EA heated for 35 h
P3HT:PC71EA heated for 0.5 h
P3HT:PC71BM heated for 35 h
P3HT:PC71BM heated for 0.5 h
P3HT:PC71BM for as-prepared
P3HT:PC71EA for as-prepared
PC71EA heated for 0.5 h
PC71BM heated for 0.5 h
5
10
15
2
20
25
30
35
5
10
15
20
25
30
35
θ
(degree)
2
θ
(degree)
P3HT:PC71PrA heated for 35 h
P3HT:PC71PrA heated for 0.5 h
P3HT:PC71BA heated for 35 h
P3HT:PC71BA heated for 0.5 h
P3HT:PC71PrA for as-prepared
PC71PrA heated for 0.5 h
P3HT:PC71BA for as-prepared
PC71BA heated for 0.5 h
5
10
15
20
25
30
35
5
10
15
20
25
30
35
2θ
(degree)
2θ
(degree)
P3HT:PC71BM
P3HT:PC71EA
P3HT:PC71BA
P3HT:PC71PrA
P3HT:PC71PeA
P3HT:PC71PeA heated for 35 h
P3HT:PC71PeA heated for 0.5 h
P3HT:PC71PeA for as-prepared
PC71PeA heated for 0.5 h
0
0.5
35
5
10
15
20
25
30
35
Time (h)
2θ
(degree)
Figure 7. XRD profiles of BHJ films based on (a) PC71BM, (b) PC71EA, (c) PC71PrA, (d) PC71BA, and (e) PC71PeA in as-prepared
stage, and upon heat treatment at 150 °C for 0.5 h and 35 h, and (f) their intensity change for the peak around 5.2 °.
The morphology change of the active layers during heat
treatment was investigated by optical microscopy (OM) and
atomic force microscopy (AFM). As seen from Figure 8, a large
amount of particle-like aggregates appeared in the
P3HT:PC71BM BHJ active layer after 5 h heat treatment. This is
consistent with many literature observations, and illustrates that
the P3HT:PC71BM blend film without crosslinking or
polymerizing function is highly instable and tends to occurrence
of large phase separation. And this should be the main account
for its device performance deterioration. In contrast, all the BHJ
blend films based on newly developed fullerene acrylates did not
observe any significant large aggregates in their OM images
even after 35 h heat treatment. This verifies that the acrylate
functionality enables fullerene derivatives to polymerize into
oligomers or polymers, which have a much lower diffusion
potential than PC71BM, and thus retard phase evolution. In AFM
images (Figure 9), no big difference can be found for C70
fullerene acrylate blend films between heat treatment for 0.5 h

7
and for 35 h. Moreover, the surface root-mean-square (RMS)
roughness was determined to be 1.06, 1.00, 1.07, 1.01, and 1.06
nm for the blend films based on PC71EA, PC71PrA, PC71BA, and
PC71PeA, respectively, upon heat treatment for 0.5 h. These data
changed into 2.21, 1.51, 1.28, and 1.31 nm, respectively, after the
films were heated for 35 h, suggesting the surface roughness did
not change significantly during heating treatment. In contrast,
the PC71BM-based blend film became extremely rough during the
similar heat treatment with a RMS roughness from 1.06 nm for
the film upon heat for 0.5 h to 9.21 nm after heat for 35 h. This is
consistent with the above OM observations, confirming again
morphology change would be the main account for instability of
PC71BM solar cells.
best performance stage and long time annealing stage during the
ACCEPTED MANUSCRIPT
heat treatment (Figure 10 and Table 1). It was found that the
electron mobility for the 0.5 h heat-treated active layer increases
following the order of P3HT:PC71EA (6.70 × 10-5 cm2 V-1 s-1) <
P3HT:PC71PrA (7.58 × 10-5 cm2 V-1 s-1) < P3HT:PC71BA (2.42 ×
10-4 cm2 V-1 s-1) < P3HT:PC71PeA (4.24 × 10-4 cm2 V-1 s-1). But
when compared with the PC71BM reference active layer with an
electron mobility of 3.11 × 10-4 cm2 V-1 s-1, only P3HT:PC71PeA
active layer can compete with. These results coincide with their
performance sequence, which also shows a trend to enhance
along with side chain elongation in C70 fullerene acrylates and an
efficiency exceeding the reference device only for PC71PeA solar
cells. On the other hand, hole mobilities of these 0.5 h heat-
treated active layers were determined to be 3.80 × 10-3, 5.00 × 10-
3, 4.93 × 10-3, 4.58 × 10-3, and 5.10 × 10-3 cm2 V-1 s-1 for those
based on PC71BM, PC71EA, PC71PrA, PC71BA, and PC71PeA,
respectively. These comparable data imply that hole mobility is
not among the factors accounting for their performance
difference. When these active layers were heated at 150 °C for
35 h, their electron mobility were changed into 1.23 × 10-5, 1.12
× 10-5, 5.14 × 10-5, 1.72 × 10-4, and 3.77 × 10-4 cm2 V-1 s-1, while
hole mobility became 1.46 × 10-3, 3.31 × 10-3, 2.19 × 10-3, 2.83 ×
10-3, and 3.27 × 10-3 cm2 V-1 s-1, respectively. Compared with
those of the 0.5 h-treated, the electron mobility decreased 96.0%,
83.3%, 32.2%, 29.0% and 11.1%, while the hole mobility
dropped 61.6%, 33.8%, 55.6%, 38.2% and 35.9%, respectively.
These data reveal that long-time heating treatment resulted in the
decrease in both electron and hole mobility, which would be one
of accounts for their device performance deterioration. Moreover,
a clear trend was found in electron mobility decreasing extent,
which follows a descending order of PC71BM-, PC71EA-,
PC71PrA-, PC71BA-, and PC71PeA-based active layers and is
coinciding with their performance changes. It is also noteworthy
that PC71BM and PC71EA active layers lost their electron
mobility larger than 80%, while the PC71PeA active layer only
had a 11.1% drop. This implies the importance of both the
crosslinking function and the appropriate side chain length in
these C70 acrylates for maintaining their device performances
during heat treatment.
Figure 8. Optical images of P3HT:PC71EA, PC71PrA, PC71BA,
PC71PeA, and PC71BM blend films (1.5/1 in wt %) after at 150 oC
for different times.
3. Conclusions
In this work, four acrylate-functionalized C70 fullerene
derivatives, PC71EA, PC71PrA, PC71BA and PC71PeA, having a
molecular shape and size similar to PC71BM but with different
length of their side chains, has been synthesized and investigated
as photovoltaic acceptor materials for efficient and thermostable
organic solar cells. As compared to PC71BM, these C70 fullerene
acrylates exhibit similar light absorption bands, redox potentials,
and HOMO and LUMO energy levels, but can polymerize upon
heating and turn into insoluble objects. In combination with
P3HT, all these fullerene acrylates demonstrated their capability
as a sole acceptor component for fabrication of efficient and
stable BHJ solar cells. The optimal device efficiencies were
ranging from 3.32% to 4.12%, which are comparable or even
slightly higher than PC71BM-based reference device (4.06%).
Benefiting their polymerizable function, their devices all showed
an enhanced thermostability than PC71BM reference cells.
Moreover, there seems the existence of a trend that both device
efficiency and thermostability increase along with side chain
elongation in their molecular structure. Consequently, the devices
based on P3HT:PC71PeA afforded the highest optimal PCE
(4.12%) and the best thermostability. Therefore, unlike PC71BM
reference cells that maintained only 11.3% efficiency of the best
value upon heat treatment at 150 °C for 5 h, PC71PeA cells
retained over 90% efficiency of their best value even upon heat
treatment for 35 h. The results demonstrate the reasonable design
Figure 9. AFM images (5×5 ꢀm2) of BHJ blend films based on
(a) P3HT:PC71BM, (b) P3HT:PC71EA, (c) P3HT:PC71PrA, (d)
P3HT:PC71BA, and (e) P3HT:PC71PeA upon heat treatment at
150 °C for 0.5 h and 35 h.
Finally, charge mobility of the studied BHJ active layers were
measured by the space-charge-limited-current (SCLC) method
according to the Mott-Gurney equation:[44-45]
9
ꢆꢇ
ꢈꢉ
ꢀ = ꢁꢂεꢃꢄ ꢅ
8
ꢊ
where J is current density, ε0 is the permittivity of free space, εr is
the relative permittivity of the material, µ is the charge mobility,
L is the thickness of the active layer, V is the applied voltage.
Two kinds of devices, electron-only and hole-only with the
structure of ITO/ ZnO/ active layer/ Ca/ Al and
ITO/PEDOT:PSS/active layer/Au respectively, were fabricated.
And two series active layers, one upon heat treatment at 150 °C
for 0.5 h and the other upon heat treatment at 150 °C for 35 h,
were chosen for the investigation of charge mobility at device

8
Tetrahedron
of these fullerene acceptors. Further applicaAtioCn CofEtPheTseECD70 MANUSCRIPT
fullerene acrylates in combination with other high performance
donor materials for fabrication of high efficient and stable solar
cells is ongoing.
Figure 10. J1/2-Va characteristics for the (a, c) electron-only and (b, d) hole-only devices upon heat treatment at 150 °C (a, b) for 0.5 h
and (c, d) for 35 h.
UV-ozone treatment for 15 min in order to increase the
hydrophilic nature of the surface and remove any residual organic
contamination. Then, PEDOT:PSS aqueous solution (Heraeus
Clevios PVP. Al 4083) was filtered through a 0.2 ꢀm filter and
spin-coated at a rate of 4000 rpm for 60 s on top of the ITO
electrode, followed by baking at 145 ºC for 15 min in air,
affording PEDOT:PSS film with a thickness of 25 nm. After the
substrate was transferred into a nitrogen-filled glovebox (O2 < 1
ppm, H2O < 1 ppm), the solution of P3HT:tested fullerene
acceptor (1.5/1 wt/wt) in o-dichlorobenzene was spin-coated onto
it at 1000 rpm for 60 s. Before completely drying, the above half-
cell was placed in a covered Petri dish to undergo solvent
annealing for desired time, and then subjected to thermal
treatment at desired temperature for certain time. Finally, a Ca
(10 nm) and then an Al layer (100 nm) were thermally
evaporated at a base pressure of 10-6 mbar to finish the device
fabrication. The active area of the device was 0.04 cm2. The film
thickness was measured by a Veeco Dektak 150 profilometer.
Current density-voltage (J-V) curve was recorded with a Keithley
2420 source meter. Photocurrent was acquired upon irradiation
using an AAA solar simulator (Oriel 94043A, 450 W) with a AM
1.5 G filter. Light intensity was adjusted to 100 mW cm-2 using a
4. Experimental
4.1 Measurements and Characterization
1H and 13C NMR were recorded on a Varian Mercury
spectrometer operating at 400 and 100 MHz, respectively, using
CDCl3 as a solvent and tetramethylsilane as an internal reference.
Matrix-assisted laser desorption/ionization Fourier transform
mass (MALDI FT-MS) spectroscopy was carried out on a
Thermo Fisher Scientific LTQ FT Ultra using 2,5-
dihydroxybenzoic acid (DHB) as a matrix. UV-vis absorption
spectroscopy was performed on
a
Hitachi U-3310
spectrophotometer. IR was recorded on a Thermo Scientific
Nicolet iS5 FT-IR spectrometer. CV measurements were
performed on a CHI 660C instrument with a three-electrode cell
using a glass carbon working electrode, a platinum wire counter
electrode, and
a
Ag/AgNO3 reference electrode in o-
dichlorobenzene/acetonitrile (4/1, v/v) solution containing 0.1 M
Bu4NPF6 electrolyte. The scan rate was 50 mV s-1. TGA was
carried out on a TGA Q500 instrument under N2 with a heating
rate of 10 °C min-1. DSC was performed on a Q2000 modulated
DSC instrument under N2 with a heating and cooling rate of 10
°C min-1. XRD was carried out on a PANanlytical X’ Pert Pro
diffractometer with Cu Kα beam (40 kV, 40 mA) in θ-2θ scans
(0.033 Å step size, 30 s per step). Film morphology was observed
by AFM on a Nanonavi E-Sweep AFM instrument in the tapping
mode.
standard NREL-certified silicon cell.
External quantum
efficiency (EQE) spectra were recorded using a home-made
system consisting of a Oriel Model 74225 monochromator, a
75W Xe lamp, an optical chopper, a lock-in amplifier, and a
NREL-calibrated crystalline silicon cell. For stability checking, a
number of the above half-cells were heated at 150 °C in dark in
the nitrogen glovebox (O2 < 1 ppm, H2O < 1 ppm) for different
times (from 0 to 35 h), and then deposited with the top Ca/Al
electrode to finish the device fabrication.
4.2 Fabrication and Characterization of organic solar cells
The polymer/fullerene solar cells having
a traditional
sandwich structure using an indium tin oxide (ITO)-coated glass
as an anode and a Ca/Al as a cathode were adopted in the work.
Commercial ITO-coated glass substrate was cleaned in
successive ultrasonic baths in detergent, deionized water, acetone
and isopropanol for 15 min each and twice, then subjected to
In order to measure charge mobility in the active layer by the
space-charge-limited-current (SCLC) method, electron-only
devices with a structure of ITO/ZnO (30 nm)/the studied active

9
layer/Ca (10 nm)/Al (100 nm) and hole-only devices with a
structure of ITO/PEDOT:PSS (25 nm)/ the studied active
layer/Au (100 nm), were fabricated following the same method
as organic solar cell preparation. In the fabrication, the ZnO
precursor solution was prepared by dissolving zinc acetate
dihydrate (Zn(CH3COO)2·2H2O, Aldrich, 99.9%, 1 g) and
ethanolamine (NH2CH2CH2OH, Aldrich, 99.5%, 0.28 g) in 2-
methoxyethanol (CH3OCH2CH2OH, Aldrich, 99.8%, 10 mL)
under vigorous stirring for 12 h for the hydrolysis in air. The
ZnO layer was obtained by spin-coating this ZnO precursor
solution at 4000 rpm and baked at 170°C for 20 min. After
device fabrication, the dark current densities were measured by
applying a voltage between -1 and 7 V using a computer-
controlled Keithley 2420 source meter in N2 atmosphere.
143.99, 143.43, 141.75, 141.24, 138.83, 138.22, 136.46, 133.91,
ACCEPTED MANUSCRIPT
132.88, 131.63, 131.22, 130.82, 130.54, 128.84, 128.56, 128.22,
71.11, 69.18, 61.38, 33.36, 30.02. Anal. calcd for C82H12O2 (%):
C, 95.71; H, 1.18; Found: C, 94.76; H, 1.09; MALDI FT-MS:
m/z 1029 (M+).
Compound PC71PrA was synthesized following the same
method as PC71EA but using compound 3-2 in place of 3-1.
Yield:40%. 1H NMR (400 MHz, CDCl3/CS2, ppm): δ 7.91 (d, J =
8 Hz, 2H), 7.53 (t, J = 8 Hz, 2H), 7.40-7.46 (m, 1H), 6.40 (dd, J1
= 20 Hz, J2 = 4 Hz, 1H), 6.08-6.15 (m, 1H), 5.81 (dd, J1 = 12 Hz,
J2 = 4 Hz, 1H), 4.31 (t, J = 8 Hz, 2H), 2.49-2.56 (m, 2H), 2.16-
2.28 (m, 2H). 13C NMR (100 MHz, CDCl3/CS2, ppm): δ 165.69,
155.97, 155.28, 152.29, 152.04, 151.56, 151.27, 151.20, 150.66,
150.59, 149.48, 149.22, 148.60, 148.19, 147.59, 147.54, 146.94,
145.96, 145.67, 144.83, 144.58, 144.20, 144.06, 143.49, 142.80,
141.83, 141.67, 141.47, 140.28, 139.46, 137.90, 137.28, 133.90,
132.91, 131.62, 131.52, 130.86, 130.65, 128.78, 128.44, 128.37,
71.61, 70.00, 63.89, 35.80, 31.55, 26.09. Anal. calcd for
C83H14O2 (%): C, 95.58; H, 1.35; Found: C, 94.43; H, 1.79;
MALDI FT-MS: m/z 1043 (M+).
4.3 Materials and Synthesis
Unless indicated, all the commercial reagents were used as
received without further purification. Reaction solvents were
dehydrated following standard methods and were freshly distilled
prior to use. Compound 1 and 2 were synthesized following the
reported methods.[42-43]
Compound PC71BA was synthesized following the same
method as PC71EA but using compound 3-3 in place of 3-1.
Compound 3-1. Under the protection of argon, NaOMe (81
mg, 1.5 mmol) was added to a mixture of compound 2-1 (413 mg,
1.3 mmol) and pyridine (30 ml), and then stirred for 20 min at
room temperature. Afterward, a solution of C70 (840 mg, 1 mmol)
in 150 ml dry o-dichlorobenzene (o-DCB) was added and the
mixture was heated to reflux for 48 h. After evaporating off the
solvents, the residue was subjected to column chromatography on
silica gel. The eluent was changed from CS2 for separating
unreacted C70, to dichloromethane for compound 3-1, and finally
to chloroform for byproducts. The yield of compound 3-1 was
1
Yield: 45%. H NMR (400 MHz, CDCl3/CS2, ppm): δ 7.89 (d, J
= 8 Hz, 2H), 7.51 (t, J = 8 Hz, 2H), 7.39-7.44 (m, 1H), 6.37 (dd,
J1 = 20 Hz, J2 = 4 Hz, 1H), 6.07-6.15 (m, 1H), 5.81 (dd, J1 = 12
Hz, J2 = 4 Hz, 1H), 4.21 (t, J = 8 Hz, 2H), 2.45-2.47 (m, 2H),
1.85-1.96 (m, 4H). 13C NMR (100 MHz, CDCl3/CS2, ppm): δ
165.61, 156.00, 155.34, 152.05, 151.56, 151.28, 149.21, 148.57,
148.41, 147.60, 146.06, 145.95, 145.79, 144.98, 140.27, 139.44,
138.04, 137.51, 132.92, 131.68, 130.86, 130.43, 128.66, 128.46,
128.31, 71.81, 69.76, 64.15, 36.27, 34.51, 28.86, 23.07. Anal.
calcd for C84H16O2 (%): C, 95.45; H, 1.53; Found: C, 94.56; H,
1.93; MALDI FT-MS: m/z 1056 (M+).
1
45%. H NMR (400 MHz, CDCl3/CS2, ppm): δ 7.95 (d, J = 4 Hz,
2H), 7.53 (t, J = 8 Hz, 2H), 7.45-7.48 (m, 1H), 4.02-4.12 (m, 2H),
2.67-2.75 (m, 2H).
Compound PC71PeA was synthesized following the same
method as PC71EA but using compound 3-4 in place of 3-1.
Compound 3-2 was synthesized by the same method as
compound 3-1 by using compound 2-2 in place of 2-1. Yield:
40%. 1H NMR (400 MHz, CDCl3/CS2, ppm): δ 7.90 (d, J = 8 Hz,
2H), 7.51 (t, J = 8 Hz, 2H), 7.40-7.43 (m, 1H), 3.79 (t, J = 8 Hz,
2H), 2.47-2.52 (m, 2H), 2.04-2.17 (m, 2H).
1
Yield: 45%. H NMR (400 MHz, CDCl3/CS2, ppm): δ 7.87 (d, J
= 8 Hz, 2H), 7.50 (t, J = 8 Hz, 2H), 7.37-7.43 (m, 1H), 6.38 (dd,
J1 = 20 Hz, J2 = 4 Hz, 1H), 6.06-6.13 (m, 1H), 5.78 (dd, J1 = 12
Hz, J2 = 4 Hz, 1H), 4.15 (t, J = 8 Hz, 2H), 2.36-2.47 (m, 2H),
1.51-1.83 (m, 6H). 13C NMR (100 MHz, CDCl3/CS2, ppm): δ
166.46, 155.94, 155.25, 151.46, 141.18, 141.10, 150.56, 150.50,
149.41, 149.37, 149.21, 149.16, 149.10, 148.56, 148.52, 148.44,
148.41, 148.29, 148.04, 147.96, 147.43, 146.83, 146.28, 145.82,
144.98, 144.11, 143.93, 143.63, 143.25, 141.71, 141.39, 140.13,
138.88, 138.02, 137.58, 132.82, 131.59, 130.75, 130.39, 128.49,
128.13, 72.00, 69.88, 64.42, 36.31, 34.77, 29.92, 28.85, 26.30,
26.16. Anal. calcd for C85H18O2 (%): C, 95.32; H, 1.69; Found: C,
95.04; H, 1.88; MALDI FT-MS: m/z 1070 (M+).
Compound 3-3 was synthesized by the same method as
compound 3-1 by using compound 2-3 in place of 2-1. Yield:
38%. 1H NMR (400 MHz, CDCl3/CS2, ppm): δ 7.90 (d, J = 8 Hz,
2H), 7.52 (t, J = 8 Hz, 2H), 7.40-7.45 (m, 1H), 3.73 (t, J = 8 Hz,
2H), 2.43-2.47 (m, 2H), 1.68-1.95 (m, 4H).
Compound 3-4 was synthesized by the same method as
compound 3-1 by using compound 2-4 in place of 2-1. Yield:
46%. 1H NMR (400 MHz, CDCl3/CS2, ppm): δ 7.89 (d, J = 8 Hz,
2H), 7.51 (t, J = 8 Hz, 2H), 7.41-7.45 (m, 1H), 3.66 (t, J = 8 Hz,
2H), 2.40-2.45 (m, 2H), 1.50-1.93 (m, 6H).
Acknowledgments
Compound PC71EA. To a solution of compound 3-1 (390 mg,
0.4 mmol) in dry chloroform (45 ml) was added triethylamine
(2.72 mL, 18.85 mmol). After it was cooled to 0 ºC, acryloyl
chloride (1.64 mL, 20 mmol) was added dropwise. The reaction
mixture was stirred for 4 h at room temperature, and then added
with saturated brine and extracted with chloroform. The
combined organic layers were dried over anhydrous NaSO4 and
concentrated with rotary evaporator. The residue was subjected to
column chromatography using chloroform/petroleum ether (1/1,
v/v) as an eluent to give the compound PC71EA in a yield 46%.
1H NMR (400 MHz, CDCl3/CS2, ppm): δ 7.92 (s, 2H), 7.53 (t, J
= 8 Hz, 2H), 7.44 (t, J = 8 Hz, 1H), 6.51 (d, J = 16 Hz, 1H), 6.18-
6.27 (m 1H), 5.88 (d, J = 8 Hz, 1H), 4.53 (t, J = 8 Hz, 2H), 2.77-
2.90 (m, 2H). 13C NMR (100 MHz, CDCl3/CS2, ppm): δ 165.78,
155.78, 155.23, 151.24, 151.17, 150.93, 150.87, 150.60, 149.49,
149.22, 148.68, 148.55, 148.38, 148.19, 148.01, 147.89, 147.55,
147.47, 147.04, 146.89, 146.35, 146.13, 145.92, 145.63, 144.51,
The work was financially supported by the National Natural
Science Foundation of China (Nos. 21674125 and 51761145043),
the Strategic Priority Research Program of Chinese Academy of
Sciences (No. XDB20020000), and Zhengzhou Institute of
Technology.
References and notes
[1] Günes, S.; Neugebauer, H.; Sariciftci, N. S. Chem. Rev. 2007, 107, 1324-
1338.
[2] Clarke, T. M.; Durrant, J. R. Chem. Rev. 2010, 110, 6736-6767.
[3] Dou, L.; You, J.; Yang, J.; Chen, C. C.; He, Y.; Murase, S.; Moriarty, T.;
Emery, K.; Li, G.; Yang, Y. Nat. Photonics. 2012, 6, 180-185.
[4] Zhang, H.; Yao, H.; Hou, J.; Zhu, J.; Zhang, J.; Li, W.; Yu, R.; Gao, B.;
Zhang, S.; Hou, J. Adv. Mater. 2018, 30, 1800613.
[5] Yuan, J.; Zhang, Y. Q.; Zhou, L. Y.; Peng, H. J.; Zou, Y. P. Joule. 2019, 3,
1-12.

10
Tetrahedron
ACCEPTED MANJU. MSaCter.RChIePmT. A. 2016, 4, 9286-9292.
[7] (a) Ye, H. Y.; Li, W. S.; Chin. S. J. Org. Chem. 2012, 32, 266-283.
(b) Dou, L.; You, J.; Hong, Z. Adv Mater. 2013, 25, 6642-6671.
(c) Lu, L.; Zheng, T.; Wu, Q.; Schneider, A. M.; Zhao, D.; Yu, L. Chem.
Rev. 2015, 115, 12666-12731.
[6] Meng, L. X.; Zhang, Y. M.; Li, C. X.; Zhang, X.; Wang, Y. B.; Ke, X.;
Chen, Y. S. Science. 2018, 361, 1094-1098.
[26] Wu S. C.; Strover L. T.; Yao X.; Chen X. Q.; Xiao W. J.; Liu L. N.;
Wang J. D.; Li W. S. ACS Appl. Mater. 2018, 10, 35430−35440.
[27] Chen, X.; Chen, L.; Chen, Y. W. J. Polym. Sci. 2013, 51, 4156-4166.
[28] Nam, C. Y.; Qin, Y.; Park, Y. S.; Hlaing, H.; Lu, X. H.; Ocko, B. M.;
Black, C. T.; Grubbs, R. B. Macromolecules. 2012, 45, 2338-2347.
[29] Carlé, J. E.; Andreasen, B.; Tromholt, T.; Madsen, M. V.; Norrman, K.;
Jørgensen, M.; Krebs, F. C. J. Mater. Chem. 2012, 22, 24417-24423.
[30] Brotasa, G.; Farinhasa, J.; Ferreira, Q.; Morgado; Charas, J. A. Synth.
Met. 2012, 162, 2052-2058.
[8] Serap, G.; Helmut, N.; Niyazi, S. S. Chem. Rev. 2007, 107, 1324-1338.
[9] Kawano, K.; Adachi, C. Adv. Funct. Mater. 2009, 19, 3934-3940.
[10] Lee, J. H.; Sagawa, T.; Yoshikawa, S. Energy Technol. 2013, 1, 85-93.
[11] Tracey, M. C.; James, R. D.; Chem. Rev. 2010, 110, 6736-6767.
[12] Liu, F.; Gu, Y.; Wang, C.; Zhao, W.; Chen, D.; Briseno, A. L.; Russell,
T. P. Adv. Mater. 2012, 24, 3947-3951.
[31] Zhang, C.; Chen, S.; Xiao, Z.; Zuo, Q.; Ding, L. Org. Lett. 2012, 14,
1508-1511.
[13] Ye, L.; Zhang, S. Q.; Ma, W.; Fan, B. H.; Guo, X.; Huang, Y.; Ade, H.;
Hou, J. H. Adv. Mater. 2012, 24, 6335-6341.
[14] Mihailetchi, V. D.; Xie, H.; Boer, B.; Koster, L. J. A.; Blom, P. W. M.
Adv. Funct. Mater. 2006, 16, 699-708.
[15] Ma, W. L.; Yang, C. Y.; Gong, X.; Lee, K.; Heeger, A. J. Adv. Funct.
Mater. 2005, 15, 1617-1622.
[16] Li, G.; Yao, Y.; Yang, H. C.; Shrotriya, V.; Yang, G. W.; Yang, Y. Adv.
Funct. Mater. 2007, 17, 1636-1644.
[17] Zhou, H. Q.; Zhang, Y.; Seifter, J.; Collins, S. D.; Luo, C.; Bazan, G. C.;
Nguyen, T. Q.; Heeger, A. J. Adv. Mater. 2013, 25, 1646-1652.
[18] Xiao, Z. G.; Yuan, Y. B.; Yang, B.; VanDerslice, J.; Chen, J. H.; Dyck,
O.; Duscher, G.; Huang, J. S. Adv. Mater. 2014, 26, 3068-3075.
[19] Wang, J. L.; Xiao, F.; Yan, J.; Wu, Z.; Liu, K. K.; Chang, Z. F.; Zhang,
R. B.; Chen, H.; Wu, H. B.; Cao, Y. Adv. Funct. Mater. 2016, 26, 1803-
1812.
[32] a) He, D.; Du, X.; Xiao, Z.; Ding, L. Org. Lett. 2014, 16, 612-615.
b) Xiao, Z.; Geng, X.; He, D.; Jia, X.; Ding, L. Energ. Environ. Sci.
2016, 9, 2114-2121.
[33] Goh, T.; Huang, J. S.; Yager, K. G.; Sfeir, M. Y.; Nam, C. Y.; Hazari, N.;
Taylor, A. D. Adv. Energy Mater. 2016, 6, 1600660
[34] Guillaume, W.; Lionel, D.; Olivier, D.; Agnès, R.; Piétrick, H.; Christine,
D. L. Polym. Int. 2014, 63, 1346-1361.
[35] Rumer, J. W.; McCulloch, I. Mater. Today. 2015, 18, 425-435.
[36] Dress, M.; Hoppe, H.; Winder, C.; Neugebauer, H.; Sariciftci, N.;
Schwinger, W. J. Mater. Chem. 2005, 15, 5158–5163.
[37] Hsieh, C. H.; Cheng, Y. J.; Li, P. J.; Chen, C. H.; Dubosc, M.; Liang, R.
M.; Hsu C. S. J. Am. Chem. Soc. 2010, 132, 4887-4893.
[38] Cheng, Y. J.; Hsieh, C. H.; Li, P. J.; Hsu, C. S. Adv. Funct. Mater. 2011,
21, 1723-1732.
[39] Chen L, Lia X, Chen Y. W. Polym. Chem. 2013, 4, 5637-5644.
[40] Diacon A, Derue L, Lecourtier C, Dautel O, Wantz G, Hudhomme P. J.
Mater. Chem. C. 2014, 2, 7163-7167.
[20] Peet, J.; Kim, J. Y.; Coates, N. E.; Ma, W. L.; Moses, D.; Heeger, A. J.;
Bazan, G. C. Nat. Mater. 2007, 6, 497-500.
[21] Sun, Y. M.; Welch, G. C.; Leong, W. L.; Takacs, C. J.; Bazan, G. C.;
Heeger, A. J. Nat. Mater. 2012, 11, 144-48.
[22] Wang, E. G.; Ma, Z. F.; Zhang, Z.; Vandewal, K.; Henriksson, P.; Ingan,
O.; Zhang, F. L.; Andersson, M. R. J. Am. Chem. Soc. 2011, 133, 14244-
14247.
[23] Li, W. W.; Roelofs, W. S. C.; Turbiez, M.; Wienk, M. M.; Janssen, R. A.
Adv. Mater. 2014, 26, 3304-3309.
[24] Griffini G.; Douglas J. D.; Piliego C.; Holcombe T. W.; Turri S.; Fréchet
J. M. J.; Mynar J. L. Adv. Mater. 2011, 23, 1660-1664.
[41] Deb N, Dasari R. R, Moudgil K, Hernandez J, Marder S. R, Sun Y,
Karimd A, Bucknal D. G. J. Mater. Chem. A. 2015, 3, 21856-21863.
[42] Wang, J. L.; Chen, X. Q.; Yao, X.; Wang, H. Y.; Li, W. S. Tetrahedron
Lett. 2017, 58, 2695-2699.
[43] Kazuhiro Watanabe, et al. Process for producing a beta-ketoalcohol: US,
5260486 [P]. 1993-11-09.
[44] Zhao, G.; He, Y.; Xu, Z.; Hou, J.; Zhang, M.; Min, J.; Chen, H. Y.; Ye,
M.; Hong, Z.; Yang, Y.; Li, Y. Adv. Funct. Mater. 2010, 20, 1480-1487.
[45] Goh, C.; Kline, R. J.; McGehee, M. D.; Kadnikova, E. N.; Fréchet, J. M.
J. Appl. Phys. Lett. 2005, 86, 122110.
[25] Chen X. Q.; Yao X.; Xiang X.; Liang L.; Shao W.; Zhao F. G.; Li W. S.

ACCEPTED MANUSCRIPT
Highlights


A series of polymerizable C70 fullerene acrylates with different methylene lengths were
designed and synthesized.
These C70 fullerene acrylates have a molecular shape and size like PC71BM and have no
difference in their light absorption, redox potential, band gap, HOMO and LUMO energy
levels.

Polymer solar cells based on these C70 fullerene acrylates exhibited similar efficiency to those
based on P3HT/PC71BM, but the former cells possessed much better thermal stability than the
latter.