Solvent washing with toluene enhances efficiency and increases reproducibility in perovskite solar cells

Abstract

We report a simple process for reproducibly fabricating perovskite solar cells. We emphasize that the solvent washing technique is the most practical method for successful uniform crystallization so it facilitates highly efficient reproducible perovskite solar cells. The critical parameter for tuning crystallinity is determined to be the type of washing solvent and the quantity dispensed. The amount of washing solvent strongly affects the particle size distribution resulting in better or worse interconnection between the crystal grains. We discovered that 20 µl of toluene is the best washing solvent for device reproducibility. Our proposed parameters result in 90% reproducible perovskite solar cells with an average efficiency around 8%.

---

Introduction

Organic–inorganic hybrid materials have been successfully employed in many electronic applications such as field effect transistors,¹ sensors,² photo detectors,³ light emitting diodes,⁴ and thin film photovoltaic devices.⁵ Among these organic–inorganic hybrid electronics, perovskite solar cells based on methylammonium lead(II) halides (MaPbX₃ X = I, Cl, Br) as a member of third generation photovoltaics, are of great interest in recent years because of their low cost and excellent properties – such as long exciton diffusion lengths and life times, high absorption coefficients, excellent charge carrier mobilities and direct band gap.⁶ As a consequence of those properties, high power conversion efficiencies of over 19% have been achieved.⁷

Many reports on highly efficient perovskite solar cells introduce mesoporous device structures.^8,9 However, such a structure requires high temperature processing to obtain mesoporous scaffold as well as a compact TiO₂ layer. It is crit]ical for perovskite crystallization to exhibit uniform and homogeneous nucleation with smooth and dense film morphology for efficient solar cells. The film quality and uniform crystallization have significant effect on device performance. Poor morphology can lead to electrical shorting and also affect charge transport/separation and recombination.¹⁰ The most efficient perovskite solar cells based on mesoporous TiO₂, have been fabricated by co-evaporation which leads to uniform and excellent film morphology.¹¹ However, co-evaporation requires high vacuum and energy, which is a disadvantage for low-cost solar cells. In contrast, solution-processed planar device architectures based on PEDOT:PSS and PCBM, can be more attractive considering that it is simple and low temperature processing advantages which gives a chance to make flexible devices.^12,13 It should be noted, however, that one step solution-processed perovskite solar cells with planar geometries have some difficulties such as inhomogeneous surface morphology, irregular crystallinity and poor reproducibility.^14–16 To solve such problems, many different methods such as vapor assisted deposition, small molecule addition or solvent washing have been employed.^10,11,15 Recent studies have also shown that the solvent washing method is one of the most effective ways for inducing crystallization and controlling morphology.^10,17,18

In this paper, we optimized a highly reproducible and efficient perovskite solar cell fabrication way via solvent washing method. We discussed that which organic solvent best on perovskite film crystallinity, morphology, easy application, reproducibility etc. and amount of washing solvent which is directly effects on crystal grain size and cell efficiency. We discovered that 20 µl of toluene is the best washing solvent for device reproducibility. Our proposed parameters result in 90% reproducible perovskite solar cells with average efficiency around 8%.

Experimental section

All experimental details related with solar cell fabrication and characterization procedures are given ESI.†

Result and discussion

In this method, different organic solvents have been used for improving surface morphology without thermal annealing.⁵ In the solvent washing method, uniform perovskite nucleation occurs after a second and subsequent spin coating step. Therefore, this method is easier, cheaper and allows one to fabricate highly reproducible perovskite solar cells than other proposed methods.^15–19 Therefore, it is possible to acquire efficient and reproducible perovskite solar cells by developing an optimized route to manage crystallization, film surface morphology, nucleation and growth. Although it has been previously reported that only a few drops of organic solvent is sufficient for nucleation and growth of crystals, the amount of organic solvent is very critical in order to obtain uniform perovskite crystals and previous reports do not give a quantitative protocol.

A literature search did not yield any reports on optimization of solvent washing parameters such as the exact quantity of solvent, solvent type, spin rate, device size etc. Therefore, introducing a detailed fabrication process based on solvent washing is a need to fabricate reproducible efficient solar cell.

Herein, we report a single step solution-processable route for optimization of solvent washing method to obtain reproducible, crystalline, planar perovskite solar cells. The simple approach involves spin coating of perovskite precursors (CH₃NH₃PbI₃) dissolved in GBL (γ-butyrolactone) or DMF (dimethyl formamide), followed immediately with a subsequent spin coating of an additive organic solvent onto the wet film to induce crystallization. A series of organic solvents less polar then GBL and DMF were tested in great detail (Fig. S1†). These organic solvents decrease the solubility of precursors in GBL/DMF and lead to fast crystallization within the film. Rapid change in film color from yellow to reddish-brown indicates crystalline perovskite formation is observed upon adding organic solvent during spinning as shown in Fig. 1.

Fig. 1 Scheme of the solvent washing process. Solar cell configuration is ITO/PEDOT:PSS/CH₃NH₃PbI₃/PCBM/Al.

Here, it is critical that GBL/DMF which are polar solvents and perovskite precursors dissolved in, have high boiling points (∼204/154 °C) and must be remove from the film surface as soon as possible to obtain uniform perovskite thin film with high crystallinity. Although many researchers offer different methods to remove the solvent such as hot casting or solvent washing, common point of these methods is removing solvent during spinning to fast crystallization. When perovskite solution get in touch with non-polar organic solvent during spinning, perovskite precursors suddenly precipitate and uniform crystals occur. Before the contacting of toluene drop, perovskite thin film is still wet and toluene drops lead to fast drying of the film which results in uniform film morphology. Annealing of wet films after spin coating mostly yield result with aggregation problems. In this method, we can imagine that washing solvent is a some kind of drying agent. That is why, washing with organic solvents pave the way for fabricating more effective film surfaces, instead of resorting to high annealing temperature treatments (90–110 °C, 1–2 h). It should be noted that the optimization process was carried out by firstly using GBL as the precursor solvent, and then those parameters were applied to DMF as a casting solvent as well. Initially we investigated which washing organic solvent was the best for the fabrication of statistically reproducible solar cells. Fig. 2a shows that both dichlorobenzene and toluene-washed perovskite films exhibit the largest current densities, while chloroform-washed and non-treated films show the lowest performance. We found that the amount of toluene is very critical for reproducible efficient perovskite solar cells. 40 µl or less is necessary for producing highly crystalline films and washing with more than 40 µl of toluene drastically decreases the efficiency as shown in Fig. 2b and c.

Fig. 2 (a–d) Current–voltage characteristics of different washing organic solvents (a) and solar cell performance as a function of amount of toluene used (b). Current density curves showing overlay performances of toluene washing quantity (c). Statistical efficiency graph based on the number of solar cells. The average cell efficiency around 8% PCE with 90% reproducibility (d).

The increase in crystal radius leads to decrease in the number of grain boundaries so in parallel with charge carriers mobility and exciton diffusion length which is directly related with cell PCE. Table 1 shows a summary of efficiency of solar cells based on the amount of washing toluene employed. Larger crystal radius and uniform crystallization is observed by decreasing the amount of toluene from 80 to 20 µl and different solvents as can be clearly seen from SEM images (Fig. S2 and S3†). While dichlorobenzene shows high efficiency when used as a washing solvent, we chose to employ toluene as this solvent exhibits highly reproducible results in comparison with dichlorobenzene (Fig. 2d). Therefore, toluene is very effective on the crystallization behavior and the best organic solvent for solvent washing method while chloroform and dichloromethane do not give reproducible cells. In addition, perovskite film surface formed after chloroform washing shows sponge-like structures as can be seen from Fig. 3b. When toluene is employed instead of chloroform, crystal grain size increases (see Fig. 3b and c and insets). Possible explanation regarding these results could be related with miscibility of washing solvent and perovskite solution.¹⁵ Since the miscibility of toluene with GBL/DMF is better than that of chloroform, nucleation and growth of crystals start very fast resulting in fine crystals. As well known, crystallinity has great effect on charge separation, transport and diffusion length. The defects in crystals behave as trap sites resulting in charge recombination.^19–22

Table 1 Summary of the photovoltaic parameters of the CH₃NH₃PbI₃-based perovskite solar cells

Toluene amount (µl)	V oc (mV)	J sc (mA cm^−2)	FF	Efficiency (PCE) %
80	900	9.46	0.53	6.43 ± 0.5
70	900	10.92	0.54	7.66 ± 0.4
60	900	11.33	0.56	8.22 ± 0.3
50	950	12.21	0.55	9.08 ± 0.5
40	900	12.31	0.59	9.16 ± 0.3
20	900	13.24	0.56	9.54 ± 0.5
0	750	7.04	0.43	3.07 ± 1.0

---

Fig. 3 (a–c) SEM images of perovskite layer without solvent washing (a), with chloroform washing (b), with toluene washing (c).

Therefore, toluene is very effective on the crystallization behavior and the best organic solvent for solvent washing method while chloroform and dichloromethane do not give reproducible cells. In addition, perovskite film surface formed after chloroform washing shows sponge-like structures as can be seen from Fig. 3b. When toluene is employed instead of chloroform, crystal grain size increases (see Fig. 3b and c and insets). Possible explanation regarding these results could be related with miscibility of washing solvent and perovskite solution.¹⁵ Since the miscibility of toluene with GBL/DMF is better than that of chloroform, nucleation and growth of crystals start very fast resulting in fine crystals. As well known, crystallinity has great effect on charge separation, transport and diffusion length. The defects in crystals behave as trap sites resulting in charge recombination.^19–22 It is clear from Fig. 3a that as-spun perovskite films exhibit large but separated crystal grains from each other and rough, rough surfaces which do not exist in uniform thin films. The solar cells fabricated without adding organic solvent (but annealed at 90 °C for 90 minutes) give low photovoltaic performance around 3% since the random accumulation of perovskite pre-cursors on the PEDOT:PSS surface during thermal treatment (Table 1). The film morphology shows major changes from solvent to solvent as can be seen from Fig. 3b and c. The surface morphology shows a sponge-like structure when chloroform used as washing solvent (Fig. 3b). On the other hand, toluene-washed films do not form large domains and, moreover, the crystal grains are larger than chloroform cast films (Fig. 3c). The addition of organic solvent improves the crystallinity and the coverage of the perovskite thin films. We believe that it facilitates the diffusion of free carriers and increasing the efficient charge carriers.¹⁴

To better understand the reasons for these observations, we evaluated XRD patterns, optical absorption spectra and electrical measurements of formed films. Fig. 4 shows XRD patterns of different perovskite films for comparison. It is clear from Fig. 4, perovskite crystals are formed in all cases. However the peak intensities show some differences indicating the degree of crystallinity is different among the different conditions. The highest peak intensity is observed for the films washed with 20 µl toluene while the lowest one for non-washed films. These results are in agreement with the solar cell efficiencies. It may be suggested that the crystallization ratio increases with solvent washing.

Fig. 4 (a and b) X-ray diffraction (XRD) patterns of perovskite films which have been obtained different washing process (a). SAXS results of three different films (b).

On the other hand, small-angle X-ray scattering (SAXS) analyses show that the narrowest average particle size distribution gives the best solar cell efficiency and the largest vise versa. As it is clear from Fig. 4b, 20 µl toluene gives the narrowest distribution. We may suggest that homogenous particle size distribution lead more ordered film resulting better interconnection between the crystal grains. Thus the efficiency increases by narrowing the average size distribution.

The absorption spectra of the same films support this approach (Fig. 5). The highest crystallized films give the highest absorption as can be clearly seen from Fig. 5. The best light harvesting of perovskite crystals is observed for the films washed with 20 µl toluene. Non-washed films show poor crystallization which result in poor photovoltaic response.

Fig. 5 (a and b) UV-Vis absorption spectra of perovskite films which have been obtained from different solvent washing processes (a). Bode phase plot: phase angle versus frequency graph of perovskite solar cells (b).

As a last step of this work, we also employed DMF, which is commonly preferred as the precursor solvent, instead of GBL. Although DMF dissolves precursors better than GBL, it is not possible to obtain fine perovskite crystals at room temperature with any amount of solvent washing with toluene. The critical parameter that we discovered for the fabrication of efficient solar cells with DMF is the pre-heating of substrate up to 80 °C for 5 minutes. Just only in this case, solvent washing works and gives good crystallization and more efficient solar cells. As it is clear from SEM images given in the Fig. 6a–c, (for AFM images see Fig. S4†) pre-heated substrates give fine crystals and better efficiency. DMF employed to perovskite precursors does not give a smooth crystal surfaces and correspondingly high efficient solar cells, unless pre-heating treatment of substrate is carried out. As can be seen in Fig. 6c, heating of substrate to 80 °C clearly influences the efficiency of solar cells. Table 2 shows the comparison of solar cell efficiency prepared on no heating and pre-heated, 80 °C substrates. After solvent washing during spinning process, the short term low temperature annealing process could be enough to get rid of residual solvents and to obtain reproducible efficient solar cells. This optimized method allows one to fabricate rapid, easy and extremely reproducible (90%) perovskite solar cells.

Fig. 6 (a–c) Difference between no heating and pre-heated, 80 °C substrates for perovskite precursors dissolved in DMF could be clearly seen from (a and b) SEM. (a) Without and (b) with pre-heating treatment of substrate. Current–voltage characteristics of perovskite solar cells, precursors dissolved in DMF (c).

Table 2 Photovoltaic parameters of perovskite solar cells obtained by using pre-heated or non-heated substrates

Substrate process	V oc (mV)	J sc (mA cm^−2)	FF	Efficiency (PCE) %
No heating	300	5.38	0.32	0.74
Pre-heated, 80 °C	700	7.05	0.58	4.67

---

Conclusions

In summary, we report a toluene washing method for producing highly efficient perovskite solar cells that lead to fast crystallization and shorter thermal annealing times. In addition, the fabrication process yields statistically reproducible devices (over 90% reproducibility). Finally, we determined that pre-heating a substrate to 80 °C enables one to use DMF as a casting solvent and combined with toluene washing, resulted in uniform and highly crystalline thin films with higher performance than the unheated counterparts. Our findings are an important contribution with regard to easy solution process and low cost procedure for fabricating perovskite solar cells.

Acknowledgements

K. Kara and C. Kırbıyık (PN: 15101011) partially contributed from their PhD thesis and they thank to TUBİTAK (2211-C program). AL. Briseno thanks the NSF Center for Hierarchical Manufacturing (CMMI-0531171), and TUBITAK for providing the fellowship (2221) for his visit to Turkey.

References

1. C. R. Kagan, D. B. Mitzi and C. D. Dimitrakopoulos, Organic–Inorganic Hybrid Materials as Semiconducting Channels in Thin-Film Field-Effect Transistors, Science, 1999, 286(5441), 945–947.
2. S. Wang, Y. Kang, L. Wang, H. Zhang, Y. Wang and Y. Wang, Organic/inorganic hybrid sensors: a review, Sens. Actuators, B, 2013, 182, 467–481.
3. X. Wang, W. Song, B. Liu, G. Chen, D. Chen, C. Zhou and G. Shen, High-Performance Organic–Inorganic Hybrid Photodetectors Based on P3HT: CdSe Nanowire Heterojunctions on Rigid and Flexible Substrates, Adv. Funct. Mater., 2013, 23(9), 1202–1209.
4. M. Sessolo and H. J. Bolink, Hybrid Organic–Inorganic Light-Emitting Diodes, Adv. Mater., 2011, 23(16), 1829–1845.
5. 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, Efficient inorganic–organic hybrid heterojunction solar cells containing perovskite compound and polymeric hole conductors, Nat. Photonics, 2013, 7(6), 486–491.
6. M. Xiao, F. Huang, W. Huang, Y. Dkhissi, Y. Zhu, J. Etheridge, A. Gray-Weale, U. Bach, Y.-B. Cheng and L. Spiccia, A Fast Deposition-Crystallization Procedure for Highly Efficient Lead Iodide Perovskite Thin-Film Solar Cells, Angew. Chem., 2014, 126(37), 10056–10061.
7. H. Zhou, Q. Chen, G. Li, S. Luo, T.-B. Song, H.-S. Duan, Z. Hong, J. You, Y. Liu and Y. Yang, Interface engineering of highly efficient perovskite solar cells, Science, 2014, 345(6196), 542–546.
8. Q. Chen, H. Zhou, Z. Hong, S. Luo, H.-S. Duan, H.-H. Wang, Y. Liu, G. Li and Y. Yang, Planar Heterojunction Perovskite Solar Cells via Vapor-Assisted Solution Process, J. Am. Chem. Soc., 2014, 136(2), 622–625.
9. K. Wojciechowski, M. Saliba, T. Leijtens, A. Abate and H. J. Snaith, Sub-150 [degree]C processed meso-superstructured perovskite solar cells with enhanced efficiency, Energy Environ. Sci., 2014, 7(3), 1142–1147.
10. N. J. Jeon, J. H. Noh, Y. C. Kim, W. S. Yang, S. Ryu and S. I. Seok, Solvent engineering for high-performance inorganic–organic hybrid perovskite solar cells, Nat. Mater., 2014, 13(9), 897–903.
11. M. Liu, M. B. Johnston and H. J. Snaith, Efficient planar heterojunction perovskite solar cells by vapour deposition, Nature, 2013, 501(7467), 395–398.
12. O. Malinkiewicz, A. Yella, Y. H. Lee, G. M. Espallargas, M. Graetzel, M. K. Nazeeruddin and H. J. Bolink, Perovskite solar cells employing organic charge-transport layers, Nat. Photonics, 2014, 8(2), 128–132.
13. P. Docampo, J. M. Ball, M. Darwich, G. E. Eperon and H. J. Snaith, Efficient organometal trihalide perovskite planar-heterojunction solar cells on flexible polymer substrates, Nat. Commun., 2013, 4, 1–6.
14. P.-W. Liang, C.-Y. Liao, C.-C. Chueh, F. Zuo, S. T. Williams, X.-K. Xin, J. Lin and A. K. Y. Jen, Additive Enhanced Crystallization of Solution-Processed Perovskite for Highly Efficient Planar-Heterojunction Solar Cells, Adv. Mater., 2014, 26(22), 3748–3754.
15. Y.-J. Jeon, S. Lee, R. Kang, J.-E. Kim, J.-S. Yeo, S.-H. Lee, S.-S. Kim, J.-M. Yun and D.-Y. Kim, Planar heterojunction perovskite solar cells with superior reproducibility, Sci. Rep., 2014, 4, 6953.
16. J. M. Ball, M. M. Lee, A. Hey and H. J. Snaith, Low-temperature processed meso-superstructured to thin-film perovskite solar cells, Energy Environ. Sci., 2013, 6(6), 1739.
17. C.-C. Chen, S.-H. Bae, W.-H. Chang, Z. Hong, G. Li, Q. Chen, H. Zhou and Y. Yang, Perovskite/polymer monolithic hybrid tandem solar cells utilizing a low-temperature, full solution process, Mater. Horiz., 2015, 2(2), 203–211.
18. J. W. Jung, S. T. Williams and A. K. Y. Jen, Low-temperature processed high-performance flexible perovskite solar cells via rationally optimized solvent washing treatments, RSC Adv., 2014, 4(108), 62971–62977.
19. G. Xing, N. Mathews, S. Sun, S. S. Lim, Y. M. Lam, M. Grätzel, S. Mhaisalkar and T. C. Sum, Long-Range Balanced Electron- and Hole-Transport Lengths in Organic–Inorganic CH₃NH₃PbI₃, Science, 2013, 342(6156), 344–347.
20. S. Sun, T. Salim, N. Mathews, M. Duchamp, C. Boothroyd, G. Xing, T. C. Sum and Y. M. Lam, The origin of high efficiency in low-temperature solution-processable bilayer organometal halide hybrid solar cells, Energy Environ. Sci., 2014, 7(1), 399–407.
21. J.-Y. Jeng, Y.-F. Chiang, M.-H. Lee, S.-R. Peng, T.-F. Guo, P. Chen and T.-C. Wen, CH₃NH₃PbI₃ Perovskite/Fullerene Planar-Heterojunction Hybrid Solar Cells, Adv. Mater., 2013, 25(27), 3727–3732.
22. G. E. Eperon, V. M. Burlakov, P. Docampo, A. Goriely and H. J. Snaith, Morphological Control for High Performance, Solution-Processed Planar Heterojunction Perovskite Solar Cells, Adv. Funct. Mater., 2014, 24(1), 151–157.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details containing device fabrication and additional characterizations. See DOI: 10.1039/c5ra27122c

---