Solution-processed benzotrithiophene-based donor molecules for efficient bulk heterojunction solar cells

Solution-processed benzotrithiophene-based donor

molecules for e

fficient bulk heterojunction solar cells†

Dhananjaya Patra,aChao-Cheng Chiang,aWei-An Chen,bKung-Hwa Wei,c Meng-Chyi Wuband Chih-Wei Chu*ad

In this study we used convergent syntheses to prepare two novel acceptor–donor–acceptor (A–D–A) small

molecules (BT4OT, BT6OT), each containing an electron-rich benzotrithiophene (BT) unit as the core, flanked by octylthiophene units, and end-capped with electron-deficient cyanoacetate units. The

number of octylthiophene units affected the optical, electrochemical, morphological, and photovoltaic

properties of BT4OT and BT6OT. Moreover, BT4OT and BT6OT possess low-energy highest occupied molecular orbitals (HOMOs), providing them with good air stability and their bulk heterojunction (BHJ)

photovoltaic devices with high open-circuit voltages (Voc). A solar cell device containing BT6OT and

[6,6]-phenyl-C71-butyric acid methyl ester (PC71BM) in a 1 : 0.75 ratio (w/w) exhibited a power

conversion efficiency (PCE) of 3.61% with a short-circuit current density (Jsc) of 7.39 mA cm
2, a value of

Vocof 0.88 V, and afill factor (FF) of 56.9%. After adding 0.25 vol% of 1-chloronaphthalene (CN) as a

processing additive during the formation of the blendfilm of BT6OT:PC71BM (1 : 0.75, w/w), the PCE

increased significantly to 5.05% with values of Jscof 9.94 mA cm
2,Vocof 0.86 V, and FF of 59.1% as a

result of suppressed nanophase molecular aggregation.

Introduction

Harvesting solar energy through the use of solution-processable organic solar cells (OSCs) appears to be an effective strategy for controlling global energy issues. During the last decade, OSCs based on polymers have gained much attention because of their low cost, exibility, light weight, and solution processability over large areas.1–4 _{The power conversion efficiencies (PCEs)}

of polymer solar cells (PSCs) have increased signicantly to over 8% as a result of recent developments of a variety of novel electron-donating polymers and device architectures.5,6

Although the PCEs of small-molecule organic solar cells (SMOSCs) are much lower than those of PSCs, a few encour-aging PCEs of greater than 6% have been reported recently, suggesting that SMOSCs might become promising alternatives to PSCs,7–9particularly because of their relatively simple puri-cation, controlled molecular weight distributions, high open-circuit voltages (Voc) and charge carrier mobilities, high purity,

and reproducible solution processing.10–12Nevertheless, several

critical issues will need to be addressed if SMOSCs are tond wider applicability, including improvements in their lm quality,ll factors (FFs), and morphologies, thereby resulting in greater PCEs. For this purpose, innovations in the design of novel molecules are both urgent and challenging.

Planar electron-donating molecules containing alternating repeating units of electron donors (D) and electron acceptors (A) have been used recently in devices exhibiting promising effi-ciencies.10,11Furthermore, such D–A architectures can lower the

material band gap and extend the absorption band toward longer wavelength, while simultaneously allowing effective manipulation of the energy levels of the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO).8,12 _{The planar D–A solution-processed OSC}

materials that have been reported previously have included dithienosilole,13_{benzodithiophene,}14_{triphenylamine,}15

diketo-pyrrolopyrrole,16 _rhodanine,17 _{and cyanoacetate}18–20 _{units as}

their electron-push and electron-pull moieties. Some A–D–A molecules reported by Chen and coworkers, featuring end-capped cyanoacetate acceptors, have been used in devices exhibiting promising PCEs.18–20Recent landmark PCEs of up to 7.4% have been achieved from a planar A–D–A molecule comprising a central donor benzodithiophene unit end-capped with rhodanine acceptors.20 _{Benzotrithiophene (BT), an}

elec-tron-donating candidate that is more planar and more sulfur-rich than benzodithiophene, has been copolymerized recently and with various accepting units to achieve devices with PCEs of up to 5.6%.21,23_{BT contains two freea-positions; its third fused} a

Research Center for Applied Sciences, Academia Sinica, Taipei, Taiwan, ROC. E-mail: gchu@gate.sinica.edu.tw

b_{Institute of Electronic Engineering, National Tsing Hua University, Hsinchu, Taiwan,}

ROC

c_{Department of Materials Science and Engineering, National Chiao-Tung University,}

Hsinchu, Taiwan, ROC

d_{Department of Photonics, National Chiao-Tung University, Hsinchu, Taiwan, ROC}

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c3ta11544e

Cite this: J. Mater. Chem. A, 2013, 1, 7767 Received 18th April 2013 Accepted 7th May 2013 DOI: 10.1039/c3ta11544e www.rsc.org/MaterialsA

Materials Chemistry A

PAPER

Published on 07 May 2013. Downloaded by National Chiao Tung University on 28/04/2014 01:59:56.

View Article Online

thiophene bearing a solubilizing side chain can extend the electron-donating capacity of the conjugated moiety.24_{The BT}

core enhances the degrees of intermolecular p–p stacking, charge transport, and mobility.21–25_{To the best of our}

knowl-edge, there have been no previous reports on BT-based solution-processed SMOSCs.

In this paper we report the synthesis and characterization of two new planar electron-donating molecules, BT4OT and BT6OT, each featuring BT as the donor unit, alkyl cyanoacetates as electron acceptor units, and octylthiophene as the bridge between them (Fig. 1). BT4OT and BT6OT differ in terms of their number of alkyl thiophene units, which affected their photo-physical, electrochemical, and photovoltaic properties. We prepared therst BT-based SMOSC to exhibit excellent photo-voltaic performance including a PCE of 5.05% when we used BT6OT as the electron donor, [6,6]-phenyl-C71-butyric acid

methyl ester (PC71BM) as the electron acceptor, and 0.25% of

1-chloronaphthalene (CN) as the processing additive.

Experimental

Materials

All chemicals and solvents were purchased in reagent grade from Aldrich, ACROS, Fluka, or Lancaster, except Pd(PPh3)4,

which was obtained from Strem Chemical. Tetrahydrofuran (THF), toluene, and diethyl ether were distilled over Na/benzo-phenone; all reagents were used as received.

Measurements and characterization

1_{H and}13_{C NMR spectra were recorded using a Varian Unity 300}

MHz spectrometer and CDCl3; chemical shis are reported as d

values (ppm) relative to an internal tetramethylsilane (TMS) standard. Elemental analyses, determined using a HERAEUS CHN-OS RAPID elemental analyzer, and mass spectra, deter-mined using matrix-assisted laser desorption ionization time-of-ight mass spectrometry (MALDI-TOF MS; Applied Bio-systems, DE-PRO, Texas, USA), were performed in the Academia Sinica mass spectrometry laboratory. UV-Vis absorption spectra were recorded using an HP G1103A apparatus from dilute

solutions in CHCl3 or from solid lms that had been

spin-coated onto a glass substrate from CHCl3 solutions (10 mg

mL
1). Cyclic voltammetry (CV) was performed in 0.1 M tetra-butylammonium hexauorophosphate (TBAPF6) in MeCN at

room temperature using a BAS 100 electrochemical analyzer with a standard three-electrode electrochemical cell operated at a scanning rate of 100 mV s
1. During the CV measurements, the solutions were purged with N2for 30 s. In each case, a C

working electrode was coated with a thin layer of copolymers, a Pt wire was used as the counter electrode, and a Ag wire was used as the quasi-reference electrode; a Ag/AgCl (3 M KCl) electrode served as the reference electrode for all potentials quoted herein. The redox couple of ferrocene/ferrocenium ion (Fc/Fc+) was used as an external standard. The corresponding energy levels of the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) were calcu-lated from the experimental values of Eox/onsetand Ered/onsetfor

the solidlms of BT4OT and BT6OT, which were formed by drop-casting lms at a similar thickness from THF solutions (ca. 5 mg mL
1). The onset potentials were determined from the intersections of two tangents drawn at the rising currents and background currents of the CV measurements. Atomic force microscopy (AFM) images of the thinlms (on glass substrates) were obtained using a Digital Instruments NS 3a controller and a D3100 stage.

Fabrication of organic solar cells (OSCs)

BHJ solar cells were prepared on a commercially available ITO-coated glass substrate in a sandwiched structure of ITO/ PEDOT:PSS (40 nm)/(small molecule):PCBM (ca. 190 nm)/Ca (50 nm)/Al (90 nm). Prior to device fabrication, ITO-coated glass substrates (1.5 1.5 cm2_{) were ultrasonically cleaned}

sequen-tially in detergent, deionized water, acetone, and isopropyl alcohol. Aer routine solvent cleaning, the ITO substrates were treated with UV ozone for 15 min and then spin-coated with the PEDOT:PSS layer (ca. 30 nm) at 4000 rpm. The active layer solution, BT4OT:PC61BM or BT4OT:PC61BM (8 mg mL
1

for donor materials in CHCl3), was then cast upon the

modi-ed ITO substrate, aer ltering through a 0.45 mm

Fig. 1 Chemical structures of the small molecules BT4OT and BT6OT.

polytetrauoroethylene (PTFE) lter with a spin rate of 7000 rpm, for 1 min; BT6OT was also blended with PC71BM

(8 mg mL
1in CHCl3) at various weight ratios in the presence of

various volumes (vol%) of CN. Finally, layers of Ca (50 nm) and Al (90 nm) were thermally evaporated through a shadow mask at a pressure below 6 10
6torr. All PSC devices were prepared and measured under ambient conditions; the active area of each device was 0.15 cm2. The active layer thickness was measured using an AlphaStep prolometer (Veeo, Dektak 150). Solar cell testing was performed inside a glove box under simulated AM 1.5G irradiation (100 mW cm
2) using a Xe lamp-based solar simulator (Thermal Oriel 1000 W). The EQE action spectrum was obtained under short-circuit conditions. The light source was a 450 W Xe lamp (Oriel Instrument, model 6266) equipped with a water-based IRlter (Oriel Instrument, model 6123NS). The light output from the monochromator (Oriel Instrument, model 74100) was focused onto the PV devices.

Fabrication of hole- and electron-only devices

Hole- and electron-only devices incorporating blend lms of BT6OT and PC71BM [1 : 0.75 (w/w) containing various amounts

of CN (v/v)] were sandwiched between the transparent ITO anode and cathode. The devices were prepared following the same procedure described for the fabrication of the BHJ devices, except that, for the hole-only devices, Ca was replaced with MoO3[work function (F) ¼ 5.3 eV] and, for the

electron-only devices, the PEDOT:PSS layer was replaced with Cs2CO3

[work function (F) ¼ 2.9 eV]. In the hole-only devices, MoO3was

thermally evaporated to a thickness of 20 nm and then capped with 50 nm of Al on top of the active layer. In the electron-only devices, Cs2CO3 was thermally evaporated to a thickness of

approximately 2 nm on top of the transparent ITO. For both

devices, annealing of the active layer was performed at 130C for 20 min. The SCLC method was used to evaluate the hole and electron mobilities of the small-molecule blend lms of BT6OT:PC71BM, at a weight ratio of 1 : 0.75 aer the addition of

various amounts of CN (0.25, 0.50, or 1 vol%), in hole- and electron-only devices. The electron and hole mobilities were determined bytting the plots of the dark current–voltage ( J–V) curves for single-carrier devices to the SCLC model. The dark current was given by

J ¼ 9303rmV2/8L3

where 303r is the permittivity of the polymer, m is the carrier

mobility, and L is the device thickness.

Results and discussion

Synthesis and structural characterization

Scheme 1 presents the synthetic routes toward compounds BT4OT and BT6OT. The syntheses of some of the compounds, performed with slight modications of reported procedures, are described in the ESI.†14,23,25 _{Friedel–Cras acylation of}

2,3-dibromothiophene gave ketone 1; Suzuki–Miyaura cross-coupling of 1 with thiophene-3-boronic acid led to assembly of trithiophene system 2, which underwent subsequent oxidative ring closure with DDQ to form BT core 3. The acyl-functional-ized BT 3 was then converted using the Huang Minlon modi-cation to the alkyl-BT 4. Compound 4 was converted to 5 through deprotonation with n-BuLi followed by addition of tri-methyltin chloride. Compounds 10 and 11 were synthesized through Knoevenagel condensations of 7 and 9, respectively, which were then symmetrically Stille coupled with 5 to obtain BT4OT and BT6OT, respectively. Both BT4OT and BT6OT are

Scheme 1 Synthesis of BT4OT and BT6OT. Reagents and conditions: (i) anhydrous AlCl3, C8H17COCl, 0C; (ii) toluene, 3-thiopheneboronic acid, Na2CO3, 90C, Pd(PPh3)4, 24 h; (iii) DDQ, BF3OEt2, CH2Cl2, 0C to RT, Zn, MeOH; (iv) NH2NH2$H2O, KOH, ethylene glycol; (v) THF,
78_{C, SnMe3Cl; (vi)}

trimethyl(4-octylthien-2-yl)-stannane, Pd(PPh3)4, toluene, reflux; (vii) NBS, CHCl3–AcOH (1 : 1), 0_{C then RT overnight; (viii) CHCl3, octylcyanoacetate, RT; (ix) Pd(PPh3)4, toluene, reflux.}

soluble in common organic solvents, including CH2Cl2, CHCl3,

THF, chlorobenzene (CB), and dichlorobenzene. Optical properties

We investigated the photophysical characteristics of BT4OT and BT6OT by recording their UV-Vis absorption spectra from CHCl3 solutions as well as solid lms on glass substrates

(Fig. 2); Table 1 lists the normalized absorption maxima and optical band gaps (Eoptg ) for both the solutions and solidlms.

In solution, BT4OT experienced two distinct absorptions with maximum wavelengths of 373 and 483 nm, respectively. The former is due to p–p* transitions and the latter due to intramolecular charge transfer (ICT) between the donor ben-zotrithiophene (BT) moiety and the acceptor cyanoacetate moieties.23,25 _{For BT6OT, in contrast, overlap of the} p–p*

transition and ICT bands led to the appearance of a new broad band at 466 nm covering more of the visible region.19,22

The attachment of two lateral alkyl thiophene units resulted in an 18 nm blue shi in the absorptions of BT6OT in solu-tion, relative to BT4OT, presumably because of increased disorder in the conjugated system, due to steric effects of the alkyl side chains.26,27 _{For their solid lms, the absorption}

maxima of BT4OT and BT6OT appeared at 569 and 561 nm, red-shied by approximately 86 and 96 nm, respectively,

relative to those in solution, attributable to the highly coplanar structures of the BT units and corresponding strong p–p interchain interactions.17–20,28 _{The optical band gaps}

calculated from the onsets of absorptions of BT4OT and BT6OT in the solid state were 1.75 and 1.72 eV, respectively; these almost-identical values presumably resulted from the similar electron-withdrawing abilities of the two terminal cyanoacetate moieties.

Electrochemical properties

We used cyclic voltammetry (CV) to investigate the redox behavior of BT4OT and BT6OT and their electronic states (i.e., HOMO/LUMO levels). Fig. 3 displays the oxidation and reduc-tion cyclic voltammograms of BT4OT and BT6OT; Table 1 summarizes the electrochemical properties of BT4OT and BT6OT, including their onset potentials for oxidation (Eonsetox )

and reduction (Eonset

red ) and their electrochemical band gaps.

BT4OT and BT6OT both underwent oxidation (p-doping), due to their electron-rich BT units, in the positive potential range and reduction (n-doping), and due to their electron-poor cya-noacetate units, in the negative potential range.27_{The values of}

Eonsetox and Eonsetred were +1.11 and
0.74 V, respectively, for BT4OT

and +1.06 and
0.78 V, respectively, for BT6OT. We estimated the HOMO and LUMO energy levels from the oxidation and

Fig. 2 Normalized UV-Vis absorption spectra of BT4OT and BT6OT as dilute solutions in CHCl3and as thinfilms on glass surfaces at room temperature.

Table 1 Optical and electrochemical properties of BT4OT and BT6OT

Small molecule

Solutiona _Solidlmb

Energy levels Band gapsd

lmax,abs(nm) lmax,abs(nm) Eoxonset(V)/HOMOc(eV) Eredonset(V)/LUMOc(eV) Eecg (eV) Eoptg (eV)

TT4OT 380, 483 396, 569, 618 1.11/
5.48
0.74/
3.61 1.85 1.75

TT6OT 465 561 1.06/
5.41
0.78/
3.57 1.84 1.72

a_{Dilute solution in CHCl}

3.bSpin-coated from CHCl3solution onto the glass surface.cEHOMO/ELUMO¼ [
(Eonset
Eonset(FC/FC+_{vs. Ag/Ag}+₎)
4.8] eV, where

4.8 eV is the energy level of ferrocene below the vacuum level and Eonset(FC/FC+_{vs. Ag/Ag}+₎¼ 0.45 eV.dElectrochemical band gap: Eec_g¼ E_ox/onset
E_red/onset;

optical band gap: Eoptg ¼ 1240/ledge.

Fig. 3 Cyclic voltammograms of BT4OT and BT6OT as thinfilms (scan rate: 100 mV s
1).

reduction potentials and the reference energy level for ferrocene (4.8 eV below the vacuum level) according to the equations15,29

EHOMO/LUMO¼ [
(Eonset
Eonset(FC/FC+_{vs. Ag/Ag}+))
4.8] eV

and

band gap¼ Eoxonset
E red

onset (1)

where Eonset(FC/FC+_{vs. Ag/Ag}+₎was equal to 0.45 eV. The calculated HOMO and LUMO energy levels were
5.48 and
3.61 eV, respectively, for BT4OT and
5.41 and
3.57 eV, respectively, for BT6OT. The HOMO and LUMO of BT6OT were slightly higher in energy than those of BT4OT, presumably because of the former featured additional electron-rich alkyl thiophene rings surrounding the BT moieties.30_{Enhancing the}

electron-donating ability on both sides of the BT core decreased the oxidation potential and decreased the electron-withdrawing ability in BT6OT.29,31_{The electrochemical band gaps, calculated}

from the oxidation and reduction potentials, were approxi-mately 1.85 eV for BT4OT and 1.84 eV for BT6OT.

The low HOMO energy levels of BT4OT and BT6OT suggested that we might obtain solar cell devices exhibiting high open-circuit voltages, because the value of Vocis directly proportional

to the difference between the HOMO energy level of the donor and the LUMO energy level of the acceptor (PCBM).29,30_{For the}

exciton binding energy of BT4OT and BT6OT to be overcome to result in efficient electron transfer from the donor to the acceptor, we required the LUMO energy level of the electron donor (BT4OT or BT6OT) to be positioned above the LUMO energy level of the acceptor by at least 0.3 eV.29 _The

electro-chemical band gaps were in good agreement with the optical band gaps; in addition, they were also in a desirable range for use in organic photovoltaic applications.

Photovoltaic properties

Our motivation for designing BT-based small molecules was to investigate their potential applications in bulk heterojunction (BHJ) solar cells; here, we used BT4OT and BT6OT as electron donors and PC61BM or PC71BM as electron acceptors in devices

having the conguration ITO/PEDOT:PSS/small mole-cule:PCBM/Ca/Al. Details of the performance of the BHJ devices incorporating BT4OT and BT6OT are presented in the ESI.† To achieve better performance in photovoltaic devices, CHCl3was

chosen as the solvent to obtain active layers of the blended small molecules with goodlm qualities. Table 2 reveals that the best performance for the photovoltaic device based on BT4OT blended with PC61BM at a weight ratio of 1 : 0.75 (w/w)

in CHCl3was a PCE of 2.98% (Voc¼ 0.88 V; Jsc¼ 6.32 mA cm
2;

FF¼ 53.6); for the device based on BT6OT, it was a PCE of 3.19% (Voc¼ 0.86 V; Jsc¼ 6.78 mA cm
2; FF¼ 54.8%) (see Fig. S6 and

S7, ESI†). The values of Vocwe obtained for both the BT4OT- and

BT6OT-based devices are higher than those reported previously for all other BT-based PCS devices.21–23 _{The presence of two}

additional octylthiophene units in the main chain of BT6OT is the reason for its enhanced value of Jscrelative to that of BT4OT.

We optimized the BT6OT-based device by blending BT6OT with different ratios of PC71BM; Fig. 4 and Table 2 summarize the

results. The highest PCE (3.61%; Voc ¼ 0.88 V; Jsc ¼ 7.39 mA

cm
2; FF¼ 56.9%) was that recorded for the device we prepared from BT6OT blended with PC71BM at a ratio of 1 : 0.75 (w/w).

An additive solvent having a boiling point higher than that of the parent solvent and exhibiting selective solubility for the donor and the acceptor can deeply impact the semiconducting properties of an organic material.10,13,32_{We further optimized}

the device incorporating BT6OT through the addition of various volumes of CN in the active layer to control the nanoscale morphology by minimizing molecular phase aggregation (Fig. 5). Adding 0.25 vol% of CN to the BT6OT:PC71BM lm

Table 2 Photovoltaic properties and electron and hole mobilities of BT4OT and BT6OT

Active layer blend Blend ratio

Additive (vol%) Voc(V) Jsc (mA cm
2) FF % PCE Hole mobility (mh, cm2V
1s
1) Electron mobility (me, cm2V
1s
1) me/mh BT4OT:PC61BM 1 : 0.75 — 0.88 6.32 53.6 2.98 — — — BT6OT:PC61BM 1 : 0.75 — 0.86 6.78 54.8 3.19 — — — BT6OT:PC71BM 1 : 0.50 — 0.88 7.23 54.1 3.44 2.53 10
4 3.71 10
4 1.46 BT6OT:PC71BM 1 : 0.75 — 0.88 7.39 56.9 3.61 2.89 10
4 3.86 10
4 1.33 BT6OT:PC71BM 1 : 1 — 0.85 6.35 50.7 2.74 2.47 10
4 6.51 10
4 2.63 BT6OT:PC71BM 1 : 0.75 0.25 0.86 9.94 59.1 5.05 6.61 10
4 15.11 10
4 2.28 BT6OT:PC71BM 1 : 0.75 0.50 0.85 7.58 54.6 3.52 4.89 10
4 10.03 10
4 2.05 BT6OT:PC71BM 1 : 0.75 0.75 0.76 0.829 30.4 0.23 2.37 10
4 6.28 10
4 2.64

Fig. 4 J–V curves of BHJ solar cell devices incorporating BT6OT:PC71BM blends of various weight ratios, recorded under AM 1.5G irradiation at 100 mW cm
2.

resulted in remarkable enhancements in the value of Jsc(from

7.39 to 9.94 mA cm
2) and the FF, while maintaining a constant value of Voc, leading to the PCE increase from 3.65 to 5.05%. We

suspect that the addition of the solvent additive improved the value of Jscthrough the generation of a greater number of

photo-charge carriers and improved transport through the formation of a favorable lm morphology.32 _{Nevertheless, addition of}

higher fractions of CN (e.g., 0.50 or 0.75 vol%) in the BT6OT:PC71BM blendlm decreased the PCEs and the

photo-voltaic parameters dramatically (Table 2); this behavior presumably resulted from morphological disorder, as evi-denced using atomic force microscopy (AFM).13,32

The surface morphology of the active layer in a solar cell device is also a key parameter affecting its performance.17,18,33

Fig. 6 displays AFM and phase images of the morphologies of the BT6OT:PC71BM (1 : 0.75) blendlms cast without CN and

with the optimized concentration of CN. The AFM image of

BT6OT:PC71BM (1 : 0.75) reveals a smooth surface with little

aggregation between BT6OT and PC71BM, with a

root-mean-square roughness (Rrms) of 1.64 nm. Even fewer aggregated

domains, resulting in a value of Rrmsof 1.54 nm, appeared in the

lm aer addition of 0.25 vol% of CN to BT6OT:PC71BM

(1 : 0.75), leading to a dramatic enhancement in PCE from 3.651 to 5.05%.14_{The PCEs decreased dramatically, however, to 3.52}

and 0.23% aer the addition of 0.50 and 0.75 vol%, respectively, of CN to the BT6OT:PC71BM (1 : 0.75) blend, leading to poorer

morphologies with highly aggregated domains and values of Rrms of 1.95 and 2.08 nm, respectively. These lower values of

Rrms increased the diffusional escape probabilities for the

mobile charge carriers, thereby minimizing their recombina-tion.15_{It has been noted previously from AFM images that lower}

degrees of aggregation of BT6OT and PCBM inhibit charge recombination, while enhanced p–p stacking improves the transport of photogenerated charges, leading to higher values of Jsc.15,27,29

Hole and electron mobilities are key parameters for both material design and device fabrication.15_{Fig. S8 (ESI†) presents}

J–V curves of the hole and electron mobilities (me and mh,

respectively) of the optimized blend BT6OT:PC71BM (1 : 0.75) in

the absence of CN and in the presence of the optimal CN concentration, as measured using the space-charge limited current (SCLC) model. Hole and electron mobilities of the device containing BT6OT:PC71BM (1 : 0.75) were 2.89  10
4

and 3.86 10
4 cm2 V
1 s
1, respectively, providing a me/mh

ratio of 1.33 (Table 2). A dramatic increase of me(3.9 fold) and

mh(2.4 fold) was observed aer we had added 0.25 vol% of the

processing additive active layer BT6OT:PC71BM (1 : 0.75) blend.

Less-balanced charge transport ratios (me/mh¼ 2.47 and 2.64,

see Table 2), with dramatically decreased values of Jsc, resulted

when we prepared devices of higher CN contents (0.5 and 0.75 vol%, respectively) compared to 0.25 vol% of CN me/mh¼ 2.05.

The highest hole mobility of the optimized device with added CN (0.25 vol%) was 6.61  10
4 cm2 _V
1 _s
1_{, which is an}

explanation of the highest value of Jsc(9.97 mA cm
2) for this

device among all of those we tested (containing 0, 0.50, or 0.75 vol% of CN).15,27

Next, we recorded external quantum efficiency (EQE) plots for the devices containing BT6OT:PC71BM (1 : 0.75) lms,

prepared in the absence of CN and in the presence of the optimal amount of CN, under monochromatic light. Fig. 7 reveals that all of the devices exhibited good EQE behavior, with a broad photoresponsive range extending from 350 to 700 nm. The solar cell device containing BT6OT:PC71BM

(1 : 0.75, w/w) with 0.25 vol% of CN provided the highest EQE (ca. 47%) among all the tested devices.15,20 _{The devices}

con-taining BT6OT:PC71BM (1 : 0.75, w/w) with 0 and 0.5 vol% of CN

displayed similar EQEs (ca. 41%), values of Jsc(7.39 and 7.58 mA

cm
2, respectively) and PCEs (3.61 and 3.52%, respectively). A dramatic decrease, however, in the photocurrent of the device containing BT6OT:PC71BM (1 : 0.75, w/w) with 0.75 vol% of CN

resulted in the lowest values of Jscand PCE, due to a low EQE of

approximately 5%. We attribute this poor device performance to a serious degree of material aggregation associated with the decrease in the area of the heterojunction interface. Thus, in

Fig. 5 J–V characteristics of solar cells incorporating BT6OT:PC71BM (1 : 0.75) and various amounts (vol%) of CN, recorded under AM 1.5G irradiation at 100 mW cm
2.

Fig. 6 AFM images of BT6OT:PC71BM (1 : 0.75) blends containing various amounts of CN as a processing additive: (a) 0, (b) 0.25, (c) 0.50, and (d) 1 vol%.

this study, the device containing BT6OT:PC71BM (1 : 0.75, w/w)

with 0.25 vol% of CN exhibited the highest values of Jsc(9.94 mA

cm
2) and PCE (5.05%), due to its greatest EQE response and superior charge transport properties and nanoscale morphology relative to all of the other tested devices.

Conclusions

We have developed a symmetrical synthetic approach toward the novel, planar BT-based A–D–A small molecules, BT4OT and BT6OT, for use in solution-processed BHJ solar cells. The number of octylthiophene units signicantly affected the optical, electrochemical, and photovoltaic properties of BT4OT and BT6OT. A photovoltaic device containing a blend of BT6OT:PC71BM at a weight ratio of 1 : 0.75 provided the highest

PCE (5.05%), with a high value of Voc(0.88 V) and a notable FF

(59.1%), aer we had added 0.25 vol% of CN as a processing additive to the blendlm. Devices incorporating BT4OT and BT6OT exhibited values of Voc higher than those of devices

featuring all other previously reported BT-containing polymer counterparts. These outstanding results suggest that small molecules might soon provide device performances comparable with those of their polymeric counterparts.

Acknowledgements

We thank the National Science Council of Taiwan (NSC 100-2120-M-009-004 and 101-2221-E-001-010), the Academia Sinica Research Project on Nano Science and Technology, and the Thematic Project of Academia Sinica, Taiwan (AS-100-TP-A05) fornancial support.

References

1 A. J. Heeger, Chem. Soc. Rev., 2010, 39, 2354.

2 A. C. Arias, J. D. MacKenzie, I. McCulloch, J. Rivnay and A. Salleo, Chem. Rev., 2010, 110, 3.

3 Y. J. Cheng, S. H. Yang and C. S. Hsu, Chem. Rev., 2009, 109, 5868.

4 X. Zhao and X. Zhan, Chem. Soc. Rev., 2011, 40, 3728. 5 Z. He, C. Zhong, X. Huang, W.-Y. Wong, H. Wu, L. Chen,

S. Su and Y. Cao, Adv. Mater., 2011, 23, 4636.

6 L. Dou, J. Gao, E. Richard, J. You, C. C. Chen, K. C. Cha, Y. He, G. Li and Y. Yang, J. Am. Chem. Soc., 2012, 134, 10071. 7 Y. Sun, G. C. Welch, W. L. Leong, C. J. Takacs, G. C. Bazan

and A. J. Heeger, Nat. Mater., 2012, 11, 44.

8 S. Loser, C. J. Bruns, H. Miyauchi, R. P. Ortiz, A. Facchetti, S. I. Stupp and T. J. Marks, J. Am. Chem. Soc., 2011, 133, 8142. 9 J. Zhang, D. Deng, C. He, Y. J. He, M. J. Zhang, Z. G. Zhang,

Z. J. Zhang and Y. F. Li, Chem. Mater., 2011, 23, 817. 10 A. Mishra and P. B¨auerle, Angew. Chem., Int. Ed., 2012, 51,

2020.

11 Y. Lin, Y. Li and X. Zhan, Chem. Soc. Rev., 2012, 41, 4245. 12 Y. Li, Q. Guo, Z. Li, J. Pei and W. Tian, Energy Environ. Sci.,

2010, 3, 1427.

13 T. S Poll, J. A. Love, T.-Q. Nguyen and G. C. Bazan, Adv. Mater., 2012, 24, 3646.

14 Y. Liu, X. Wan, F. Wang, J. Zhou, G. Long, J. Tian and Y. Chen, Adv. Mater., 2011, 23, 5387.

15 D. Sahu, C. H. Tsai, H. Y. Wei, K. C. Ho, F. C. Chang and C. W. Chu, J. Mater. Chem., 2012, 22, 7945.

16 M. Chen, W. Fu, M. Shi, X. Hu, J. Pan, J. Ling, H. Li and H. Chen, J. Mater. Chem. A, 2013, 1, 105.

17 Z. Li, G. He, X. Wan, Y. Liu, J. Zhou, G. Long, Y. Zuo, M. Zhang and Y. Chen, Adv. Energy Mater., 2012, 2, 74. 18 Y. Liu, X. Wan, F. Wang, J. Zhou, G. Long, J. Tian, J. You,

Y. Yang and Y. Chen, Adv. Energy Mater., 2011, 1, 771. 19 J. Zhou, X. Wan, Y. Liu, G. Long, F. Wang, Z. Li, Y. Zuo, C. Li

and Y. Chen, Chem. Mater., 2011, 23, 4666.

20 J. Zhou, X. Wan, Y. Liu, Y. Zuo, G. He, G. Long, W. Ni, C. Li, X. Su and Y. Chen, J. Am. Chem. Soc., 2012, 134, 16345. 21 C. B. Nielsen, B. C. Schroeder, A. Hadipour, B. P. Rand,

S. E. Watkins and I. McCulloch, J. Mater. Chem., 2011, 20, 17642.

22 C. B. Nielsen, R. S. Ashraf, B. C. Schroeder, P. D'Angelo, S. E. Watkins, K. Song, T. D. Anthopoulos and I. McCulloch, Chem. Commun., 2012, 48, 5832.

23 S. C. Lan, P. A. Yang, M. J. Zhu, C. M. Yu, J. M. Jiang and K. H. Wei, Polym. Chem., 2013, 4, 1132.

24 B. C. Schroeder, C. B. Nielsen, Y. J. Kim, J. Smith, Z. Huang, J. Durrant, S. E. Watkins, K. Song, T. D. Anthopoulos and I. McCulloch, Chem. Mater., 2011, 23, 4025.

25 C. B. Nielsen, J. M. Fraser, B. C. Schroeder, J. Du, A. J. White, W. Zhang and I. McCulloch, Org. Lett., 2011, 13, 2414. 26 H. C. Chu, D. Sahu, Y. C. Hsu, H. Padhy, D. Patra, J. T. Lin,

D. Bhattacharya, K. L. Lu, K. H. Wei and H. C. Lin, Dyes Pigm., 2012, 93, 1488.

27 D. Patra, D. Sahu, H. Padhy, D. Kekuda, C. W. Chu, K. H. Wei and H. C. Lin, Macromol. Chem. Phys., 2011, 212, 1960. 28 A. Tang, L. Li, Z. Lu, J. Huang, H. Jia, C. Zhan, Z. Tan, Y. Li

and J. Yao, J. Mater. Chem. A, 2013, 1, 5747.

29 D. Patra, M. Ramesh, D. Sahu, H. Padhy, C. W. Chu, K. H. Wei and H. C. Lin, J. Polym. Sci., Part A: Polym. Chem., 2012, 50, 967.

Fig. 7 EQE curves of BHJ solar cells incorporating BT6OT:PC71BM blends con-taining various amounts of CN as a processing additive.

30 I. H. Jung, J. Yu, E. Jeong, S. Kwon, H. Kong, K. Lee, H. Y. Woo and H. K. Shim, Chem.–Eur. J., 2010, 16, 3743. 31 D. Patra, M. Ramesh, D. Sahu, H. Padhy, C. W. Chu,

K. H. Wei and H. C. Lin, Polymer, 2012, 53, 1219.

32 J. K. Park, C. Kim, B. Walker, T. Q. Nguyen and J. H. Seo, RSC Adv., 2012, 2, 2232.

33 Y. Lin, Z.-G. Zhang, Y. Li, D. Zhu and X. Zhan, J. Mater. Chem. A, 2013, 1, 5128.