UV‑Ozone‑Treated Reduced Graphene Oxide Enhances PEDOT:PSS Hole Transport Layer in Inverted Perovskite Solar Cells

Abstract

Inverted planar perovskite solar cells (PSCs) offer low‑temperature, low‑cost fabrication and reduced hysteresis compared to conventional n‑i‑p architectures. The performance of PSCs hinges on the quality of the hole transport layer (HTL). Here, we demonstrate a green UV‑ozone treatment that improves the hydrophilicity of reduced graphene oxide (rGO) while preserving its high electrical conductivity. The treated rGO is blended into PEDOT:PSS to form a modified HTL. Devices fabricated with this rGO/PEDOT:PSS layer achieve a power conversion efficiency (PCE) of 10.7 %, with a short‑circuit current density (J_SC) of 16.75 mA cm^–2, open‑circuit voltage (V_OC) of 0.87 V, and fill factor (FF) of 75 %. This represents a 27 % improvement over PSCs using pristine PEDOT:PSS, attributed to the smoother surface and enhanced hole mobility of the rGO‑doped layer.

Background

Hybrid organic‑inorganic perovskites have emerged as a leading photovoltaic material, with PCEs climbing from 3.8 % to 22.1 % in the past decade—surpassing crystalline silicon cells. However, the high‑temperature annealing required for traditional n‑i‑p PSCs limits their application on flexible substrates. The inverted p‑i‑n structure, first reported by Guo et al. in 2013, places a hole‑transporting material (HTM) beneath the perovskite layer, followed by an electron‑transporting material (ETM). This architecture simplifies processing, reduces hysteresis, and improves fill factor. Despite progress in HTL and ETL development, the efficiency of inverted PSCs still trails that of conventional devices.

Graphene and its derivatives offer exceptional conductivity, transparency, and environmental stability, making them attractive for HTL modification. Yet, commercial reduced graphene oxide (rGO) tends to aggregate in aqueous media due to limited hydrophilic groups, whereas graphene oxide (GO) suffers from low conductivity. A solution‑processable rGO that balances dispersibility and conductivity is therefore desirable. In this work, we present a facile, environmentally friendly UV‑ozone protocol that introduces hydroxyl and carboxyl groups onto rGO, enabling stable aqueous dispersion without compromising its electrical properties. Incorporating this treated rGO into PEDOT:PSS yields a high‑performance HTL for inverted PSCs.

Methods/Experimental

Chemicals

PEDOT:PSS (Clevios™ PVP. Al 4083), CH₃NH₃I (MAI), PbI₂ (99 %), anhydrous N,N‑dimethylformamide (DMF, 99.8 %), and chlorobenzene (CB, 99.8 %) were purchased from Sigma‑Aldrich. [6,6]‑Phenyl‑C₆₁‑butyric acid methyl ester (PC₆₁BM, > 99 %) and 2,9‑dimethyl‑4,7‑diphenyl‑1,10‑phenanthroline (BCP, > 99 %) were obtained from Xi’an Polymer Light Technology Corp. rGO was synthesized following Y.F. Chen’s protocol.

Solution Preparation

Five milligrams of rGO were placed in a quartz Petri dish and subjected to UV‑ozone exposure (270 W) for 2 h. The treated rGO was collected and dispersed in deionized water to a concentration of 1 mg mL^–1 using an ultrasonic bath. To prepare the HTL, rGO solutions (volume ratios 0.1, 0.2, and 0.3 v / v) were mixed with PEDOT:PSS at room temperature, stirred overnight, and filtered through 0.45 µm PTFE filters.

The perovskite precursor was prepared by dissolving MAI and PbI₂ (1:1 molar ratio) in DMF (40 wt %) at 60 °C for 24 h, then filtering through 0.45 µm PTFE filters.

Device Fabrication

Inverted PSCs were assembled on ITO (1.5 × 1.5 cm²) following a spin‑coating sequence: ITO/PEDOT:PSS/rGO/CH₃NH₃PbI₃/PC₆₁BM/BCP/Ag. The ITO substrates were cleaned with acetone, isopropanol, and water. The rGO/PEDOT:PSS layer was spin‑coated at 4000 rpm for 40 s, then annealed at 150 °C for 10 min in air. The perovskite layer was deposited via a one‑step spin‑coating of the 40 wt % DMF solution (4000 rpm, 40 s), followed by a rapid drop of 70 µL CB 6 s into the wet film. The films were annealed at 110 °C for 30 min under N₂. PC₆₁BM (20 mg mL^–1 in CB) was spin‑coated at 3000 rpm for 40 s, then BCP (saturated solution in IPA) at 2000 rpm for 30 s. Finally, 100 nm Ag was evaporated.

Characterization

XPS was performed using an ESCALAB 250 spectrometer to assess functional groups. XRD patterns of perovskite films were recorded on a Bede high‑resolution diffractometer. AFM (SPI3800) provided surface morphology. Current–voltage (J–V) characteristics were measured with a Keithley 2400 under AM 1.5 G (100 mW cm^–2) illumination from an ABET SUN 3000 solar simulator.

Results and Discussion

Figure 1 illustrates the optical contrast between untreated and UV‑ozone‑treated rGO dispersions. The treated rGO disperses uniformly, evidencing the introduction of hydrophilic –OH and –COOH groups without full oxidation, as confirmed by the preserved black color versus the brown GO.

XPS spectra (Figure 2) show increased intensities of C–O and C–(O)–OH peaks after UV‑ozone treatment, confirming successful functionalization while maintaining the sp² network.

Perovskite crystallinity was probed by XRD (Figure 3). Both films on pristine PEDOT:PSS and rGO/PEDOT:PSS display characteristic peaks at 14.14°, 28.08°, and 31.86° (110, 220, 310 planes). The sharper peaks for the rGO‑modified HTL indicate enhanced crystallinity, which correlates with improved charge transport.

AFM (Figure 4) reveals a slight increase in RMS roughness from 1.15 nm (pristine PEDOT:PSS) to 1.27 nm (rGO/PEDOT:PSS). This modest roughening can promote perovskite grain growth, further supporting the XRD observations.

Device performance (Figure 5) shows that the 0.2 v / v rGO/PEDOT:PSS HTL delivers the best metrics: V_OC = 0.87 V, J_SC = 16.75 mA cm^–2, FF = 75 %, and PCE = 10.75 %. Compared to pristine PEDOT:PSS (PCE = 8.48 %), this represents a 27 % increase. Dark J–V measurements (Figure 5b) confirm reduced leakage current, which enhances both V_OC and J_SC by lowering series resistance and suppressing shunts.

Statistical analysis of 60 devices (Figure 6) confirms the reproducibility: mean PCE of 9.7 ± 1.04 % for rGO‑doped devices versus 7.65 ± 0.48 % for pristine PEDOT:PSS. The improvement is largely driven by higher J_SC (15.25 ± 1.8 mA cm^–2) and FF (72.37 ± 2.03 %) rather than V_OC.

Conclusions

We have introduced a simple, eco‑friendly UV‑ozone process that yields water‑dispersible rGO with preserved conductivity. When blended into PEDOT:PSS, this rGO acts as a high‑mobility additive, enhancing surface morphology and reducing device leakage. Inverted PSCs incorporating 0.2 v / v rGO/PEDOT:PSS achieve a PCE of 10.75 %, a 27 % boost over conventional PEDOT:PSS HTLs. This strategy demonstrates a viable route to high‑efficiency, solution‑processable perovskite solar cells and could be extended to other optoelectronic devices.