Lead halide perovskites are at the forefront of optoelectronic materials due to their high absorption coefficient, tunable bandgap, long carrier diffusion length, and small exciton binding energy^[1-3], yielding high-performance optoelectronic devices. The power conversion efficiency (PCE) of organic–inorganic hybrid perovskite solar cells has exceeded 25%^[4]. But the organic–inorganic hybrid perovskites present low thermal stability. All-inorganic perovskites (CsPbX₃, X = I, Br, Cl) prepared by replacing the organic ions (MA⁺, FA⁺) with inorganic Cs⁺ show better thermal stability^[5]. But the photoactive phase α-CsPbI₃ is easy to change to the non-photoactive phase (δ-CsPbI₃) at room temperature^{[6, 7]}. The bromide-containing inorganic perovskite CsPbI₂Br is becoming popular due to better phase stability^[8-10].

Compared with the organic–inorganic hybrid perovskite solar cells, CsPbI₂Br solar cells exhibited a relatively large energy loss (>0.8 eV) (E_loss = E_g – qV_oc)^[11-13], which is mainly caused by defects in the perovskite film and the large energy level offset between the perovskite and charge-transport layers^{[14, 15]}. A feasible strategy is to modify the charge-transport materials to reduce the energy level offset. Modification of PTAA hole-transport layer (HTL) was performed to obtain suitable energy levels and improve the hole-transporting capability^[16]. Seo et al. developed a series of F-containing PTAA derivatives with the highest occupied molecular orbital (HOMO) energy levels from –5.14 to –5.63 eV to match the energy levels of perovskites^[17]. However, these derivatives need tedious chemical synthesis. In this work, the energy level offset was reduced from 0.79 to 0.60 eV by blending 7.5% PVK into PTAA layer. Significantly improved V_oc (from 1.10 to 1.19 V) and PCE (from 11.1% to 13.6%) were obtained.

CsPbI₂Br films were characterized by SEM (Fig. S1), XRD (Fig. S2), and optical measurements (Fig. S3). Solar cells with a structure of ITO/SnO₂/CsPbI₂Br/HTL/MoO₃/Ag (Figs. 1(a) and 1(b)) were fabricated. The performance of perovskite solar cells changing with PVK content in PTAA is summarized in Figs. S4 and S5 and Table S1. The device with pristine PTAA exhibits a J_sc of 14.4 mA cm^–2, a V_oc of 1.10 V, a FF of 70%, and a PCE of 11.1%. The device with PTAA (7.5% PVK) gave a PCE of 13.6%, with a J_sc of 14.5 mA cm^–2, a V_oc of 1.19 V, and a FF of 79% (Fig. 1(c)). The integrated current density from EQE spectrum is 13.8 mA cm^–2, which is consistent with the J_sc from J–V measurements (Fig. S6). The cells presented slight J–V hysteresis (Fig. S7). The steady-state current and stabilized PCE at 0.99 V are 13.2 mA cm^–2 and 13.0%, respectively (Fig. S8).

The difference in charge extraction for pristine PTAA and PTAA (7.5% PVK) was investigated by conducting the transient photocurrent (TPC) measurements under the short circuit condition (Fig. 1(d)). The TPC decay curves were fitted with a single exponential function of I₀exp(−t/τ), where I₀ is steady-state photocurrent and τ represents the charge extraction time. τ was estimated to be 0.26 and 0.83 μs for the cells with PTAA (7.5% PVK) and pristine PTAA, respectively. The decrease of τ indicates that adding 7.5% PVK into PTAA layer increases the charge extraction efficiency and reduces the charge recombination, which is consistent with the electrochemical impedance spectroscopy (EIS) measurements (Fig. S9 and Table S2).

The changes of J–V curves, V_oc and J_sc with light intensity were investigated (Figs. S10–S12). The slope of the semilogarithmic plot for V_oc versus light intensity is equal to nkT/q, where n is the diode quality factor, k is the Boltzmann constant, T is the temperature, and q is the elementary charge. n was calculated to be 1.72 and 1.21 for the solar cells with pristine PTAA and PTAA (7.5% PVK), respectively. The reduction of n indicates reduced trap-induced recombination. Linear fitting for J_sc versus light intensity plots gave slopes of 0.94 and 0.97 for the cells with pristine PTAA and PTAA (7.5% PVK), respectively. The slope for the cell with PTAA (7.5% PVK) is closer to 1, indicating less bimolecular recombination^[18].

The energy levels for pristine PTAA and PTAA (7.5% PVK) were measured by ultraviolet photoelectron spectroscopy (UPS) (Fig. 1(e)). The workfunction (WF) for pristine PTAA is –3.92 eV, with a HOMO level of –5.29 eV, which is consistent with the reported value (–5.23 eV)^[19]. The HOMO level for PTAA (7.5% PVK) film is –5.48 eV, which is closer to the valence band maximum (VBM) (–6.08 eV) of CsPbI₂Br^[20], leading to a better energy level alignment at CsPbI₂Br/HTL interface. The energy level offset being reduced from 0.79 to 0.60 eV (Fig. 1(f)) accounts for the increase of V_oc.

In summary, PVK was introduced into PTAA HTL to optimize the energy level alignment at the CsPbI₂Br/HTL interface. The energy level offset decreased from 0.79 to 0.60 eV. The resulting solar cells delivered a PCE of 13.6% and a high V_oc of 1.19 V. This work provides an effective approach for developing efficient all-inorganic perovskite solar cells.

B. Yang thanks National Natural Science Foundation of China (62004066) and Hunan Provincial Science and Technology Department (2019GK2101) for financial support. L. Ding thanks the National Key Research and Development Program of China (2017YFA0206600) and the National Natural Science Foundation of China (51773045, 21772030, 51922032, 21961160720) for financial support.

Supplementary materials to this article can be found online at https://doi.org/1674-4926/42/3/030501.