Efficient ternary polymer solar cells with dihydronaphthyl-based C60 bisadduct as an third component material

Solar Energy 170 (2018) 164–173

Contents lists available at ScienceDirect

Solar Energy journal homepage: www.elsevier.com/locate/solener

Efficient ternary polymer solar cells with dihydronaphthyl-based C60 bisadduct as an third component material ⁎

Guanxiong Maa,b, Zhiyong Liuc, , Ning Wangd,

T

⁎

a

Department of Materials Physics and Chemistry, School of Materials Science and Engineering, Northeastern University, Shenyang 110819, China Key Laboratory for Anisotropy and Texture of Materials, Ministry of Education, Northeastern University, Shenyang 110819, China c College of Science, Shenyang Agricultural University, Shenyang 110866, China d Key Laboratory of Physics and Technology for Advanced Batteries (Ministry of Education), College of Physics, Jilin University, Changchun 130012, China b

A R T I C LE I N FO

A B S T R A C T

Keywords: Photocurrent Charge carrier mobility Charge carrier recombination Photoluminescence

A dihydronaphthyl-based C60 bisadduct (NCBA) acceptor is introduce as a third component to typical PTB7: PC71BM binary polymer solar cells (PSCs). NCBA play bridging role between the lowest unoccupied molecular orbital (LUMO) of PTB7 and PC71BM and provides more routes for charge carrier transfer at the interface between PTB7 and PC71BM, a higher open-circuit voltage (VOC) realize upon addition of NCBA can be achieve relative to the neat PC71BM as an electron acceptor. The strong visible light absorption in the range from 350 to 700 nm of NCBA molecule compared with PC71BM molecule, it has the effect of apparently complementary visible light absorption compare with the binary PTB7:PC71BM thin films. The current density–voltage (J-V) characteristics dependent on incident light intensity were employed and to analyze the effect of NCBA concentration on the charge carrier recombination process of ternary PSCs. The crystallinity and surface morphology of the ternary PTB7:NCBA:PC71BM thin films with NCBA concentration is 15% of blend acceptor is similar to that of the binary PTB7:PC71BM thin films, which guarantees suitable efficient exciton dissociation and charge carrier transport. The transient photovoltage (TPV) and transient photocurrent (TPC) were measured, and the results illustrate the effect of NCBA as the third component materials in terms of higher charge carrier density and long charge carrier lifetime and weaken charge carrier recombination.

1. Introduction The last decade has witnessed rapid progress in the development of polymer solar cells (PSCs), and now power-conversion efficiencies (PCE) exceeding 13% have been realized in different types of PSCs (Cui et al., 2017; Arı et al., 2016). The performance of PSCs is strongly correlated with the excitonic nature of photoactive materials. However, despite the enormous strides achieved in past years, insufficient partial absorption in solar irradiation is still an inevitable issue restraining the photovoltaic performance of PSCs (Chen et al., 2017; Armin et al., 2015). Therefore, to overcome the absorption and open-circuit voltage (VOC) limitations of PSCs, various strategies at different levels of maturity are currently in the exploration phase, e.g., the addition of some appropriate complementary absorbers and the shallower energy-level electron-acceptor materials into the binary PSCs, realizing so-called ternary PSCs (An et al., 2016b; Lu et al., 2015). A ternary strategy with two donors and one acceptor or two acceptors and one donor has been demonstrated as an effective method to improve the photovoltaic performance of PSCs by enhancing photogenerated charge carrier

⁎

Corresponding authors. E-mail addresses: [email protected] (Z. Liu), [email protected] (N. Wang).

https://doi.org/10.1016/j.solener.2018.05.056 Received 5 March 2018; Received in revised form 27 April 2018; Accepted 14 May 2018 0038-092X/ © 2018 Elsevier Ltd. All rights reserved.

generation and collection efficiency (An et al., 2016b, 2016c). Ternary PSCs present some advantages: the simplicity of single-step processing for photoactive layers rather than the more complex tandem cells compared with tandem PSCs and higher VOC and JSC compared with binary PSCs. Nowadays, the most representative single-layer PSCs are those with a blend thin films of PTB7 and PC71BM as the photoactive layer, and the corresponding best power-conversion efficiency (PCE) has exceeded 10% under AM 1.5G illumination at an illumination intensity of 100 mW cm−2 (He et al., 2015; Lin et al., 2015). For most PSCs devices, the photoactive layer consists of a blend thin films of conjugated polymer donor and fullerene acceptor materials. The ideal polymer donor materials should have the advantage of a broad and strong absorption band in the visible and near-infrared regions, which can match well the solar spectrum (Brabec et al., 2002; Peet et al., 2007). The ideal acceptor materials should have the advantages of shallower lowest unoccupied molecular orbital (LUMO) energy level, electron mobility and electron affinity, which can ensure the acquisition of high VOC in real devices. The [6,6]-phenyl-C71-butyric acid methyl ester (PC71BM)

Solar Energy 170 (2018) 164–173

G. Ma et al.

0

NCBA in Acceptor 0 wt% 15 wt% 30 wt%

b 80

60

-8

EQE (%)

-2

JSC (mA cm )

-4

-12

0.0

100 10 -2

40

20

-16

Current Density (mA cm )

Fig. 1. (a) J-V curves of inverted structure “ITO/ ZnO/PTB7:NCBA:PC71BM/MoO3/Ag” with different weight ratios of NCBA as an acceptor under illumination of an AM 1.5G solar simulator at an illumination intensity of 100 mW cm−2. (b) EQE spectra of devices with the same specifications as (a). (c) J–V curves under dark conditions of (a).

100

a

1 0.1

0.2

0.4 0.6 Voltage (V)

0.8

0 300

NCBA in Acceptor 0 wt% 15 wt% 30 wt%

400

500

600

700

800

Wavelength (nm)

c NCBA in Acceptor 0 wt% 15 wt% 30 wt%

0.01 1E-3 1E-4 1E-5 1E-6 1E-7 1E-8 -1.0

-0.5

0.0 Voltage (V)

0.5

1.0

PTB7:NCBA thin films poses a difficulty for the elimination of disrupted molecular ordering and blocked charge carrier transport pathways, which can reduce charge carrier collection efficiency and photovoltaic performance (Lu et al., 2013; Cheng et al., 2014). However, the ternary PSCs may provide an alternative approach to achieve high performance by mixing the second fullerene derivative acceptor (Ferenczi et al., 2011; Li et al., 2011). In this work, NCBA is employed as an third component material in PTB7:PC71BM photoactive layer and a partial replacement of PC71BM. Owing to the shallower LUMO energy level of NCBA compared to PC71BM, and to the fact that NCBA plays a bridging role between PTB7 and PC71BM, substantial improvement in VOC and efficient exciton dissociation at the PTB7/acceptor interface with the addition of NCBA are realized. Analysis of the effect of NCBA concentration on photogenerated charge carrier generation and on collection efficiency and photovoltaic performance are carried out in detail.

comprises typical electron acceptor materials and matches the LUMO of most low-bandgap conjugated polymers for most PSC devices (Foertig et al., 2014; Liu et al., 2014). Chunru’s group has synthesized fullerene derivatives of the dihydronaphthyl-based C60 bisadduct (NCBA) as an electron acceptor (Meng et al., 2012, 2014). NCBA has the advantages of improved absorption in the 400–500 nm region and a shallower LUMO energy level, which leads to a high VOC and PCE [0.82 V and 5.85% and the similar photovoltaic parameter compared with indeneC60 bisadduct (ICBA) as an electron acceptor, respectively] based on poly(3-hexylthiophene) (P3HT) as donor materials (Meng and Zhang, 2012). NCBA, then, has a higher thermally driven crystallization, over 280 °C, which illustrates better thermal stability compared to PCBM and PC71BM (Meng et al., 2012, 2014). The PSCs with P3HT:NCBA and PTB7:PC71BM as photoactive layers exhibit better photovoltaic performance, and the best value exceeds 6% and 10%, respectively (Ganesamoorthy et al., 2017). Unfortunately, binary PSCs based on PTB7 as the donor material and NCBA as acceptors yielded lower PCE relative to the common low-bandgap PTB7:PC71BM system, which is attributed to two main reasons: low charge carrier generation efficiency for the PTB7:NCBA interface and a correspondingly small short-circuit current density (JSC), and poor charge carrier collection efficiency compared with the PTB7:PC71BM thin films as a photoactive layer and a corresponding decrease of JSC and fill factor (FF) Zhang et al., 2017a. Compared with PTB7/PC71BM thin films, the PTB7/NCBA interface exhibits a lower free-energy driving force and energy-level difference, which cannot guarantee efficient photogenerated charge carrier generation and exciton dissociation (Foertig et al., 2014; Hawks et al., 2013). Charge carrier transport and collection is related to the morphology and crystallization of the photoactive layer (Chang et al., 2013; Kim et al., 2012). Owing to the nature of the weaker crystallinity of PTB7, fine intermixing of

2. Experimental details The fabrication method and parameters are provided in the Supplementary Materials. The PSC devices were fabricated with the following inverted structure: ITO/ZnO/PTB7:NCBA:PC71BM/MoO3/Ag. The charge carrier mobility was measured by the space charge limited current (SCLC) method. Hole-only and electron-only diodes were fabricated using the architectures “ITO/PEDOT:PSS/PTB7:NCBA:PC71BM/ Au” and “Al/PTB7:NCBA:PC71BM/Al,” respectively. The chemical structures of PTB7, NCBA, and PC71BM and the energy-level diagram of the device are shown in Fig. S1. The average photovoltaic parameters were obtained from five PSCs fabricated in parallel.

165

Solar Energy 170 (2018) 164–173

G. Ma et al.

PSCs, respectively. The leakage current in PSCs can be considered undesirable current that is injected from the electrodes prior to voltage turn-on that result in lower FF and JSC values (Yang and Ma, 2013). The higher rectification ratio and weakened dark current indicate that the photoactive layer without and with the addition of a small amount of NCBA contains a very low density of defects compared with higher NCBA concentration PSCs (Zhou et al., 2014). To further clarify the effect of adding NCBA into PTB7:PC71BM thin films, the absorption spectra of neat PTB7, NCBA, and PC71BM and blended thin films of PTB7:PC71BM and PTB7:NCBA:PC71BM (1:0.225:1.275) are shown in Fig. S2(a) and (b), respectively. In order to avoid the experimental error, the thin films with same sample concentration and spin-coating parameter were fabricated, and the same thickness of neat thin films, binary and ternary thin films were achieved (the thickness of neat thin films is 30 nm, the thickness of binary and ternary thin films is 50 nm). In the short wavelengths region from 300 to 350 nm, the absorption peaks of NCBA are slight weaker than that of PC71BM. Interestingly, in the medium and long wavelengths region from 350 to 700 nm, the absorbance of NCBA is much stronger than those of PC71BM, as shown in the inset of Fig. S2(a). After small amount of NCBA to replace PC71BM, the absorption spectra of ternary thin films during 350–640 is slight enhance, which is attribute to the NCBA is much stronger than those of PC71BM during medium and long wavelengths region. Apparently complementary absorption spectra are exhibited after adding a small amount of NCBA into binary PTB7:PC71BM thin films, indicating that the photon harvesting of ternary PTB7:NCBA:PC71BM thin films could be optimized compared with binary PTB7:PC71BM thin films. The photon-harvesting ability of ternary PTB7:NCBA:PC71BM thin films can be enhanced at short and medium wavelengths (in the range from 300 to 650 nm) and corresponding to the NCBA absorption range, which is consistent with the EQE curve. The visible light range absorption ratio of ternary PTB7:NCBA:PC71BM (1:0.225:1.275) thin films was slightly enhanced, by 8%, compared with binary PTB7:PC71BM thin films. X-ray diffraction (XRD) curves were measured to investigate the effect of NCBA concentration on PTB7:PC71BM thin films crystallinity, and the results are shown in Fig. S3. The XRD curves of PTB7:PC71BM thin films with NCBA concentrations of 0, 15, and 30 wt% are shown in Fig. S3. The angles at which the peak intensities occur are related to the PTB7 molecule and PC71BM molecule inter-planer distances, to the crystallinity of PTB7:NCBA:PC71BM thin films, and by Bragg’s law (Höfle et al., 2014):

Table 1 Photovoltaic performance of PTB7:NCBA:PC71BM-based ternary PSCs with different NCBA concentration under the same optimized conditions. NCBA (wt%)

VOC (V)

JSC (mA cm−2)

FF (%)

0 5 10 15 20 25 30 100

0.76 0.76 0.77 0.78 0.78 0.79 0.79 0.80

17.4 17.6 17.7 18.6 17.3 16.5 15.2 10.2

64.8 65.4 68.7 67.9 67.2 65.7 64.4 53.2

± ± ± ± ± ± ± ±

0.25 0.24 0.21 0.24 0.19 0.25 0.21 0.18

± ± ± ± ± ± ± ±

PCE (%) 2.84 2.58 2.48 1.98 1.87 1.78 1.79 2.41

8.57 8.74 9.36 9.85 9.06 8.56 7.73 4.34

± ± ± ± ± ± ± ±

0.24 0.21 0.18 0.14 0.18 0.19 0.22 0.14

3. Results and discussion The inverted PSC devices with the structure of “ITO/ZnO/ PTB7:NCBA:PC71BM/MoO3/Ag” have a weight ratio of PTB7:blend acceptor of 1:1.5 and a weight concentration of NCBA in the blend acceptor that was changed from 0 to 30 wt% and 100 wt%. Fig. 1(a) presents J-V curves with 0, 15, and 30 wt% of NCBA concentration under AM 1.5G illumination at an illumination intensity of 100 mW cm−2. Table 1 shows all the performance parameters as a function of NCBA concentration. The NCBA concentration has an intense influence on the VOC, JSC, and FF of PSCs. VOC increases with NCBA concentration from 0.76 V (without NCBA) to 0.78 V (15 wt% of NCBA concentration) and further increased to 0.79 V (30 wt% of NCBA concentration) and 0.8 V (neat NCBA). However, the device exhibits lower photovoltaic performance at a NCBA concentration over 15 wt%, and the worst photovoltaic performance at a neat NCBA as acceptor even with a significantly enhanced VOC value. These results can be attributed to the comprehensive decrease in FF and JSC when NCBA concentration exceeds 15 wt%. FF and JSC increase when the NCBA concentration is less than 15 wt%, and reach their highest values at 15 wt% NCBA concentration, with a VOC of 0.78 V, a JSC of 18.6 mA cm−2, a FF of 67.9% and a corresponding PCE of 9.85%. Then, photovoltaic performance decreases significantly with a further increase of NCBA concentration from 15 to 100 wt%, with a lowest PCE of 4.34%. Compared with binary PTB7:PC71BM PSCs, a ternary device with 15 wt% NCBA concentration exhibits comprehensively enhanced photovoltaic parameters (0.76 V versus 0.78 V for VOC, 17.4 mA cm−2 versus 18.6 mA cm−2 for JSC, 64.8% versus 67.9% for FF, and 8.57% versus 9.58% for PCE). The external quantum efficiency (EQE) curves of devices with 0, 15, and 30 wt% NCBA concentration are shown in Fig. 1(b). The EQE spectrum of PTB7:NCBA:PC71BM (1:0.225:1.275)-based PSCs is slightly enhanced in the range from 350 to 650 nm compared with binary PSCs, while exhibiting a slight decrease at long wavelength. The calculated JSC values from the EQE spectrum are 16.7, 17.6, and 14.6 mA cm−2, corresponding to NCBA concentrations of 0, 15, and 30 wt%, respectively (average error of both calculated and measured JSC values is lower than 5%). This clearly illustrates the agreement with the higher JSC of the PSCs mentioned above. The higher EQE spectrum of PTB7:NCBA:PC71BM (1:0.225:1.275)-based PSCs is attribute to the together better visible light spectra and improve charge carrier kinetics process (del Pozo et al., 2017). We will to analyze the reason from absorption spectrum and charge carrier kinetics process of ternary photoactive layers after the addition of NCBA. Fig. 1(c) shows the current density-voltage (J-V) characteristics of the three PSCs in the dark. In the dark, the PTB7:PC71BM (1:1.5)-based and PTB7:NCBA:PC71BM-based (1:0.225:1.275) PSCs show low reverse bias current (leakage currents) and a large rectification ratio of over 104 at ± 1 V (the current rectification ratio is defined as the ratio of the current at 1 V and the current at −1 V based on the dark J-V curves). The rectification ratios are 3.52 × 104, 3.28 × 104, and 1.88 × 104 corresponding to the PTB7:PC71BM (1:1.5)-based, PTB7:NCBA:PC71BM (1:0.225:1.275)-based, and PTB7:NCBA:PC71BM (1:0.45:1.05)-based

λ = 2dsinθ, where λ is the wavelength of X-ray radiation used, i.e., 0.154 nm; θ is the peak position half-angle; and d is the inter-planer distance. The obvious diffraction peaks at 2θ is 20.86°, 21.25°, and 22.13°, can be observed from PTB7:NCBA:PC71BM thin films with NCBA concentrations of 0, 15, and 30 wt%, respectively, which corresponds to the PTB7 molecule. The corresponding d values are 0.425, 0.418, and 0.401 nm, respectively. The intensity of the (1 0 0) diffraction peak was significantly enhanced for the ternary thin films with NCBA replacing the 15 and 30 wt% PC71BM molecules, indicating the enhancement of the PTB7 molecular arrangement. The PTB7 peak intensity was increased from 0.34 (without NCBA) to 0.49 (with 15 wt% NCBA), and further increased to 0.62 (with 30 wt% NCBA). NCBA has a higher crystallinity than PC71BM, which improves the conformation of PTB7 chains and enhances the crystallinity of PTB7 molecules (Meng et al., 2012, 2014). The diffraction peak of PC71BM at 98° strongly decreases, while the intensity of NCBA increases at 105° with increasing NCBA concentration, indicating that individual phase of NCBA molecules has been formed and the phase of PC71BM molecule has decreased in ternary PTB7:NCBA:PC71BM thin films. The big diffraction angle shift of acceptor materials diffraction peak is attribute to the formation of the NCBA phase and the decrease of the PC71BM phase. At higher NCBA concentration, the crystallinity of PTB7:PC71BM thin films were 166

Solar Energy 170 (2018) 164–173

G. Ma et al.

does hinder photocurrent generation. To further verify the energy transfer from PTB7 to the acceptor, the TRPL spectra of neat PTB7, binary PTB7:PC71BM and PTB7:NCBA thin films, and ternary PTB7:NCBA:PC71BM (1:0.225:1.275) thin films, are measured by monitoring the 720 nm emission corresponding to the PTB7 absorption peak, as shown in Fig. S2(a). The fluorescence lifetime (η) of neat PTB7, PTB7:PC71BM thin films, PTB7:NCBA thin films, and PTB7:NCBA:PC71BM (1:0.225:1.275) thin films are 12.8, 4.52, 5.54, and 3.85 ns, respectively. The short fluorescence lifetime of the PTB7:NCBA:PC71BM (1:0.225:1.275) thin films is attributed to the higher efficiency of exciton dissociation and of energy transfer at the PTB7/NCBA, PTB7/PC71BM, and NCBA/PC71BM interfaces, and avoidance of photogenerated charge carrier recombination (Cowan et al., 2010; Moritomo et al., 2015). However, the fluorescence lifetime of PTB7:NCBA thin films increased to 5.54 ns, slightly higher than that of PTB7:PC71BM thin films and significantly higher than that of PTB7:NCBA:PC71BM (1:0.225:1.275) thin films, implying the lower efficiency of exciton dissociation and energy transfer at the PTB7/NCBA interfaces and strong photogenerated charge carrier recombination (Yao et al., 2016; Marinins et al., 2016). The decrease η value of PTB7:NCBA:PC71BM (1:0.225:1.275) thin films indicates the existence of energy transfer from PTB7 to NCBA, and then transfer to PC71BM, which may provide a potential route for improving NCBA exciton utilization efficiency, in addition to the exciton dissociation at the PTB7/ NCBA and PTB7/PC71BM interfaces (Su et al., 2017; Baran et al., 2016). The optimized result is a photogenerated exciton dissociated into the free charge carrier at the donor/acceptor interfaces and was transported to the individual electrode. However, the un-dissociated exciton on PTB7 can also transfer its energy to the blend acceptor, and can then be dissociated into free charge carriers at the NCBA/PC71BM interface (Cheng et al., 2014, 2016; Ameri et al., 2013a). After excess NCBA molecule partially replaced the PC71BM molecule, the crystallinity and surface morphology of ternary PTB7:NCBA:PC71BM thin films were significantly changed, and the charge carrier mobility was possibly affected. The space charge limited current (SCLC) method was used to evaluate the charge carrier mobility dependence on NCBA concentration (Li et al., 2016). Hole-only and electron-only SCLC devices were fabricated with the structures of ITO/ PEDOT:PSS/PTB7:NCBA:PC71BM/Au and Al/PTB7:NCBA:PC71BM/Al, respectively. The SCLC method will to used and to evaluated the charge carrier mobility dependent on NCBA concentration. All the curves follow Ohm’s law in the low voltage regime with slopes value is lower than 1, followed by slopes value is higher than 2 in the high-voltage regime. The slopes greater than 2 in the high-voltage regime resulted from the

completely changed, and the large phase of NCBA may form some electron traps in the PTB7/NCBA and NCBA/PC71BM interfaces, resulting in decreased charge carrier transport and photovoltaic parameters for ternary PTB7:NCBA:PC71BM (1:0.45:1.05)-based PSCs. The atomic force microscopy (AFM) were employed to measure the surface morphology of photoactive layer respectively. The AFM images of thin films with different NCBA doping ratios are shown in Fig. S4a–d, the slimier and slight rough morphology and short grain size with a relatively lower root mean square (RMS) value both binary PTB7:PC71BM thin films and ternary PTB7:NCBA:PC71BM (1:0.225:1.275) thin films and the value of 5.42 nm and 5.75 nm respectively, while the slight enhancement of RMS value for ternary PTB7:NCBA:PC71BM (1:0.45:1.05) thin films (the RMS value is 7.25 nm) and strong increase RMS value of binary PTB7:NCBA thin films (the RMS value is 23.4 nm). It means that the surface morphology of the ternary PTB7:NCBA:PC71BM (1:0.225:1.275) thin films has not been drastically disrupted compared with binary PTB7: PC71BM thin films. Thus, while the small amount of NCBA to replace PC71BM, it has the effect of protect the surface morphology of photoactive layer and hinders the aggregation of PTB7. However, the enhancement of PTB7 molecule aggregate and increase the grain size of ternary PTB7:NCBA:PC71BM (1:0.45:1.05) thin film, while the excess NCBA replace to PC71BM. Generally, smooth surface of ternary PTB7:NCBA:PC71BM (1:0.225:1.275) thin films will provide full and complete contact between photoactive layer and modified electrodes layer, which is improved charge carrier transfer between photoactive layer and modified electrodes layer (Xiao et al., 2016). Fig. 2(a) and (b) show the photoluminescence (PL) and time-resolved photoluminescence (TRPL) spectra of neat PTB7, binary PTB7:PC71BM and PTB7:NCBA thin films, and ternary PTB7:NCBA:PC71BM (1:0.225:1.275) thin films, respectively, which are used to investigate the working mechanism and charge carrier dynamic process of ternary thin films. The neat PTB7 thin film exhibits strong PL emission with an emission peak at approximately 770 nm. However, for the three types of thin films, namely PTB7:PC71BM (1:1.5), PTB7:NCBA (1:1.5), and PTB7:NCBA:PC71BM (1:0.225:1.275) thin films, PL emission is markedly quenched. The PL intensities of PTB7:NCBA (1:1.5), PTB7:PC71BM (1:1.5) and PTB7:NCBA:PC71BM (1:0.225:1.275) thin films decreased by 91%, 94%, and 97%, respectively, relative to that of neat PTB7 thin films, suggesting an efficient exciton dissociation at the PTB7/acceptor material interface and an energy transfer from PTB7 to acceptor molecules compared with binary PTB7/acceptor material (Nian et al., 2016; Lin et al., 2013). Simultaneously, the lower PL quenching ratio of PTB7:NCBA (1:1.5) thin film indicates that neat NCBA as an electron acceptor does not favor exciton dissociation and

10000

η (ns) neat PTB7 12.8 PTB7:NCBA:PC71BM 3.85

PTB7:NCBA PTB7:PC71BM

700

Counts

PL Intensity (a.u.)

1000

neat PTB7 PTB7:NCBA:PC71BM PTB7:NCBA PTB7:PC71BM

750 800 Wavelength (nm)

5.54 4.52

100

10

850

1

0

4

8

12

16

20

Time (ns)

Fig. 2. (a) PL spectra. (b) TRPL spectra of neat PTB7, binary thin films of PTB7:PC71BM and PTB7:NCBA, and ternary thin films of PTB7:NCBA:PC71BM (1:0.225:1.275). 167

Solar Energy 170 (2018) 164–173

G. Ma et al.

electric field dependence of the charge carrier mobility. The Poole–Frenkel equation describing the effect of the energetic disorder of randomly oriented polymer matrix on the charge carrier mobility is mentioned below (Cheon et al., 2014).

Table 2 Electron and hole mobility based on the SCLC method with different NCBA concentrations.

μ (E ) = μ0 exp(β E ) Hole-only and electron-only SCLC devices were fabricated with the structure of ITO/PEDOT:PSS/PTB7:NCBA:PC71BM/Au and Al/ PTB7:NCBA:PC71BM/Al respectively. The charge carrier mobility were determined by dark current model and is described as (Cho et al., 2017):

J=

NCBA (wt%)

μe (cm2 V−1 s−1)

μh (cm2 V−1 s−1)

0 15 30 100

1.58 × 10−4 1.78 × 10−4 1.34 × 10−4 0.75 × 10−4

4.41 × 10−3 4.28 × 10−3 4.15 × 10−3 3.52 × 10−3

9 V2 εr ε0 μ 3 8 d

10 -2

Jph (mA cm )

where JD is the current density, ε0 is the permittivity of free space (8.85 × 10−14 F/cm), εr is the dielectric constant of used materials and assumed to be 3 and μe is the zero-field mobility, which is a typical value for organic materials, V is the effective voltage and L is the thickness of the active layer. The thickness of the film (L) was measured using a step profiler. The SCLC curve of electron-only and hole-only devices are shown in Fig. 3. The hole mobility (μh) and electron mobility (μe) values are summarized in Table 2. The μe value exhibits a significant change trend with increasing NCBA concentration. It is apparent that the μe value increased from 1.58 × 10−4 cm2 V−1 s−1 at 0 wt% NCBA to 1.78 × 10−4 cm2 V−1 s−1 at 15 wt% NCBA and strongly decreased to 1.34 × 10−3 cm2 V−1 s−1 at 30 wt% NCBA and 0.75 × 10−4 cm2 V−1 s−1 at neat NCBA. The μh value exhibited a slight nearly linear decrease dependent on NCBA concentration, with values of 4.41 × 10−3, 4.28 × 10−3, 4.15 × 10−3 cm2 V−1 s−1 and 3.52 × 10−3 cm2 V−1 s−1 corresponding to NCBA concentrations of 0, 15, 30 and 100 wt%, respectively. The enhancement of μe could be ascribed to the enhanced molecular crystallinity and bi-continuous interpenetrating network that are well maintained in the optimized ternary blend thin films (An et al., 2016a). The excess NCBA (the NCBA concentration is 30 and 100 wt%) damages the uniform intermixing of PTB7 and PC71BM (Fig. S3) and the interface contact between photoactive layer and modified electrode layer (Fig. S4), which causes the lower charge carrier mobility. When the NCBA concentration was below 15 wt%, the slightly enhanced crystallinity and similar surface morphology of both the ternary PTB7:NCBA:PC71BM (1:0.225:1.275) and binary PTB7:PC71BM thin films guaranteed exciton dissociation and charge carrier transport. To gain more insight into light absorption and the exciton dissociation process, we measured the photocurrent density (Jph), saturation current density (Jsat), maximum exciton generation rate (Gmax), and charge carrier collection efficiency (Jph/Jsat) under short-circuit

1

and maximum power-point conditions. Fig. 4 shows the photocurrent density versus effective voltage (Jph-Veff) curves of PSCs dependent on NCBA concentration. The detailed Jph, Jsat, Gmax, and Jph/Jsat values under short-circuit and maximum power-point conditions were calculated, and the results are listed in Table 3. Jph and Veff are defined as JL − JD and V0 − Va, respectively, where JD and JL are current density in the dark and under a standard solar simulated light source, respectively; V0 is the voltage at Jph = 0; and Va is the applied voltage (Ma et al., 2017). Jph has a nearly linear dependence on the voltage at a low value of Veff, and reaches a saturated state (where the saturation photocurrent density [Jsat] was obtained) at a Veff higher than 2 V. The saturation current density (Jsat, which is Jph values reach a saturated state) is only limited by the maximum exciton generation rate (Gmax), and so the Gmax value is only governed by the amount of absorbed photons. As a result, Jsat can been defined as (Nian et al., 2016)

Jsat = qLGmax,

-2

Current Density (A cm )

-2

Current Density (A cm )

1

0.01

1E-3

1E-5 0.0

NCBA in Acceptor 0 wt% 15 wt% 30 wt% 100 wt%

0.5

1.0 Voltage (V)

1.5

1

Fig. 4. Jph-Veff curves of PSCs with different NCBA concentrations.

Electron-only

1E-4

0.1 Veff (V)

a 0.1

NCBA in Acceptor 0 wt% 15 wt% 30 wt%

2.0

b Hole-only

0.1

0.01 NCBA in Acceptor 0 wt% 15 wt% 30 wt% 100 wt%

1E-3

1E-4 0.0

0.5

1.0 Voltage (V)

Fig. 3. J-V curves of (a) electron-only and (b) hole-only devices. 168

1.5

2.0

Solar Energy 170 (2018) 164–173

G. Ma et al.

Jphoto (I ,V ) = edG (I ) PC (I ,V )

Table 3 Jph, Jsat, Gmax and Jph/Jsat under the short circuit and maximum power point conditions of PSCs with different NCBA concentration. −3

NCBA (wt %)

Jph (mA cm−2)

Jsat (mA cm−2)

Gmax (m

s

0 15 30

17.4 18.6 15.2

19.3 20.2 18.7

1.20 × 1028 1.26 × 1028 1.17 × 1028

−1

)

Jph/Jsat (%)

Jph/Jsat (max) (%)

90.2 91.8 81.3

72.5 73.4 67.4

where G(I) is the generation rate of free electron-hole pairs per unit volume, and PC(I,V) is the charge carrier collection probability. As is evident from Fig. S5(a), (c), and (e), the current density becomes insensitive to the reverse bias of the applied voltage region and independent of the applied voltage of approximately −0.5 V. The reverse saturation current is calculated as

G (I ) = Jphoto (V = −0.5 V). At an applied voltage of −0.5 V, a maximum number of charge carriers can be swept out and collected by the individual electrodes. The charge carrier collection efficiency can be calculated by the formula (Xu et al., 2017; Cheng et al., 2017)

where q is elementary charge, L the distance between anode and cathode, and Gmax a measure of the maximum number of photons absorbed. The noticeable enhancement of Jph and Gmax suggests the occurrence of enhanced visible light absorption and harvesting in the photoactive layer (Chen et al., 2015; Tseng et al., 2016). For the PTB7:NCBA:PC71BM-based PSCs with NCBA concentrations of 0, 15, and 30 wt%, the Jph values are 17.4, 18.6, and 15.2 mA cm−2, respectively. At Veff = 2 V, the corresponding Jsat values are 19.3, 20.2, and 18.7 mA cm−2, respectively. The corresponding Jph/Jsat values under short-circuit conditions are then 90.2%, 91.8%, and 81.3% for PSCs, respectively, while the respective Jph/Jsat values are 72.5%, 73.4%, and 67.4% under maximum power-point conditions. The higher Jph value under low Veff is means a better charge carrier extraction efficiency, leading to higher FF and JSC values. The similar Jph/Jsat value for both 15 wt% NCBA-based PSCs and without NCBA-based PSCs implies that the efficient charge carrier transport and collection of photoactive layers are achieved in PSCs containing 15 wt% NCBA and without NCBA. However, the lower Jph/Jsat value under short-circuit conditions in 30 wt% NCBA-based PSCs is attributed to the lower efficiency of charge carrier transport and collection when excess NCBA replaces PC71BM. Meanwhile, the higher Jph/Jsat values under maximum powerpoint conditions for the PTB7:NCBA:PC71BM-based PSCs with NCBA concentrations of 0 and 15 wt% is attributed to the weaker bimolecular recombination and better charge carrier transport and collection ability of both 15 wt%-NCBA-based PSCs and without NCBA-based PSCs under maximum power-point conditions. As Jph/Jsat is the charge carrier collection efficiency, the poor Jph-Veff characteristic is attributed to the lower efficiency of photogenerated exciton dissociation and charge carrier transport and collection, which usually leads to a lower JSC and FF (Zhang et al., 2015). The Gmax values were 1.2 × 1028 (0 wt% NCBA), 1.23 × 1028 (15 wt% NCBA), and 1.17 × 1028 m−3 s−1 (30 wt % NCBA), respectively. This result means that adding 15 wt% NCBA to an blend acceptor based on ternary PTB7:NCBA:PC71BM thin films as a photoactive layer can generate a significant amount of photogenerated excitons. The high photogenerated exciton generation rate, efficient exciton dissociation, and better charge carrier transport and collection ability result in the high JSC and FF (Zhong et al., 2017; Yoon et al., 2017). And then, the increment in photovoltaic performance, which is attributed to a lowest charge carrier generation, transport and collection resulting from higher trap states density in both the photoactive layer contact interface and the bulk properties of the photoactive layer should manifest in PSCs J-V dark characteristics. Thus, for the binary PSCs and ternary PSCs, it has shown the similar charge carrier recombination in dark and in illumination condition. Figs. S5(a), (c), and (e) show the J-V characteristic curves of PTB7:PC71BM (1:1.5)-based PSCs, PTB7:NCBA:PC71BM (1:0.225:1.275)-based PSCs, and PTB7:NCBA:PC71BM (1:0.45:1.05)based PSCs, respectively, for incident-light intensities ranging from 3 to 100 mW cm−2. Fig. S5(b), (d), and (f) show the normalized photocurrent densities for Fig. S5(a), (c), and (e), respectively. The current density (J) of the PSCs are defined as (Cowan et al., 2010; C. Liu et al., 2017):

PC (I ,V ) =

Jphoto (I ,V ) Jphoto (I ,−0.5 V)

.

This confirms that all the free electron-hole pairs can be swept out, and geminate recombination (monomolecular recombination and bimolecular recombination) can be avoided within the photoactive layer and at the photoactive layer/electrode interface, so that PC is close to 1 at the short-circuit condition (Cowan et al., 2010; Park et al., 2009). While the applied voltages are in the range from the maximum power point condition (Vmax, which is approximately 0.6 V) to −0.5 V, namely the short-circuit condition, the charge carrier collection probability is dependent on the incident-light intensity, which is the maximum number of photogenerated charge carriers (higher than 90%) swept out of the photoactive layer and collected by the individual electrode, and is suitable for the study of charge carrier recombination. However, while the applied voltages are higher than Vmax (in the range from Vmax to VOC) or at the open-circuit condition, the maximum number of photogenerated charge carriers underwent recombination (nearly 100%) within the photoactive layer, and PC(I,Vapplied) ≈ PC(V). This result is not suitable for the study of charge carrier recombination. For all PSCs, the together monomolecular recombination and bimolecular recombination or single bimolecular recombination dominates for applied voltages in the range from Vmax to VOC. The variation in PC with incident-light intensity is most prominent at the open-circuit voltage; that is, at the externally applied bias at which J (I,V) is 0. At a given voltage, the competition between charge carrier transport and charge carrier recombination determines the charge carrier concentration (n) and photogenerated charge carrier lifetime (τ) within the photoactive layer. Therefore, the higher n during the lower internal voltage, the lower the photogenerated charge carrier transport and collection ability, which is expected to change the magnitude of the photogenerated charge carrier recombination rate, and cause the geminate recombination within the photoactive layer (Cowan et al., 2010). Thus, the photogenerated charge carrier recombination mechanism, which is dependent on the applied bias and evolves from monomolecular recombination to bimolecular recombination kinetics at the open-circuit condition, were researched by analyzing the J-V characteristic curves and normalized photocurrent curves dependent on incident-light intensity (Kyaw et al., 2013). From Fig. S5(b), (d) and (f), at the short-circuit condition, this photogenerated charge carrier recombination kinetics with applied voltages arises from the variation of incident-light intensity, evolving from geminate recombination at the open-circuit condition and from bimolecular recombination at the short-circuit condition (Cowan et al., 2010; Chaturvedi et al., 2016). The J-V characteristics and normalized photocurrents of three PSCs indicate that the charge carrier collection probability is strongly dependent on the NCBA concentration. The normalized photocurrent curve shows a subtle spread at the short-circuit condition dependent on applied voltages, especially for the higher NCBA concentration PSCs. These results indicate that bimolecular recombination is observed at any applied voltage in the three PSCs, and in the more significant phenomenon of higher NCBA concentration-based

J (I ,V ) = Jphoto (I ,V ) + Jdark The photocurrent density is defined as: 169

Solar Energy 170 (2018) 164–173

G. Ma et al.

a

b

NCBA in acceptor 0 wt% n=1.11 15 wt% n=1.13 30 wt% n=1.57

0.78 0.76 VOC (V)

-2

JSC (mA cm )

10

0.80

NCBA in acceptor 0 wt% α=0.952 15 wt% α=0.948 30 wt% α=0.917

0.74 0.72 0.70

1

0.68 0.66

10 Light Intensity (mW cm-2)

10 -2 Light Intensity (mW cm )

100

100

Fig. 5. Charge carrier recombination study of the ternary PSCs. Dependence of (a) JSC and (b) VOC on light intensity for ternary PSCs.

significantly deviates by a unity of 1(kT/q), and geminate recombination is the dominant mechanism (Lu et al., 2015; Z. Liu et al., 2017). In our cases, the PTB7:NCBA:PC71BM (1:0.45:1.05)-based PSCs showed a slope of 1.57(kT/q), while, for PTB7:PC71BM (1:1.5)-based and PTB7:NCBA:PC71BM (1:0.225:1.275)-based PSCs, unity slopes close to 1, of 1.11(kT/q) and 1.13(kT/q), were attained, implying that monomolecular recombination is almost negligible and bimolecular recombination is dominant in the devices processed with PTB7:PC71BM (1:1.5)-based and PTB7:NCBA:PC71BM (1:0.225:1.275)-based PSCs. These results indicate that, without NCBA and with an addition of a small amount of NCBA to the PTB7:PC71BM thin films, interfacial surface trap densities were reduced and trap-assisted charge carrier recombination was suppressed, which is perfectly consistent with enhanced JSC and FF. However, the higher n value of PTB7:NCBA:PC71BM (1:0.45:1.05)-based ternary PSCs implies that the excess NCBA concentration produced a significant degree of interfacial surface trapping at the PTB7/blend acceptor interface, and caused strong monomolecular recombination at the open-circuit condition. In order to further assess the charge carrier recombination dynamics and to corroborate the degree of suppressed charge carrier recombination, the charge carrier lifetime (τ) and charge carrier density (p) were measured by transient photovoltage (TPV) measurements from Fig. 6(a) and by transient photocurrent (TPC) measurements from Fig. 6(b) at the open-circuit condition over a range of different incident light intensities (Ameri et al., 2013b). p and τ in PSCs at the open-circuit condition under a given steady-state light intensity can be calculated from the TPV and TPC transients, respectively, using a differential capacitance technique proposed by Shuttle et al. (2008). When the PSC output reached a steady state, a pulsed laser (a nitrogen-laser-pumped dye laser as an excitation source with a wavelength of 620 nm, frequency of 2 Hz, and pulse duration of 1 ns) generated a small perturbation. The charge carrier concentration induced by a small-amplitude laser pulse (Δp), Δp ≪ p , at different levels of steady-state bias light, was used to probe the density of states by measuring the change in voltage (ΔV) Hamilton et al., 2010. At the open-circuit condition, excess photogenerated charge carriers are recombined within the photoactive layer and the photoactive layer/electrode modified layer. The VOC value dependent on incident-light intensity shows that the recombination process of PSCs is predominantly influenced by bimolecular recombination and weakens the monomolecular recombination. The charge concentration (p), obtained from fitting the TPV to an exponential, is defined as (Shuttle et al., 2008; Credgington et al., 2011)

PSCs. Meanwhile, for the PTB7:NCBA:PC71BM (1:0.45:1.05)-based PSCs at the same incident light intensity, it is implied that the same normalized photocurrent density will result in a higher applied voltage. At any given applied voltage, as long as it was the same, lower photocurrent intensities were observed when a significant amount of NCBA replaced PC71BM. The dependence of JSC and VOC on incident light intensity offers deeper insight into the effect of NCBA concentration on the photogenerated charge carrier recombination process. JSC is dependent on incident light intensity for PSCs, and photogenerated charge carriers can be efficiently transported to the individual electrodes at short-circuit conditions. This charge carrier recombination can be assessed by the following equation and used to research the effect of NCBA concentration on bimolecular recombination (Zhang et al., 2017a; Sun et al., 2017): α JSC ∝ Plight ,

where α is an exponential factor (Zhang et al., 2011; Lenes et al., 2009). Fig. 5(a) illustrates JSC as a function of incident light intensity for PTB7:PC71BM (1:1.5)-based, PTB7:NCBA:PC71BM (1:0.225:1.275)based, and PTB7:NCBA:PC71BM (1:0.45:1.05)-based PSCs. The extracted α values are 0.952, 0.948, and 0.917 for the three devices, respectively. For the PTB7:PC71BM (1:1.5)-based and PTB7:NCBA:PC71BM (1:0.225:1.275)-based PSCs, the α values are similar and close to 1, indicating the weaker bimolecular recombination and the effective photogenerated charge carrier transported and collected by the individual electrodes (Ameri et al., 2013a; An et al., 2015). However, the significance of α values lower than 1 for the PTB7:NCBA:PC71BM (1:0.45:1.05)-based PSCs suggests that bimolecular recombination is dominant within the photoactive layer compared with PTB7:PC71BM (1:1.5)-based PSCs and PTB7:NCBA:PC71BM (1:0.225:1.275)-based PSCs. At the open-circuit condition, all the photogenerated charge carriers will be recombined in the photoactive layers. Fig. 5(b) shows the relationship between VOC and Plight in PSCs. The VOC values are dependent on incident light intensity (Plight) in PSCs, which helps determine the degree of charge carrier recombination in the devices. The charge carrier recombination mechanism is defined by the following equation and used to research the effect of NCBA concentration on single bimolecular recombination or geminate recombination (Cowan et al., 2010; Zhang et al., 2017b):

VOC ∝ n

kT ln(I ), q

p=

1 Aed

V

∫0 OC CdV ,

C = ΔQ/ΔV0,

where k is Boltzmann’s constant. A slope at unity 1(kT/q) implies that bimolecular recombination is the dominant mechanism, while the slope

t

ΔQ = ∫0 Idt , 170

Solar Energy 170 (2018) 164–173

G. Ma et al.

a

10

b NCBA in Acceptor 0 wt% 15 wt% 30 wt%

0.2

6

JSC (mA)

ΔV (mV)

8

NCBA in Acceptor 0 wt% 15 wt% 30 wt%

4

0.1

2 0

0

50

100 Time (μs)

150

0.0

0

5

10 15 Time (μs)

20

25

Charge Carrier Lifetime (ms)

c

10 NCBA in Acceptor 0 wt% 15 wt% 30 wt% 2

4

6 16

8

10

-3

Charge Carrier Density *10 (cm ) Fig. 6. Transient (a) photovoltage and (b) photocurrent. (c) τ vs. n of ternary PSCs with 0, 15, and 30 wt% NCBA concentrations under 0.1, 0.15, 0.25, 0.4, 0.6, and 1 Sun illumination.

ternary PTB7:NCBA:PC71BM (1:0.225:1.275)-based PSCs. Addition of 15 wt% NCBA as an blend acceptor causes only minor changes in the charge carrier concentration and charge carrier lifetimes of PTB7:PC71BM, while the p value of ternary PTB7:NCBA:PC71BM (1:0.45:1.05)-based PSCs is reduced by 50–60% compared to binary PTB7:PC71BM (1:1.5)-based PSCs and ternary PTB7:NCBA:PC71BM (1:0.225:1.275)-based PSCs (pPTB7:PC71BM is 7.4 × 1016 cm−3, pPTB7:NCBA:PC71BM (1:0.225:1.275) is 8.2 × 1016 cm−3, and pPTB7:NCBA:PC71BM 16 cm−3). τ under 1 Sun irradiation continuously (1:0.45:1.05) is 3.5 × 10 decreases upon addition of NCBA (τPTB7:PC71BM is 9.9 ms, τPTB7:NCBA:PC71BM (1:0.225:1.275) is 7.8 ms, and τPTB7:NCBA:PC71BM (1:0.45:1.05) is 5.1 ms). These τ versus p results are perfectly consistent with the JSC and FF trend as a function of NCBA concentration of the PSCs. Furthermore, λ also gradually increases with increasing NCBA concentration, and the λ values are 1.58, 2.05, and 2.95, corresponding to binary PTB7:PC71BM (1:1.5)-based PSCs and ternary PTB7:NCBA:PC71BM (1:0.225:1.275)-based and PTB7:NCBA:PC71BM (1:0.45:1.05)-based PSCs respectively. These values causes strong photogenerated charge carrier recombination and a higher-order dependency on p when excess NCBA replaces PC71BM, leading to a significant decrease of JSC and FF. Simultaneously, for the PSCs under with and without light illuminated, it has the same charge carrier recombination process (specifically in the diode region). The higher p values and lower λ values of ternary PTB7:NCBA:PC71BM (1:0.225:1.275)-based PSCs, which is attribute to the lower trap states density in both the photoactive layer contact interface and within photoactive layer. These results are attributed to the effect of trapping and release in energetic traps, as well as trapping due to morphological traps.

where A is the device area, e the electronic charge, d the device thickness, C the capacitance, and no the concentration of photogenerated charge carriers due to the laser pulse. The differential C is defined as the voltage change when a small number of photogenerated charge carriers is added to the device, which was extrapolated back to V = 0. Small-perturbation charge carrier lifetimes are found to follow an exponential dependent on VOC of the form (Hawks et al., 2013; Kirchartz and Nelson, 2012; Foertig et al., 2012) q

τΔp = τΔpo e− ϑkT VOC , q

p = p0 e mkT VOC , δ=

m ϑ

+ 1,

τΔp = τΔn0

p0 λ

( ), p

where m and ϑ are the slopes of τ(VOC) and p(VOC), respectively, and δ is the reaction order. τΔpo is the charge carrier lifetime achieved by laserpulse illumination, which is constant; τΔp is the small-perturbation charge carrier lifetime; p is the average excess charge carrier concentration; po is the concentration of photogenerated charge carriers due to the laser pulse; and λ is the recombination exponent (Hawks et al., 2013). Fig. 6(c) shows the charge carrier lifetime as a function of the charge carrier density for the binary PTB7:PC71BM (1:1.5)-based PSCs and the ternary PTB7:NCBA:PC71BM (1:0.225:1.275)-based and PTB7:NCBA:PC71BM (1:0.45:1.05)-based PSCs. We found that τΔp versus p behaves completely differently in PTB7:NCBA:PC71BM (1:0.45:1.05)based PSCs compared to binary PTB7:PC71BM (1:1.5)-based PSCs and 171

Solar Energy 170 (2018) 164–173

G. Ma et al.

Simultaneously, the higher energy-level disorder in a photoactive layer causes charge carrier recombination between free charge carriers and trapped charge carriers that can decrease FF and a significantly reduced charge carrier injection that can decrease JSC (Baumann et al., 2011; Hong et al., 2016). However, adding a small amount of NCBA into the PTB7:PC71BM thin films did not cause a significant change in the photogenerated charge carrier recombination dynamics. Indeed, PTB7 as an efficient transport matrix allowed us to limit and overcome the otherwise dominant photogenerated charge carrier recombination losses of the NCBA:PC71BM thin films with a small amount of NCBA.

optics of conventional and inverted thick junction organic solar cells. ACS Photonics 2, 1745–1754. Baran, D., Ashraf, R.S., Hanifi, D.A., Abdelsamie, M., Gasparini, N., Rohr, J.A., Holliday, S., Wadsworth, A., Lockett, S., Neophytou, M., Emmott, C.J., Nelson, J., Brabec, C.J., Amassian, A., Salleo, A., Kirchartz, T., Durrant, J.R., McCulloch, I., 2016. Reducing the efficiency-stability-cost gap of organic photovoltaics with highly efficient and stable small molecule acceptor ternary solar cells. Nat. Mater. Baumann, A., Savenije, T.J., Murthy, D.H.K., Heeney, M., Dyakonov, V., Deibel, C., 2011. Influence of phase segregation on recombination dynamics in organic bulk-heterojunction solar cells. Adv. Funct. Mater. 21, 1687–1692. Brabec, C.J., Winder, C., Sariciftci, N.S., Hummelen, J.C., Dhanabalan, A., van Hal, P.A., Janssen, R.A.J., 2002. A low-bandgap semiconducting polymer for photovoltaic devices and infrared emitting diodes. Adv. Funct. Mater. 12, 709–712. Chang, M., Choi, D., Fu, B., Reichmanis, E., 2013. Solvent based hydrogen bonding: impact on poly (3-hexylthiophene) nanoscale morphology and charge transport characteristics. ACS Nano 7, 5402–5413. Chaturvedi, N., Swami, S.K., Dutta, V., 2016. Electric field assisted spray deposited MoO3 thin films as a hole transport layer for organic solar cells. Sol. Energy 137, 379–384. Chen, J.D., Cui, C., Li, Y.Q., Zhou, L., Ou, Q.D., Li, C., Li, Y., Tang, J.X., 2015. Singlejunction polymer solar cells exceeding 10% power conversion efficiency. Adv. Mater. 27, 1035–1041. Chen, Y., Qin, Y., Wu, Y., Li, C., Yao, H., Liang, N., Wang, X., Li, W., Ma, W., Hou, J., 2017. From binary to ternary: improving the external quantum efficiency of smallmolecule acceptor-based polymer solar cells with a minute amount of fullerene sensitization. Adv. Energy Mater. 1700328–1700336. Cheng, P., Li, Y., Zhan, X., 2014. Efficient ternary blend polymer solar cells with indeneC-60 bisadduct as an electron-cascade acceptor. Energy Environ. Sci. 7, 2005–2011. Cheng, P., Yan, C., Wu, Y., Wang, J., Qin, M., An, Q., Cao, J., Huo, L., Zhang, F., Ding, L., Sun, Y., Ma, W., Zhan, X., 2016. Alloy acceptor: superior alternative to PCBM toward efficient and stable organic solar cells. Adv. Mater. 28, 8021–8028. Cheng, P., Zhang, M.Y., Lau, T.K., Wu, Y., Jia, B.Y., Wang, J.Y., Yan, C.Q., Qin, M., Lu, X.H., Zhan, X.W., 2017. Realizing small energy loss of 0.55 eV, high open-circuit voltage > 1 V and high efficiency > 10% in fullerene-free polymer solar cells via energy driver. Adv. Mater. 29. Cheon, K.H., Cho, J., Lim, B.T., Yun, H.-J., Kwon, S.-K., Kim, Y.-H., Chung, D.S., 2014. Analysis of charge transport in high-mobility diketopyrrolopyrole polymers by space charge limited current and time of flight methods. RSC Adv. 4, 35344–35347. Cho, H.J., Kim, Y.J., Chen, S., Lee, J., Shin, T.J., Park, C.E., Yang, C., 2017. Over 10% efficiency in single-junction polymer solar cells developed from easily accessible random terpolymers. Nano Energy 39, 229–237. Cowan, S.R., Roy, A., Heeger, A.J., 2010. Recombination in polymer-fullerene bulk heterojunction solar cells. Phys. Rev. B 82, 245207–245217. Credgington, D., Hamilton, R., Atienzar, P., Nelson, J., Durrant, J.R., 2011. Non-geminate recombination as the primary determinant of open-circuit voltage in polythiophene: fullerene blend solar cells: an analysis of the influence of device processing conditions. Adv. Funct. Mater. 21, 2744–2753. Cui, Y., Yao, H., Gao, B., Qin, Y., Zhang, S., Yang, B., He, C., Xu, B., Hou, J., 2017. Finetuned photoactive and interconnection layers for achieving over 13% efficiency in a fullerene-free tandem organic solar cell. J. Am. Chem. Soc. 139, 7302–7309. del Pozo, G., Arredondo, B., Romero, B., Susanna, G., Brunetti, F., 2017. Degradation of PEIE interlayer in PTB7:[70]PCBM based solar cells characterized by impedance spectroscopy. Sol. Energy 144, 105–110. Ferenczi, T.A.M., Mueller, C., Bradley, D.D.C., Smith, P., Nelson, J., Stingelin, N., 2011. Organic semiconductor: insulator polymer ternary blends for photovoltaics. Adv. Mater. 23, 4093–4097. Foertig, A., Rauh, J., Dyakonov, V., Deibel, C., 2012. Shockley equation parameters of P3HT:PCBM solar cells determined by transient techniques. Phys. Rev. B 86, 115302. Foertig, A., Kniepert, J., Gluecker, M., Brenner, T., Dyakonov, V., Neher, D., Deibel, C., 2014. Nongeminate and geminate recombination in PTB7: PCBM solar cells. Adv. Funct. Mater. 24, 1306–1311. Ganesamoorthy, R., Sathiyan, G., Sakthivel, P., 2017. Review: fullerene based acceptors for efficient bulk heterojunction organic solar cell applications. Sol. Energy Mater. Sol. Cells 161, 102–148. Hamilton, R., Shuttle, C.G., O’Regan, B., Hammant, T.C., Nelson, J., Durrant, J.R., 2010. Recombination in annealed and nonannealed polythiophene/fullerene solar cells: transient photovoltage studies versus numerical modeling. J. Phys. Chem. Lett. 1, 1432–1436. Hawks, S.A., Deledalle, F., Yao, J., Rebois, D.G., Li, G., Nelson, J., Yang, Y., Kirchartz, T., Durrant, J.R., 2013. Relating recombination, density of states, and device performance in an efficient polymer: fullerene organic solar cell blend. Adv. Energy Mater. 3, 1201–1209. He, Z., Xiao, B., Liu, F., Wu, H., Yang, Y., Xiao, S., Wang, C., Russell, T.P., Cao, Y., 2015. Single-junction polymer solar cells with high efficiency and photovoltage. Nat. Photonics 9, 174–179. Höfle, S., Lutz, T., Egel, A., Nickel, F., Kettlitz, S.W., Gomard, G., Lemmer, U., Colsmann, A., 2014. Influence of the emission layer thickness on the optoelectronic properties of solution processed organic light-emitting diodes. ACS Photonics 1, 968–973. Hong, J., Kim, Y.J., Park, T., Park, C.E., 2016. The importance of the polymer molecular weight and the processing solvent in PBDTTT-C:PCBM bulk heterojunction solar cells: their effects on the nanostructural active texture. Sol. Energy 140, 27–33. Kim, Y., Kim, G., Lee, J., Lee, K., 2012. Morphology controlled bulk-heterojunction layers of fully electro-spray coated organic solar cells. Sol. Energy Mater. Sol. Cells 105, 272–279. Kirchartz, T., Nelson, J., 2012. Meaning of reaction orders in polymer:fullerene solar cells. Phys. Rev. B 86, 165201. Kyaw, A.K.K., Wang, D.H., Gupta, V., Leong, W.L., Ke, L., Bazan, G.C., Heeger, A.J., 2013.

4. Conclusions In summary, the blend acceptor material studied in this work (NCBA:PC71BM) exhibits some advantages over the neat PC71BM. First, the energy level can be tuned by changing the ratio of NCBA:PC71BM blend acceptor material. The more efficient exciton dissociation and charge carrier transport of the PTB7/NCBA, NCBA/PC71BM, and PTB7/ PC71BM interfaces can increase photovoltaic performance. The VOC values of PSCs can also be increased due to the shallower LUMO energy level of the NCBA:PC71BM blend acceptor material. Second, the crystallinity of the PTB7:NCBA:PC71BM (1:0.225:1.275) thin films can be slightly enhanced and keep the surface morphology compared with binary PTB7:PC71BM thin films, which is beneficial to efficient exciton dissociation and charge carrier transport and collection at individual electrodes. Third, PTB7:NCBA:PC71BM (1:0.225:1.275)-based PSCs exhibit a higher visible-light absorbance and exciton generation rate, and the PTB7:NCBA:PC71BM (1:0.225:1.275)-based ternary PSCs exhibit weakened monomolecular and bimolecular recombination under opencircuit and short-circuit conditions, respectively. These results demonstrate that NCBA as third component material and plays a bridging role in the PTB7:NCBA:PC71BM (1:0.225:1.275)-based ternary PSCs, which is a promising characteristic for future development of high-performance PSCs. Acknowledgment This work was financially supported by the Scientific Research Foundation of Shenyang Agricultural University (Grant No. 880415039). We would like to thank LetPub (www.letpub.com) for providing linguistic assistance during the preparation of this manuscript. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.solener.2018.05.056. References Ameri, T., Khoram, P., Min, J., Brabec, C.J., 2013a. Organic ternary solar cells: a review. Adv. Mater. 9, 1–22. Ameri, T., Heumüller, T., Min, J., Li, N., Matt, G., Scherf, U., Brabec, C.J., 2013b. IR sensitization of an indene-C60 bisadduct (ICBA) in ternary organic solar cells. Energy Environ. Sci. 6, 1796–1801. An, Q., Zhang, F., Sun, Q., Wang, J., Li, L., Zhang, J., Tang, W., Deng, Z., 2015. Efficient small molecular ternary solar cells by synergistically optimized photon harvesting and phase separation. J. Mater. Chem. A 3, 16653–16662. An, Q., Zhang, F., Yin, X., Sun, Q., Zhang, M., Zhang, J., Tang, W., Deng, Z., 2016a. Highperformance alloy model-based ternary small molecule solar cells. Nano Energy 30, 276–282. An, Q., Zhang, F., Sun, Q., Zhang, M., Zhang, J., Tang, W., Yin, X., Deng, Z., 2016b. Efficient organic ternary solar cells with the third component as energy acceptor. Nano Energy 26, 180–191. An, Q., Zhang, F., Zhang, J., Tang, W., Deng, Z., Hu, B., 2016c. Versatile ternary organic solar cells: a critical review. Energy Environ. Sci. 9, 281–322. Arı, M., Kanat, Z., Dinçer, H., 2016. Design, computational screening and synthesis of novel non-peripherally tetra hexylthio-substituted phthalocyanines as bulk heterojunction solar cell materials. Sol. Energy 134, 1–8. Armin, A., Yazmaciyan, A., Hambsch, M., Li, J., Burn, P.L., Meredith, P., 2015. Electro-

172

Solar Energy 170 (2018) 164–173

G. Ma et al.

Park, S.H., Roy, A., Beaupre, S., Cho, S., Coates, N., Moon, J.S., Moses, D., Leclerc, M., Lee, K., Heeger, A.J., 2009. Bulk heterojunction solar cells with internal quantum efficiency approaching 100%. Nat. Photon. 3, 297–302. Peet, J., Kim, J.Y., Coates, N.E., Ma, W.L., Moses, D., Heeger, A.J., Bazan, G.C., 2007. Efficiency enhancement in low-bandgap polymer solar cells by processing with alkane dithiols. Nat. Mater. 6, 497–500. Shuttle, C.G., O’Regan, B., Ballantyne, A.M., Nelson, J., Bradley, D.D.C., de Mello, J., Durrant, J.R., 2008. Experimental determination of the rate law for charge carrier decay in a polythiophene: fullerene solar cell. Appl. Phys. Lett. 92, 093311. Su, W., Fan, Q., Guo, X., Meng, X., Bi, Z., Ma, W., Zhang, M., Li, Y., 2017. Two compatible nonfullerene acceptors with similar structures as alloy for efficient ternary polymer solar cells. Nano Energy 38, 510–517. Sun, Q., Zhang, F., An, Q., Zhang, M., Ma, X., Zhang, J., 2017. Simultaneously enhanced efficiency and stability of polymer solar cells by employing solvent additive and upside-down drying method. ACS Appl. Mater. Interfaces 9, 8863–8871. Tseng, Z.-L., Chen, L.-C., Chiang, C.-H., Chang, S.-H., Chen, C.-C., Wu, C.-G., 2016. Efficient inverted-type perovskite solar cells using UV-ozone treated MoOx and WOx as hole transporting layers. Sol. Energy 139, 484–488. Xiao, Y., Wang, H., Zhou, S., Yan, K., Xie, W., Guan, Z., Tsang, S.-W., Xu, J.-B., 2016. Efficient ternary bulk heterojunction solar cells with PCDTBT as hole-cascade material. Nano Energy 19, 476–485. Xu, W., Yao, X., Meng, T., Wang, K., Huang, F., Gong, X., Cao, Y., 2017. Perovskite hybrid solar cells with a fullerene derivative electron extraction layer. J. Mater. Chem. C 5, 4190–4197. Yang, D., Ma, D., 2013. 1,1-Bis[(di-4-tolylamino)phenyl]cyclohexane for fast response organic photodetectors with high external efficiency and low leakage current. J. Mater. Chem. C 1, 2054–2060. Yao, N., Huang, J., Fu, K., Deng, X., Ding, M., Zhang, S., Xu, X., Li, L., 2016. Reduced interfacial recombination in dye-sensitized solar cells assisted with NiO: Eu3+, Tb3+ coated TiO2 film. Sci. Rep. 6. Yoon, S., Jo, J.W., Yu, S.H., Yung, J.H., Son, H.J., Chung, D.S., 2017. Development of novel conjugated polyelectrolytes as water-processable interlayer materials for highperformance organic photodiodes. ACS Photonics 4, 703–709. Zhang, Y., Dang, X.-D., Kim, C., Nguyen, T.-Q., 2011. Effect of charge recombination on the fill factor of small molecule bulk heterojunction solar cells. Adv. Energy Mater. 1, 610–617. Zhang, Q., Kan, B., Liu, F., Long, G., Wan, X., Chen, X., Zuo, Y., Ni, W., Zhang, H., Li, M., Hu, Z., Huang, F., Cao, Y., Liang, Z., Zhang, M., Russell, T.P., Chen, Y., 2015. Smallmolecule solar cells with efficiency over 9%. Nat. Photon. 9, 35–41. Zhang, M., Wang, J., Zhang, F., Mi, Y., An, Q., Wang, W., Ma, X., Zhang, J., Liu, X., 2017a. Ternary small molecule solar cells exhibiting power conversion efficiency of 10.3%. Nano Energy 39, 571–581. Zhang, M., Zhang, F., An, Q., Sun, Q., Wang, W., Ma, X., Zhang, J., Tang, W., 2017b. Nematic liquid crystal material as morphology regulator for ternary small molecule solar cells with power conversion efficiency exceeding 10%. J. Mater. Chem. A 5, 3589–3598. Zhong, L., Gao, L., Bin, H., Hu, Q., Zhang, Z.-G., Liu, F., Russell, T.P., Zhang, Z., Li, Y., 2017. High efficiency ternary nonfullerene polymer solar cells with two polymer donors and an organic semiconductor acceptor. Adv. Energy Mater. 7, 1602215–1602221. Zhou, Y., Khan, T.M., Shim, J.W., Dindar, A., Fuentes-Hernandez, C., Kippelen, B., 2014. All-plastic solar cells with a high photovoltaic dynamic range. J. Mater. Chem. A 2, 3492–3497.

Intensity dependence of current–voltage characteristics and recombination in highefficiency solution-processed small-molecule solar cells. ACS Nano 7, 4569–4577. Lenes, M., Morana, M., Brabec, C.J., Blom, P.W.M., 2009. Recombination-limited photocurrents in low bandgap polymer/fullerene solar cells. Adv. Funct. Mater. 19, 1106–1111. Li, N., Machui, F., Waller, D., Koppe, M., Brabec, C.J., 2011. Determination of phase diagrams of binary and ternary organic semiconductor blends for organic photovoltaic devices. Sol. Energy Mater. Sol. Cells 95, 3465–3471. Li, H.-W., Guan, Z., Cheng, Y., Lui, T., Yang, Q., Lee, C.-S., Chen, S., Tsang, S.-W., 2016. On the study of exciton binding energy with direct charge generation in photovoltaic polymers. Adv. Electron. Mater. 2, 1600200–1600209. Lin, Y., Ma, L., Li, Y., Liu, Y., Zhu, D., Zhan, X., 2013. A solution-processable small molecule based on benzodithiophene and diketopyrrolopyrrole for high-performance organic solar cells. Adv. Energy Mater. 3, 1166–1170. Lin, Y., Wang, J., Zhang, Z.G., Bai, H., Li, Y., Zhu, D., Zhan, X., 2015. An electron acceptor challenging fullerenes for efficient polymer solar cells. Adv. Mater. 27, 1170–1174. Liu, F., Zhao, W., Tumbleston, J.R., Wang, C., Gu, Y., Wang, D., Briseno, A.L., Ade, H., Russell, T.P., 2014. Understanding the morphology of PTB7: PCBM blends in organic photovoltaics. Adv. Energy Mater. 4, 1–9. Liu, C., Li, Z., Zhang, X., Guo, W., Zhang, L., Ruan, S., 2017. Annealing-free ZnO:PEI composite cathode interfacial layer for efficient organic solar cells. ACS Photonics 4, 2952–2958. Liu, Z., Niu, S., Wang, N., 2017. Light illumination intensity dependence of current-voltage characteristics in polymer solar cells with solution-processed titanium chelate as electron extraction layer. Sol. Energy 155, 1044–1051. Lu, L., Xu, T., Chen, W., Lee, J.M., Luo, Z., Jung, I.H., Park, H.I., Kim, S.O., Yu, L., 2013. The role of N-doped multiwall carbon nanotubes in achieving highly efficient polymer bulk heterojunction solar cells. Nano Lett. 13, 2365–2369. Lu, L., Chen, W., Xu, T., Yu, L., 2015. High-performance ternary blend polymer solar cells involving both energy transfer and hole relay processes. Nat. Commun. 6, 7327–7333. Ma, X., Zhang, F., An, Q., Sun, Q., Zhang, M., Miao, J., Hu, Z., Zhang, J., 2017. Liquid crystal material as the third component for ternary polymer solar cells with efficiency of 10.83% and enhanced stability. J. Mater. Chem. A 5, 13145–13153. Marinins, A., Yan, Z., Chen, H., Linnros, J., Veinot, J.G.C., Popov, S., Sychugov, I., 2016. Photostable polymer/Si nanocrystal bulk hybrids with tunable photoluminescence. ACS Photonics 3, 1575–1580. Meng, X., Zhang, W., Tan, Z.a., Li, Y., Ma, Y., Wang, T., Jiang, L., Shu, C., Wang, C., 2012. Highly efficient and thermally stable polymer solar cells with dihydronaphthyl-based [70] fullerene bisadduct derivative as the acceptor. Adv. Funct. Mater. 22, 2187–2193. Meng, X., Zhang, W., Tan, Z.a., Du, C., Li, C., Bo, Z., Li, Y., Yang, X., Zhen, M., Jiang, F., 2012. Dihydronaphthyl-based [60] fullerene bisadducts for efficient and stable polymer solar cells. Chem. Commun. 48, 425–427. Meng, X., Zhao, G., Xu, Q., Tan, Z.a., Zhang, Z., Jiang, L., Shu, C., Wang, C., Li, Y., 2014. Effects of fullerene bisadduct regioisomers on photovoltaic performance. Adv. Funct. Mater. 24, 158–163. Moritomo, Y., Yonezawa, K., Yasuda, T., 2015. Carrier density effect on recombination in PTB7-based solar cell. Sci. Rep. 5, 13648. Nian, L., Gao, K., Liu, F., Kan, Y., Jiang, X., Liu, L., Xie, Z., Peng, X., Russell, T.P., Ma, Y., 2016. 11% efficient ternary organic solar cells with high composition tolerance via integrated near-ir sensitization and interface engineering. Adv. Mater. 28, 8184–8190.

173