Improved fullerene-free polymer solar cells using a rationally designed binary mixed solution of an electron extracting layer

Abstract

In continuation of the progressive development of cathode interfacial engineering for successfully improving the power conversion efficiency (PCE) of organic solar cells (OSCs), a breakthrough strategy of combining two small organic materials named Alq₃ (aluminum tris(8-hydroxyquinolinate)) and QPhPBr (tetraphenylphosphonium bromide) in one solution acting as a cathode interfacial layer (CIL) has been performed. The strong electron extracting property of Alq₃ determined previously and the crucial process of vacuum deposition led to our concept to propose this idea of a mixed binary solution processing technique. We successfully delivered a high performance of PCE above 11% with an increased V_oc (0.860 V), J_sc (18.38 mA cm⁻²) and FF (69.85%) using the combination of Alq₃ with QPhPBr as a CIL fabricated using a conventional device structure with PBDB-T:ITIC-Th as the blended active layer. The PCE is 2.5-fold higher than the reference device of bare Al cathode (4.38% PCE). Moreover, the binary mixed solution of CIL has shown a better performance than the Ca/Al based device with a PCE value 1.5 times higher than the value of 7.32% of the Ca/Al device. Their attractive performance has promoted a new challenge for new upcoming bulk-heterojunction OSCs in future prospects.

---

Introduction

The photoactive layer, the most specific and attractive part of bulk heterojunction (BHJ) organic solar cells (OSCs) based on a mixed solution of donor and acceptor materials, is the major concern of scientists and researchers for developing effectiveness and achieving a high power conversion efficiency comparable to inorganic perovskite¹ solar cells. It is quite well known that suitable protection of the photoactive layer, sandwiched between two electrodes, is highly recommended. Several problems have been faced, for example, classical diffusion of oxygen and water, exciton degradation, low lifetime, stability etc. from the direct fabrication of the photoactive layer between electrodes, which causes low performance.² As one of the most important conditions, the environmental parameters, which greatly influence the stability and lifetime of these OSCs, are critical for the diffusion of oxygen and water into the photoactive layer upper electrode. Thus, it is very important that the permeability of these electrodes should be minimized towards attraction with oxygen and water molecules. To overcome these major deficiencies of OSCs, barrier layers³ which have the ability to protect the photoactive layer from degradation and corrosion processes are used. Such an additional barrier layer atop the active layer in the normal device structure of an OSC has been proposed as a carrier blocking layer because it acts to avoid quenching of carriers at the interface of the active layer/cathode electrode. Until now, various names have been given to this effective barrier layer with its unique properties. Since this barrier layer shows great qualities and proves to have good characteristics for achieving high performance OSCs, we would like to denote it a “cathode interfacial material (CIM).” Its attractive properties include: avoiding the corrosion of the blend photoactive layer, negligible interference in the Vis-IR range of absorption, a low highest occupied molecular orbital (HOMO) energy level, the creation of a good dipole interaction with the cathode electrode in order to adjust the work function with a low barrier space for enriching electron collection, easy fabrication with optimal thickness and semi-transparency for selectively large electron transportation, and improvement of stability and lifetime of BHJ OSCs. Various types of cathode buffer layers (CBLs) have been proposed, for example, metal oxides⁴ ZnO, TiO_x, Nb₂O₅ and SnO_x were identified especially for an inverted device structure. Then, water soluble conjugated polymer⁵ molecules including PFN, PEI, and PAA were developed and successfully used. Later on, fullerene based salts⁶ with better matching of the lowest unoccupied molecular orbital (LUMO) energy level to the corresponding fullerene electron acceptor were introduced. In a similar trend, small molecule perylene diimide based neutral PDIN and ionic PDINO⁷ were also successfully designed. Further, many more metal salts/complexes, sol–gel derivatives, carbon-based materials, organic–inorganic hybrid⁸ materials, graphene⁹ and its derivative based quantum dots as interlayers have been systemically reported. Mostly either single layer¹⁰ or bilayer compositions¹¹ of such CBLs were applied in earlier reports. Recently, our group has introduced a new function of an electron extraction layer (EEL) composed of a mixed solution of two small molecules containing an ionic environment which creates an ion-exchange framework that possesses a high electron affinity and promotes easy extraction and transportation of electrons towards the cathode electrode.¹² Their realistic advantages in the improvement of (PCE) power conversion efficiency results suggest that a mixed blend solution of such an EEL may have a different mechanism strategy for efficient charge extraction. The possible mechanism proposed is that once the two materials are mixed together in solution, a coulombic force attraction is produced between their ionic species. Their continuous interaction and ion-exchange push all the cationic species to one side and their prospective anions on the other side. This layered complex binding structure may preferentially creates a strong electron affinity towards the acceptor material of the active layer interface and the resultant higher possibility of electron extraction taking place. The schematic mechanism is presented in Fig. 1. On one side negative ions passivate the dipole interaction with the positive cathode and adjust it into the proper position for the collection of electrons. During this process, whenever free positive carriers come across the EEL from the exciton dissociated interface, the deep HOMO level of the EEL pushes them back.

Fig. 1 Schematic diagram of the mechanistic pathway of the cathode buffer layer.

Thus, overall our strategy of a mixed blend solution of an EEL fabricated on top of an active layer potentially provides a good platform and has the benefits of improving the open-circuit voltage (V_oc), the short-circuit current density (J_sc), and fill factor (FF), and finally succeeds in achieving a large output of PCE. Previously, He et al. demonstrated the effect of a PFN CBL for the significant enhancement of V_oc, J_sc, and FF in polymer solar cells (PSCs), achieving high efficiency.¹³ Based on the above important facts and the proposed mechanism of a new function of a blend binary mixed solution of M-EEL (mixed electron extraction layer), we have progressed towards a new combination of two small molecules, aluminium tris(8-hydroxyquinoline) (Alq₃) and tetraphenyl phosphonium bromide (QPhPBr). For instance, it has been already justified that Alq₃ has efficient properties as a CBL and provides improvement in the lifetime of OSCs.¹⁴ Alq₃ was previously deposited under vacuum conditions, adopting both an inverted and layer by layer device configuration where control of the layer thickness is very crucial and challenging. Moreover Alq₃ was employed using a single layer or bilayer deposition technique with metal or metal halides.^11d,15 We observed that Alq₃ has good solubility in polar solvents like methanol and can easily mix well with QPhPBr. Hence, after considering all the conditions stated above of these two small molecules acting as an EEL, in this article we proposed the progressive aspects of the binary function of an EEL fabricated on a non-fullerene based blend active layer substrate. Protective encapsulation of the photoactive layer with a binary blend EEL in order to achieve an efficient performance of BHJ OSCs was systemically presented with relevant experimental studies. The photoactive layer was composed of a well-known polymer donor PBDB-T blended with a recently developed thiophene based non-fullerene electron acceptor material ITIC-Th in the presence of 1% chloronaphthalene (CN) additive. Note that using ITIC-Th as an electron acceptor for blend binary devices achieves a PCE of ∼10%¹⁶ under inverted device configurations. There has been a ternary solar cell composition reported with a PCE of above 11%.¹⁷ Until now, there have been no reports of using a PBDB-T polymer donor combined with ITIC-Th in a binary solar cell system fabricated under conventional configurations. Of the three reference device configurations with CBLs of Ca/Al, only Al and methanol/Al were also taken into consideration. To explain our satisfactory results from the binary behaviour of EEL, the comparative performance of their single molecules was also predicted. All the EEL conditions were fabricated using a standard conventional device configuration: ITO/PEDOT:PSS/active layer/CBL/Al. The molecular structure of the polymer donor, acceptor and cathode buffer layer material are shown in Fig. 2.

Fig. 2 Molecular structures (a) and HOMO and LUMO energy alignments (b) of the materials used in this study: donor PBDB-T, acceptor ITIC-Th, CIL substrates Alq₃ and QPhPBr.

Results and discussion

The conventional device structures of the PBDB-T:ITIC-Th blend active layer with five different CBL encapsulation treatments were fabricated and their respective J–V characteristics are shown in Fig. 3 and Table 1. The optimal Alq₃–QPhPBr weight ratio is 1 : 1.5 (Table S1, ESI†). All the device parameters were examined in series starting from bare Al and moving towards a single solution (Alq₃ or QPhPBr) followed by a blend mixture of a binary solution of Alq₃:QPhPBr as the CBL. The optimized thickness condition (∼12 nm) of each CBL material was used in fabrication (Table S2, ESI†). It was determined that a thickness of CBL of less than 12 nm is not sufficient for protection of the active layer to avoid direct contact of the Al cathode and with a thickness of above 12 nm, hindrance of electron transport from the electron acceptor to the Al cathode takes place and causes high series resistance. Data are given in the ESI† documents. Moving from the bare Al cathode to binary Alq₃:QPhPBr, a large improvement in the V_oc from 0.634 V to 0.860 V was observed. A large improvement in V_oc from the new blend buffer layer is attributed to the low energy loss and proper tuning of the low cathode work function via dipole interaction, which induces a band bending mechanism between the cathode interfacial layer and the Al cathode, creating excellent ohmic contact and preventing space charge under the device working condition. The relative increase of V_oc from the initial to the highest CBL was calculated as 0.634 < 0.731 < 0.815 < 0.844 < 0.860 for bare Al, methanol treatment, Alq₃ solution, QPhPBr solution and Alq₃:QPhPBr mixture, respectively. Next, parameter J_sc also showed effective variation by replacing the bare Al cathode condition to insertion of the mixed blend buffer layer. The highest J_sc of 18.38 mA cm⁻² was measured using a blend mixture which is comparatively 35% higher than the bare Al cathode observed at 13.67 mA cm⁻². Also, the single buffer layer material presented a lower response of J_sc as 15.78 mA cm⁻² from methanol treatment, 16.57 mA cm⁻² from Alq₃ and 17.41 mA cm⁻² from QPhPBr buffer layer.

Table 1 Summary of photovoltaic data of the blend active layer PBDB-T:ITIC-Th using different CBL environments

ETL	V oc (V)	J sc (mA cm^−2)	J cacl (mA cm^−2)	FF (%)	PCEMax^b (%)	PCEAve^c (%)	μ e (10^−4) cm^2 V^−1 s^−1	R s (Ω cm^2)	R sh (kΩ cm^2)
a Calculated from the EQE spectrum. b The best performing device was prepared under the optimal conditions as provided in the methods section. c The statistical results were obtained from 25 cells.
Al	0.634 ± 0.004	13.67 ± 0.42	12.11	50.61 ± 1.16	4.38	3.95 ± 0.43	2.19 ± 0.15	25.55 ± 0.20	0.68 ± 0.11
MeOH	0.731 ± 0.006	15.78 ± 0.38	14.68	64.44 ± 1.64	7.43	7.08 ± 0.35	3.79 ± 0.11	18.86 ± 0.25	0.76 ± 0.07
Alq3	0.815 ± 0.002	16.57 ± 0.41	15.81	67.92 ± 1.38	9.17	8.92 ± 0.25	3.85 ± 0.25	15.68 ± 0.18	0.92 ± 0.08
QPhPBr	0.844 ± 0.003	17.41 ± 0.45	16.74	69.11 ± 1.55	10.15	9.74 ± 0.41	4.10 ± 0.12	8.78 ± 0.20	1.25 ± 0.10
Alq3:QPPBr	0.860 ± 0.004	18.38 ± 0.31	17.5	69.85 ± 0.77	11.05	10.83 ± 0.22	4.17 ± 0.18	6.22 ± 0.14	1.87 ± 0.12
Ca/Al	0.792 ± 0.002	15.26 ± 0.35	—	60.73 ± 1.75	7.32	6.89 ± 0.43	—	—	—

---

J _sc depends upon the full spectrum range of absorption from the blend photoactive layer and morphological characteristics. Without CBL encapsulation, degradation of its surface generally occurs due to exposure of moisture and air which can be easily understood from the bare Al condition response. However, the significant increase of J_sc from the blend mixture of CBL compared to the single buffer layer suggests that their combination would have provided a better tendency of the number of excitons absorbed within the lifetime period and makes them more stable. The increase of J_sc performance occurs in the order 18.38 > 17.41 > 16.57 > 15.78 > 13.67 from Alq₃:QPhPBr > QPhPBr > Alq₃ > methanol > bare Al, respectively. Moreover, the high J_sc from such a new buffer layer can be also explained due to the fact that the optimized thickness behaviour over the active layer accompanies the reduction in the electron trap state during Al deposition, while the decrease in J_sc from thicknesses of CBL over 12 nm gives a suitable explanation (see ESI† documents). Additionally, the higher J_sc output stimulated under the light J–V curve was further configured with external quantum efficiency (EQE) measurements (Fig. 3b). EQE spectra from different CBL conditions show good absorption in a wide range of 360–800 nm. The EQE spectrum is further improved while changing CBL from the bare Al cathode to the binary buffer layer. Notably, the calculated J_sc from the product of the EQE integration demonstrated for the devices with different condition of the interlayer perfectly matches with the value measured from its J–V characteristics.

Fig. 3 (a) Under light illumination, J–V characteristic curves of PBDB-T:ITIC-Th based blend active layer under different coating of CBL substrates. (b) Their corresponding EQE spectra.

The third important parameter to take into consideration is the FF. The large built in potential resulting in greater exciton dissociation creates a number of free charge carriers followed by improved mobility of the carriers in proper channels and avoids recombination loss, corresponding to a high FF of the given devices. For instance, better density packing and surface morphology acquire easy charge transportation and collection at the respective electrodes, which is also a beneficial process. Thus, a proper methodology of fabrication with a suitable thickness and film forming ability of the active layer followed by suitable protection coating with optimized thickness of CBL are collectively responsible for achieving a high FF. Effective coating of a new blend mixture of CBL Alq₃:QPhPBr over the active layer shows an FF of 69.85%, which is much higher than the value of 50.61% without CBL coating from bare Al as well as with methanol treatment condition (64.44%). A further slight improvement in the FF was observed from single CBL coating 67.92% from Alq₃ and 69.11% from QPhPBr, which proves that the combined behaviour of a mixed solution reaches the top limit of electron extraction, transportation and collection. Additionally, they were capable of pushing back hole carriers to the opposite electrode and improved the electron to hole mobility ratio due to the existence of the deep HOMO energy level. Hence the total PCE which is defined by multiplication of three controlling parameters, V_oc, J_sc and FF is attributed to 11.05, 10.15, 9.17, 7.43 and 4.38%, corresponding to Alq₃:QPhPBr, QPhPBr, Alq₃, methanol and bare Al, respectively. The total 2.5 fold increase in PCE was demonstrated from the binary mixture of Alq₃:QPhPBr CBL in comparison to the bare Al cathode. Moreover, the comparative performance with the Ca/Al CBL condition was also measured and observed at V_oc of 0.79 V, J_sc of 15.26 mA cm⁻² and FF of 60.73% (Table 1). These results proved that our new binary mixture solution of CBL has the capability to replace the well-known standard CBL condition of vacuum deposition Ca/Al process to attain highly efficient photovoltaic cells. The absorption properties using UV-Vis absorption spectroscopy were characterized for CBL. Their negligible or transparent behaviour in the full visible and near IR region supports them as ideal substrates for new buffer layer treatments. The absorption spectrum of each buffer substrate and blend active layer thin film is presented in Fig. 4a. Their corresponding HOMO, LUMO energy (Fig. 2b) and band gap were estimated from ultraviolet photoelectron spectroscopy (UPS) measurements (Fig. S1 and Tables S3, S4, ESI†). The deep high HOMO energy level of these new buffer substrates below the level of the active layer promotes hole blocking activity and lowers the recombination process.

Fig. 4 (a) Absorption spectra of thin films of QPhPBr, Alq₃ and the blend active layer, (b) normalized work function plot of different CBL conditions using the UPS method.

Again, the ultraviolet photoelectron spectroscopy (UPS) based work function characteristics of different coatings of the CBL layer were estimated to give a clear explanation of the high V_oc and the data are shown in Fig. 4b. The decrease in the kinetic energy (KE) value was observed and accordingly the work function was calculated. The resultant experimental data clearly shows that the binary mixture of CBL Alq₃:QPhPBr shows an upshift in the vacuum level of 3.96 eV. However, from the single material treatment Alq₃ was observed at 3.6 eV and 4.06 eV from QPhPBr. All the three conditions were upshifted from the bare Al electrode occurring at 4.25 eV. The estimated value of the work function from the new binary buffer properly matching with the LUMO energy level of the electron acceptor promotes easy accessibility of the electron transport towards the cathode and improved band bending property. However, from treatment with Alq₃ or QPhPBr, a small increase in the distance of energy matching with the LUMO electron acceptor would be the reason for the lower performance compared to the binary mixture. Furthermore, the dark current–density study with the different condition of the CBL coating was also monitored to understand the detailed origin of the improved device performance shown in Fig. 5. The low current density appears at a negative bias, corresponding to a reduction in the leakage of current of the devices, which is closely related to the blocking effect of Al diffusion. It was observed that the binary buffer layer of Alq₃:QPhPBr has the lowest leakage current response than other CBL conditions. On the other hand, bare Al possesses a high level of leakage current at negative bias, revealing that the effective blocking phenomenon from the new binary buffer layer is highly recommended. The highest shift in the voltage of the Alq₃:QPhPBr based device compared to that of the bare Al cathode confirms the best ohmic contact formation and perfect band bending effect at the interface between the interlayer and cathode and consequently agrees well for the improved V_oc of the device. Hence it can be explained that band bending is one of the reasons for the high V_oc in the Alq₃:QPhPBr based device.

Fig. 5 Dark J–V curves in the absence and presence of different CBLs.

Next, we move to explore a more in-depth investigation of our J–V data supported by the light source dependence on J_sc study and the J_phvs. V_int measurements. The experimental power law dependence of the photocurrent on the incident light intensity and the resultant plot of the net photocurrent (J_ph = J_L − J_D) dependence on the effective applied voltage V_eff are shown in Fig. 6a and b. Logarithmic plots of different CBL encapsulations indicate that the power exponent from both QPhPBr and the binary buffer interlayer Alq₃:QPhPBr reaches 1.00 which is much higher than the exponent power result of the bare Al cathode appearing at 0.84, confirming that there is a rare possibility of a build-up of net space charges even at the highest illumination intensity.¹⁸ This describes the appearance of a high electric field and high charge carrier motilities even at low effective voltage and consequently prevents bimolecular recombination losses.

Fig. 6 (a) Light intensity dependent short circuit current density plots, (b) J_phvs. V_eff plots under different optimized condition of CBLs.

Another experimental study of the net photocurrent dependence (J_ph = J_D − J_L) vs. effective applied voltage (V_eff) applied for different cathode buffer interlayer conditions also proves that incorporation of such a binary mixture new buffer interlayer provides an improved charge collection efficiency under working conditions. The J_ph − V_eff characteristics of the Alq₃:QPhPBr based device approaches a 94% value at a net photocurrent density of saturation, while the single buffer layer coated device shows a lower response. For example, at maximal power output conditions, J_ph/J_sat for bare Al, methanol, Alq₃ and QPhPBr based CBL devices reaches 90%, 93%, 93.3% and 93.5%, respectively. A drastic improvement in the J_ph/J_sat value by incorporation of the Alq₃:QPhPBr interlayer indicates a superior exciton dissociation into free carriers and subsequently the collection of all the free carriers at the electrodes with minimized bimolecular losses.¹⁸ Such a bimolecular recombination reduction in the new scenario of the cathode buffer layer based device is responsible for the high charge mobility therein. Since charge transport¹⁹ is described by the given Einstein equation D = (K_BT/e) × μ. Where D = diffusion coefficient, K_B is the Boltzmann constant, T is the temperature; e is the electron charge and μ is the mobility. Thus, the enhancement in charge diffusion followed by the charge transport is collectively responsible for the distinctly different J_ph/J_sat values when compared with the reference device.

In general, the large electron extraction and collection at the cathode electrode can be observed due to the high mobility of carriers as demonstrated by using the space charge limited current (SCLC)²⁰ model. The hole mobility of the active layer was estimated to be 1.59 × 10⁻⁴ cm² V⁻¹ s⁻¹ (Fig. S2, ESI†). Considering the central role of the CBL for efficient hole blocking and charge mobility influencing the device performance, moving from the bare Al cathode to the new binary mixture cathode buffer layer, successive improvement in electron mobility from 2.19 × 10⁻⁴ to 4.17 × 10⁻⁴ guarantees efficient electron transport during the charge collection process for achieving high efficiency. However, for other CBL conditions, the electron mobility was calculated as 3.79 × 10⁻⁴, 3.85 × 10⁻⁴, 4.10 × 10⁻⁴ for methanol, Alq₃ and QPhPBr, respectively (Fig. 7a).

Fig. 7 (a) Electron mobility characteristics of different CBL encapsulations, (b) variation with time of the normalized efficiency of OSCs with different CBLs.

It is a well-known fact that the degradation probability of solar cell devices without encapsulation of any cathode buffer layer depends on two severe conditions:²¹ (1) air and moisture exposure into the active layer causes instability and reduces lifetime and (2) the phenomenon involving diffusion of the cathode metal into the active layer during deposition. In this regard, we have demonstrated how the long stability of the proposed device can be achieved in the presence of a new CBL substrate introduced between the active layer and cathode. In order to estimate the lifetime stability of the device in different CBL fabrication, we kept each cell in continuous air exposure for 2 h. The performance of the J–V curve was monitored after 30 minute intervals and the results are presented in Fig. 7b. Without encapsulation of CBL, rapid degradation within one hour occurred, whereas, the binary buffer layer provides high stability with slow degradation in the power conversion efficiency. It was shown that the Alq₃:QPhPBr based device decreased by 14% while the devices based on Alq₃ and QPhPBr were attributed to a degradation rate of 18% and 25%, respectively. However, the bare Al cathode and methanol treated device degraded much faster by 50% and 70%, respectively, in a similar time duration of air exposure. For instance, it was also observed that if these devices were kept under a nitrogen environment, they would be stable for several days. Thus it is expected that encapsulation to shield from water and oxygen can make the lifetime of such OSCs even longer.

Additionally, we performed AFM surface morphology analysis for differently encapsulated ETL. It was found that the mixed binary layer of Alq₃:QPhPBr shows a large reduction in the surface roughness appearing at 3.95 nm which is much smaller than that of the control device occurring at 7.95 nm as well as from single CBL treatment (5.26 nm and 4.12 nm from Alq₃ and QPhPBr, respectively). This suggests that we need to gain a moderate roughness size by incorporating a suitable ETL to improve device efficiency. Detailed discussion and images are provided in Fig. S3 (ESI†).

X-ray diffraction experiments of Alq₃, QPhPBr, and their binary mixture films were conducted and their data are shown in Fig. 8. All three films are highly crystalline with multi diffraction bands clearly shown. Compared to the pure Alq₃ and QPhPBr films, the mixture film exhibits a much different diffraction spectrum, in which the band positions are shifted and the band intensity becomes intense or weak. The comparisons of the XRD data clearly indicated that Alq₃ and QPhPBr mixed to form a new species, for example, through the intermolecular ππ-interaction between Alq₃ 8-hydroxyquinolinate and QPhPBr benzene and dipolar-static interactions between Alq₃ and QPhPBr, which may explain the different tuning in the metal work function and different output of the device performance compared to their individual counterpart based CILs.

Fig. 8 The XRD data of the CBL materials, Alq₃, QPhPBr and their binary mixture on the silica substrate.

Conclusions

The effective roles of five different conditions of cathode buffer interlayers using a PBDB-T:ITIC-Th blend active layer were demonstrated. By comparing their drastic features in different CBL fabrications, it was evident that insertion of a blend binary mixture of Alq₃:QPhPBr under optimized thickness and weight ratio conditions is beneficial for improving device performance with V_oc, 0.860 V; J_sc, 18.38 mA cm⁻²; FF, 69.85% and total PCE of 11.05% being achieved. The purpose of such a binary layer was also to mitigate the slow degradation process under air exposure to achieve long term stability and lifetime. In contrast, without encapsulation of CBL between the active layer and the metal cathode, rapid degradation occurs as serious damage of the active layer leads to reduced diode property, a large space charge barrier and finally lowers solar cell efficiency. However, based on a systematic study of the correlation between the OSC performance and the mixed binary layer of the organic cathode buffer material properties, this provides basic guidelines for further application of small molecule organic CBL materials. These beneficial findings and proposed mechanism may in turn have commercial interest for developing new display technology that offers alternatives to the current flat panel display. Their relatively highly efficient performance intends to focus on more challenging and superior development of new ideas for CBL promotion.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors gratefully acknowledge the financial support funded by the National Natural Science Foundation of China (NSFC, No. 21773262, 91433202, and 21221002), Chinese Academy of Sciences (CAS, XDB12010200) and CAS President's International Fellowship Initiative (Grant No. 2016PM054).

References

1. (a) D. Angmo, X. Peng, J. Cheng, M. Gao, N. Rolston, K. Sears, C. Zuo, J. Subbiah, S.-S. Kim, H. C. Weerasinghe, R. H. Daustkardt and D. Vak, ACS Appl. Mater. Interfaces, 2018, 10, 22143; (b) M.-H. Li, H.-H. Yeh, Y.-H. Chiang, U. S. Jeng, C.-J. Su, H.-W. Shiu, Y.-J. Hsu, N. Kosugi, T. Ohigashi, Y.-A. Chen, P.-S. Shen, P. Chen and T.-F. Guo, Adv. Mater., 2018, e1801401; (c) C. Ma, D. Shen, M. F. Lo and C.-S. Lee, Angew. Chem., Int. Ed., 2018, 57, 9941; (d) P.-Y. Gu, N. Wang, C. Wang, Y. Zhou, G. Long, M. Tian, W. Chen, X. W. Sun, M. G. Kanatzidis and Q. Zhang, J. Mater. Chem. A, 2017, 5, 7339; (e) L. Zhang, X. Liu, J. Li and S. McKechnie, Sol. Energy Mater. Sol. Cells, 2018, 175, 1; (f) Z. Jia, T. Jiu, Y. Li and Y. Li, Mater. Chem. Front., 2017, 1, 2261.
2. H. Ju, K. M. Knesting, W. Zhang, X. Pan, C.-H. Wang, Y. W. Yang, D. S. Ginger and J. Zhu, ACS Appl. Mater. Interfaces, 2016, 8, 2125.
3. (a) B. Walker, H. Choi and J. Y. Kim, Curr. Appl. Phy., 2017, 17, 370; (b) J. Min, Y. N. Luponosov, Z.-G. Zhang, S. A. Ponomarenko, T. Ameri, Y. Li and C. J. Brabec, Adv. Energy Mater., 2014, 4, 1400816; (c) N. Wang, K. Zhao, T. Ding, W. Liu, A. S. Ahmed, Z. Wang, M. Tian, X. W. Sun and Q. Zhang, Adv. Energy Mater., 2017, 7, 1700522; (d) L. Lv, X. Wang, T. Dong, X. Wang, X. Wu, L. Yang and H. Huang, Mater. Chem. Front., 2017, 1, 1317.
4. (a) M. Tountas, Y. Topal, A. Verykios, A. Soultati, A. Kaltzoglou, T. A. Papadopoulos, F. Auras, K. Seintis, M. Fakis, L. C. Palilis, D. Tsikritzis, S. Kennou, A. Fakharuddin, L. Schmidt-Mende, S. Gardelis, M. Kus, P. Falaras, D. Davazoglou, P. Argitis and M. Vasilopoulou, J. Mater. Chem. C, 2018, 6, 1459; (b) T. Van-Huong, R. Khan, I.-H. Lee and S.-H. Lee, Sol. Energy Mater. Sol. Cells, 2018, 179, 260; (c) Y. Choi, G. Kim, H. Kim, S. H. Kim and K. Lee, Thin Solid Films, 2015, 583, 86.
5. (a) D. Zhou, X. Cheng, H. Xu, H. Yang, H. Liu, F. Wu, L. Chen and Y. Chen, J. Mater. Chem. A, 2016, 4, 18478; (b) H. Wang, W. Zhang, C. Xu, X. Bi, B. Chen and S. Yang, ACS Appl. Mater. Interfaces, 2013, 5, 26; (c) J.-S. Yeo, M. Kang, Y.-S. Jung, R. Kang, S.-H. Lee, Y.-J. Heo, S.-H. Jin, D.-Y. Kim and S.-I. Na, Nano Energy, 2016, 21, 26; (d) Y. Zhou, C. Fuentes-Hernandez, J. Shim, J. Meyer, A. J. Giordano, H. Li, P. Winget, T. Papadopoulos, H. Cheun, J. Kim, M. Fenoll, A. Dindar, W. Haske, E. Najafabadi, T. M. Khan, H. Sojoudi, S. Barlow, S. Graham, J.-L. Bredas, S. R. Marder, A. Kahn and B. Kippelen, Science, 2012, 336, 327; (e) B. A. E. Courtright and S. A. Jenekhe, ACS Appl. Mater. Interfaces, 2015, 7, 26167.
6. (a) C. Cui, Y. Li and Y. Li, Adv. Energy Mater., 2017, 7; (b) X. Liu, W. Jiao, M. Lei, Y. Zhou, B. Song and Y. Li, J. Mater. Chem. A, 2015, 3, 9278; (c) W. Jiao, D. Ma, M. Lv, W. Chen, H. Wang, J. Zhu, M. Lei and X. Chen, J. Mater. Chem. A, 2014, 2, 14720; (d) J. W. Jung, J. W. Jo and W. H. Jo, Adv. Mater., 2011, 23, 1782.
7. (a) Z.-G. Zhang, B. Qi, Z. Jin, D. Chi, Z. Qi, Y. Li and J. Wang, Energy Environ. Sci., 2014, 7, 1966; (b) J.-l. Wu, W.-K. Huang, Y.-C. Chang, B.-C. Tsai, Y.-C. Hsiao, C.-Y. Chang, C.-T. Chen and C.-T. Chen, J. Mater. Chem. A, 2017, 5, 12811.
8. Z. Zheng, Q. Hu, S. Zhang, D. Zhang, J. Wang, S. Xie, R. Wang, Y. Qin, W. Li, L. Hong, N. Liang, F. Liu, Y. Zhang, Z. Wei, Z. Tang, T. P. Russell, J. Hou and H. Zhou, Adv. Mater., 2018, 30, 1801801.
9. (a) H. B. Yang, Y. Q. Dong, X. Wang, S. Y. Khoo, B. Liu and C. M. Li, Sol. Energy Mater. Sol. Cells, 2013, 117, 214; (b) L. Zhang, Z. C. Ding, T. Tong and J. Liu, Nanoscale, 2017, 9, 3524; (c) H. B. Yang, Y. Q. Dong, X. Wang, S. Y. Khoo and B. Liu, ACS Appl. Mater. Interfaces, 2014, 6, 1092.
10. (a) L. Zhang, C. Liu, T. Lai, H. Huang, X. Peng, F. Huang and Y. Cao, J. Mater. Chem. A, 2016, 4, 15156; (b) D.-X. Yuan, X.-D. Yuan, Q.-Y. Xu, M.-F. Xu, X.-B. Shi, Z.-K. Wang and L.-S. Liao, Phys. Chem. Chem. Phys., 2015, 17, 26653.
11. (a) Z. Huai, L. Wang, Y. Sun, R. Fan, S. Huang, X. Zhao, X. Li, G. Fu and S. Yang, ACS Appl. Mater. Interfaces, 2018, 10, 5682; (b) J. Min, H. Zhang, T. Stubhan, Y. N. Luponosov, M. Kraft, S. A. Ponomarenko, T. Ameri, U. Scherf and C. J. Brabec, J. Mater. Chem. A, 2013, 1, 11306; (c) H. I. Kim, B. Thi Thu Trang, G.-W. Kim, G. Kang, W. S. Shin and T. Park, ACS Appl. Mater. Interfaces, 2014, 6, 15875; (d) A. Singh, A. Dey, D. Das and P. K. Iyer, ACS Appl. Mater. Interfaces, 2016, 8, 10904; (e) A. Singh, A. Dey, D. Das and P. K. Iyer, J. Mater. Chem. C, 2017, 5, 6578.
12. (a) W. Li, D. Yan, W. liu, J. Chen, W. Xu, C. Zhan and J. Yao, Sol. RRL, 2017, 1, 1700014; (b) M. Gupta, D. Yan, J. Xu, J. Yao and C. Zhan, ACS Appl. Mater. Interfaces, 2018, 10, 5569; (c) Y. D. Monika Gupta, Shen Fugang, Xu Jianzhong and Zhan Chuanlang, Acta Phys. Chim. Sin, 2018, 34(X) DOI:10.3866/PKU.WHXB201805101; (d) W. Liu, W. Li, J. Yao and C. Zhan, Chin. Chem. Lett., 2018, 28, 381; (e) W. Li, D. Yan, F. Liu, T. Russell, C. Zhan and J. Yao, Sci. China: Chem., 2018, 62 DOI:10.1007/s11426-018-9320-3.
13. Z. He, C. Zhong, X. Huang, W.-Y. Wong, H. Wu, L. Chen, S. Su and Y. Cao, Adv. Mater., 2011, 23, 4636.
14. (a) Y. Lare, B. Kouskoussa, K. Benchouk, S. O. Djobo, L. Cattin, M. Morsli, F. R. Diaz, M. Gacitua, T. Abachi, M. A. del Valle, F. Armijo, G. A. East and J. C. Bernede, J. Phys. Chem. Solids, 2011, 72, 97; (b) Q. L. Song, F. Y. Li, H. Yang, H. R. Wu, X. Z. Wang, W. Zhou, J. M. Zhao, X. M. Ding, C. H. Huang and X. Y. Hou, Chem. Phys. Lett., 2005, 416, 42; (c) C. Zhang, P. Zhang, X. Xu, Y. Dang, X. Chen and B. Kang, Mater. Lett., 2016, 164, 591.
15. Z. El Jouad, L. Barkat, N. Stephant, L. Cattin, N. Hamzaoui, A. Khelil, M. Ghamnia, M. Addou, M. Morsli, S. Bechu, C. Cabanetos, M. Richard-Plouet, P. Blanchard and J. C. Bernede, J. Phys. Chem. Solids, 2016, 98, 128.
16. (a) F. Zhao, S. Dai, Y. Wu, Q. Zhang, J. Wang, L. Jiang, Q. Ling, Z. Wei, W. Ma, W. You, C. Wang and X. Zhan, Adv. Mater., 2017, 29, 1700144; (b) Z. Li, K. Jiang, G. Yang, J. Y. L. Lai, T. Ma, J. Zhao, W. Ma and H. Yan, Nat. Commun., 2016, 7, 13094; (c) Y. Lin, F. Zhao, Q. He, L. Huo, Y. Wu, T. C. Parker, W. Ma, Y. Sun, C. Wang, D. Zhu, A. J. Heeger, S. R. Marder and X. Zhan, J. Am. Chem. Soc., 2016, 138, 4955.
17. (a) Q. An, F. Zhang, W. Gao, Q. Sun, M. Zhang, C. Yang and J. Zhang, Nano Energy, 2018, 45, 177; (b) Y. Chen, P. Ye, X. Jia, W. Gu, X. Xu, X. Wu, J. Wu, F. Liu, Z.-G. Zhu and H. Huang, J. Mater. Chem. A, 2017, 5, 19697; (c) W. Chen and Q. Zhang, J. Mater. Chem. C, 2017, 5, 1275.
18. M. Lenes, M. Morana, C. J. Brabec and P. W. M. Blom, Adv. Funct. Mater., 2009, 19, 1106.
19. V. D. Mihailetchi, L. J. A. Koster, J. C. Hummelen and P. W. M. Blom, Phys. Rev. Lett., 2004, 93, 216601.
20. G. G. Malliaras, J. R. Salem, P. J. Brock and C. Scott, Phys. Rev. B: Condens. Matter Mater. Phys., 1998, 58, 13411.
21. P. Peumans, V. Bulovic and S. R. Forrest, Appl. Phys. Lett., 2000, 76, 2650.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c8qm00311d

---