Enhanced Performance of CsPbI₂Br Perovskite Solar Cells via ZnCl₂–MnCl₂ Doping

Introduction

Hybrid organic–inorganic perovskites have attracted significant attention due to their high carrier mobilities and tunable bandgaps, driving PCEs from 3.8 % to 23.3 % through cation‑exchange strategies.[12–17] However, environmental stability remains a critical bottleneck.[18] All‑inorganic cesium lead halides, such as CsPbI₂Br, offer a compelling alternative: a favorable bandgap of ~1.91 eV and robust black‑phase stability in ambient air.[19–20] Grain size and defect density critically influence device efficiency; larger grains reduce grain‑boundary recombination while low trap densities facilitate charge extraction.[27–33] Compositional doping—introducing foreign ions into the lattice—has emerged as a powerful tool to tailor film crystallinity and electronic properties.[34–36] Building on this foundation, we systematically investigate the impact of ZnCl₂–MnCl₂ co‑doping on CsPbI₂Br PSCs, demonstrating marked improvements in morphology, carrier dynamics, and long‑term stability.

Results and Discussion

We synthesized a 1.0 M CsBr/PbI₂ precursor solution in a 1.4:1 DMF/DMSO mixture, incorporating ZnCl₂ and MnCl₂ at 0 %, 0.25 %, and 0.50 % molar ratios (CsPbI₂Br‑0%, CsPbI₂Br‑0.25%, CsPbI₂Br‑0.50%). A one‑step spin‑coating protocol produced ~350 nm thick films annealed at 150 °C. SEM micrographs (Fig. 1a–c) reveal that the 0.25 % doped film displays a highly uniform, pinhole‑free surface, whereas the 0.50 % film shows sparse pinholes that compromise device performance.

Top‑view SEM images of CsPbI₂Br films: a undoped, b 0.25 % doped, c 0.50 % doped.

XRD analysis (Fig. 2a) shows that the 0.25 % doped film exhibits a shift of the (100) peak toward higher angles, indicating lattice contraction due to partial substitution of Pb²⁺ by Zn²⁺/Mn²⁺. XPS spectra (Fig. 2c–f) confirm shifts in Pb 4f, I 3d, and Br 3d binding energies, further supporting lattice incorporation of dopants.

X‑ray diffraction (XRD) patterns (a) and expanded (100) peaks (b). XPS spectra of Cs 3d (c), Pb 4f (d), I 3d (e), and Br 3d (f) for the doped films.

J–V characteristics (Fig. 3a) demonstrate that the 0.25 % doped device achieves a champion PCE of 14.15 % (Jsc = 15.66 mA cm⁻², Voc = 1.23 V, FF = 73.37 %). EQE measurements (Fig. 3b) corroborate the high Jsc values. Electrochemical impedance spectroscopy (EIS) reveals a recombination resistance (Rrec) of 1,016 Ω for the doped film versus 620 Ω for the undoped, reflecting reduced defect‑mediated recombination. Minimal hysteresis is observed in the forward and reverse J–V scans (Fig. 3d).

Light‑weight J–V curves (a), EQE spectra (b), Nyquist plots (c), and J–V hysteresis (d) for the doped and undoped devices.

Long‑term stability tests (Fig. 4a) show that the 0.25 % device retains 87 % of its initial PCE after 30 days in N₂, with Voc remaining stable. Thermal cycling at 80 °C for 150 min preserves 96 % of the original PCE (Fig. 4c). UV–vis absorption (Fig. 4d) confirms that the bandgap remains unchanged, with an onset near 600 nm, indicating that doping does not adversely affect light‑harvesting capabilities.

Normalized device parameters over time (a), efficiency histogram for 30 devices (b), thermal stability (c), and UV–vis absorption spectra (d) for the 0.25 % doped film.

Experimental Section

Materials and Methods

Materials

SnO₂, CsBr, ZnCl₂, MnCl₂, DMSO, DMF, spiro‑OMeTAD, and PbI₂ were purchased from Alfa Aesar, Sigma‑Aldrich, and Xi’an Polymer Light Technology Corp., respectively.

Device Fabrication

ITO glass substrates were sequentially cleaned with detergent, isopropyl alcohol, acetone, and deionized water, followed by 10 min oxygen plasma. A 1:6 SnO₂ aqueous dispersion was spin‑coated at 3,000 rpm (40 s) and annealed at 150 °C (30 min). The perovskite precursor (1.0 M) was filtered (0.22 µm) and stirred at 70 °C for 2 h. Spin‑coating involved 1,000 rpm (12 s) then 5,000 rpm (30 s) with a 100 µL chlorobenzene drip. Films were annealed at 50 °C (1 min) then 150 °C (5 min). A spiro‑OMeTAD HTL (72.3 mg mL⁻¹, with Li‑TFSI, TBP, and chlorobenzene) was spin‑coated at 4,000 rpm (30 s), followed by 80 nm Au deposition via thermal evaporation.

Characterization

XRD was performed on a Rigaku‑2500 diffractometer. SEM imaging used a HITACHI 2100. Solar‑cell performance was measured with a Keithley 2420 under AM 1.5 (100 mW cm⁻²) illumination, calibrated by a silicon reference cell. EIS was conducted on a Zahner Zennium system. EQE was recorded with a Newport Oriel IQE‑200 setup. Device area was 0.044 cm².

Conclusions

Co‑doping CsPbI₂Br perovskites with 0.25 % ZnCl₂–MnCl₂ markedly improves film quality, reducing trap densities and suppressing recombination. The resulting PSC achieves a PCE of 14.15 %, with Jsc = 15.66 mA cm⁻², Voc = 1.23 V, and FF = 73.37 %. These results underscore the efficacy of compositional engineering in advancing the performance and stability of all‑inorganic perovskite solar cells.

The Table of Contents Entry

We present a cost‑effective, reproducible doping strategy that enhances CsPbI₂Br film morphology and device performance, yielding a record 14.15 % efficiency and robust long‑term stability.

Abbreviations

DMF

N,N‑dimethylformamide

DMSO

Dimethyl sulfoxide

EQE

External quantum efficiency

SEM

Scanning electron microscope

XPS

X‑ray photoelectron spectroscopy

XRD

X‑ray diffraction