Identifying the Soft Nature of Defective Perovskite Surface Layer and Its Removal Using a Facile Mechanical Approach

•Perovskite defective surface layers are mechanically softer than crystalline regions•Adhesive tape can remove defective surface layers and stabilize perovskite devices•Tape treatment is compatible with the scaling up of perovskite solar modules The defective perovskite surface layer composed of nanocrystals and amorphous regions has been shown to initialize the degradation of perovskites. To stabilize perovskite materials and devices, it is essential to remove such a defective surface layer. Herein, we studied the mechanical properties of the perovskite surface layer and found that such a defective layer is softer than the crystalline regions and has weaker bonding to the crystalline layer underneath. Based on this discovery, we report a simple strategy to remove the soft, defective surface layer with low-cost adhesive tapes. The adhesive tape has appropriate bonding with perovskites and, thus, removes the defective surface layer without damaging the underlying crystalline regions. This method has also shown the universality in commonly used perovskite compositions and the compatibility with scaling up of perovskite photovoltaics. The presence of a defective layer composed of nanocrystals and amorphous regions at the surface of perovskite films has been shown to initialize the degradation of perovskites and cause nonradiative recombination. Here, we report the discovery that these defective surface layers are mechanically softer than the crystalline regions. The defective surface layer has a weaker bonding with the crystalline layer underneath it, which enables a facile approach to mechanically peel-off these defective layers using adhesive tapes. The chosen low-cost tape has an appropriate bonding force with perovskites, so the peeling does not damage the crystalline region and embedded interfaces underneath. The tape-treated devices retained 97.1% of the initial efficiency after operation at the near maximum power point under one sun illumination for 1,440 h at 65°C. This method is universally effective in enhancing the stability of various commonly used perovskite compositions and is compatible with the scaling up of perovskite solar modules. The presence of a defective layer composed of nanocrystals and amorphous regions at the surface of perovskite films has been shown to initialize the degradation of perovskites and cause nonradiative recombination. Here, we report the discovery that these defective surface layers are mechanically softer than the crystalline regions. The defective surface layer has a weaker bonding with the crystalline layer underneath it, which enables a facile approach to mechanically peel-off these defective layers using adhesive tapes. The chosen low-cost tape has an appropriate bonding force with perovskites, so the peeling does not damage the crystalline region and embedded interfaces underneath. The tape-treated devices retained 97.1% of the initial efficiency after operation at the near maximum power point under one sun illumination for 1,440 h at 65°C. This method is universally effective in enhancing the stability of various commonly used perovskite compositions and is compatible with the scaling up of perovskite solar modules. Metal halide perovskites (MHPs) have demonstrated remarkable progress in photovoltaic power conversion efficiency (PCE) from 3.8% to 25.2% in less than one decade,1NRELBest Research-cell efficiency chart.https://www.nrel.gov/pv/cell-efficiency.htmlDate: 2020Google Scholar showing their great potential as the next-generation low-cost solar technology. Among all the remaining issues to be addressed, poor stability is the most outstanding challenge that needs both fundamental understanding and engineering improvement toward the commercialization of perovskite photovoltaics.2Christians J.A. Habisreutinger S.N. Berry J.J. Luther J.M. Stability in perovskite photovoltaics: A paradigm for newfangled technologies.ACS Energy Lett. 2018; 3: 2136-2143Crossref Scopus (85) Google Scholar In the past few years, the stability of perovskite solar cells (PSCs) has been dramatically improved with global efforts dedicated to tuning perovskite compositions,3Kim H.S. Hagfeldt A. Park N.G. Morphological and compositional progress in halide perovskite solar cells.Chem. Commun. 2019; 55: 1192-1200Crossref PubMed Google Scholar passivating perovskite defects,4Chen B. Rudd P.N. Yang S. Yuan Y. Huang J. Imperfections and their passivation in halide perovskite solar cells.Chem. Soc. Rev. 2019; 48: 3842-3867Crossref PubMed Google Scholar constructing stable heterostructures,5Wang Y. Wu T. Barbaud J. Kong W. Cui D. Chen H. Yang X. Han L. Stabilizing heterostructures of soft perovskite semiconductors.Science. 2019; 365: 687-691Crossref PubMed Scopus (238) Google Scholar and developing better encapsulation methods compatible with perovskite materials.6Boyd C.C. Cheacharoen R. Leijtens T. McGehee M.D. Understanding degradation mechanisms and improving stability of perovskite photovoltaics.Chem. Rev. 2019; 119: 3418-3451Crossref PubMed Scopus (532) Google Scholar Meanwhile, the perovskite community is gaining a deeper understanding of the structural features that cause the degradation of perovskites, which, in turn, can significantly guide the stabilization of PSCs. Recently, we reported the discovery of a defective layer composed of disconnected nanocrystals and amorphous phases at the surface of apparent single crystalline grains in polycrystalline perovskite films by solution deposition methods and such a defective layer was observed to initialize and accelerate the degradation of PSCs (unpublished data). The accelerated degradation is not caused by excess PbI2 because red light, which PbI2 does not absorb, also accelerates the degradation when the defective layer is present. Therefore, eliminating such a defective surface layer is an essential step for stabilizing PSCs and modules. Currently, removing the defective perovskite surfaces by vacuum-assisted solid-phase recrystallization or polishing processes has been demonstrated to enhance the stability and efficiency of PSCs.7Back H. Kim G. Kim H. Nam C.-Y. Kim J. Kim Y.R. Kim T. Park B. Durrant J.R. Lee K. Highly stable inverted methylammonium lead tri-iodide perovskite solar cells achieved by surface re-crystallization.Energy Environ. Sci. 2020; 13: 840-847Crossref Google Scholar,8Kong W. Zhao C. Xing J. Zou Y. Huang T. Li F. Yang J. Yu W. Guo C. Enhancing perovskite solar cell performance through femtosecond laser polishing.Sol. RRL. 2020; 4: 42-45Crossref Scopus (8) Google Scholar However, these strategies are appropriate for small-area PSCs but are not compatible with the scalable, high-speed manufacturing of large-area perovskite photovoltaic modules, particularly those made by roll-to-roll processes and have strict requirements for substrate flatness and perovskite material thickness. Therefore, it is important to develop an effective stabilization method that is compatible with the high-throughput manufacturing of perovskite modules without significantly increasing production cost, as low-cost and high-throughput processability is one major advantage of perovskite photovoltaics in competing with conventional counterparts.9Green M.A. Ho-Baillie A. Snaith H.J. The emergence of perovskite solar cells.Nat. Photonics. 2014; 8: 506-514Crossref Scopus (4302) Google Scholar, 10Zhao Y. Zhu K. Organic–inorganic hybrid lead halide perovskites for optoelectronic and electronic applications.Chem. Soc. Rev. 2016; 45: 655-689Crossref PubMed Google Scholar, 11Rong Y. Hu Y. Mei A. Tan H. Saidaminov M.I. Seok S.I. McGehee M.D. Sargent E.H. Han H. Challenges for commercializing perovskite solar cells.Science. 2018; 361: eaat8235Crossref PubMed Scopus (635) Google Scholar, 12Jung E.H. Jeon N.J. Park E.Y. Moon C.S. Shin T.J. Yang T.Y. Noh J.H. Seo J. Efficient, stable and scalable perovskite solar cells using poly(3-hexylthiophene).Nature. 2019; 567: 511-515Crossref PubMed Scopus (1166) Google Scholar, 13Min H. Kim M. Lee S.U. Kim H. Kim G. Choi K. Lee J.H. Seok S.I. Efficient, stable solar cells by using inherent bandgap of α-phase formamidinium lead iodide.Science. 2019; 366: 749-753Crossref PubMed Scopus (444) Google Scholar, 14Chen W. Wu Y. Yue Y. Liu J. Zhang W. Yang X. Chen H. Bi E. Ashraful I. Grätzel M. et al.Efficient and stable large-area perovskite solar cells with inorganic charge extraction layers.Science. 2015; 350: 944-948Crossref PubMed Scopus (1657) Google Scholar In this work, we report a facile method for removing defective surface layers using low-cost adhesive tapes that is compatible with scalable manufacturing of perovskite photovoltaic modules. This method is based on our observation that MHP polycrystalline films with various compositions show an interruption of lattice continuity on the film surfaces. The top surfaces have a typical morphology of many nanocrystals surrounded by an amorphous phase. Such defective surfaces with higher density dangling chemical bonds and vacancies than crystalline regions in the grain interior are found to have weaker bonding with underlying crystalline grains. The weaker mechanical strength at the defective surface regions compared with that at the crystalline regions and the relatively soft nature of perovskites allow the mechanical peeling off of these defective surface layers by low-cost adhesive tapes—demostrated herein for the first time. This simple, high-throughput, and low-cost tape treatment significantly enhanced the stability of perovskites. In addition, the adhesive residual is shown to passivate the perovskites, boosting the efficiency of p-i-n structure solar cells to 22.0%. The mechanical properties of the defective perovskite surface have barely been studied in the past. To characterize the mechanical properties of the defective surface layer on the top of grains, we performed a nanoscratch test on both pristine and polished MAPbI3 films.15Yuan Y. Huang J. Ion migration in organometal trihalide perovskite and its impact on photovoltaic efficiency and stability.Acc. Chem. Res. 2016; 49: 286-293Crossref PubMed Scopus (898) Google Scholar, 16Aristidou N. Eames C. Sanchez-Molina I. Bu X. Kosco J. Islam M.S. Haque S.A. Fast oxygen diffusion and iodide defects mediate oxygen-induced degradation of perovskite solar cells.Nat. Commun. 2017; 8: 15218Crossref PubMed Scopus (543) Google Scholar, 17Tomastik J. Ctvrtlik R. Nanoscratch test—a tool for evaluation of cohesive and adhesive properties of thin films and coatings.EPJ Web Conf. 2013; 48: 00027Crossref Scopus (30) Google Scholar The MAPbI3 films with a thickness of ∼600 nm were fabricated by spin-coating and were verified to yield a device efficiency over 19.0% using a p-i-n device structure before the nanoscratch test.18Bai Y. Lin Y. Ren L. Shi X. Strounina E. Deng Y. Wang Q. Fang Y. Zheng X. Lin Y. et al.Oligomeric silica-wrapped perovskites enable synchronous defect passivation and grain stabilization for efficient and stable perovskite photovoltaics.ACS Energy Lett. 2019; 4: 1231-1240Crossref Scopus (74) Google Scholar In the nanoscratch test, a tip was used to scratch the perovskite films from the top surface to the film interior with a penetration depth of 30 nm while moved laterally for ∼500 nm, as illustrated in Figure 1A. The lateral force of the tip was recorded, which should reflect the bonding strength at different probing depths. The measurement has been conducted on four groups of pristine and surface-polished films, and the results were statistically analyzed. As shown in Figure 1B, the lateral force measured in the pristine MAPbI3 film increases almost linearly with the scratching depth, which was controlled by gradually increasing the normal load. After the nanocrystal/amorphous phase top layer was polished off, the lateral force was larger than that in the pristine film at all scratch depths, indicating that the defective surface layers have less adhesion to the region underneath than the crystalline grain interiors. The averaged lateral force on four different polished films was ∼50% larger than that on the pristine films, and this result verifies the weaker adhesion of the defective surface layers to the underlining crystalline regions compared with inside the crystalline region. Based on this observation, we designed a method to remove the defective layer with adhesive tapes, while the underlying crystalline region remains unaffected, by utilizing the weaker adhesion of the defective region to crystalline regions and the relatively soft nature of MHPs, which is due to the weak ionic bonding between the large-size metal cations and halide anions.5Wang Y. Wu T. Barbaud J. Kong W. Cui D. Chen H. Yang X. Han L. Stabilizing heterostructures of soft perovskite semiconductors.Science. 2019; 365: 687-691Crossref PubMed Scopus (238) Google Scholar,19Yu J. Wang M. Lin S. Probing the soft and nanoductile mechanical nature of single and polycrystalline organic–inorganic hybrid perovskites for flexible functional devices.ACS Nano. 2016; 10: 11044-11057Crossref PubMed Scopus (55) Google Scholar,20Stavrakas C. Zelewski S.J. Frohna K. Booker E.P. Galkowski K. Ji K. Ruggeri E. Mackowski S. Kudrawiec R. Plochocka P. et al.Influence of grain size on phase transitions in halide perovskite films.Adv. Energy Mater. 2019; 9Crossref PubMed Scopus (17) Google Scholar As illustrated in Figure 1C, an adhesive tape made of soft adhesive on a flexible polymer substrate is pressed onto the rough and defective perovskite surfaces to form an intimate contact with appropriate bonding strength. After peeling off this adhesive material, nanocrystals and the amorphous phase were retained on the tape and were, thus, removed from perovskite films. One prerequisite for the success of this method is that the bonding of perovskite large grains with the charge-transport-layer-covered indium tin oxide (ITO) substrate should be strong enough so that the large grains are not peeled off. To demonstrate this, tape treatment was performed by pressing a low-cost 3M Temflex 1700213M. 3M™ Temflex™ Vinyl Electrical Tape 1700.https://www.3m.com/3M/en_US/company-us/all-3m-products/∼/3M-Temflex-Vinyl-Electrical-Tape-1700/?N=5002385+3294355723&rt=rudGoogle Scholar adhesive tape onto pristine MAPbI3 films and subsequently separating the tape from the MAPbI3 surfaces at an angle of ∼180° to the film surface, as schematically illustrated in Figure 1C. Video S1 shows this peeling process. The cross-sectional scanning electron microscopy (SEM) image in Figure S1 shows that the tape forms gapless intimate contact with perovskite films after pressing the tape at a small pressure of 1 MPa at room temperature. Using the 180° peel method22Ebnesajjad S. Landrock A.H. Adhesives Technology Handbook.Third Edition. William Andrew Publishing, 2015: 339-352Crossref Google Scholar, the adhesive strength between the tape and pristine perovskite surfaces was measured to be 2.3 ± 0.1 N cm−1. The peeling-off force was increased to 2.6 ± 0.1 N cm−1 when the tape was applied a second or third time, indicating the successful peeling of the defective region during the first peeling-off process. To verify that the tape treatment can effectively remove the nanocrystals and amorphous phase on the perovskite film surfaces, we conducted high-resolution transmission electron microscopy (HRTEM) measurement of tape-treated MAPbI3 solar cells that contain both an electron transport layer and a metal electrode. One representative image is shown in Figure 2B and twelve others at randomly selected locations in a 6 μm-long focused ion beam (FIB) lamellae of the same device are shown in Figure S2, respectively. No nanocrystal or amorphous phase was observed at the perovskite top surface in any of these HRTEM images. The perovskite lattices inside the grains were found to be continuous and extended directly to the perovskite/C60 interface without the presence of obvious extended defects. To illustrate this, we zoomed-in on region 2, which is 50 nm away from perovskite/C60 interface and region 1, which is at the perovskite-C60 interface, and we marked the (211) plane in the figure. The same trend was also observed at twelve other locations, shown in Figure S2. This result confirms the effectiveness of adhesive tape in removing the defective surface layers. In addition, we carried out energy dispersive spectroscopy (EDS) measurement to analyze the perovskite residuals on the tape after peeling. Both lead and iodide were detected on the tape with an atomic ratio of ∼1:3 (Figure S3 and Table S1). The surface topography changes of the MAPbI3 films caused by tape treatment were characterized via both atomic force microscopy (AFM) and top-view SEM. As shown in Figure S4, AFM measurement results show that the overall topography of the MAPbI3 film did not change notably, while the film became slightly smoother after tape treatment with the root mean square (RMS) roughness reduced from 11.39 to 9.15 nm after peeling. This also indicates an advantage of tape treatment over polishing; the soft adhesive can have conformal contact with the non-flat film surfaces, which allows for the removal of the defective layers without the need to flatten the whole films. This is particularly attractive for applications where the retention of the surface textures is preferred, such as perovskite/silicon tandem solar cells. No pinhole in the film was observed in the SEM image of tape-treated MAPbI3 surfaces (Figure S5), indicating the bonding between large grains and poly(bis(4-phenyl)(2,4,6-trimethylphenyl)amine) (PTAA)-coated ITO is stronger than that between the defective layer and the crystalline region. We further investigated the effectiveness and uniformity of tape treatment by carrying out multiple structural, optoelectrical, and mechanical characterizations. First, the crystallinity of surface perovskites was examined via grazing incidence X-ray diffraction (GIXRD) on the same MAPbI3 film before and after tape treatment. As shown in Figure 3A, the diffraction peak intensity of all crystallographic planes increased significantly after tape treatment, confirming the exposure of the crystalline region by successful removal of nanocrystals or amorphous surfaces. Subsequently, the activation energy (Ea) for ion migration of the MAPbI3 films was measured via temperature-dependent dark conductivity. Lateral structure devices were fabricated by evaporating two Au electrodes on glass/PTAA/MAPbI3 films, which is shown in the inset of Figure 3B. The Ea can be extracted from the Nernst-Einstein equation: σ(T) = (σ0/T)exp(-Ea/kBT), where kB is Boltzmann constant, σ0 is a constant, and T is temperature. The applied electric field was 0.4 V μm−1, which is close to that in the operational solar cells. As shown in Figure 3B, the conductivity can be well separated into electronic conduction and ionic conduction regions for both cases, which is typical for MAPbI3.23Yuan Y. Chae J. Shao Y. Wang Q. Xiao Z. Centrone A. Huang J. Photovoltaic switching mechanism in lateral structure hybrid perovskite solar cells.Adv. Energy Mater. 2015; 5: 572-574Crossref Scopus (426) Google Scholar,24Xing J. Wang Q. Dong Q. Yuan Y. Fang Y. Huang J. Ultrafast ion migration in hybrid perovskite polycrystalline thin films under light and suppression in single crystals.Phys. Chem. Chem. Phys. 2016; 18: 30484-30490Crossref PubMed Google Scholar For the film without tape surface treatment, the ionic conductivity, with an Ea of 0.31 ± 0.02 eV, started to dominate the total conductivity when the temperature increased to 241 K. In striking contrast, the tape-treated film exhibited an increased transition temperature of 273 K and a larger Ea of 0.71 ± 0.03 eV, indicating that ion migration was greatly suppressed through the elimination of defective surfaces that acted as the high-speed ion-migration channel. The surface mechanical properties of pristine and tape-treated MAPbI3 films were studied with a nanoindenter. The indentation depth of a Berkovich tip used in this measurement was fixed within 10% of the total film thickness to avoid the impact from the substrate.25Mamun A.A. Mohammed Y. Ava T.T. Namkoong G. Elmustafa A.A. Influence of air degradation on morphology, crystal size and mechanical hardness of perovskite film.Mater. Lett. 2018; 229: 167-170Crossref Scopus (10) Google Scholar As shown in Figure 3C, the surface of the pristine MAPbI3 exhibited a hardness of 0.54 ± 0.03 GPa, and the surface hardness increased to 0.64 ± 0.05 GPa after tape treatment. This again proves that pristine perovskite surfaces with defective surface layers are softer than the underlying high-crystallinity grain interior, and the softer surface should result from the dangling chemical bonds and high-density defects on the film surface.15Yuan Y. Huang J. Ion migration in organometal trihalide perovskite and its impact on photovoltaic efficiency and stability.Acc. Chem. Res. 2016; 49: 286-293Crossref PubMed Scopus (898) Google Scholar,16Aristidou N. Eames C. Sanchez-Molina I. Bu X. Kosco J. Islam M.S. Haque S.A. Fast oxygen diffusion and iodide defects mediate oxygen-induced degradation of perovskite solar cells.Nat. Commun. 2017; 8: 15218Crossref PubMed Scopus (543) Google Scholar We then mapped steady-state photoluminescence (PL) of the same MAPbI3 film before and after tape treatment to examine the uniformity of tape treatment. As shown in Figure 3D, the tape-treated MAPbI3 film exhibited higher PL intensity in comparison to the pristine film, indicating less nonradiative recombination after eliminating the defective surfaces. In addition, almost the entire measured area exhibited enhanced PL intensity and no dark areas were observed after tape treatment, suggesting good uniformity of tape treatment in removing the defective surfaces without introducing any physical damage to the films, such as peeling off any grains from the film surface. PSCs were fabricated with a p-i-n planar heterojunction configuration of ITO/PTAA/perovskites/C60/bathocuproine (BCP)/Cu. MAPbI3 films were deposited onto the PTAA-covered ITO substrate with a one-step spin-coating process.18Bai Y. Lin Y. Ren L. Shi X. Strounina E. Deng Y. Wang Q. Fang Y. Zheng X. Lin Y. et al.Oligomeric silica-wrapped perovskites enable synchronous defect passivation and grain stabilization for efficient and stable perovskite photovoltaics.ACS Energy Lett. 2019; 4: 1231-1240Crossref Scopus (74) Google Scholar As shown in Figure 4A, the control device without tape treatment exhibited an open-circuit voltage (Voc) of 1.09 V, a short-circuit current density (Jsc) of 22.5 mA cm−2, a fill factor (FF) of 78.9%, and thus a PCE of 19.3%, which are consistent with previous results.18Bai Y. Lin Y. Ren L. Shi X. Strounina E. Deng Y. Wang Q. Fang Y. Zheng X. Lin Y. et al.Oligomeric silica-wrapped perovskites enable synchronous defect passivation and grain stabilization for efficient and stable perovskite photovoltaics.ACS Energy Lett. 2019; 4: 1231-1240Crossref Scopus (74) Google Scholar After tape treatment, the device delivered an enhanced Voc of 1.13 V and a larger FF of 81.0% and a comparable Jsc of 22.4 mA cm−2, yielding a higher PCE of 20.5% from J-V curve and a stabilized efficiency of 20.3% (Figure 4B). The mean PCEs are 18.8% ± 0.3% and 20.1% ± 0.3% for perovskite devices without and with tape treatment, respectively. The photovoltaic parameters of the devices with and without tape treatment were analyzed and the statistical distribution is shown in Figure S6. No noticeable hysteresis was observed under forward and reverse photocurrent sweeps (Figure S7A and Table S2), and the Jsc calculated from the external quantum efficiency (EQE) spectrum agreed well with that derived from the J-V curves (Figure S7B). The enhanced Voc can be attributed to suppressed nonradiative recombination, which is supported by the enhanced PL intensity shown in Figure 3D and tripled charge recombination lifetime measured by transient photovoltage (TPV) at a light bias of 1 sun intensity (Figure 4C). A reduced trap density in the tape-treated perovskite device was also discovered by the thermal admittance spectroscopy measurement result, shown in Figure 4D. The tape-treated device exhibited a much lower trap density of states (tDOS) in shallow trap depth region (0.25 to 0.35 eV). On the one hand, the reduced trap density can be attributed to the removal of defective surfaces, and one the other hand, we found that further pressing the tape onto the perovskite surface might leave a thin layer of polymer (rubber) adhesive on the top of the films, which was also observed to have passivation effects on devices. The adhesive residuals on the perovskite surfaces were confirmed by the X-ray photoelectron spectroscopy (XPS) measurement. As shown in Figure S8A, the intensity of the O1s peak on the surface of a MAPbI3 film increased significantly after tape treatment, which should originate from the polymer resin residues—a key component in the adhesive of tapes.213M. 3M™ Temflex™ Vinyl Electrical Tape 1700.https://www.3m.com/3M/en_US/company-us/all-3m-products/∼/3M-Temflex-Vinyl-Electrical-Tape-1700/?N=5002385+3294355723&rt=rudGoogle Scholar The XPS spectra of the tape before and after surface treatment of a MAPbI3 film show both Pb and I signals (Figures S8B and S8C), which further confirms that the defective perovskite surface layer was removed by the adhesive tape. It is also very encouraging that the adhesive residual had no detrimental effect on the device efficiencies, which is consistent with our previous work that an insulating layer of many compositions also has passivation effects on perovskites due to the various type of passivation functional groups.26Wang Q. Dong Q. Li T. Gruverman A. Huang J. Thin insulating tunneling contacts for efficient and water-resistant perovskite solar cells.Adv. Mater. 2016; 28: 6734-6739Crossref PubMed Scopus (403) Google Scholar,27Zhao Y. Wei J. Li H. Yan Y. Zhou W. Yu D. Zhao Q. A polymer scaffold for self-healing perovskite solar cells.Nat. Commun. 2016; 7: 10228Crossref PubMed Scopus (425) Google Scholar To quantify the passivation effect of the adhesive, we further tested the performance of the tape-treated devices with and without toluene washing. It was found the Voc decreased slightly after washing off the adhesive layer with toluene (Figure S9 and Table S3). Therefore, the reduced trap density is attributable to both the removal of defective surfaces and the passivation effect of adhesive residuals. Overall, the longer carrier recombination lifetime and lower trap density of the tape-treated devices demonstrate that the tape treatment was able to effectively reduce the perovskite surface defects, accounting for the suppressed charge recombination and, thus, enhanced device efficiency. In addition to nanocrystals and amorphous particles, we found that the adhesive tape was also very effective in removing excessive PbI2 particles which have been reported to cause the degradation of perovskites due to photolysis effects.28Liu F. Dong Q. Wong M.K. Djurišić A.B. Ng A. Ren Z. Shen Q. Surya C. Chan W.K. Wang J. et al.Is excess PbI2 beneficial for perovskite solar cell performance?.Adv. Energy Mater. 2016; 6: 7849Crossref Google Scholar,29Roose B. Dey K. Chiang Y.H. Friend R.H. Stranks S.D. Critical assessment of the use of excess lead iodide in lead halide perovskite solar cells.J. Phys. Chem. Lett. 2020; 11: 6505-6512Crossref PubMed Scopus (36) Google Scholar As shown in Figure S10, the MAPbI3 film processed from a PbI2-excess precursor solution (PbI2:MAI molar ratio of 1.05:1) shows white PbI2 particles on the top surface, which were then effectively removed after treatment with the adhesive tape. Our study further provides a simple strategy to remove the excessive PbI2, thus enhancing the light stability of perovskite films. To further test the broad applicability of this surface treatment method, we applied this method to a set of mixed cation and halide perovskite compositions containing cesium (Cs), rubidium (Rb), or tin (Sn), as well as MAPbI3 films prepared via different methods, including one-step and two-step spin-coating methods and blade-coating. The GIXRD spectra in Figure S11 show enhanced XRD peak intensity after tape treatment on all these perovskite films, indicating that there are defective layers residing on the surface of all these perovskites. This appears to be a common phenomenon for perovskite films, which is independent of compositions or preparation methods; however, the increases in peak intensity after tape treatment varied with perovskite compositions. MAPbI3 films exhibited the largest XRD intensity increase after tape treatment, indicating that defective regions on the film surface form most easily on MAPbI3 film. This is in accordance with the HRTEM results on a widely adopted stable composition of Cs0.05FA0.81MA0.14PbI2.55Br0.45. Figure S12 shows the co-existence of an amorphous phase and high-crystallinity regions on the surface of Cs0.05FA0.81MA0.14PbI2.55Br0.45 (CsFAMA) grains, while most of the MAPbI3 film surfaces are covered by amorphous phase. This might be because MA cations are more volatile and escape more easily from MAPbI3 surfaces during the thermal annealing process.30Ciccioli A. Latini A. Thermodynamics and the intrinsic stability of lead halide perovskites CH3NH3PbX3.J. Phys. Chem. Lett. 2018; 9: 3756-3765Crossref PubMed Scopus (89) Google Scholar,31Turren-Cruz S.H. Hagfeldt A. Saliba M. Methylammonium-free, high-performance, and stable perovskite solar cells on a planar architecture.Science. 2018; 362: 449-453Crossref PubMed Scopus (484) Google Scholar Subsequently, we fabricated the PSCs based on these films and detailed device performance parameters are summarized in Table 1, and the corresponding J-V curves are depicted in Figure S13. Obviously, the tape treatment works well on all perovskite compositions for improving efficiencies, and the highest PCE of 22.0% (stabilized PCE of 21.8%, Figure S14) was realized for Rb0.05Cs0.05FA0.85MA0.05PbI2.85Br0.15 (RbCsFAMA)-based p-i-n structure devices. It is also noted that the tape treatment increased PCEs of Sn-containing narrow-band-gap (1.21 eV) PSCs with a composition of Cs0.20FA0.80Pb0.50Sn0.50I3 from 18.6% to 20.2%, which is one of the highest PCEs reported for Pb-Sn binary PSCs to date.32Yang Z. Yu Z. Wei H. Xiao X. Ni Z. Chen B. Deng Y. Habisreutinger S.N. Chen X. Wang K. et al.Enhancing electron diffusion length in narrow-bandgap perovskites for efficient monolithic perovskite tandem solar cells.Nat. Commun. 2019; 10: 4498Crossref PubMed Scopus (110) Google Scholar, 33Kapil G. Bessho T. Ng C.H. Hamada K. Pandey M. Kamarudin M.A. Hirotani D. Kinoshita T. Minemoto T. Shen Q. et al.Strain relaxation and light management in tin-lead perovskite solar cells to achieve high efficiencies.ACS Energy Lett. 2019; 4: 1991-1998Crossref Scopus (57) Google Scholar, 34Lin R. Xiao K. Qin Z. Han Q. Zhang C. Wei M. Saidaminov M.I. Gao Y. Xu J. Xiao M. et al.Monolithic all-perovskite tandem solar cells with 24.8% efficiency exploiting comproportionation to suppress Sn(II) oxidation in precursor ink.Nat. Energy. 2019; 4: 864-873Crossref Scopus (356) Google ScholarTable 1Photovoltaic Parameters of the PSCs with and without Tape TreatmentCompositionTape TreatmentVoc (V)Jsc (mA cm−2)FFPCE (%)MAPbI3(one step)w/o1.0922.50.78919.3w/1.1322.40.81120.5MAPbI3(two step)w/o1.0721.70.76217.8w/1.0921.90.78418.7MAPbI3(doctor bladed)w/o1.1021.70.79018.8w/1.1321.50.80619.6Cs0.05FA0.81MA0.14PbI2.55Br0.45(one step)w/o1.0922.30.78819.2w/1.1322.50.80820.5Rb0.05Cs0.05FA0.85MA0.05PbI2.85Br0.15(one step)w/o1.1223.30.78820.6w/1.1523.40.81722.0Cs0.40FA0.60PbI1.94Br1.06(one step)w/o1.1917.50.82017.1w/1.2418.60.76317.6Cs0.20FA0.80Pb0.50Sn0.50I3(one step)w/o0.8131.00.74318.6w/0.8431.00.77620.2 Open table in a new tab The stability of MAPbI3 films prepared by a one-step method was tested under light-soaking conditions with simulated AM 1.5G irradiation. Here, only half the area of each film was treated with tape so that we could exclude the impact of film-to-film quality variation on the stability study. As presented in Figure 5A, the tape-treated right half still remained black after 8 h of light soaking, while the control half already decomposed into yellow phases in less than 4 h, showing that tape treatment can improve film light stability. On the one hand, the improved film stability can be attributed to the removal of defective surface layers. On the other hand, the aforementioned adhesive residuals that remain on the top of the films can further encapsulate the perovskite films. This is supported by the worst light stability of the film after using toluene to wash off the adhesive residue, as shown in Figure S15. This simple surface treatment method can be used to stabilize large-area perovskite modules. To demonstrate it, we pressed tape from a wide tape roller onto a large-area (∼100 cm2) blade-coated MAPbI3 film (Figure S16) and then light-soaked it under simulated AM 1.5 G intensity for 3.5 h. As shown in Figure 5B, the tape-treated half exhibited much better light stability, and the uniformity of the tape treatment is good, since no yellow spot was observed. We tested whether the speed of pressing and peeling was fast enough so that this can be integrated into a roll-to-roll fabrication line with a linear process speed in excess of 500 mm s−1, the fast coating speed reported for scalable perovskite coating.35Deng Y. Van Brackle C.H. Dai X. Zhao J. Chen B. Huang J. Tailoring solvent coordination for high-speed, room-temperature blading of perovskite photovoltaic films.Sci. Adv. 2019; 5: eaax7537Crossref PubMed Scopus (130) Google Scholar Therefore, the tape treatment holds great promise in scalable fabrication of stable and efficient perovskite solar modules. Long-term operational stability of encapsulated RbCsFAMA perovskite devices were tested under a plasma lamp with a light intensity equivalent to AM 1.5 G in air with a relative humidity of ∼60 ± 10%.36Yang S. Chen S. Mosconi E. Fang Y. Xiao X. Wang C. Zhou Y. Yu Z. Zhao J. Gao Y. et al.Stabilizing halide perovskite surfaces for solar cell operation with wide-bandgap lead oxysalts.Science. 2019; 365: 473-478Crossref PubMed Scopus (362) Google Scholar The light source contained a substantial ultraviolet component and no ultraviolet filter was used during stability testing. All devices were loaded with a resistor so that they worked at maximum power point (MPP) at the beginning of the tests. Instead of monitoring device stability at a reduced temperature of ∼25°C, which may cause an overestimation of solar cell stability,37Chen B. Song J. Dai X. Liu Y. Rudd P.N. Hong X. Huang J. Synergistic effect of elevated device temperature and excess charge carriers on the rapid light-induced degradation of perovskite solar cells.Adv. Mater. 2019; 31: e1902413Crossref PubMed Scopus (41) Google Scholar we conducted the stability test at real operation temperature. Light illumination also heated the perovskite devices to ∼65°C measured at the glass surface. The PCE of an encapsulated control device degraded rapidly from 19.1% to 14.7% after testing for 324 h (Figure 5C). On the contrary, the efficiency of a tape-treated device dropped only slightly to 97.1% of its initial efficiency after light soaking for 1,440 h. This is one of the most stable PSCs reported for those tested at real operation conditions.12Jung E.H. Jeon N.J. Park E.Y. Moon C.S. Shin T.J. Yang T.Y. Noh J.H. Seo J. Efficient, stable and scalable perovskite solar cells using poly(3-hexylthiophene).Nature. 2019; 567: 511-515Crossref PubMed Scopus (1166) Google Scholar,36Yang S. Chen S. Mosconi E. Fang Y. Xiao X. Wang C. Zhou Y. Yu Z. Zhao J. Gao Y. et al.Stabilizing halide perovskite surfaces for solar cell operation with wide-bandgap lead oxysalts.Science. 2019; 365: 473-478Crossref PubMed Scopus (362) Google Scholar,38Christians J.A. Schulz P. Tinkham J.S. Schloemer T.H. Harvey S.P. Tremolet de Villers B.J. Sellinger A. Berry J.J. Luther J.M. Tailored interfaces of unencapsulated perovskite solar cells for >1,000 hour operational stability.Nat. Energy. 2018; 3: 68-74Crossref Scopus (551) Google Scholar, 39Arora N. Dar M.I. Hinderhofer A. Pellet N. Schreiber F. Zakeeruddin S.M. Grätzel M. Perovskite solar cells with CuSCN hole extraction layers yield stabilized efficiencies greater than 20%.Science. 2017; 358: 768-771Crossref PubMed Scopus (993) Google Scholar, 40Bai S. Da P. Li C. Wang Z. Yuan Z. Fu F. Kawecki M. Liu X. Sakai N. Wang J.T.-W. et al.Planar perovskite solar cells with long-term stability using ionic liquid additives.Nature. 2019; 571: 245-250Crossref PubMed Scopus (556) Google Scholar, 41Bi D. Li X. Milić J.V. Kubicki D.J. Pellet N. Luo J. LaGrange T. Mettraux P. Emsley L. Zakeeruddin S.M. et al.Multifunctional molecular modulators for perovskite solar cells with over 20% efficiency and high operational stability.Nat. Commun. 2018; 9: 4482Crossref PubMed Scopus (168) Google Scholar, 42Wang L. Zhou H. Hu J. Huang B. Sun M. Dong B. Zheng G. Huang Y. Chen Y. Li L. et al.A Eu3+-Eu2+ ion redox shuttle imparts operational durability to Pb-I perovskite solar cells.Science. 2019; 363: 265-270Crossref PubMed Scopus (434) Google Scholar Moreover, we measured the operational stability of another five tape-treated devices to check the reproductivity of our strategy. The stability data in Figure S17 show that all tape-treated devices retained 95.2% ± 1.9% of their initial efficiencies after light soaking for over 1,400 h. This indicates that the tape treatment method has a very good reproducibility and is, thus, promising for scalable manufacturing of perovskite modules. In summary, we demonstrated a strategy to remove defective perovskite surfaces with low-cost adhesive tapes to stabilize perovskite films. This strategy also increased the PCEs of the perovskite devices up to 22.0%. The tape-treated device exhibited a long operational time of >1,400 h with negligible efficiency loss. Our studies have also shown the universality of this method in other commonly used perovskite compositions and the compatibility of this method with scaling up of perovskite photovoltaics. This method can also be broadly used in other thin-film perovskite electronic devices and eventually in all perovskite compositions.