High‑Performance Flexible Photodetectors Based on Solution‑Processed CsPbI3‑xBrx Inorganic Perovskites

Introduction

Over the past two decades, inorganic semiconductors such as InGaAs, GaN, ZnO, and Si have dominated photodetector research because of their superior optical and electrical properties, yielding high detectivity and rapid response times. However, their complex growth protocols and high equipment costs hinder large‑scale commercialization. Consequently, there is a growing demand for cost‑effective, scalable alternatives that retain high performance.

Hybrid halide perovskites (HHPs) have attracted considerable attention in photovoltaics, with power conversion efficiencies climbing from 3.8 % to over 23 % in just a decade, driven by optimal bandgaps, high absorption coefficients, and ambipolar carrier transport. Despite their photodetection potential, HHPs suffer from poor environmental stability, limiting practical deployment. In contrast, inorganic cesium lead halide perovskites (IHPs) exhibit superior air stability, making them ideal for photodetector applications. While CsPbI₃ degrades above 300 °C, partial substitution of iodide with bromide (forming CsPbI_3‑xBr_x) stabilizes the perovskite phase and tailors the bandgap.

In this study, we fabricated flexible photodetectors using solution‑processed CsPbI_3‑xBr_x films (x = 1, 2). The resulting devices exhibit fast response (90 µs/110 µs for CsPbI₂Br and 100 µs/140 µs for CsPbIBr₂), high on/off ratios (~10⁴), and remarkable detectivities (~10¹² Jones) under a 10 mV bias. They also demonstrate outstanding mechanical flexibility and environmental resilience, positioning CsPbI_3‑xBr_x as a powerful candidate for flexible photodetector technologies.

Method

Materials

Lead iodide (PbI₂, 99.99 %), lead bromide (PbBr₂, 99.99 %), cesium iodide (CsI, 99.99 %), and cesium bromide (CsBr, 99.99 %) were sourced from Xi’an Polymer Light Technology Corporation. Diethyl ether (DEE), acetone, absolute ethanol, N,N‑dimethylformamide (DMF), and dimethyl sulfoxide (DMSO) were purchased from Sigma‑Aldrich.

Perovskite precursors were prepared under inert atmosphere. For CsPbI₂Br, 1 M CsBr, 1 M PbBr₂, 2 M CsI, and 2 M PbI₂ were dissolved in a 9:1 volume mixture of anhydrous DMSO and DMF, then stirred at 75 °C for >2 h. CsPbIBr₂ followed a similar protocol with 2 M CsBr, 2 M PbBr₂, 1 M CsI, and 1 M PbI₂.

Preparation

Polyimide (PI) flexible substrates were sequentially cleaned in acetone, absolute ethanol, and deionized water for 15 min, then oven‑dried. Interdigitated gold electrodes (80 nm) were thermally evaporated onto the PI. Substrates were UV‑ozone treated for 20 min before transfer to a nitrogen glovebox for film deposition. CsPbI_3‑xBr_x films were spin‑coated (80 µL precursor, 2000 rpm, 60 s) with a 0.5 mL DEE anti‑solvent drop 10 s before spin completion. Films were annealed at 65 °C for 5 min, then 135 °C for 15 min.

Measurements and Characterizations

SEM imaging was performed with a FEI‑INSPECT F50 field‑emission SEM. XRD was collected on a Bede D1 using Cu Kα radiation. UV‑vis absorption spectra were recorded with a Shimadzu UV‑3101 PC. Current–voltage characteristics were measured using a Keithley 2636 under 520 nm laser illumination. Transient photocurrent was captured with an Agilent DOS5012A oscilloscope and an optical chopper. All measurements were conducted at room temperature in ambient conditions.

Results and Discussion

To achieve high‑quality perovskite films, a one‑step spin‑coating protocol with DEE anti‑solvent was employed. Figure 1 shows plan‑view SEM images of CsPbI_3‑xBr_x films with and without DEE treatment. Without DEE, films exhibit numerous pinholes and small grains; with DEE, the surface becomes continuous and grains enlarge, confirming that DEE markedly improves morphology.

SEM images of CsPbI_3‑xBr_x films: (a) CsPbI₂Br, (b) CsPbIBr₂ without DEE; (c) CsPbI₂Br, (d) CsPbIBr₂ with DEE.

XRD patterns (Fig. 2a) confirm the formation of orthorhombic CsPbIBr₂ (peaks at 14.75°, 20.94°, 29.96°, 34.93°) and tetragonal CsPbI₂Br (peaks at 14.44°, 20.30°). UV‑vis absorption spectra (Fig. 2b) reveal a modest blue‑shift for CsPbIBr₂ relative to CsPbI₂Br, attributable to bromide incorporation. Tauc analysis yields bandgaps of 1.91 eV (CsPbI₂Br) and 2.05 eV (CsPbIBr₂), in agreement with literature values.

(a) XRD patterns; (b) absorption spectra; (c) bandgap of CsPbI₂Br; (d) bandgap of CsPbIBr₂.

Device architecture (Fig. 3a) consists of PI/Au interdigitated electrodes coated with CsPbI_3‑xBr_x. Under 520 nm illumination, the photocurrent is generated via charge carrier transport across the perovskite layer (Fig. 3b). Photocurrent measurements (Fig. 3c,d) show maximum currents of 180 µA (CsPbI₂Br) and 120 µA (CsPbIBr₂) at an illumination intensity of 8.23 mW cm⁻², achieved at a bias of only 10 mV. Transient response (Fig. 3e,f) yields rise/decay times of 90 µs/110 µs (CsPbI₂Br) and 100 µs/140 µs (CsPbIBr₂), outperforming comparable flexible devices reported previously.

(a) Device structure; (b) charge transport schematic; (c) I–t of CsPbI₂Br; (d) I–t of CsPbIBr₂; (e) rise/fall times for CsPbI₂Br; (f) rise/fall times for CsPbIBr₂.

Electrical characterization (Fig. 4) demonstrates symmetrical I–V behavior from –5 V to +5 V, indicating ohmic contacts and negligible interfacial barriers. Under low bias, the devices achieve an on/off ratio of ~10⁴. Responsivity and detectivity (Fig. 4c,d) peak at 10¹² Jones at 10 mV, surpassing previous CsPbBr₃ flexible PDs (10¹⁰ Jones at 2 V) and rigid CsPbBr₃ single‑crystal PDs (10¹¹ Jones).

Device I–V curves (a) CsPbI₂Br, (b) CsPbIBr₂; responsivity and detectivity (c) CsPbI₂Br, (d) CsPbIBr₂ under 10 mV.

Stability tests (Fig. 5) reveal that after 30 days in ambient air (35–45 % RH), the I–t curves change by <3 %. Mechanical bending (100 cycles at 9.12 mm radius) induces a negligible <2–3 % drop in performance. These results confirm the structural integrity and environmental resilience of the CsPbI_3‑xBr_x photodetectors.

Reproducible I–t curves after 30 days in air: (a) CsPbI₂Br, (b) CsPbIBr₂; after 100 bends: (c) CsPbI₂Br, (d) CsPbIBr₂.

Conclusion

We have fabricated flexible photodetectors using solution‑processed CsPbI_3‑xBr_x (x = 1, 2) films treated with DEE. The devices achieve a 10⁴ on/off ratio, 90–140 µs response times, and a detectivity of ~10¹² Jones under a 10 mV bias. Importantly, the photodetectors retain performance after 30 days in ambient conditions and after 100 bending cycles, demonstrating both environmental stability and mechanical flexibility. These findings establish CsPbI_3‑xBr_x perovskites as a robust, low‑cost platform for high‑performance flexible photodetection.

Availability of Data and Materials

The datasets used in this study are available from the corresponding author upon reasonable request.

Abbreviations

DEE

Diethyl ether

DMF

N,N‑dimethylformamide

DMSO

Dimethyl sulfoxide

HHPs

Hybrid halide perovskites

IHPs

Inorganic cesium lead halide perovskites

PDs

Photodetectors

SEM

Scanning electron microscope

UV-vis

Ultraviolet‑visible

XRD

X‑ray diffraction