Spontaneous configurational evolution induced by an in situ self-formed p-type CuI interface layer in perovskite solar cells

Abstract

The undesirable corrosion reaction between CH₃NH₃PbI₃ and Cu cathode can form a p-type CuI layer in situ. The effect can promote an amazing evolution of the configuration of the hole-transport-material-free perovskite solar cells. The evolution can boost the cell efficiency by a dozen-fold.

---

Organolead halide perovskites such as CH₃NH₃PbX₃ (X = I and Br) have attracted considerable attention since they were first used in dye-sensitized solar cells in 2009.¹ Park et al.² conducted a pioneer work on solid-state solar cells with CH₃NH₃PbI₃ perovskite nanocrystals in 2012. In the same year, Snaith et al.³ also gained a breakthrough achievement on solid-state solar cells with an insulating Al₂O₃ layer. Afterwards, the power conversion efficiency (PCE) of solar cells with CH₃NH₃PbX₃, called perovskite solar cells (PSCs), rapidly increased to more than 22%.⁴ Most of these efficient PSCs contained the organic hole transporting materials (HTMs), such as 2,2′,7,7′-tetrakis-(N,N-di-4-methoxyphenylamino)-9,9′-spirobifluorene (Spiro-OMeTAD), poly(3,4-ethylenedioxythiophene) (PEDOT:PSS), and poly[bis(4-phenyl) (2,4,6-trimethylphenyl)-amine] (PTAA). Despite the high PCEs of the cells, the use of those organic HTMs partially hinders their future commercialization because of their relatively high cost and chemical instability. To solve this problem, PSCs without HTM or with inorganic HTM have been explored. Typically, Etgar et al. first reported HTM-free PSCs and the PCE of the cells can be optimized to more than 10% at present.^5–8 Correspondingly, the few available inorganic p-type materials, including CuSCN,^9,10 Cu₂O,¹¹ NiO,¹² and CuI,^13,14 have been employed as HTMs in PSCs. On the other hand, noble metals, such as gold or silver, served as the cathode in typical PSCs. However, it is well known that the use of noble metals inevitably increases the cost of device fabrication. Therefore, relatively cheaper alternatives, such as nickel,¹⁵ carbon^16,17 and copper,¹⁸ have been reported.

Recently, it was found that some metal cathodes reacted easily with CH₃NH₃PbI₃ and formed metal iodides.^19,20 This undesirable phenomenon is similar to the situation in dye-sensitized solar cells wherein the iodide/triiodide redox couple was prone to corroding metal the cathode and led to degraded cell performance.^19,21,22 Inspired by this, it is expected that CuI can form spontaneously and serve as HTM in principle if a thin Cu layer is deposited on the CH₃NH₃PbI₃ film directly. Therefore, we attempted to fabricate inorganic CuI HTM in PSCs with a cheaper Cu cathode, utilizing the unfavorable corrosive effect of CH₃NH₃PbI₃ to metal materials.

Herein, the HTM-free PSCs with a low-cost Cu cathode were first fabricated, which initially showed low PCE partially due to the relative small work function of Cu. However, this dilemma can be easily broke by simply storing the cells in dry air. It was found that the PCE of the cells showed approximately thirty times enhancement after storing for eight days. This is because of a layer of CuI in situ self-formed during storage caused by the corrosion of CH₃NH₃PbI₃ on the Cu cathode. The self-formed CuI served as a HTM, leading to the spontaneous configurational transformation of the cells from the HTM-free one to the HTM-contained one, and hindered the recombination of electrons and holes at the interface between CH₃NH₃PbI₃ and Cu cathode. This study indicates a possible way to engineer low-cost and efficient PSCs and other photoelectric devices.

We fabricated Cu-cathode PSCs (Cu-PSCs) with a pristine configuration of FTO/TiO₂/CH₃NH₃PbI₃/Cu. However, a layer of CuI was produced between CH₃NH₃PbI₃ and the Cu cathode several days later without extra processing. In contrast, we also fabricated Au-cathode PSCs (Au-PSCs) with the same configuration. The schematic is shown in Fig. S1(a) in ESI.†

Fig. 1(a) presents the schematic of the configurational self-transforming process occurring in our Cu-PSCs. We thought that a layer of CuI was produced in situ between the CH₃NH₃PbI₃ and Cu cathode by a reaction of those two layers. It was the spontaneously formed CuI that made pristine Cu-PSCs change with elevated performance parameters.

Fig. 1 (a) Schematic of the configurational self-transforming process of Cu-PSCs. (b) J–V curves of Cu-PSCs aged for different days. (c) V_oc of Cu-PSCs and Au-PSCs aged for different days. (d) PCE of Cu-PSCs and Au-PSCs aged for different days. (e) J_sc of Cu-PSCs and Au-PSCs aged for different days. (f) FF of Cu-PSCs and Au-PSCs aged for different days.

Fig. 1(b) shows the J–V curves of Cu-PSCs after aging for different days; “0 day” represents the data collected as soon as the Cu cathode was deposited on CH₃NH₃PbI₃. The J–V curves of Au-PSCs are shown in Fig. S1(b).† It was clear that the fresh Cu-PSCs showed poor performance, whereas the performance gradually improved and reached a peak when aged for 8 days. The J–V curves under the forward and reverse scans for the 10-day aged Cu-PSCs are shown in Fig. S6,† revealing the existence of hysteresis. Fig. 1(c) exhibits the change in the open circuit voltage (V_oc) of Cu-PSCs aged for different days with Au-PSCs as a contrast. The V_oc of Cu-PSCs improved tremendously from 0.13 V to 0.60 V in 12 days. From Fig. 1(d), the PCE of Cu-PSCs improved from 0.1% to more than 3.0% in the first 8 days and decreased to about 2.0% in the next 4 days. Fig. 1(e) and (f) display the changes in the short circuit current density (J_sc) and fill factor (FF) of Cu-PSCs, respectively, also with Au-PSCs for comparison. J_sc improved from the pristine 2.63 mA cm⁻² to the maximum 10.51 mA cm⁻² when aged for 8 days. The FF improved from 27.9% to the maximum 51.7% when aged for 10 days. However, from Fig. 1(c)–(f), the four performance parameters of the Au-PSCs were relatively stable in 12 days with only a slight decline. Therefore, it could be concluded that optimization of Cu-PSCs came from the cooperation of CH₃NH₃PbI₃ and Cu cathode.

To demonstrate the spontaneous corroding process as well as CuI interface layer produced as a HTM during the interaction of CH₃NH₃PbI₃ and the Cu cathode, X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM) and Raman scattering spectroscopy measurements were performed.

Fig. 2(a) shows the XRD patterns of the 0-day aged and 8-day aged Cu-PSCs, in which the main diffraction peaks located at 14.1° were assigned to the (110) lattice planes of CH₃NH₃PbI₃. The diffraction peaks located at 26.5° and 37.7° were assigned to FTO. We can see that a new peak located at 25.5° appeared after the cells were aged for 8 days. The new diffraction peak was assigned to the (111) lattice planes of CuI crystal, in agreement with the reported literatures.^23,24 Fig. 2(b) displays the XPS spectra of the interface between CH₃NH₃PbI₃ and Cu cathode on Cu-PSCs aged for 8 days. To measure the interface XPS information, a sticky tape was used to peel off the copper cathode. Both sides of the copper cathode and the CH₃NH₃PbI₃ film were tested. The Cu 2p spectra of Cu-PSCs aged for 8 days with Cu cathode exfoliated before the measurement is shown in Fig. 2(b); XPS spectra of the exfoliated Cu cathode is shown in Fig. S2(c).† We also measured the XPS spectra of the Cu film made by thermal evaporation as well as commercial CuI powder as a reference, as shown in Fig. S2(a) and (b).† From Fig. S2(a),† the binding energy of Cu 2p_3/2 and 2p_1/2 peak were 932.5 and 952.6 eV respectively. From Fig. S2(b),† binding energy of CuI 2p_3/2 and 2p_1/2 peak was 932.0 and 952.0 eV respectively. In Fig. 2(b), we can see that the two main peaks of Cu 2p spectra exhibit asymmetric shapes and the overlapped peaks can be fitted with the 2p peaks of CuI and Cu, which demonstrated that CuI was indeed produced between CH₃NH₃PbI₃ and Cu. Raman spectra were obtained to further certify the existence of CuI. The inset in Fig. 2(c) shows the Raman spectra of the CuI powder, where the TO and LO phonon modes were at 123.5 and 140.2 cm⁻¹, respectively, corresponding to the reported literature.²⁵ Fig. 2(b) shows the Raman spectra of Cu-PSCs with the laser cast on the Cu cathode, where Cu was made slightly thinner to ensure they were penetrated by the laser. From Fig. 2(b), the TO and LO phonon mode was to 128.3 and 145.0 cm⁻¹, respectively; we observed a slight shifting towards higher wavenumber relative to the CuI powder. This shift may come from the increased length of the chemical bond of CuI induced by coupling with CH₃NH₃PbI₃. However, the distance between the TO and LO mode was the same as the CuI powder; this demonstrated the formation of CuI between CH₃NH₃PbI₃ and the Cu cathode. A more distinct clue could be displayed by the change in the SEM images of Cu-PSCs with time. Fig. 2(d) displays the cross-sectional SEM views of Cu-PSCs aged for 8 days, the cross-sectional SEM views of the pristine Cu-PSCs are shown in Fig. S3.† Comparing the two SEM photographs, it was clear that the pristine single-layer Cu cathode transformed to a bilayer structure when the cells were aged for 8 days. The extra layer between CH₃NH₃PbI₃ and Cu cathode was the spontaneously formed CuI produced by the reaction of CH₃NH₃PbI₃ and Cu, in accordance with the XRD, XPS and Raman scattering measurements. Although the efficiency of Cu-PSCs began to decay after 8 days and almost decreased to zero 20 days later, the structure of the CuI/Cu bilayer was very stable. As shown in Fig. S5,† the 30-day aged Cu-PSCs still displayed the clear bilayer structure of CuI/Cu.

Fig. 2 (a) XRD pattern of fresh and 8-day aged Cu-PSCs. (b) XPS spectra of 8-day aged Cu-PSCs with the Cu cathode peeled off before being measured; the curve fit (black line) was obtained by the superposition of CuI (red line) and Cu (blue line), compared to the experimental data (gray dots). (c) Raman scattering spectra of 8-day aged Cu-PSCs, the curve fit (black line) was the superposition of TO (red line) and LO (blue line), compared to the experimental data (gray dots). The inset shows the Raman scattering spectra of the CuI powder. (d) Cross-section SEM image of 8-day aged Cu-PSCs. The self-formed CuI interface layer was located between the layer of CH₃NH₃PbI₃ and Cu cathode.

The spontaneous formation of CuI optimized the configuration of Cu-PSCs, as well as the energy band structure, yielding amazing improvement of the performance. Fig. 3(a) shows the schematic energy band diagrams of 0-day aged Cu-PSCs and 8-day aged Cu-PSCs. In our pristine Cu-PSCs, on account of the low work function of Cu, the electrons generated by the incident photons could recombine easily at the interface of perovskite and Cu due to the large diffusion length, which explained why there was almost no efficiency in fresh Cu-PSCs. However, when aged for several days, the CuI interface layer was self-formed in situ between CH₃NH₃PbI₃ and Cu, serving as HTM, which accelerated the transport of holes and suppressed the injection of electrons into the Cu cathode. The change in energy band structure of fresh Cu-PSCs and aged Cu-PSCs can be proven by the different dark J–V curves in Fig. 3(b). We can see the curve of the pristine Cu-PSCs shows an onset of the dark current at approximately 0.10 V. The onset shifted to about 0.60 V when the cells were aged for about 8 days, indicating a decline in the flow of electrons from the conduction band of CH₃NH₃PbI₃ to the Cu cathode. Compared to Cu, Au has a relatively higher work function, which may form a higher potential barrier with CH₃NH₃PbI₃ to block the electrons injecting into the Au cathode. Therefore, Au-PSCs have a lower recombination rate at the interface between the cathode and CH₃NH₃PbI₃ than Cu-PSCs, which indicates that our Au-PSCs have good performance. In Fig. 3(b), the Au-PSCs show an onset of the dark current at about 0.70 V, which is far larger than that of the fresh Cu-PSCs. Considering that the fabrication of Cu-PSCs was the same as the pristine Au-PSCs except for the metal cathode, the larger onset of the dark current indicated a smaller flow of electrons from CH₃NH₃PbI₃ to the cathode.

Fig. 3 (a) Schematic energy band diagrams of 0-day aged Cu-PSCs and 8-day aged Cu-PSCs. (b) Dark J–V curves of Au-PSCs, the 0-day aged Cu-PSCs and 8-day aged Cu-PSCs.

Regarding the possible mechanism of spontaneous formation of CuI, a similar process to the degradation of CH₃NH₃PbI₃ occurred,²⁶ as shown in eqn (1)–(4). Eqn (5) displays the overall reaction between CH₃NH₃PbI₃ and Cu. We stored PSCs in a drying cabinet with a relative humidity of 5%; CH₃NH₃PbI₃ degraded gradually and produced iodine with the effect of moisture and oxygen. Humidity may play an important role in the formation of CuI. The Cu-PSCs were also aged in a dry nitrogen-filled glovebox under ambient conditions with a humidity of ∼60%. More details are shown in ESI.†

CH₃NH₃I → CH₃NH₂ + HI (2)

4HI + O₂ → 2H₂O + 2I₂ (3)

2Cu + I₂ → 2CuI (4)

In summary, the performance of HTM-free PSCs with low-cost Cu cathode showed unexpected improvement after being aged for several days in dry air. This was attributed to the spontaneous configurational transformation of the cells from the HTM-free ones to the HTM-contained ones due to the self-formation of p-type CuI in situ at the interface of CH₃NH₃PbI₃ and Cu. The self-formed CuI layer effectively hindered the recombination of electrons and holes at the interface of CH₃NH₃PbI₃ and Cu cathode. Therefore, it was proved that the undesirable corrosion phenomenon of CH₃NH₃PbI₃ to the metal cathode in typical PSCs is in turn positive in the cells with the Cu cathode, which might be utilized to design other photoelectric devices based on organolead halide perovskites.

Acknowledgements

This study was supported primarily by the National Natural Science Foundation of China (61377051), the National Basic Research Program of China (2013CB632404), and the Natural Science Foundation and Key Research Project of Jiangsu Province, China (BK20130053, BK20141233, BE2015090). The authors thank Mr Lei Zhang and Prof. Xuejin Zhang from the College of Engineering and Applied Sciences at Nanjing University for Raman spectroscopy measurements.

Notes and references

1. A. Kojima, K. Teshima, Y. Shirai and T. Miyasaka, J. Am. Chem. Soc., 2009, 131, 6050–6051.
2. H.-S. Kim, C.-R. Lee, J.-H. Im, K.-B. Lee, T. Moehl, A. Marchioro, S.-J. Moon, R. Humphry-Baker, J.-H. Yum, J. E. Moser, M. Grätzel and N.-G. Park, Sci. Rep., 2012, 2, 591.
3. M. M. Lee, J. Teuscher, T. Miyasaka, T. N. Murakami and H. J. Snaith, Science, 2012, 338, 643–647.
4. W. S. Yang, J. H. Noh, N. J. Jeon, Y. C. Kim, S. Ryu, J. Seo and S. I. Seok, Science, 2015, 348, 1234–1237.
5. L. Etgar, P. Gao, Z. Xue, Q. Peng, A. K. Chandiran, B. Liu, M. K. Nazeeruddin and M. Graetzel, J. Am. Chem. Soc., 2012, 134, 17396–17399.
6. W. Abu Laban and L. Etgar, Energy Environ. Sci., 2013, 6, 3249–3253.
7. B.-E. Cohen, S. Gamliel and L. Etgara, Appl. Mater., 2014, 2, 081502.
8. S. Aharon, A. Dymshits, A. Rotem and L. Etgar, J. Mater. Chem. A, 2015, 3, 9171–9178.
9. P. Qin, S. Tanaka, S. Ito, N. Tetreault, K. Manabe, H. Nishino, M. K. Nazeeruddin and M. Grätzel, Nat. Commun., 2014, 5, 3834.
10. S. Ye, W. Sun, Y. Li, W. Yan, H. Peng, Z. Bian, Z. Liu and C. Huang, Nano Lett., 2015, 15, 3723–3728.
11. B. A. Nejand, V. Ahmadi, S. Gharibzadeh and H. R. Shahverdi, ChemSusChem, 2016, 9, 302–313.
12. X. Xu, Z. Liu, Z. Zuo, M. Zhang, Z. Zhao, Y. Shen, H. Zhou, Q. Chen, Y. Yang and M. Wang, Nano Lett., 2015, 15, 2402–2408.
13. J. A. Christians, R. C. M. Fung and P. V. Kamat, J. Am. Chem. Soc., 2014, 136, 758–764.
14. G. A. Sepalage, S. Meyer, A. Pascoe, A. D. Scully, F. Huang, U. Bach, Y. B. Cheng and L. Spiccia, Adv. Funct. Mater., 2015, 25, 5650–5661.
15. Q. Jiang, X. Sheng, B. Shi, X. Feng and T. Xu, J. Phys. Chem. C, 2014, 118, 25878–25883.
16. Z. Ku, Y. Rong, M. Xu, T. Liu and H. Han, Sci. Rep., 2013, 3, 3132.
17. Z. Li, S. A. Kulkarni, P. P. Boix, E. Shi, A. Cao, K. Fu, S. K. Batabyal, J. Zhang, Q. Xiong, L. H. Wong, N. Mathews and S. G. Mhaisalkar, ACS Nano, 2014, 8, 6797–6804.
18. Y. Deng, Q. Dong, C. Bi, Y. Yuan and J. Huang, Adv. Energy Mater., 2016, 6, 1600372.
19. J. Xu, A. Buin, A. H. Ip, W. Li, O. Voznyy, R. Comin, M. Yuan, S. Jeon, Z. Ning, J. J. McDowell, P. Kanjanaboos, J. P. Sun, X. Lan, L. N. Quan, D. H. Kim, I. G. Hill, P. Maksymovych and E. H. Sargent, Nat. Commun., 2015, 6, 7081.
20. H. Back, G. Kim, J. Kim, J. Kong, T. K. Kim, H. Kang, H. Kim, J. Lee, S. Lee and K. Lee, Energy Environ. Sci., 2016, 9, 1258–1263.
21. S. D. Standridge, G. C. Schatz and J. T. Hupp, J. Am. Chem. Soc., 2009, 131, 8407–8409.
22. B. E. Hardin, H. J. Snaith and M. D. McGehee, Nat. Photonics, 2012, 6, 162–169.
23. M. R. Johan, K. Si-Wen, N. Hawari and N. A. K. Aznan, Int. J. Electrochem. Sci., 2012, 7, 4942–4950.
24. B. L. Zhu and X. Z. Zhao, Phys. Status Solidi A, 2011, 208, 91–96.
25. J. Potts, R. Hanson, C. Walker and C. Schwab, Solid State Commun., 1973, 13, 389–392.
26. G. Niu, W. Li, F. Meng, L. Wang, H. Dong and Y. Qiu, J. Mater. Chem. A, 2014, 2, 705–710.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra19702g

---