Origin of the Hysteresis in I–V Curves for Planar Structure Perovskite Solar Cells Rationalized with a Surface Boundary-induced Capacitance Model

Abstract

For efficient hybrid solar cells based on organometal halide perovskites, the real origin of the IV hysteresis became a big issue and has been discussed widely. In this study, simulated IV curves of different equivalent circuit models were validated with experimental IV curves of a planar perovskite solar cell with the power conversion efficiency (PCE) of 18.0% and 8.8% on reverse scan (from open circuit to short circuit) and forward scan (from short circuit to open circuit), respectively. We found that an equivalent circuit model with a series of double diodes, capacitors, shunt resistances, and single series resistance produces simulated IV curves with large hysteresis matching with the experimentally observed curves. The electrical capacitances generated by defects due to the lattice mismatch at the TiO 2 /CH 3 NH 3 PbI 3 and CH 3 NH 3 PbI 3 /spiro-OMeTAD interface are truly responsible for the hysteresis in perovskite solar cells.Perovskite solar cells based on CH 3 NH 3 PbI 3 have attracted enormous attention in the last few years due to their outstanding performance as photovoltaics. The power conversion efficiency (PCE) of the devices has dramatically improved to over 20% in a relatively short duration.1,2 Despite the unique properties and higher efficiencies, several important issues, e.g., mysterious hysteresis in IV curves and durability of stabilized performance, still remained for commercialization.3,4 It has been found the hysteresis strongly depends on the device architecture, where the planar structure and Al 2 O 3 mesoscopic perovskite cells show relatively large hysteresis than TiO 2 mesoporous structure devices. 5 The typical planar structure of SnO 2 :F(FTO)/compact TiO 2 /CH 3 NH 3 PbI 3 /spiro-OMeTAD/Au 6 suffers from severe hysteresis in the IV measurement.3,4 The reverse scan (from the open circuit to the short circuit) always shows higher PCE than the forward scan (from short circuit to the open circuit). Hence, such hysteresis in the IV curves creates ambiguity about the actual performance of the device, which is being suspected to be over-estimated. 7,8 The origin of hysteresis has been discussed on the intrinsic proprieties like ferroelectric polarization 9 and/or ionic migration 10 of the perovskite to date. However, there was no direct evidence that could support the above claims. It has been reported that passivation of TiO 2 layer by C60 or use of phenyl-C61-butyric acid methyl ester (PCBM) instead of TiO 2 in inverted device structure reduces the hysteresis. 1113 The passivation could minimize the trap states and improve electron transfer through the interface of TiO 2 /CH 3 NH 3 PbI 3 , resulting in the reduction of hysteresis.11 On the other hand, PCBM in the inverted cell could extract the carriers (electrons) more efficiently than TiO 2 without accumulation at the interface, and the hysteresis was eliminated. In another standpoint, lattice mismatch of the interfaces containing organic compounds could be ignored and consequently the hysteresis was reduced. The importance of la...

Effect of TiO2 Surface Treatment on the Current–Voltage Hysteresis of Planar‐Structure Perovskite Solar Cells Prepared on Rough and Flat Fluorine‐Doped Tin Oxide Substrates

“…[8,[11][12][13] In one of our previous reports,w e also found that the origin of hysteresis in I-V curvesc an be explained by an equivalent-circuit modeli ncluding ac apacitance as ac omponent.…”

“…[1][2][3] The I-V hysteresis creates an ambiguity about the actual power of the devices and persists as ac hallenge in this field. [4][5] To elucidate its origin,s everalh ypothesesi ncluding ion migration/displacement, ferroelectric polarization, trappingo fe lectronsa t the interfaces,a nd capacitive effects [6][7][8] were proposed. Migrationo fi ons such as CH 3 NH 3 + ,I À ,P b 2 + ,a nd H + [9] was also suggested as origin of hysteresis.D iffusion of intrinsic ionic defects in CH 3 NH 3 PbI 3 layer implies ah igh long-term stability and performance of the solar cells.I on migration can create defects in the crystal, and these defects can affect the physical and charge-transfer properties at interfaces, eventually leading to charge accumulation.…”

“…[10] Va riationo ft emperature alters the lattice parameters,a nd the difference in thermal expansion coefficient combined with the lattice mismatch between TiO 2 and perovskite could result in interfacial defects. [8] Weak contacts and defects between two layers may hinder charge transfer at this interface, resultingi nI-V hysteresis.T op rovide further clarification, many reports have also shownc apacitive effects in I-V hysteresis phenomena. [8,[11][12][13] In one of our previous reports,w e also found that the origin of hysteresis in I-V curvesc an be explained by an equivalent-circuit modeli ncluding ac apacitance as ac omponent.…”

“…[8] Weak contacts and defects between two layers may hinder charge transfer at this interface, resultingi nI-V hysteresis.T op rovide further clarification, many reports have also shownc apacitive effects in I-V hysteresis phenomena. [8,[11][12][13] In one of our previous reports,w e also found that the origin of hysteresis in I-V curvesc an be explained by an equivalent-circuit modeli ncluding ac apacitance as ac omponent. [8] Experimental I-V curves were successfully reproduced by simulating the I-V behavior of the equivalent circuit.…”

Hysteresis Analysis of Organolead Halide Perovskite Solar Cells by Transient Current Measurement

“…Several components are considered for the origin of hysteresis such as ferroelectricity, [10][11][12][13] ion migration, [14][15][16][17][18][19][20][21][22] carrier trapping 15,26 and capacitance component at the interfaces. [27][28][29][30][31][32] The effect of ferroelectricity is analyzed as the response of bias poling. The effect of ion migration and charge trapping is investigated based on the transient photocurrent analysis and device simulation.…”