Two-acceptor one-donor random terpolymers comprising thiophene- and phenyl-capped diketopyrrolopyrrole for organic photovoltaics

Article abstract of DOI:10.1039/c8nj03536a

A series of random terpolymers comprising two electron deficient phenyl (PDPP) and thiophene (ThDPP)-capped diketopyrrolopyrrole (DPP) in conjugation with the electron-donating thiophene moiety are synthesised using Stille coupling. Their optical properties, energy levels, hole mobility, crystallinity and solar cell device performance can be systematically fine-tuned by controlling the molar ratio between ThDPP/PDPP (30/70, 50/50, 70/30, and 90/10) contents in the polymer backbone. Herein, we find that the crystalline properties and hole mobility of the terpolymer are enhanced by increasing ThDPP content in the polymer backbone. However, increasing PDPP content leads to low hole mobility and weak crystalline features. These characteristic features afford remarkable effect on the solar cell device performance. Bulk heterojunction (BHJ) solar cells are constructed by using these random terpolymers as donor materials and [6,6]-phenyl-C₇₁-butyric acid methyl ester (PC₇₁BM) as the acceptor. The best device performances are obtained for polymer P5T5P with the ThDPP/PDPP ratio of 50/50 and power conversion efficiency (PCE) of 2.9% due to balanced charge carrier mobility and optimized crystallinity in addition to good miscibility and favorable surface morphology with the fullerene acceptor. This study demonstrates that improved control of the crystallinity of the polymer donor through structural engineering can greatly help in improving device performance.

Full text of DOI:10.1039/c8nj03536a

NJC
View Article Online
View Journal | View Issue
PAPER
Two-acceptor one-donor random terpolymers
comprising thiophene- and phenyl-capped
diketopyrrolopyrrole for organic photovoltaics†
Cite this: New J. Chem., 2018,
42, 20113
ab
abc
ab
B. Sambathkumar,
N. Somanathan
*
E. Varathan,
V. Subramanian
and
ab
*
A series of random terpolymers comprising two electron deficient phenyl (PDPP) and thiophene (ThDPP)-capped
diketopyrrolopyrrole (DPP) in conjugation with the electron-donating thiophene moiety are synthesised using
Stille coupling. Their optical properties, energy levels, hole mobility, crystallinity and solar cell device
performance can be systematically fine-tuned by controlling the molar ratio between ThDPP/PDPP
(30/70, 50/50, 70/30, and 90/10) contents in the polymer backbone. Herein, we find that the crystalline
properties and hole mobility of the terpolymer are enhanced by increasing ThDPP content in the polymer
backbone. However, increasing PDPP content leads to low hole mobility and weak crystalline features. These
characteristic features afford remarkable effect on the solar cell device performance. Bulk heterojunction
(BHJ) solar cells are constructed by using these random terpolymers as donor materials and [6,6]-phenyl-C71-
butyric acid methyl ester (PC71BM) as the acceptor. The best device performances are obtained for polymer
P5T5P with the ThDPP/PDPP ratio of 50/50 and power conversion efficiency (PCE) of 2.9% due to balanced
charge carrier mobility and optimized crystallinity in addition to good miscibility and favorable surface
morphology with the fullerene acceptor. This study demonstrates that improved control of the crystallinity of
the polymer donor through structural engineering can greatly help in improving device performance.
Received 15th July 2018,
Accepted 29th October 2018
DOI: 10.1039/c8nj03536a
rsc.li/njc
low band gap-conjugated polymers consisting of electron-rich
(donor, D) and electron-poor units (acceptor, A) arranged alter-
1. Introduction
There has been significant research interest in organic bulk nately in a polymer chain with a push–pull effect.15–17 One key
heterojunction (BHJ) polymer solar cells (PSCs) as they are challenge for improving PCE in PSCs is to develop new low-band-
flexible and light in weight; the fabrication of large-area devices gap polymers with desired optoelectronic properties. Therefore
using various printing methodologies has led to BHJ as a conjugated polymers should exhibit broad and strong absorption
promising next-generation renewable energy resource.1–4 for high short-circuit current density ( Jsc), suitable HOMO and
In general, BHJs have a photoactive layer made up of a solid LUMO energy levels with fullerene derivatives for efficient exciton
blend of conjugated polymers as the electron donor and full- dissociation for high open-circuit voltage (Voc) and favourable
erene derivatives such as [6,6]-phenyl-C71-butyric acid methyl morphology and high charge transport for high FF. Hence, con-
ester (PC71BM) as the electron acceptor.5–7 This combination jugated polymers have to balance these photovoltaic parameters
provides efficient charge separation and charge transport from for high PCE.18–23 However, only a limited number of D–A
the photo-generated exciton at the donor/acceptor interface.8 copolymers exhibit a high PCE value. In addition, all successful
Recently, the power conversion efficiency (PCE) of PSCs has D–A alternate copolymers have a fixed D–A ratio (1 : 1), which
remarkably increased over 13% in single junctions and 17% in further complicates the modification of the optoelectronic proper-
tandem structures.9–14 This is possibly due to the development of ties and device performance.24,25
Recently, random Terpolymer design has been explored as a
a CSIR-Central Leather Research Institute, Sardar Patel Road, Adyar,
Chennai 600020, Tamil Nadu, India. E-mail: sambaathkumarb@gmail.com
b CSIR-Network of Institutes for Solar Energy, New Delhi, India
c Department of Chemistry, Indian Institute of Technology Madras,
Chennai 600036, Tamil Nadu, India
promising strategy for building ideal donor materials for BHJ
solar cells.26 The attractive advantage of terpolymer design is
that the inclusion of additional comonomer units in the con-
jugated backbone provides complementary light absorption,
which facilitates the harvesting of more photons, fine-tuning the
energy level, band gap, charge transport ability, packing nature,
and controlling solubility, crystallinity and miscibility with
† Electronic supplementary information (ESI) available: This section contains
detailed synthetic procedures for the monomers and copolymers, 1H and 13C NMR,
and MALDI mass spectra. See DOI: 10.1039/c8nj03536a
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018 New J. Chem., 2018, 42, 20113--20122 | 20113

View Article Online
Paper
NJC
fullerene.27–31 Moreover, in random copolymerization, it is possible N-bromosuccinimide (NBS), 2,5-bis(trimethylstannyl)thiophene,
to fine-tune optical and electrical properties by varying donor or tris(dibenzylideneacetone)dipalladium(0), Aliquat, anhydrous
acceptor monomer composition stoichiometrically in the polymer toluene, chloroform-d and 1,1,2,2-tetrachloroethane-d2 were
backbone.32–37 However, the irregular arrangement of comonomer purchased from Aldrich Chemical Company and used without
units in the polymer backbone raises the fundamental question any further purification.
about its molecular packing and charge transport nature.38,39
For instance, Janssen et al. synthesized random and regular
2.2 Instrumentation methods
1H and 13C NMR spectra were recorded on a Bruker 400 MHz
terpolymers by copolymerizing DPP and thienopyrrolodione
(TPD) in an alternate and random manner, and they reported
that regular terpolymers exhibit high device performance
compared to random polymers.40 Therefore selecting a proper
comonomer pair without disturbing its molecular packing and
charge transport is crucial for designing a random terpolymer.
For example, Jo et al. reported a series of 2A/1D-based random
terpolymers by varying the comonomer compositions of Diketo-
pyrrolopyrrole (DPP, A1) and Isoindigo (A2) randomly copolymer-
ized with the thiophene unit in different ratios. They found that
their carrier mobility and PCE was higher than that of the reference
parent alternate regular polymer.41 In addition, the same group
synthesized random terpolymers by tuning the composition of
thiophene- and pyridine-flanked DPP copolymerized with
bithiophene. The compositional terpolymer showed higher
device performance with PCE of 8.11% and high hole mobility
compared to the corresponding parent copolymer.42 Similarly,
Kim et al. reported a series of terpolymers by randomly
copolymerizing thiophene (Th) and selenophene (Se) with
DPP; they systematically studied the effect of Th/Se ratio, and
they found that increasing Se content enhanced the crystalline
nature, which can lead to high mobility with the best PCE of
7.2% obtained for Se/Th (10/90).43
Among various acceptor units, DPP units are widely investigated
due to their high extension coefficient, electron-deficient lactam
group, rigid planar molecular skeleton favouring strong intermole-
cular p–p interaction, absorption in the near-infrared (NIR) region
and high charge transport nature. Moreover, thiophene-flanked
DPP (ThDPP) polymers exhibit low band gaps with absorption in
the near-infrared (NIR) region and show high Jsc and favourable
morphology for charge transport.44–46 Phenyl-flanked DPP (PDPP)
units show medium band gaps with strong absorption in the visible
region and exhibit deep HOMO levels, which favour higher Voc.47–50
Based on the above consideration, here, we synthesized and
characterized a series of random terpolymers comprising both
ThDPP and PDPP units copolymerized randomly with thiophene
units using Stille polymerization. By systematically varying the
composition of ThDPP and PDPP units in a polymer backbone, it
is possible to fine-tune the optical and electrical properties of
terpolymers. A systematic analysis is reported for their structure–
property relationship using optical, electrochemical, thermal, elec-
trical, density functional theory calculations and photovoltaic study.
NMR spectrometer. UV-Visible absorption spectra were
recorded on a Varian Carey 50 Bio UV-Visible spectrophoto-
meter. Elemental analyses for the polymers were performed on
a Euro Vector S.P.A, Euro EA 3000 CHNS Elemental Analyzer.
Cyclic voltammetry (CV) measurements were performed using a
CHI 600D electrochemical workstation. CV measurements were
carried out in 0.1 M tetrabutylammonium hexafluorophosphate
(TBAPF6) acetonitrile solution using a platinum disc electrode
as the working electrode, an Ag/AgCl electrode as the reference
electrode and a platinum wire electrode as the counter elec-
trode. For CV measurements, the polymer was dissolved in
chloroform solution and coated on a platinum disc with a
concentration of 1 mg mLꢀ1. Thermal analysis was carried out
using Mettler TOLEDO TGA/SDTA 851e at a heating rate
of 10 1C minꢀ1. X-ray diffraction (XRD) measurements were
carried out using a Bruker AXS D8 Advance X-ray diffractometer
at a wavelength of (1.5464 Å). The surface morphologies were
observed by atomic force microscopy (AFM) on a Nova 1.0.26
RC1 in semi contact mode with NT-MDT solver software.
2.3 Solar cell and mobility measurement
The ITO-coated glass substrates were cleaned by ultrasonication
sequentially in detergent, deionized water, acetone and isopropyl
alcohol for 15 minutes and dried in an oven at 80 1C for 12 hours.
A thin layer of poly(3,4-ethylenedioxythiophene):poly(styrene sulfo-
nate) (PEDOT:PSS; Clevios P VP Al 4083) as a hole transport layer
was spin-coated at 3000 rpm and dried at 110 1C for 20 minutes.
The blending solution was prepared by taking the polymer and
PCBM in the ratio of 1 : 2 (18 mg mLꢀ1) in chloroform/ortho-
dichlorobenzene (CF/o-DCB; 90 : 10) as the co-solvent mixture. The
blending solution was spin-coated onto the ITO/PEDOT:PSS layer at
800 rpm for 60 s under N2 environment. The active layer thickness
was approximately B100 nm. Finally, 1 nm of LiF and 100 nm
of Al were thermally deposited at a pressure of approximately
1 ꢁ 10ꢀ6 Torr. The effective device area of the cell was 0.08 cm2. The
current density–voltage characterization was studied using a Keithley
2400 source measure unit. The photocurrent was measured using a
Xe lamp source under 1 sun AM 1.5 solar illumination.
The hole-only device was fabricated using a device configu-
ration of ITO/PEDOT:PSS (30 nm)/polymer (50 nm)/Au (100 nm).
Hole mobility was calculated from the dark current density–voltage
( J–V) curve by fitting the space-charge-limited current (SCLC)
model using the Mott–Gurney equation
2
2. Experimental section
2.1 Materials
9
ðVÞ
J ¼ me0er
8
d3
2-Thiophenecarbonitrile, 4-bromobenzonitrile, dimethyl succinate, where J is the current density, d is the film thickness of the active
potassium tert-butoxide, 2-methyl-2-butanol, tri(o-tolyl)phosphine, layer, m is the hole mobility, er is the relative dielectric constant
20114 | New J. Chem., 2018, 42, 20113--20122 This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
NJC
Paper
(assumed as 2.75 for conjugated polymers) of the transport medium, synthetic procedures of the corresponding monomer and
eo is the permittivity of free space (8.854 ꢁ 10ꢀ12 F mꢀ1), d is the terpolymer are discussed in the ESI.† All the terpolymers
thickness of the polymer and V is the effective voltage.
exhibited good solubility in common organic solvents such as
tetrahydrofuran, dichloromethane, chloroform, toluene, chloro-
benzene and 1,2-dichlorobenzene. Gel permeation chromato-
graphy (GPC) was performed to determine the weight-average
molecular weight (Mw), number-average molecular weight (Mn),
and polydispersity index (PDI) in tetrahydrofuran solution relative
to that of the polystyrene standard, and the results are tabulated
in Table 1.
3. Results and discussion
3.1 Synthesis
The chemical structures of the four terpolymers are shown in
Fig. 1, and their synthetic routes are outlined in Scheme 1. The
terpolymers were synthesized by copolymerizing ThDPP, PDPP
and 2,5-bis(trimethylstannyl)thiophene using palladium-catalyzed
Stille polymerization at 100 1C for 24 hours in toluene. The
3.2 Optical properties
different molar feed ratios of comonomers 50 : 50, 70 : 30, 30 : 70 The photophysical properties of random terpolymers were
and 90 : 10 (ThDPP/PDPP) and their respective terpolymers are investigated using UV-Visible (UV-Vis) spectroscopy. Fig. 2 displays
named as P5T5P, P7T3P, P3T7P and P9T1P. The feed ratio of the the absorption spectra of terpolymers in dilute chloroform solution
terpolymer was determined by NMR spectra. The detailed and thin films, and their corresponding optical parameters are
Fig. 1 Chemical structures of random terpolymers P3T7P, P5T5P, P7T3P and P9T1P.
Scheme 1 Synthesis route of monomer and terpolymer.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018 New J. Chem., 2018, 42, 20113--20122 | 20115

View Article Online
Paper
NJC
Table 1 Optical and thermal properties and molecular weight characterization of terpolymers
Eg(opt) (eV) (thin film) Td (1C) Mn (kg molꢀ1
)
Mw (kg molꢀ1
)
Dispersity (Ð)
a
b
c
d
e
f
Polymer lmax (nm) (solution) lmax (nm) (thin film)
P3T7P
P5T5P
P7T3P
P9T1P
346, 399, 563, 639
401, 696
399, 722
367, 403, 580, 641, 697 1.47
463
454
429
431
32.7
92.4
49.4
38.0
103.0
204.1
182.0
111.0
3.14
2.21
3.68
2.92
407, 716
403, 753
410, 823
1.38
1.31
1.30
419, 793
a
b
c
d
Absorption in CHCl3 solution. Absorption for thin films on quartz. Optical band gap calculated from the onset of the thin film. 5% weight
e
f
loss temperature measured by TGA under N2. Number-average molecular weight. Weight-average molecular weight.
summarized in Table 1. In both the solution and the thin film, all temperatures (Td) of terpolymer are summarized in Table 1. The
the terpolymers exhibited two distinct absorption bands. The Td values of polymers P3T7P, P5T5P, P7T3P and P9T1P are 463,
absorption band in the lower wavelength region (350 to 450 nm) 454, 429 and 431 1C, respectively. As determined from the Td value,
corresponded to localized p–p* transition (LT), and the band in the increasing PDPP monomer content on the polymer backbone
higher wavelength region (500–1000 nm) was ascribed to intra- enhances the thermal stability of the terpolymer.53 In addition,
molecular charge transfer (ICT) from the donor to the acceptor. In all the terpolymers exhibit good thermal stability, which is essen-
thin films, lmax for polymer P9T1P was observed at 823 nm, which tial for device fabrication and photovoltaic application.
was B240 nm longer than that of P3T7P (580 nm), B70 nm longer
3.4 Electrochemical properties
than that of P7T3P (753 nm) and B100 nm longer than that of
P5T5P (716 nm). This finding illustrates that increasing the ThDPP
comonomer unit in the polymer backbone results in a red shift in
the long wavelength region. This is probably due to heteroatom
interaction (SꢂꢂꢂO) between the lactam oxygen and the adjacent
sulphur unit in thiophene, facilitating a high degree of coplanar
conformation and electron donating nature of the thiophene unit
in ThDPP when compared to that for PDPP.51 This feature helps
enhance strong intermolecular p–p interactions and highly-ordered
packing, which is the reason for this dramatic red shift in
absorption.52 Furthermore, compared to the solution, in the thin
film, all the terpolymers show broad absorption as well as red
shifts in both absorption maxima and onset, which are due to the
strong aggregating nature in the thin film. From the onset of thin
film absorption, the optical band gaps Eogpt are 1.47, 1.38, 1.31
and 1.30 eV for P3T7P, P5T5P, P7T3P and P9T1P, respectively.
Thus, it is possible to fine-tune the absorption maxima and band
gap by regulating the molar ratio of PDPP and ThDPP units in a
polymer chain.
To study the HOMO, LUMO and electrochemical band gap of
the copolymer, cyclic voltammetry (CV) analysis was performed
on a copolymer coated onto a Pt electrode as the working
electrode in 0.1 M tetrabutylammonium hexafluorophosphate
(Bu4NPF6) solution of acetonitrile as a supporting electrolyte.
For calibration, the ferrocene/ferrocenium ion (Fc/Fc+) redox
couple was used as an external standard with a potential of
0.40 V against the (Ag/Ag+) potential, assuming the redox
potential of Fc/Fc+ to have an absolute energy level of ꢀ4.80 eV
in vacuum. Based on their first onset oxidation potential (Eox)
and the reduction potential (Ered), the corresponding HOMO and
LUMO energy levels of the copolymer were calculated.54 The CV
curves are shown in Fig. 4, and their corresponding HOMO,
LUMO and electrochemical band gap are summarized in
Table 2. The HOMO energy levels of copolymers P3T5P, P5T5P,
P7T3P and P9T1P were ꢀ5.44, ꢀ5.36, 5.31, and ꢀ5.25 eV,
respectively. These values clearly imply that increasing the
phenyl-flanked DPP unit in the polymer backbone lowers the
HOMO value due to the weak electron-donating ability of phenyl
compared to that of the thiophene unit.55
3.3 Thermal properties
The thermogravimetric (TGA) thermograms of this terpolymer
However, there is not much variation in the LUMO energy
are shown in Fig. 3, and the corresponding decomposition levels of all the copolymers, suggesting that LUMO is mainly
Fig. 2 Absorption spectra of terpolymers in (a) dilute chloroform solution (b) thin film.
20116 | New J. Chem., 2018, 42, 20113--20122 This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
NJC
Paper
Table 2 Electrochemical characterization of polymers
Polymer
Eox/Ered
HOMOa (eV)
LUMOa (eV)
Eegc b (eV)
P3T7P
P5T5P
P7T3P
P9T1P
ꢀ1.04/ꢀ0.55
ꢀ0.96/ꢀ0.51
ꢀ0.91/ꢀ0.48
ꢀ0.85/ꢀ0.48
ꢀ5.44
ꢀ5.36
ꢀ5.31
ꢀ5.25
ꢀ3.85
ꢀ3.89
ꢀ3.92
ꢀ3.92
1.59
1.47
1.39
1.33
a
Calculated according to the formula HOMO/LUMO = ꢀe(Eox/red + 4.40)
b
(eV). Eegc = (LUMOꢀHOMO).
polymer backbone, it is possible to adjust the HOMO/LUMO
energy levels and band gaps of copolymers.
3.5 Theoretical calculation
To understand the composition effect of ThDPP/PDPP on the
molecular geometry and electronic properties of copolymers,
three model systems were studied with different arrangements
in monomer sequence. The ground state geometries were opti-
mized using density functional theory (DFT) at the B3LYP/6-31G*
level of theory, as implemented in the Gaussian 09 package.58 All
the optimized structures were characterized by frequency analysis
and were shown to be positive frequencies. To simplify the
simulation, the long alkyl group on the DPP unit was replaced
by a methyl group.59 The optimized geometries, chemical struc-
tures, dihedral angles and electron density contour plots of HOMO
and LUMO are displayed in Fig. 5. It can be noted from Fig. 5 that
for the PDPP-Th-ThDPP model system, the dihedral angle between
PDPP and Th unit is 191, whereas a value of 101 was observed for
Th and ThDPP. Therefore, the interfacial angle between the ThDPP
and PDPP units is B301. Upon adding one more ThDPP unit into
Fig. 3 TGA thermograms of terpolymers recorded at a heating rate of
10 1C minꢀ1 under N2 atmosphere.
controlled by the DPP unit. The LUMO energy levels of
copolymers P3T5P, P5T5P, P7T3P and P9T1P are ꢀ3.85,
ꢀ3.89, ꢀ3.92 and ꢀ3.92 eV, respectively, which are significantly
larger than the LUMO of PCBM (ca. 4.2 eV). These LUMO offsets
(B0.3 eV) provide a sufficient driving force for efficient exciton
dissociation and charge transfer with fullerene derivatives.56
The electrochemical band gaps (Eegc) of the copolymers are
calculated based on whether their LUMO and HOMO difference
is in the range from 1.59 to 1.33 eV, which is slightly larger than
the optical band gap Eogpt probably due to the high exciton
binding energy of the organic conjugated material.57 The above
results reveal that by varying the copolymer composition in the
Fig. 4 Cyclic voltammograms of terpolymers in thin films.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018 New J. Chem., 2018, 42, 20113--20122 | 20117

View Article Online
Paper
NJC
the PDPP-Th-ThDPP core, the dihedral angle between the ThDPP in their lamellar spacing arrangements, the first-order peak for
and PDPP units in the resultant modelled system (PDPP-Th- P9T1P shows high intensity compared to that of the other three
ThDPP-Th-ThDPP) was marginally altered (from B301 to B251). polymers. Similarly, second- and third-order diffraction patterns
Similarly, by adding one more PDPP unit into the PDPP-Th-ThDPP also follow a similar trend. In addition, in the wide-angle region,
core, the dihedral angle between two adjacent PDPP units in the additional diffraction peaks are observed at 2y = 24.001, 23.791,
resultant modelled system (PDPP-Th-PDPP-Th-ThDPP) exhibited a 23.511 and 23.241 for P5T5P, P7T3P, P3T7P and P9T1P, which
high value of B401, which was two times larger than that for two correspond to d spacings of 3.70 Å, 3.73 Å, 3.78 Å and 3.83 Å,
adjacent ThDPP (B201). The preceding discussion clearly reveals indicating p–p stacking between the polymer backbone. The p–p
that increasing the ThDPP unit in the polymer backbone increases stacking distances of the copolymers can be ranked in the order of
molecular planarity in the co-polymer, which facilitates inter- P9T1P o P7T3P o P5T5P o P3T7P, indicating that copolymer
molecular interaction and conjugation strength. These results P9T1P has a shorter p–p distance compared to others. From the
are in good agreement with the results of absorption and XRD XRD peak intensities, we infer that P9T1P displays an intense
analyses. Electron density contour plots (Fig. 5) indicate that for diffraction pattern in both the small-angle and wide-angle regions,
PDPP-Th-ThDPP, the electron density of HOMO is mainly deloca- and the diffraction intensities of the copolymers are ranked in the
lized on Th and ThDPP units with negligible contribution from order of P9T1P 4 P7T3P4 P5T5P 4 P3T7P. Thus, the XRD result
PDPP. On the other hand, the electron density distribution of clearly explains that increasing ThDPP content in the polymer
LUMO is completely delocalized on whole units, favouring LUMO backbone enhances crystalline behaviour and p–p stacking. To
stabilization. These features confirm an intramolecular charge- further understand the nanostructure of the blend films, XRD
transfer transition between ThDPP and PDPP units. Similar analysis is carried out. It is clear from Fig. 6 that intermixing
HOMO and LUMO electron density distributions are observed PC71BM moieties with the polymer disrupts blend crystallization
in PDPP-Th-PDPP-Th-ThDPP, which indicates that the addition due to the effect of PC71BM intercalation. However, their lamellar
of PDPP does not affect the electron density distribution. In case of spacing arrangements and p–p stacking distances are nearly the
the PDPP-Th-ThDPP-Th-ThDPP model system, both HOMO and same as those of pristine terpolymers.
LUMO electron densities are delocalized on ThDPP units.
3.7 Hole mobility
3.6 XRD analysis
To further explore the composition effect of ThDPP/PDPP
To evaluate the effect of the PDPP : ThDPP ratio in a polymer, content in the terpolymers on their electrical performance, hole
we performed lamellar ordering and p–p stacking arrangement mobility measurements were obtained for pristine copolymers
X-ray diffraction (XRD) analysis for the copolymers in thin using the space charge limit current (SCLC) method. The hole
films. Their XRD patterns are shown in Fig. 6. At small angle mobility values of pristine copolymers are calculated from the
values, all the copolymers show first-order, second-order and current density–voltage ( J–V) curves. Fig. 7 depicts their log–log
third-order diffraction peaks at a ratio of 1 : 2 : 3, indicating current ( J–V) plots, and their mobility data are summarized in
long-range-ordered lamellar arrangement.60–62 The first-order Table 3. The hole motilities of P3T7P, P5T5P, P7T3P, P9T1P are
lamellar stacking peak for the terpolymer exists at around 1.9 ꢁ 10ꢀ6, 7.8 ꢁ 10ꢀ6, 2.1 ꢁ 10ꢀ5 and 3.4 ꢁ 10ꢀ5 cmꢀ2 Vꢀ1 sꢀ1
,
2y = 4.6–4.46, which corresponds to the lamellar spacing distance respectively. From the values, it is clear that increasing ThDPP
from 19 to 19.78 Å. Even though there are no significant variations content in the polymer backbone helps increase the mobility.
Fig. 5 Chemical structures and their corresponding dihedral angles (a), optimized geometries (b), pictorial representations of the optimized molecular
orbitals of LUMO (c) and HOMO (d) of the modeled oligomers obtained at the B3LYP/6-31G* level of theory.
20118 | New J. Chem., 2018, 42, 20113--20122 This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
NJC
Paper
Fig. 6 XRD patterns of (a) pristine terpolymers and (b) blend films.
terpolymers, and their photovoltaic parameters are tabulated in
Table 3. The Voc values of the random copolymers are ranked in
the order of P9T1P (0.51V) o P7T3P (0.59 V) o P5T5P (0.60 V) o
P3T7P (0.70 V). The variations observed in the Voc values of the
terpolymers are in good agreement with the HOMO energy level of
the terpolymers since Voc is directly proportional to the HOMO
energy level of the polymer donor and the LUMO energy level of
the PC71BM acceptor unit.63,64 Therefore, polymers with lower
HOMO energy levels show higher Voc. Moreover a nonlinear
composition dependence was observed in the PCE of the
terpolymers from 0.54% to 2.9% in the order of P5T5P (2.9%) 4
P7T3P (1.38%) 4 P9T1P (1.18%) 4 P3T7P (0.54%). Specifically,
P5T5P showed the highest PCE value of 2.9% (Voc = 0.59 V;
Jsc = 12.3 mA cmꢀ2; and FF = 39) among all the terpolymers.
However, polymers with higher ThDPP (P7T3P and P9T1P)
content, which show higher hole mobility, provided poor device
performance. Similarly, P3T7P with higher PDPP content
showed poor PCE probably due to its lower hole mobility and
Fig. 7 Hole mobility of pristine terpolymers in thin films.
Notably, P9T1P shows a higher mobility value compared to their poor crystallinity, which is in accordance with SCLC measure-
analogue copolymer probably due to its stronger crystalline beha- ment and XRD analysis.
viour and more rigid characteristics, which is in good agreement
with the results of XRD and DFT characterizations.
To determine the variation in the current density of the polymer/
PC71BM-based device, external quantum efficiencies (EQEs) were
measured under optimized conditions and are shown in Fig. 8b.
The measured EQE spectra are in good agreement with the optical
absorption and J–V curves. Notably, the P5T5P/PC71BM device
showed a good spectral coverage from 300 to 900 nm with max-
imum EQE values of B63% at 370 nm, B41% at 550 nm and 33%
at 760 nm due to its high Jsc value. P7T3P and P9T1P showed low
EQE values compared to P5T5P, whereas P3T7P showed extremely
low EQE values with poor spectral coverage, indicating their poor
device performance. Thus, higher EQE with broad spectral coverage
3.8 Photovoltaic properties and morphological behaviour
The photovoltaic performances of the terpolymers were investi-
gated by fabricating BHJ PSCs with the conventional device
structure of ITO/PEDOT:PSS (40 nm)/polymer:PC71BM (100 nm)/
LiF (1 nm)/Al (100 nm). The optimal weight ratio of polymer :
PC71BM was 1 : 2 (wt%) with a total concentration of 18 mg mLꢀ1
in a mixed co-solvent of CF : ODCB (9 : 1). Fig. 8 illustrates the J–V
characteristics and external quantum efficiency (EQE) of the
Table 3 Photovoltaic properties of polymer/PC71BM under optimized conditions and hole mobility measurements of the pristine polymer under dark
conditions
Polymer
Voc (V)
Jsc (mA cmꢀ2
)
FF
PCE
mh (cmꢀ2 Vꢀ1 sꢀ1
)
P3T7P
P5T5P
P7T3P
P9T1P
0.71 ꢃ 0.03
0.60 ꢃ 0.02
0.59 ꢃ 0.01
0.51 ꢃ 0.02
4.1 ꢃ 0.2
12.3 ꢃ 0.3
5.2 ꢃ 0.1
6.3 ꢃ 0.1
19 ꢃ 3
39 ꢃ 2
45 ꢃ 1
36 ꢃ 2
0.55 ꢃ 0.1
2.9 ꢃ 0.3
1.9(ꢃ0.5) ꢁ 10ꢀ6
7.8(ꢃ0.4) ꢁ 10ꢀ6
2.1(ꢃ0.1) ꢁ 10ꢀ5
3.4(ꢃ0.3) ꢁ 10ꢀ5
1.38 ꢃ 0.08
1.16 ꢃ 0.15
The weight ratio of polymer/PC71BM was 1 : 2, and the average values (4 devices; 16 individual pixels) are shown with standard deviation.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018 New J. Chem., 2018, 42, 20113--20122 | 20119

View Article Online
Paper
NJC
Fig. 8 Current–voltage (J–V) curves (a) and EQE spectra (b) of the polymer:PC71BM blend under optimized conditions.
promoted P5T5P/PC71BM-based devices for higher Jsc values com- roughness compared to P3T7P and P9T1P; in particular,
pared to others.
P5T5P shows finer phase separation compared to P7T3P.65,66
To deeply examine discrepancies in the photovoltaic perfor- This distinctive morphology favours higher device performance
mances of the polymers/PC71BM blend films, atomic force and Jsc values, whereas P3T7P shows ill-defined morphology
microscopy (AFM) analyses are performed based on their with about B300 nm-sized spherical structures that are
device configuration. The AFM images of the blend films are unfavorable for exciton migration and separation toward the
shown in Fig. 9. The root-mean-square (RMS) surface rough- polymer/fullerene interface, eventually decreasing PCE and
ness values of the P5T5P, P7T3P, P3T7P and P9T1P blend films Jsc (4.1 mA cmꢀ2).67 In contrast, P9T1P shows fibril bundle
are 0.96 nm, 1.89 nm, 6.51 nm and 7.16 nm, respectively. P5T5P with macro-phase separation and is more homogenously dis-
and P7T3P show lower RMS values with smooth surface tributed with a high RMS value probably due to its highly
Fig. 9 AFM height images of (a) P3T7P, (b) P5T5P, (c) P7T3P, and (d) P9T1P blend films under optimized conditions.
20120 | New J. Chem., 2018, 42, 20113--20122 This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
NJC
Paper
3 H. F. Dam, T. R. Andersen, M. V. Madsen, T. K. Mortensen,
M. F. Pedersen, U. Nielsen and F. C. Krebs, Sol. Energy
Mater. Sol. Cells, 2015, 140, 187–192.
crystalline nature, which favours polymer chain aggregation,
resulting in poor miscibility and lower device performance.68
4 S. Berny, N. Blouin, A. Distler, H. J. Egelhaaf, M. Krompiec,
A. Lohr, O. R. Lozman, G. E. Morse, L. Nanson, A. Pron,
T. Sauermann, N. Seidler, S. Tierney, P. Tiwana, M. Wagner
and H. Wilson, Adv. Sci., 2016, 3, 1500342.
5 A. J. Heeger, Adv. Mater., 2014, 26, 10–28.
6 J. Nelson, Mater. Today, 2011, 14, 462–470.
7 G. Li, R. Zhu and Y. Yang, Nat. Photonics, 2012, 6, 153–161.
8 C. M. Proctor, M. Kuik and T.-Q. Nguyen, Prog. Polym. Sci.,
2013, 38, 1941–1960.
9 W. Zhao, S. Li, S. Zhang, X. Liu and J. Hou, Adv. Mater.,
2017, 29, 1604059.
10 Z. Xiao, X. Jia and L. Ding, Sci. Bull., 2017, 62, 1562–1564.
11 H. Li, Z. Xiao, L. Ding and J. Wang, Sci. Bull., 2018, 63,
340–342.
12 J. Zhang, L. Zhu and Z. Wei, Small Methods, 2017, 1, 1700258.
13 X. Che, Y. Li, Y. Qu and S. R. Forrest, Nat. Energy, 2018, 3,
422–427.
14 L. Meng, Y. Zhang, X. Wan, C. Li, X. Zhang, Y. Wang, X. Ke,
Z. Xiao, L. Ding, R. Xia, H.-L. Yip, Y. Cao and Y. Chen,
Science, 2018, 361, 1094–1098.
15 H. Zhou, L. Yang and W. You, Macromolecules, 2012, 45,
607–632.
16 L. Bian, E. Zhu, J. Tang, W. Tang and F. Zhang, Prog. Polym.
Sci., 2012, 37, 1292–1331.
4. Conclusions
In conclusion, a series of random terpolymers comprising two
electron deficient phenyl- and thiophene-capped DPP units of
various compositions were synthesised. By using random
terpolymer strategy, we fine-tune the physical properties of the
terpolymers such as the absorption spectrum, HOMO/LUMO
energy levels, band gap, mobility, crystallinity and morphology,
which are directly related to photovoltaic device performance.
Increasing ThDPP content in the polymer backbone drastically
enhances crystallinity and mobility, which are major factors for
device performance, whereas increasing PDPP causes poor hole
mobility due to its non-planar conformation between the phenyl
and DPP moieties. The DFT studies also reveal that a large
dihedral angle is observed between the phenyl and DPP units; in
addition, the XRD patterns also provide weak crystalline character-
istics for higher PDPP content-based terpolymers. The enhanced
crystallinities and hole mobilities of the terpolymers are ranked in
the order of P9T1P 4 P7T3P 4 P5T5P 4 P3T7P. However, a
non-linear trend is observed in solar cell device performance
since the crystallinity of the terpolymer strongly influences the
morphology; the polymers self-aggregate, leading to poor mis-
cibility with fullerene, which results in macro-phase separation
as the ThDPP content is increased. Therefore, polymer P5T5P
(PDPP : ThDPP = 50 : 50) with optimal device morphology and
balanced hole mobility shows better device performance with
PCE of 2.9% with an open-circuit voltage (Voc) of 0.60 V, a short-
circuit current (Jsc) of 12.3 mA cmꢀ2, and a fill factor (FF) of 39%.
Thus, the random terpolymer approach affords a potential strategy
to optimize the physical properties and device performances of
terpolymers.
17 T. Xu and L. Yu, Mater. Today, 2014, 17, 11–15.
18 Y. Li, Acc. Chem. Res., 2012, 45, 723–733.
19 B. Qi and J. Wang, J. Mater. Chem., 2012, 22, 24315–24325.
20 X. Guo, M. Baumgarten and K. Mu¨llen, Prog. Polym. Sci.,
2013, 38, 1832–1908.
21 L. Dou, Y. Liu, Z. Hong, G. Li and Y. Yang, Chem. Rev., 2015,
115, 12633–12665.
22 C. L. Chochos and S. A. Choulis, Prog. Polym. Sci., 2011, 36,
1326–1414.
Conflicts of interest
23 B. Hemavathi, T. N. Ahipa and R. K. Pai, Eur. Polym. J., 2015,
72, 309–340.
24 R. Po and J. Roncali, J. Mater. Chem. C, 2016, 4, 3677–3685.
25 T. E. Kang, K. H. Kim and B. J. Kim, J. Mater. Chem. A, 2014,
2, 15252–15267.
26 L. Ye, X. Jiao, S. Zhang, H. Yao, Y. Qin, H. Ade and J. Hou,
Adv. Energy Mater., 2016, 1601138.
27 M. L. Keshtov, A. R. Khokhlov, S. A. Kuklin, S. A. Osipov,
N. A. Radychev, M. I. Buzin and G. D. Sharma, Org. Electron.,
2017, 41, 1–8.
28 R. Hou, B. Zhao, F. Wu, G. Wang, T. Shen, H. Guo, J. Zhang,
H. Chen and S. Tan, Org. Electron., 2015, 20, 142–149.
29 S. Ko, R. Mondal, C. Risko, J. K. Lee, S. Hong, M. D.
McGehee, J. L. Bredas and Z. Bao, Macromolecules, 2010,
43, 6685–6698.
30 B. Sambathkumar, P. S. V. Kumar, F. S. Deepakrao, S. S. K.
Iyer, V. Subramanian, R. Datt, V. Gupta, S. Chand and
N. Somanathan, J. Phys. Chem. C, 2016, 120, 26609–26619.
31 S. Goker, G. Hizalan, E. Aktas, S. Kutkan, A. Cirpanabde and
L. Toppare, New J. Chem., 2016, 40, 10455–10464.
There are no conflicts to declare.
Acknowledgements
The authors thank CSIR-NWP 54 for their financial support
through the TAPSUN project. E. V. thanks SERB for NPDF
(PDF/2017/000059). The authors would also like to thank
Dr A. Ajayaghosh, J. D. Sudha, and R. Ramakrishnan (NIIST,
Trivandrum, India) for the AFM measurements. The authors
sincerely thank Prof. Dr R. Damodaran and Mr E. Ramachandran
(Department of Chemistry, IIT Madras) for their timely help.
References
1 L. Dou, J. You, Z. Hong, Z. Xu, G. Li, R. A. Street and Y. Yang,
Adv. Mater., 2013, 25, 6642–6671.
2 M. C. Scharber and N. S. Sariciftci, Prog. Polym. Sci., 2013,
38, 1929–1940.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018 New J. Chem., 2018, 42, 20113--20122 | 20121

View Article Online
Paper
NJC
32 T. E. Kang, H. H. Cho, H. J. Kim, W. Lee, H. Kang and 52 Z. Yi, S. Wang and Y. Liu, Adv. Mater., 2015, 27, 3589–3606.
B. J. Kim, Macromolecules, 2013, 46, 6806–6813.
53 C. Kanimozhi, P. Balraju, G. D. Sharma and S. S. Patil,
33 V. Tamilavan, Y. Liu, R. Agneeswari, D. Y. Lee, S. Cho,
J. Phys. Chem. B, 2010, 114, 3095–3103.
Y. Jin, S. H. Park and M. H. Hyun, Polymer, 2016, 95, 54 P. Shen, H. Bin, L. Xiao and Y. Li, Macromolecules, 2013, 46,
18–25. 9575–9586.
34 T. Jiang, Y. Yang, Y. Tao, C. Fan, L. Xue, Z. Zhangrandom, 55 M. Brumbach, D. Placencia and N. R. Armstrong, J. Phys.
H. Li, Y. Li and W. Huang, Polym. Chem., 2016, 7, 926–932. Chem. C, 2008, 112, 3142–3151.
35 Y. Lim, S.-G. Ihn, X. Bulliard, S. Yun, Y. Chung, Y. Kim, 56 Y. Zhu, R. D. Champion and S. A. Jenekhe, Macromolecules,
H. Chang and Y. S. Choi, Polymer, 2012, 53, 5275–5284. 2006, 39, 8712–8719.
36 L. Ye, X. Jiao, S. Zhang, H. Yao, Y. Qin, H. Ade and J. Hou, 57 C. B. Nielsen, M. Turbiez and I. McCulloch, Adv. Mater.,
Adv. Energy Mater., 2016, 6, 1601138. 2013, 25, 1859–1880.
37 H. Li, F. Liu, X. Wang, C. Gu, P. Wang and H. Fu, Macro- 58 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
molecules, 2013, 46, 9211–9219.
38 Y. T. Chen, T. W. Huang, C. L. Wang and C. S. Hsu, Polym.
Chem., 2015, 6, 3728–3736.
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci and G. A. Petersson, et al., GAUSSIAN 09,
Revision A.02, Gaussian, Inc., Wallingford, CT, 2009.
39 H. Xu, J. Li, J. Mai, T. Xiao, X. Lu and N. Zhao, J. Phys. Chem. 59 M. E. Reish, S. Nam, W. Lee, H. Y. Woo and K. C. A. Gordon,
C, 2014, 118, 5600–5605. J. Phys. Chem. C, 2012, 116, 21255–21266.
40 K. H. Hendriks, G. H. L. Heintges, M. M. Wienk and 60 Y. Li, P. Sonar, S. P. Singh, M. S. Soh, M. V. Meurs and
R. A. J. Janssen, J. Mater. Chem. A, 2014, 2, 17899–17905. J. Tan, J. Am. Chem. Soc., 2011, 133, 2198–2204.
41 J. W. Jung, F. Liu, T. P. Russell and W. H. Jo, Energy Environ. 61 J. S. Ha, K. H. Kim and D. H. Choi, J. Am. Chem. Soc., 2011,
Sci., 2013, 6, 3301–3307. 133, 10364–10367.
42 J. W. Lee, H. Ahn and W. H. Jo, Macromolecules, 2015, 48, 62 M. Shahid, R. S. Ashraf, Z. Huang, A. J. Kronemeijer,
7836–7842.
T. C. Ward, I. McCulloch, J. R. Durrant, H. Sirringhausb
and M. Heeney, J. Mater. Chem., 2012, 22, 12817–12823.
63 M. C. Scharber, D. Mu¨hlbacher, M. Koppe, P. Denk,
C. Waldauf, A. J. Heeger and C. J. Brabec, Adv. Mater.,
2006, 18, 789–794.
43 K. H. Kim, S. Park, H. Yu, H. Kang, I. Song, J. H. Oh and
B. J. Kim, Chem. Mater., 2014, 26, 6963–6970.
44 J. W. Jung, J. W. Jo, E. H. Jung and W. H. Jo, Org. Electron.,
2016, 31, 149–170.
45 W. Li, K. H. Hendriks, M. M. Wienk and R. A. J. Janssen, 64 T. E. Kang, J. Choi, H. H. Cho, S. C. Yoon and B. J. Kim,
Acc. Chem. Res., 2016, 49, 78–85.
Macromolecules, 2016, 49, 2096–2105.
65 H. Choi, S. J. Ko, T. Kim, P. O. Morin, B. Walker, B. H. Lee,
M. Leclerc, J. Y. Kim and A. J. Heeger, Adv. Mater., 2015, 27,
3318–3324.
46 S. Qu and H. Tian, Chem. Commun., 2012, 48, 3039–3051.
47 L. Chen, D. Deng, Y. Nan, M. Shi, P. K. L. Chan and H. Chen,
J. Phys. Chem. C, 2011, 115, 11282–11292.
48 J. Yue, J. Liang, S. Sun, W. Zhong, L. Lan, L. Ying, W. Yang 66 S. K. Son, Y. S. Kim, H. J. Son, M. J. Ko, H. Kim, D. K. Lee,
and Y. Cao, Dyes Pigm., 2015, 123, 64–71.
49 V. Govindan and C. G. Wu, RSC Adv., 2017, 7, 28788–28796.
J. Y. Kim, G. H. Choi, K. Kim and B. Kim, J. Phys. Chem. C,
2014, 118, 2237–2244.
50 J. Kuwabara, N. Takase, T. Yasuda and T. Kanbara, J. Polym. 67 A. Gopal, A. Saeki, M. Ide and S. Seki, ACS Sustainable Chem.
Sci., Part A: Polym. Chem., 2016, 54, 2337–2345. Eng., 2014, 2, 2613–2622.
51 C. B. Nielsen, M. Turbiez and I. McCulloch, Adv. Mater., 68 C. V. Hoven, X. D. Dang, R. C. Coffin, J. Peet, T.-Q. Nguyen
2013, 25, 1859–1880. and G. C. Bazan, Adv. Mater., 2010, 22, E63–E66.
20122 | New J. Chem., 2018, 42, 20113--20122 This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018