Date Published: January 05, 2018
Publisher: John Wiley and Sons Inc.
Author(s): Fan Fu, Stefano Pisoni, Thomas P. Weiss, Thomas Feurer, Aneliia Wäckerlin, Peter Fuchs, Shiro Nishiwaki, Lukas Zortea, Ayodhya N. Tiwari, Stephan Buecheler.
Compositional grading has been widely exploited in highly efficient Cu(In,Ga)Se2, CdTe, GaAs, quantum dot solar cells, and this strategy has the potential to improve the performance of emerging perovskite solar cells. However, realizing and maintaining compositionally graded perovskite absorber from solution processing is challenging. Moreover, the operational stability of graded perovskite solar cells under long‐term heat/light soaking has not been demonstrated. In this study, a facile partial ion‐exchange approach is reported to achieve compositionally graded perovskite absorber layers. Incorporating compositional grading improves charge collection and suppresses interface recombination, enabling to fabricate near‐infrared‐transparent perovskite solar cells with power conversion efficiency of 16.8% in substrate configuration, and demonstrate 22.7% tandem efficiency with 3.3% absolute gain when mechanically stacked on a Cu(In,Ga)Se2 bottom cell. Non‐encapsulated graded perovskite device retains over 93% of its initial efficiency after 1000 h operation at maximum power point at 60 °C under equivalent 1 sun illumination. The results open an avenue in exploring partial ion‐exchange to design graded perovskite solar cells with improved efficiency and stability.
Thin film perovskite (ABX3, A = Cs+, [CH3NH3]+ (MA+), [CH(NH2)2]+ (FA+); B = Pb, Sn; X = Cl, Br, I) solar cells have gained considerable attention due to the high performance, easy processing and potentially low‐cost manufacturing.1 Benefiting from outstanding optoelectronic properties, such as high absorption coefficient,2 large carrier mobility and long carrier diffusion length,3, 4 and unique defect properties,5 the power conversion efficiency of perovskite solar cells has rapidly increased from 3.8% to over 22% by engineering charge transporting layers,6, 7, 8, 9 optimizing the absorber composition and deposition techniques.10, 11, 12, 13, 14, 15, 16, 17, 18, 19 The wide bandgap with flexibility to tune over a broad energy range renders perovskite solar cells ideal candidates for top cells in tandem applications with narrow bandgap bottom cells,20, 21 such as crystalline‐Si,22, 23, 24, 25 thin film Cu(In,Ga)Se2 (CIGS),26, 27, 28, 29, 30, 31 Sn‐based halide perovskites,32, 33 to realize highly efficient and cost‐effective multi‐junction devices. Currently, the perovskite based tandem efficiencies are still lower than the highest efficiency single junction bottom device,34, 35 primarily limited by the NIR‐ transparent perovskite top cells. Generally, efficient NIR‐transparent perovskite solar cells adopt a planar structure processed at low‐temperature (≈100 °C),31, 36, 37, 38 which usually delivers lower open circuit voltage VOC (< 1.1 V) when compared to mesoporous structures employing high‐temperature (≈500 °C) processed TiO2.11, 12, 13 The loss in VOC is particularly severe in planar devices using thick absorbers that are desired for current‐matching monolithic tandems. Novel concepts to improve VOC while maintaining high short circuit current density (JSC) in NIR‐transparent planar perovskite solar cells are essential to achieve over 30% tandem efficiency in combination with the well‐established c‐Si or CIGS solar cells. In conclusion, we have developed a facile partial ion‐exchange strategy to fabricate compositionally graded perovskite absorber to simultaneously improve the efficiency and operational stability in NIR‐transparent perovskite solar cells. Spin coating of organic bromide (MABr or FABr) solution onto starting MAPbI3 or FAPbI3 absorber induced halide ion‐exchange and subsequent ions diffusion, which results in a perovskite absorber with a Br composition gradient as verified by ToF‐SIMS element depth profiling. Incorporating compositional grading improved charge collection and suppressed interface recombination, enabling us to achieve improved VOC and JSC, and a steady state efficiency of 16.8% in NIR‐transparent perovskite solar cells with thick absorber (520 nm). In addition to improved efficiency, non‐encapsulated NIR‐transparent perovskite device with graded absorber retained over 93% of its initial efficiency after 1000 h operation at MPP condition at 60 °C under equivalent 1 sun illumination. When mechanically stacked on CIGS solar cells, we demonstrated 22.7% efficiency in 4‐terminal tandem configuration with 3.3% absolute efficiency gain compared to the highest single junction cell (19.4% CIGS). The PIE approach offers a viable way to tailor the composition and morphology of the mixed‐perovskite absorbers which is not easily accessible by other methods, and our results provide new direction in exploring compositional grading via partial ion‐exchange to perovskite solar cells. Perovskite Solar Cells Fabrication: Perovskite solar cells were grown on commercial ITO coated glass (sheet resistance: 8 Ω sq.−1, Zhuhai Kaivo Optoelectronics, P. R. China). The ITO glass was washed by hand first and then subjected to soap and deionized water sonification bath at 85 °C each for 15 min. The ITO glasses were then dried by compressed nitrogen gun and used for solar cells processing without additional ozone treatment. The hole transporting layer was prepared by spin coating 30 µL of PTAA (Sigma‐Aldrich) solution (5 mg mL−1 in toluene doped with 1 wt% F4‐TCNQ (97%, Sigma‐Aldrich)) at 6000 r.p.m. for 45 s on 2.5 × 2.5 cm2 substrate, followed by thermal annealing at 100 °C for 10 min. Afterward, around 200 nm PbI2 (ultradry, 99.999%, Alfa Aesar) compact film was thermally evaporated on rotating PTAA/ITO/glass without intentional heating. The deposition rate was controlled within 1–1.5 Å s−1, and the deposition pressure was between 3 and 6 × 10−8 mbar. After the PbI2 deposition, samples were transferred into glovebox for further processing. The mixed‐perovskite layers were formed by partial ion exchange reaction which contains mainly three stages. Firstly, the starting absorber, such as MAPbI3 (or FAPbI3) were prepared as follows: 300 µL of MAI (or FAI) solution (65 mg mL−1 in isopropanol) was first spread onto PbI2 surface, and then immediately started the rotation at 6000 r.p.m. for 45 s. The as‐deposited films were annealed at 100 (or 150 °C) for 10 min for MAPbI3 (or for FAPbI3). The starting absorber (MAPbI3 or FAPbI3) were subjected to spin coating (6000 r.p.m. for 45 s) of organic bromide (MABr, FABr) to induce the ion‐exchange. Afterward, the absorbers were annealed at 100 °C for 1 h under chlorobenzene vapor atmosphere. For electron transporting layer, 30 µL PCBM (PC61BM, 99.5%, Solenne BV, Netherland) solution (20 mg mL−1 in chlorobenzene) was spin coated at 5000 r.p.m. for 45 s followed by 60 min annealing at 100 °C covered by petri dish with presence of 10 µL chlorobenzene. After cooling down, 30 µL undoped ZnO nanoparticles (2.7 wt% (crystalline ZnO dissolved in isopropanol), Sigma‐Aldrich) was spin coated on top of PCBM at 4000 r.p.m. for 45 s. The ZnO nanoparticles were dried at 100 °C for 60 s to evaporate the solvent. Finally, the samples were coated with ZnO:Al front contact by RF‐magnetron sputtering and Ni/Al (50 nm/4000 nm) metallic grid by e‐beam evaporation. Finally, all cells were covered with a 105 nm MgF2 antireflection coating deposited by e‐beam evaporation and single cells were defined by mechanical scribing down to ITO back contact. The authors declare no conflict of interest. Source: http://doi.org/10.1002/advs.201700675