Efficient, high yield perovskite/fullerene planar-heterojunction solar cells via one-step spin-coating processing

Abstract

In this paper, we report the fabrication of high crystalline perovskite film planar-heterojunction solar cells by a facile one-step spin-coating technique with improved control of thermal annealing time and solution concentration. We used an indium-doped tin oxide glass/poly(3,4-ethylenedioxythiophene):poly(styrene-sulfonate) (PEDOT:PSS) substrate as the hole transport layer, a PHJ of CH₃NH₃PbI₃ perovskite fabricated by one-step spin-coating processing as the active layer and fullerene structure as the electron transport layer, a thin bathocuproine (BCP) film as an hole-blocking layer (HBL), and an aluminum (Al) layer as the negative electrode. The optimized device under AM 1.5 (100 mW cm⁻²) radiation achieved a high efficiency of 12.21% with an open circuit voltage of 0.83 V and FF of 0.68. Meanwhile, the devices do not show obvious hysteresis photovoltaic response, which has been a fundamental bottleneck for perovskite devices. The effects of MAI concentration and annealing time on the solar cells were also discussed on the basis of experimental observations.

---

Introduction

Organometal trihalide perovskites are emerging as a new generation of solution processable and low-cost materials. Owing to their broad spectral absorption,¹ high charge-carrier mobility,² small exciton binding energy (≈50 meV),³ and long exciton diffusion length,^4,5 they have recently attracted enormous attention for thin-film photovoltaics.^6–10 High-performance perovskite solar cells in both mesoporous scaffold and planar heterojunction (PHJ) architectures have been reported to show power conversion efficiencies (PCEs) over 19% with some dedicated energy level engineering.^{4,7,8,11–20} However, in traditional PHJ structure perovskite solar cells, a high-quality condensed TiO₂ requires high temperature treatment above 450 °C, which limits future development.²¹ Meanwhile, the TiO₂ layer might result in unstable device performance under ultraviolet light, which could be attributed to instability resulting from the UV light-induced desorption of surface-adsorbed oxygen.⁷ Some groups have attempted to fabricate all low-temperature (<150 °C) processed perovskite solar cells without condensed TiO₂ layer using new PHJ structure.^22–25 Guo et al.²³ reported a PHJ of CH₃NH₃PbI₃ perovskite/fullerene solid-state solar cells fabricated by spin-coating process can deliver a 3.9% PCE. Dai et al.²⁴ reported a layer-by-layer growth of CH₃NH₃PbI_3−xCl_x for planar heterojunction (PHJ) perovskite solar cells with a PCE as high as 15.12%.

Spin-coating, one of the cheapest film production methods, is widely used in solution-processed planar heterojunction perovskite solar cells. However, films produced by the conventional spin-coating methods were found to be composed of large CH₃NH₃PbI₃ grains and many uncovered pin-hole areas, which could produce the shunting in such planar heterojunction devices.²⁶ Therefore, a nonporous homogenous perovskite film must be fabricated to obtain an excellent performance of planar device. To date, the CH₃NH₃PbI₃ layer in the most efficient planar solar cells has been fabricated by either vapor deposition, a two-step sequential solution deposition, or a vapor-assisted two-step reaction process.^26–28 The vapor deposition process is likely to increase the manufacturing cost.²⁶ Moreover, the sequential two-step deposition on planar substrates has two problems: one is the incomplete conversion of PbI₂ and the other is the uncontrolled perovskite crystal sizes as well as surface morphology.^26,29 Therefore, a facile one-step spin-coating technique that can produce high-quality films with controlled morphology is highly desirable for the construction of planar devices. Seok et al.¹³ reported the solution-process fabrication of efficient perovskite solar cells comprised of bilayer architecture and MAPb(I_1−xBr_x)₃ perovskites on the mesoporous TiO₂ scaffold, with a 16.2% power-conversion efficiency (PCE) achieved under 1 sun illumination. Cheng et al.³⁰ also reported a one-step, solvent-induced, fast crystallization-deposition method that results in flat, highly uniform CH₃NH₃PbI₃ thin films. Inspired by their work and our previous works,^25,31 highly efficient perovskite/fullerene planar-heterojunction solar cells were attempted to be fabricated by a facile one-step spin-coating technique.

Herein, we report the fabrication of ordered perovskite films planar-heterojunction solar cells by a facile one-step spin-coating technique. We used an indium-doped tin oxide glass (ITO)/poly(3,4-ethylenedioxythiophene):poly(styrene-sulfonate) (PEDOT:PSS) substrate as hole transport layer, a PHJ of CH₃NH₃PbI₃ perovskite fabricated by one-step spin-coating processing as the active layer and fullerene structure as electron transport layer, a thin bathocuproine (BCP) film as an hole-blocking layer (HBL) and an aluminum (Al) layer as negative electrode. The optimized device under AM 1.5 (100 mW cm⁻²) radiation achieved a high efficiency of 12.21% with an open circuit voltage of 0.83 V and FF of 0.68. Meanwhile, the devices do not show obvious hysteresis photovoltaic response, which has been a fundamental bottleneck for perovskite devices.

Experimental

Materials

PEDOT:PSS (Levios P VP. AL 4083) was obtained from Heraeus-(Precious Metals GmbH & Co. KG) Inc. Fullerene (99.5%); CH₃NH₃I (99.5%), bathocuproine (BCP) (98.0%) and PbI₂ (99.999%) were obtained from Alfa Aesar. Dimethyl sulfoxide (DMSO) (≥99.0%); 4-hydroxybutanoic acid lactone (GBL) (≥99.0%) and toluene (≥99.0%) were purchased from Sinopharm Chemical. Reagent. Co., Ltd. Absolute ethanol (99.7%) and acetone (99.5%) were obtained from Nanjing chemical reagent co., Ltd.

Fabrication of CH₃NH₃PbI₃ perovskite PHJ solar cells

The structure of the devices in this study is illustrated in Fig. 1. ITO glass substrates were cleaned in detergent, deionized (DI) water, acetone, and ethanol for 15 min in ultrasonicator (Shumei KQ300DE). After being dried with nitrogen gas, the ITO surface was treated with UV–ozone for 15 min. Poly(3,4-ethylenedioxythiophene):poly(styrene sulfonate) (PEDOT:PSS) layer was spin-coated on ITO at 3500 rpm for 1 min, followed by baking at 120 °C for 60 min in air, and then the PEDOT:PSS-coated substrates were transferred into a glove box filled with high-purity N₂. The 1.0 M CH₃NH₃I (MAI) and 1.0 M PbI₂ for MAPbI₃ solution were stirred in a mixture of GBL and DMSO (7 : 3 v/v) at 60 °C for 12 h. The resulting solution was coated onto the PEDOT:PSS substrates by a one-step spin-coating process. During the spin-coating step, the substrate (around 1.5 cm × 1.5 cm) was treated with toluene drop-casting. A detailed time-rotation profile for the spin-coating is shown in ESI Fig. S1.† The film becomes darker after annealing at 100 °C for 5 min. The C₆₀ (30 nm), BCP (10 nm) and Al (120 nm) were thermally deposited on the substrate inside a vacuum chamber. The evaporation rates were monitored by a quartz oscillator system, and the film thickness was calibrated by a surface profiler (Veeco Dektak 6M). Finally, the devices with an active area of 0.11 cm² were strictly encapsulated with UV-curable epoxy before being taken out from the glove box.

Fig. 1 (a) Device architecture of the perovskite planar-heterojunction solar cells with the configuration of ITO/PEDOT:PSS/MAPbI₃/C₆₀ (30 nm)/BCP (10 nm)/Al (120 nm). (b) Scheme of the energy levels of each layer in the device.

Characterization

The current density–voltage (J–V) curves of the solar cells were measured with a computer-programmed Keithley 2400 source/meter under 100 mW cm⁻² illumination of simulated AM 1.5G sunlight (Newport's Oriel class A), which was calibrated by the JIS C 8912 standard. The incident photon to electron conversion efficiency IPCE spectra were measured by QTest Station 500AD Solar Cell Quantum Efficiency System (CROWNTECH, INC). The morphology of CH₃NH₃PbI films was measured by using scan electron microscope (SEM) (Hitachi S-4800). UV-Vis spectra were taken using a lambda 35 PerkinElmer ultraviolet-visible (UV-Vis) spectrophotometer. The crystallographic properties of CH₃NH₃PbI₃ perovskite on the glass/PEDOT:PSS substrate were characterized by X-ray diffraction (XRD) with data recorded in the 2θ range of 10–50° at a step of 0.02°. All measurements were carried out at room temperature. Time-resolved photoluminescent (PL) decay were measured using a FLS920 fluorescence spectroscopy (probed wavelength 550 nm, excitation wavelength 700 nm).

Results and discussion

The devices configuration used in this study is illustrated in Fig. 1a. Detailed preparation procedures and materials used can be found in the Experimental section. The cell is the simplest bilayer planar hetero-junction structure generally used in organic solar cells. In this device, the perovskite is used as the light absorber and hole transport layer, whereas C₆₀ is the electron transport layer. A high quality PEDOT:PSS dense film can be obtained when the ITO substrate is preheated to 120 °C for 5 min.¹⁹ Therefore, all devices discussed in this article were fabricated by depositing PEDOT:PSS films on pre-heated ITO substrates via spin-coating.

The CH₃NH₃I (MAI) and PbI₂ for MAPbI₃ solution were stirred in a mixture of GBL and DMSO (7 : 3 v/v) as precursor solutions at varied concentrations. The resulting solution was coated onto the PEDOT:PSS substrates by a one-step spin-coating process. The film becomes darker after annealing at 100 °C for varied durations.

Fig. 1b shows the energy level of each layer in the device. The lowest unoccupied molecular orbital (LUMO) and the highest occupied molecular orbital (HOMO) levels of MAPbI₃ perovskite are −3.9 and −5.4 eV, respectively, and those of C₆₀ are −4.5 and −6.2 eV.^32,33 C₆₀ with a LUMO of −4.5 eV could be an electron acceptor for MAPbI₃ perovskite to create a donor–acceptor PHJ system. Under irradiation, excitons are generated by the absorption of light in the perovskite layer. The oppositely charged holes and electrons in excitons are then separated at the donor–acceptor interface, and are transported by perovskite and C₆₀, respectively, resulting in the photovoltaic effect.

The surface morphologies of the MAPbI₃ perovskite films were examined by SEM, and the results are shown in Fig. 2. In Fig. 2a, we can see that the thin film is in fact very smooth and dense with some cracks in the surface. In Fig. 2b, the magnified SEM picture, we can observe that the surface exhibits a dense-grained uniform morphology with grain sizes in the range of 60–400 nm (Fig. 2c). The entire film is composed of a homogenous, well-crystallized perovskite layer (Fig. 4), which also enables the devices to show high efficiency and easy reproducibility.

Fig. 2 (a) Typical top view SEM image of the annealed perovskite layer, (b) magnified SEM image of (a), and size distribution histogram (c) of perovskite particles prepared by one-step spin coating annealing at 100 °C for 5 min.

Fig. 3 shows typical absorption spectra of a MAPbI₃ film before and after 5 min annealing at the temperature of 100 °C. The absorption at wavelength of around 740 nm from the thin film increased during the 5 min of thermal annealing. And the perovskite film after annealing has an absorption profile that extends to NIR (800 nm) with a higher absorption at 550 nm, which indicates that after annealing the perovskite film is denser.^16,19

Fig. 3 UV/Vis absorption spectra of PEDOT:PSS/perovskite films before and after annealing at 100 °C for 5 min. The inset shows the perovskite thin film after 5 min of annealing.

The crystalline structure of MAPbI₃ perovskite film before and after annealing on the glass/PEDOT:PSS substrate was characterized by X-ray diffraction and shown in the Fig. 4. It could be pointed out that the XRD pattern of lead iodide disappeared after spin-coating without annealing. The annealing step is just a crystal growth process which improves the crystal quality of perovskites as the relative intensity of perovskite XRD peaks increases, which also made the power conversion efficiency increase after thermal treatment (Table 1).

Fig. 4 XRD patterns of CH₃NH₃PbI₃ film before and after annealing 100 °C for 5 min (MAIPbI₃: 1.0 M MAI and 1.0 M PbI₂).

Table 1 Device parameters for solar cells based on perovskite films prepared with varying MAI concentrations or annealing at 100 °C for varied times

MAI conc. (M)	Annealing time (min)	Jsc (mA cm^−2)	Voc (V)	FF	PCE (%)
0.8	5	20.80	0.86	0.56	10.03
1.0	5	21.70	0.83	0.68	12.21
1.2	5	22.10	0.77	0.64	10.94
1.0	0	0.92	0.70	0.15	0.09
1.0	3	19.76	0.82	0.66	11.13
1.0	10	19.33	0.77	0.64	9.46
1.0	15	14.50	0.78	0.30	3.37

---

The devices have a structure of ITO/PEDOT:PSS/MAPbI₃/C₆₀ (30 nm)/BCP (10 nm)/Al (120 nm). The device performance is very sensitive to thermal annealing time, MAI or PbI₂ concentrations. Fig. 5a presents the current density–voltage (J–V) curves of the devices based on perovskite films prepared with varying MAI concentrations from 0.8 M to 1.2 M, demonstrating how the device performance was optimized. The photovoltaic parameters of devices are summarized in Table 1. A perovskite film prepared by higher MAI concentrations could increase the thickness of film. A thicker perovskite film absorbs more light and thus yields a larger photocurrent, while a thicker film causes loss of photovoltage, which might be due to the increased charge recombination. The highest PCE devices were obtained by annealing of MAPbI₃ prepared by 1.0 M MAI at 100 °C for 5 min. The device was measured under AM 1.5 simulated one sun illumination. The best performing device has a J_sc of 21.7 mA cm⁻², a FF of 0.68, an open circuit voltage (V_oc) of 0.83 V and a PCE of 12.21%.

Fig. 5 (a) J–V curves of CH₃NH₃PbI₃ perovskite solar cells annealing at 100 °C for 5 min while varied MAI concentrations from 0.8–1.0 M measured under simulated AM 1.5 sun light. (b) J–V curves of CH₃NH₃PbI₃ perovskite solar cells prepared by 1.0 M MAI solution while annealing at 100 °C for 5–15 min. (c) IPCE spectra (red circles and line) and integrated photocurrent density J_sc (blue line) of one optimized device.

From the previous report,¹⁸ the morphology of a perovskite film can be controlled by annealing time, which has a great impact on the performance of the PHJ perovskite solar cells. Fig. 5b and Table 1 present the current density–voltage (J–V) curves of the devices based on perovskite films annealing of different periods. The highest PCE devices were obtained by annealing of MAPbI₃ films at 100 °C for 5 min. Perovskite film coverage is large with a number of small pores after spin-coating. The small pores either increase in size or close up until the final crystalline phase is reached after different periods of the annealing process.¹⁸ The planar thin-film architecture's lower performance after annealing more than 5 min may arise from pin-hole formation, incomplete coverage of the perovskite resulting in low-resistance shunting paths and lost light absorption in the solar.²⁶ Fig. 5c shows the incident photon to electron conversion efficiency (IPCE) spectrum of one optimized device, which indicates that the device shows a spectral response in the region from the visible to near infrared (400–800 nm) with a peak IPCE of over 90% at approximately 550 nm. The integrated J_sc of 21 mA cm⁻² from the IPCE agree well with measured.

It is known that some perovskite solar cells show photocurrent hysteresis at scan direction and certain voltage scanning rate (or sweep delay time).³⁴ The origin of photo current hysteresis was ascribed to either the traps, ferroelectric properties of the perovskite material and/or the electromigration of ions in the perovskites. Here we change the scanning rate from very fast to very slow, with a delay between measurement voltage points increasing from 0 to 500 ms. The J–V curves and the results measured by different sweep times and two opposite scan directions are shown in Fig. 6. Our device does not show obvious photocurrent hysteresis of photocurrent by changing the sweep rates or the directions in our devices. The main cause of hysteretic behavior in perovskite solar cells can be found in the formation of an internal field upon ion migration, and the role of the electron transport layer is of paramount importance in determining the device response to such a transient phenomenon.³⁵ The C₆₀ used as electron transport layer in our devices might improve charge extraction with respect to interfaces involving compact TiO₂ and make the short-circuit current density insensitive to the transient phenomena to ions migration.

Fig. 6 Photocurrents of a high performance perovskite device measured with different delays between measurement points (a) and different sweep directions (b).

The time-resolved photoluminescent (PL) decay curve of planar-heterojunction solar cell was also measured as shown in Fig. S2.† It can be seen that the photoluminescent (PL) decay curve was shorter than perovskite solar cell used TiO₂ as electron transport layer from previous report,³⁶ which confirmed the faster transfer of the charge carriers in our device. So our device might inject the charge carriers from MAPbI₃ perovskite into C₆₀ more efficiently than that from MAPbI₃ perovskite into TiO₂.

Conclusion

In summary, we report the fabrication of high crystalline perovskite films planar-heterojunction solar cells by one-step spin-coating processing with better control of thermal annealing time and solution concentration. We used an indium-doped tin oxide glass/poly(3,4-ethylenedioxythiophene):poly(styrene-sulfonate) (PEDOT:PSS) substrate as hole transport layer, a PHJ of CH₃NH₃PbI₃ perovskite fabricated by one-step spin-coating processing as the active layer and fullerene structure as electron transport layer, a thin bathocuproine (BCP) film as an hole-blocking layer (HBL) and an aluminum (Al) layer as negative electrode. The optimized device under AM 1.5 (100 mW cm⁻²) radiation achieved a high efficiency of 12.21% with an open circuit voltage of 0.83 V and FF of 0.68. The devices do not show obvious hysteresis photovoltaic response, which has been a fundamental bottleneck for perovskite devices. By optimizing the thickness of the perovskite layer and suitable active layer or perovskite materials, we expect that a higher efficiency can be achieved, and this part of the research is proceeding in our lab.

Acknowledgements

This work was supported by the National High Technology Research and Development Program (“973” Program) of China (Grant No. 2015CB932203) and sponsored by Scientific Starting Fund from Nanjing University of Posts and Telecommunications (NUPTSF) (Grant No. NY215015 and Grant No. NY215106), Natural Science Foundation of Jiangsu Province (BM2012010) National Natural Science Foundation of China (GZ215073) and Synergetic Innovation Center for Organic Electronics and Information Displays.

References

1. M. M. Lee, J. Teuscher, T. Miyasaka, T. N. Murakami and H. J. Snaith, Science, 2012, 338, 643–647.
2. C. Wehrenfennig, G. E. Eperon, M. B. Johnston, H. J. Snaith and L. M. Herz, Adv. Mater., 2014, 26, 1584–1589.
3. V. D'Innocenzo, G. Grancini, M. J. P. Alcocer, A. R. S. Kandada, S. D. Stranks, M. M. Lee, G. Lanzani, H. J. Snaith and A. Petrozza, Nat. Commun., 2014, 5, 3586.
4. S. D. Stranks, G. E. Eperon, G. Grancini, C. Menelaou, M. J. P. Alcocer, T. Leijtens, L. M. Herz, A. Petrozza and H. J. Snaith, Science, 2013, 342, 341–344.
5. G. Xing, N. Mathews, S. Sun, S. S. Lim, Y. M. Lam, M. Grätzel, S. Mhaisalkar and T. C. Sum, Science, 2013, 342, 344–347.
6. O. Malinkiewicz, A. Yella, Y. H. Lee, G. M. Espallargas, M. Graetzel, M. K. Nazeeruddin and H. J. Bolink, Nat. Photonics, 2014, 8, 128–132.
7. P. Docampo, J. M. Ball, M. Darwich, G. E. Eperon and H. J. Snaith, Nat. Commun., 2013, 4, 2761.
8. J. H. Heo, S. H. Im, J. H. Noh, T. N. Mandal, C.-S. Lim, J. A. Chang, Y. H. Lee, H.-j. Kim, A. Sarkar, K. NazeeruddinMd, M. Gratzel and S. I. Seok, Nat. Photonics, 2013, 7, 486–491.
9. B. V. Lotsch, Angew. Chem., Int. Ed., 2014, 53, 635–637.
10. A. Marchioro, J. Teuscher, D. Friedrich, M. Kunst, R. van de Krol, T. Moehl, M. Gratzel and J.-E. Moser, Nat. Photonics, 2014, 8, 250–255.
11. B. Conings, L. Baeten, C. De Dobbelaere, J. D'Haen, J. Manca and H. G. Boyen, Adv. Mater., 2014, 26, 2041–2046.
12. J.-H. Im, I.-H. Jang, N. Pellet, M. Grätzel and N.-G. Park, Nat. Nanotechnol., 2014, 9, 927–932.
13. N. J. Jeon, J. H. Noh, Y. C. Kim, W. S. Yang, S. Ryu and S. I. Seok, Nat. Mater., 2014, 13, 897–903.
14. D. Y. Liu and T. L. Kelly, Nat. Photonics, 2014, 8, 133–138.
15. G. Hodes, Science, 2013, 342, 317–318.
16. Z. Xiao, C. Bi, Y. Shao, Q. Dong, Q. Wang, Y. Yuan, C. Wang, Y. Gao and J. Huang, Energy Environ. Sci., 2014, 7, 2619–2623.
17. N. Yantara, D. Sabba, F. Yanan, J. M. Kadro, T. Moehl, P. P. Boix, S. Mhaisalkar, M. Gratzel and C. Gratzel, Chem. Commun., 2015, 51, 4603–4606.
18. G. E. Eperon, V. M. Burlakov, P. Docampo, A. Goriely and H. J. Snaith, Adv. Funct. Mater., 2014, 24, 151–157.
19. C. H. Chiang, Z. L. Tseng and C. G. Wu, J. Mater. Chem. A, 2014, 2, 15897–15903.
20. A. Listorti, E. J. Juarez-Perez, C. Frontera, V. Roiati, L. Garcia-Andrade, S. Colella, A. Rizzo, P. Ortiz and I. Mora-Sero, J. Phys. Chem. Lett., 2015, 1628–1637.
21. J. B. You, Z. R. Hong, Y. Yang, Q. Chen, M. Cai, T. B. Song, C. C. Chen, S. R. Lu, Y. S. Liu, H. P. Zhou and Y. Yang, ACS Nano, 2014, 8, 1674–1680.
22. T. Liu, Q. Hu, J. Wu, K. Chen, L. Zhao, F. Liu, C. Wang, H. Lu, S. Jia, T. Russell, R. Zhu and Q. Gong, Adv. Energy Mater., 2016, 6, 1501890.
23. J.-Y. Jeng, Y.-F. Chiang, M.-H. Lee, S.-R. Peng, T.-F. Guo, P. Chen and T.-C. Wen, Adv. Mater., 2013, 25, 3727–3732.
24. Y. Chen, T. Chen and L. Dai, Adv. Mater., 2015, 27, 1053–1059.
25. Y. N. Chen, B. B. Li, W. Huang, D. Q. Gao and Z. Q. Liang, Chem. Commun., 2015, 51, 11997–11999.
26. J. Burschka, N. Pellet, S. J. Moon, R. Humphry-Baker, P. Gao, M. K. Nazeeruddin and M. Gratzel, Nature, 2013, 499, 316–319.
27. M. Z. Liu, M. B. Johnston and H. J. Snaith, Nature, 2013, 501, 395–398.
28. F. Hao, C. C. Stoumpos, D. H. Cao, R. P. H. Chang and M. G. Kanatzidis, Nat. Photonics, 2014, 8, 489–494.
29. Q. Chen, H. Zhou, Z. Hong, S. Luo, H.-S. Duan, H.-H. Wang, Y. Liu, G. Li and Y. Yang, J. Am. Chem. Soc., 2014, 136, 622–625.
30. M. Xiao, F. Huang, W. Huang, Y. Dkhissi, Y. Zhu, J. Etheridge, A. Gray-Weale, U. Bach, Y.-B. Cheng and L. Spiccia, Angew. Chem., Int. Ed., 2014, 126, 10056–10061.
31. D. B. Wang, Y. W. Wu, R. Bi, H. M. Zhang and D. W. Zhao, J. Mater. Chem. C, 2015, 3, 3922–3927.
32. L. Etgar, P. Gao, Z. Xue, Q. Peng, A. K. Chandiran, B. Liu, M. K. Nazeeruddin and M. Grätzel, J. Am. Chem. Soc., 2012, 134, 17396–17399.
33. P. Peumans and S. R. Forrest, Appl. Phys. Lett., 2001, 79, 126–128.
34. H. J. Snaith, A. Abate, J. M. Ball, G. E. Eperon, T. Leijtens, N. K. Noel, S. D. Stranks, J. T.-W. Wang, K. Wojciechowski and W. Zhang, J. Phys. Chem. Lett., 2014, 5, 1511–1515.
35. M. De Bastiani, G. Dell'Erba, M. Gandini, V. D'Innocenzo, S. Neutzner, A. R. S. Kandada, G. Grancini, M. Binda, M. Prato, J. M. Ball, M. Caironi and A. Petrozza, Adv. Energy Mater., 2016, 6, 1501453.
36. J. H. Heo, D. H. Song, H. J. Han, S. Y. Kim, J. H. Kim, D. Kim, H. W. Shin, T. K. Ahn, C. Wolf, T.-W. Lee and S. H. Im, Adv. Mater., 2015, 27, 3424–3430.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra07068j

---