Exploring the potential application of binary 3 d transition metal oxides as electron transport layer in perovskite solar cell devices

Abstract

Perovskite solar cells (PSCs) are among the most extensively studied photovoltaic technologies due to their excellent performance. A PSC device consists of multiple layers, including a perovskite absorber layer, electron transport layer (ETL), hole transport layer (HTL), and top and bottom conducting layers. The appropriate combination of these layers is crucial for achieving excellent PSCs. In this study, we explore the use of unconventional materials as ETLs for PSCs, specifically binary 3d transition metal oxides (TMOs) such as α-Fe₂O₃, ZnO, and Cr₂O₃. These materials exhibit excellent optoelectronic and semiconducting properties, making them promising candidates for ETLs in PSCs. We conduct a detailed analysis of the effects of α-Fe₂O₃, ZnO, and Cr₂O₃ as ETLs, focusing on how their structural and electronic properties influence PSC performance. Our investigation includes extensive characterization using X-ray diffraction (XRD), scanning electron microscope (SEM), Fourier transform infrared (FTIR) spectroscopy, atomic force microscopy (AFM), ultraviolet-visible (UV-Vis) spectroscopy, and solar simulator. We believe this study will provide valuable insights into the exploration of 3d TMOs as effective ETLs for PSCs. Furthermore, our findings could open new possibilities for utilizing unconventional layer combinations in PSCs, ultimately contributing to the development of high-performance PSCs.

Keywords

Perovskite TMO ETL solar cells photovoltaic

Introduction

The rapid development of solar cell technology has the potential to address global energy and environmental challenges. As a result, various types of solar cells have been developed, including perovskite solar cells (PSCs). PSCs are known for their excellent performance; however, their stability remains a major challenge, limiting their widespread utilization as a reliable energy source. PSCs are primarily composed of perovskite materials such as methylammonium lead iodide (MAPbI₃).¹ This material has a direct bandgap of approximately 1.6 eV, a high light absorption coefficient of around 10⁴ cm⁻¹, a high dielectric constant, and an exciton binding energy of about 10 meV, lower than the thermal energy available at room temperature (26 meV). These properties make MAPbI₃ material highly suitable for photovoltaic applications, as it facilitates efficient electron-hole pair generation.²

In the configuration of PSCs, the MAPbI₃ layer is combined with an electron transport layer (ETL), a hole transport layer (HTL), and top- and bottom-conducting layers, all deposited on a substrate.^2,3 The proper selection and combination of these layers are very important for obtaining high-performance PSCs. In this study, we explore the potential of binary 3d transition metal oxides (TMOs): α-Fe₂O₃, ZnO, and Cr₂O₃ as alternative ETL materials. These materials possess excellent optoelectronic and semiconducting properties, as summarized in Table 1, making them promising candidates for ETLs in PSCs. However, detailed investigations into their effectiveness as ETLs remain limited. To date, high-efficiency PSCs have primarily relied on a narrow selection of ETL materials, such as SnO₂ and TiO₂. Therefore, further exploration of alternative ETLs, particularly binary 3d TMOs with properties comparable to SnO₂ and TiO₂, is essential.

Table 1.

The characteristics of α-Fe₂O₃, ZnO, and Cr₂O₃.

Materials	Bandgap (eV)	Absorption coefficient (cm⁻¹)	Carrier mobility (cm²V⁻¹s⁻¹)	Carrier density (cm⁻³)	References
α-Fe₂O₃	1.9–2.2	10³–10⁵	0.01–1	10¹⁷–10²⁰	7 –9
ZnO	3–3.3	10⁵	115–155	10¹⁹	10 –12
Cr₂O₃	3–3.4	10⁴	0.14	10¹⁸	13 –17

To the best of our knowledge, although ZnO is less commonly used as an ETL compared to TiO₂ and SnO₂, previous reports have demonstrated its potential application in PSCs. However, PSCs utilizing ZnO as the ETL have only achieved a power conversion efficiency (PCE) of around 1.44%.⁴ Various modifications have been explored to improve ZnO's suitability as an ETL, including interfacial engineering,⁴ doping,⁵ and composite formulation.⁶ In this study, we systematically compare ZnO with two rarely explored binary 3d TMOs, α-Fe₂O₃, and Cr₂O₃, as alternative ETL materials for PSCs. Through this comparison, we aim to identify key factors that determine the suitability of materials as ETLs in PSC devices.

Here, we conduct a detailed analysis of the effects of α-Fe₂O₃, ZnO, and Cr₂O₃ as ETLs in PSCs using X-ray diffraction (XRD), scanning electron microscope (SEM), Fourier transform infrared spectroscopy (FTIR), atomic force microscopy (AFM), and ultraviolet-visible (UV-Vis) measurements. We thoroughly examine how the crystal and electronic structures of these binary 3d TMOs influence their physical properties and, in turn, the overall performance of PSCs. Based on J-V analysis using a solar simulator, the PCEs of PSCs incorporating α-Fe₂O₃, ZnO, and Cr₂O₃ as ETLs were found to be 0.43%, 1.07%, and 0.63%, respectively. Although these efficiencies remain relatively low, we believe this study provides valuable insights that could drive further exploration and discovery of novel semiconductor materials, particularly binary 3d TMOs, as potential ETLs for PSCs. Additionally, this study could open new opportunities for incorporating unconventional ETL materials into PSC configurations, ultimately contributing to the development of high-performance perovskite solar cells (Table 1).

Experiments

Materials: Fe metal powder, Cr metal powder, and ZnCl₂ powder 99% were purchased from Merck. HCl ≥37% was purchased from Smart Lab. Ammonia 25% and ethanol were obtained from Supelco. Fluorine-doped tin oxide (FTO) substrate and methylammonium iodide (MAI/CH₃NH₃I) were purchased from Greatcell. Lead iodide (PbI₂) 99.99% (trace metals basis) was purchased from Tokyo chemical industry. Anhydrous dimethyl sulfoxide (DMSO), anhydrous dimethylformamide (DMF), and anhydrous chlorobenzene were obtained from Sigma Aldrich. Carbon paste (DN-CP01) was purchased from Dyenamo.

Synthesis of TMOs: α-Fe₂O₃, ZnO, and Cr₂O₃ were synthesized using the coprecipitation method. To synthesize α-Fe₂O₃, 3 g of Fe metal powder was dissolved in 15 mL of HCl (12.07 M). The resulting solution was filtered, and 23 mL NH₄OH 25% was added until a precipitate formed. For ZnO synthesis, 1.5 g of ZnCl₂ powder was dissolved in 20 mL of deionized water, followed by the dropwise addition of 5 mL of NH₄OH 25% while stirring for 60 minutes until a milky white solution formed. Similarly, Cr₂O₃ was synthesized by dissolving 3 g of Cr metal powder in 20 mL of HCl (12.07 M) until a dark green solution was obtained. All resulting solutions were centrifuged to separate the precipitates. The obtained precipitates were then washed with deionized water until the pH was neutral and subsequently calcined at 500°C for 2 hours. Finally, the samples were grinded into powder form. After preparation, the samples were analyzed using a UV-Vis spectrophotometer (Thermoscientific Orion Aquamate 8100), XRD (Rigaku MiniFlex 600-C), and SEM–energy dispersive X-ray (EDS) (Thermoscientific Phenom ProX). The average particle size was determined using ImageJ through image segmentation of SEM images. The original SEM images were first converted to 8-bit grayscale to simplify the analysis. Thresholding was then applied to isolate individual particles from the background. Following this, the “Analyze Particles” function in ImageJ was used to extract particle size data. The resulting measurements were compiled and plotted using Origin software to generate the particle size distribution histogram.

Device Fabrication: The fabrication process began with cleaning a 2 × 2 cm² conductive FTO substrate. α-Fe₂O₃, ZnO, and Cr₂O₃ were prepared as ETLs by dissolving 0.05 g of each precursor in 5 mL of ethanol and stirring at 700 rpm for 24 hours. Meanwhile, the MAPbI₃ solution was prepared by dissolving 0.115 g of MAI and 0.363 g of PbI₂ in a mixture of DMF and DMSO (4:1) under continuous stirring at 700 rpm for 3 hours at 70°C inside a glove box. Layer deposition was carried out using the spin coating method inside a glove box. Each ETL layer was obtained by deposition of 100 µL of α-Fe₂O₃, ZnO, or Cr₂O₃ solution onto the substrate, followed by spinning at 1000 rpm for 3 seconds and then at 2000 rpm for 30 seconds. The layers were then annealed at 150°C for 30 minutes. The surface quality of all ETL samples were then characterized using AFM (N8 NEUS BRUKER). Subsequently, the MAPbI₃ layer was deposited by dropping 100 µL of the MAPbI₃ solution onto the substrate and spinning at 1000 rpm for 10 seconds, followed by 3000 rpm for 30 seconds, with an antisolvent added during the last 15 seconds. The MAPbI₃ layer was annealed at 70°C until it turned dark brown, followed by further heating at 100°C for 10 minutes. Next, a carbon top electrode was deposited using the doctor blade method and annealed at 100°C for 15 minutes. The schematic of the fabricated solar cell devices is shown in Figure 1. Finally, the performance of the fabricated PSCs was evaluated through J-V analysis using a calibrated Oriel solar simulator (Newport) under AM 1.5G illumination at 100 mW/cm². Data acquisition was performed with a high-precision Keithley SMU2400 source meter unit.

Figure 1.

Illustration of a fabricated perovskite solar cell device with α-Fe₂O₃, ZnO, and Cr₂O₃ as ETL and carbon electrode as top contact. ETL: electron transport layer.

Results and discussions

It is important to note that the optoelectronic properties of α-Fe₂O₃, ZnO, and Cr₂O₃, as conventional semiconductor materials, depend on their crystal structures, which in turn influence their effectiveness as ETLs in perovskite solar cells. Therefore, a detailed structural analysis of α-Fe₂O₃, ZnO, and Cr₂O₃ is essential. To investigate their crystal structures, the synthesized α-Fe₂O₃, ZnO, and Cr₂O₃ samples were characterized using XRD. As shown in the full XRD scan in Figure 2(a), all three materials exhibit high purity, with no detectable secondary phases or impurities. Additionally, Figure 2(b) shows the Williamson-Hall (W-H) plot derived from the XRD measurements. The W-H plot is utilized to determine the grain size and lattice strain in the samples based on equation (3).^18–20 Here, β represents the full width half maximum of the XRD peaks, θ is the Bragg angle, K a constant typically set to 0.9, and λ corresponds to the X-ray source wavelength, which is approximately 1.5406 Å.

β_{h k l} \cos θ = \frac{K λ}{D} + 4 ε \sin θ

(1)

Figure 2.

(a) Full scan of XRD measurement of α-Fe₂O₃, ZnO, and Cr₂O₃ precursor samples. (b) W-H plot derived from XRD results. XRD: X-ray diffraction.

The intercept $\frac{K λ}{D}$ provides information about the grain or crystallite size (D), while the slope 4ε represents the strain (ε). Additionally, the results were compared with those obtained using the Scherrer method (equation (2)) for further validation.

D = \frac{K}{β} . \frac{λ}{\cos θ}

(2)

Furthermore, the lattice constant of the samples can be determined using Bragg's law in equations (3) and (4) (equation for hexagonal structure crystal). Here, d represents the interplanar spacing, while h, k, and l the Miller indices corresponding to the lattice plane. Additionally, a and c denote the lattice constants along the a-axis and c-axis, respectively. The calculated values were also compared with standard database values from previous reports for validation.

n λ = 2 d \sin θ

(3)

\frac{1}{d_{h k l}^{2}} = \frac{4}{3} (\frac{h^{2} + h k + l^{2}}{a^{2}}) + \frac{l^{2}}{c^{2}}

(4)

The calculation results and detailed information on the synthesized samples are presented in Table 2. It shows that the grain sizes of α-Fe₂O₃, ZnO, and Cr₂O₃, as calculated using both the W-H and Scherer methods, fall within the range of 30 to 50 nm. Additionally, the dislocation density calculated using the Scherrer method reveals that Cr₂O₃ and α-Fe₂O₃ exhibit higher values compared to ZnO, indicating that these materials experience greater microstrain compared to ZnO. This suggests that Cr₂O₃ and α-Fe₂O₃ undergo crystal deformation during the synthesis process. These results are consistent with the W-H plot analysis and the lattice parameter calculations using Bragg's equation, which show that the lattice constants of Cr₂O₃ and α-Fe₂O₃ deviate significantly from their standard bulk values. As conventional semiconductors, the optoelectronic properties of α-Fe₂O₃, ZnO, and Cr₂O₃ are influenced by their crystal structure. Our ZnO sample likely retains similar optoelectronic properties to those reported in previous studies. However, the presence of dislocations and defects in the crystal lattices of Cr₂O₃ and α-Fe₂O₃ may result in altered optoelectronic behavior.

Table 2.

Detail calculation results using Scherrer and W-H methods derived from XRD data, as well as the structure information α-Fe₂O₃, ZnO, and Cr₂O₃ samples.

Parameters		Binary 3d TMO
Parameters		α-Fe₂O₃	ZnO	Cr₂O₃
Phase		Hematite	Wurtzite	Eskolaite
Crystal structure		Hexagonal	Hexagonal	Hexagonal
Scherrer method
Crystallite size	D (nm)	32.48	36.85	31.05
Dislocation density	δ × 10⁻³ (nm⁻²)	0.95	0.74	1.04
Williamson-Hall method
Crystallite size	D (nm)	51	51	38,4
Microstrain	ɛ × 10⁻³	1.03	0.59	0.69
Lattice constant	c (Å)	11.07	5.2	13.01
	Standard	13.72	5.2	13.58

TMO:

transition metal oxide.

SEM analysis was conducted to examine the microstructure of the synthesized α-Fe₂O₃, ZnO, and Cr₂O₃ samples, as shown in Figure 3(a)–(c). The SEM images show that all three materials exhibit relatively similar angular shapes (Figure 3, left-hand side panels). This morphological similarity can be attributed to their shared hexagonal crystal structure, which promotes the formation of comparable particle shapes. Additionally, the use of the same synthesis method for all samples likely contributed to the uniformity in particle formation of these three samples. The similarity in synthesis conditions is also reflected in the average particle size, which was analyzed using ImageJ. The results indicate that the particle size of the obtained α-Fe₂O₃, ZnO, and Cr₂O₃ samples is nearly identical, ranging from approximately 230 to 250 nm, as shown in Figure 3 (middle side panels). Furthermore, the EDX analysis (Figure 3, right-hand side panels) confirms that all samples are composed solely of their expected elemental constituents, with no detectable impurities from other elements.

Figure 3.

SEM measurement showing the morphology (left-hand sides), particle size distribution using ImageJ (right-hand side), and EDX of (a) α-Fe₂O₃, (b) ZnO, and (c) Cr₂O₃ precursor samples. EDX: energy-dispersive X-ray; SEM: scanning electron microscope.

The FTIR measurement results, shown in Figure 4, provide insights into the chemical structures of the α-Fe₂O₃, ZnO, and Cr₂O₃ samples. The spectra indicate the presence of identical organic functional groups: O-H, C-H, and H-O-H at higher wavenumbers, which originate from the solvents used during synthesis. Meanwhile, the characteristic peaks corresponding to Fe-O, Zn-O, and Cr-O bonds appear at lower wavenumbers.^21–23 Interestingly, ZnO exhibits additional peaks associated with Zn-OH functional groups, which are absent in the α-Fe₂O₃ and Cr₂O₃ samples. This suggests that the surface states of ZnO differ significantly from the other two materials, as ZnO more readily adsorbs organic compounds on its surface. These surface characteristics strongly affect the solubility of the samples, with ZnO we found demonstrates a greater tendency to dissolve in ethanol compared to α-Fe₂O₃ and Cr₂O₃. The enhanced solubility of ZnO in ethanol results in better wetting on the FTO substrate, which in turn affects the surface roughness of the thin films formed via the spin-coating method.

Figure 4.

FTIR measurement of α-Fe₂O₃, ZnO, and Cr₂O₃ precursor samples. FTIR: Fourier transform infrared spectroscopy.

UV-Vis spectroscopy was performed to analyze the optoelectronic properties of the synthesized α-Fe₂O₃, ZnO, and Cr₂O₃ samples. The absorption spectra, shown in Figure 5(a), reveal distinct optical characteristics for each material. The synthesized α-Fe₂O₃ sample exhibits three small and broad absorption peaks at 280, 440, and 541 nm, confirming its semiconducting nature.²⁴ These peaks correspond to different electronic transitions^25,26: the 541 nm arises from the Fe3d → Fe3d transition, while the 440 nm peak is attributed to the O2p → Fe3d.²⁷ Additionally, the 280 nm peak is likely associated with a double exciton transition involving magnetically coupled Fe³⁺ cations or pair excitation.^28,29 In contrast, ZnO displays a single sharp absorption peak at 372 nm, which is related to exciton absorption.^30,31 The Cr₂O₃ sample shows four broad absorption peaks. The peaks at 380, 488, and 630 nm correspond to the 3d electronic transition of Cr³⁺ ions within an octahedral coordination geometry, specifically ⁴A_2g → ⁴T_1g, ⁴A_2g → ⁴T_2g, and ⁴A_2g → ²E_2g, respectively.^32–34 Meanwhile, the peak at 228 nm is likely associated with the electronic transition of Cr⁴⁺ ions.³⁴

Figure 5.

UV-Vis measurement results: (a) absorption spectra, (b) Tauc-plot, and (c) Urbach energy calculations of the α-Fe₂O₃, ZnO, and Cr₂O₃ precursor samples. UV-Vis: ultraviolet-visible.

Figure 5(b) shows the Tauc plot, which is derived from the absorption spectra and provides insights into the optical bandgap of the samples using equation (5). In this equation, A is a constant, α the absorption coefficient, h Planck's constant, v the frequency of the incident light, and E_g corresponds to the optical bandgap of the material.

α h ν = A (h ν - E_{g})^{\frac{1}{2}}

(5)

It is important to note that the absorption spectra of α-Fe₂O₃ and Cr₂O₃ exhibit several peaks, making the determination of their bandgap values more complex. To ensure accuracy, we excluded lower energy regions in the Tauc plot, as these may correspond to false electronic transitions. For instance, in the α-Fe₂O₃, the absorption peak at approximately 280 nm is associated with pair excitation, while in Cr₂O₃, the 228 nm peak is linked to the Cr⁴⁺ electronic transition. In this case, we plot the bandgap according to 3d transition of the sample, resulting in the calculated bandgap for α-Fe₂O₃ is 2.7 eV, which is slightly higher than values reported in previous studies.^7–9 Meanwhile, the Tauc plot for ZnO, determined based on its single exciton absorption peak, yields a bandgap of 2.9 eV, which is slightly lower than previous reports.^10–12 In contrast, Cr₂O₃ exhibits a relatively high bandgap of 3.4 eV.^13–17 The bandgap values are strongly related to the presence of defects within the samples. As shown in Table 2, the defect density and microstrain in the synthesized samples are relatively high, which likely influences their optoelectronic properties, making them different from their bulk counterparts. Relatively higher bandgap of the obtained Cr₂O₃ and α-Fe₂O₃ due to internal defects, in turn, may be beneficial to enhance light absorption of the active layer and increase photocurrent of the PCS device.

We further analyzed the optoelectronic properties of the synthesized samples in detail using Urbach energy (UE) calculations. Figure 5(c) presents the UE values for α-Fe₂O₃, ZnO, and Cr₂O₃, extracted from their absorption spectra using equation (6), where E_u represents the Urbach energy. It is important to note that the slope used in UE calculation corresponds to the correct electronic transitions within the samples, following the same approach as in the bandgap determination, which was previously discussed. The slope in UE for α-Fe₂O₃ and Cr₂O₃ samples is determined according to the 3d electronic transition of Fe³⁺ and Cr³⁺, respectively.

α = α_{0} \exp (\frac{h v}{E_{u}})

(6)

The UE provides insight into sub-bandgap energy or band tailing caused by defects.^35–37 A higher UE value indicates more significant band tailing, which is associated with greater crystal disorder. In this case, α-Fe₂O₃ exhibits the highest UE of 1.35 eV, suggesting a high degree of crystal disorder, which agrees with the structural analysis presented in Table 2. Similarly, Cr₂O₃ also shows a relatively high UE of 1.28 eV, indicating notable crystal disorder. In contrast, ZnO has the lowest UE of 1.04 eV, indicating that its crystal disorder is significantly lower compared to the other samples.

Figure 6(a)–(c) shows the AFM measurement results of α-Fe₂O₃, ZnO, and Cr₂O₃ ETL on top of the indium tin oxide (ITO) substrate, respectively. It shows that they have different surface conditions, wherein both ZnO and Cr₂O₃ layers typically have almost the same surface roughness, 35 to 40 nm on average, though the ZnO layer exhibits better distribution compared to the Cr₂O₃ layer. Whereas, α-Fe₂O₃ exhibits the highest surface roughness up to 0.8 μm. Similar to our previous assumption from FTIR, ZnO, which is easier to dissolve in ethanol compared to α-Fe₂O₃ and Cr₂O₃, could support their wettability on the ITO substrate, thus resulting in a smoother layer surface. A smoother layer surface could initiate good contact on each layer in the PSC device, thus improving the photovoltaic process.³⁸ This supports the performance measurement results, wherein, PSC device using ZnO as ETL shows high PCE compared to other PSC devices using α-Fe₂O₃ and Cr₂O₃ ETL, as explained in the next section.

Figure 6.

AFM measurement of (a) α-Fe₂O₃, (b) ZnO, and (c) Cr₂O₃ ETL on ITO substrate. ETL: electron transport layer.

Moreover, the device performance of PSCs employing α-Fe₂O₃, ZnO, and Cr₂O₃ as ETLs was evaluated, as shown in Figure 7. The J-V measurement results of the fabricated device are tabulated in Table 3. Among the tested ETLs, the PSC using ZnO exhibited the highest PCE of 1.07%, which is consistent with previous reports.⁴ Additionally, the ZnO-based device demonstrated higher PCE, FF, and V_OC compared to devices using α-Fe₂O₃ and Cr₂O₃ as ETLs. The enhanced solar cell performance of the ZnO-based PSC can be attributed to its higher carrier mobility, as shown in Table 1. Moreover, ZnO exhibits significant differences in crystal structure, electronic properties, and optoelectronic characteristics, as confirmed by our previous measurements, which likely contribute to its superior photovoltaic performance. In contrast, the α-Fe₂O₃-based PSC showed the lowest performance, with a PCE of 0.43% and the lowest V_OC (277.2 mV) and FF (27.5%) among the tested devices. This could be due to higher defect density and microstrain in α-Fe₂O₃, as suggested by Table 2, which might lead to increased charge recombination and reduced carrier transport efficiency. Additionally, α-Fe₂O₃ has a relatively higher UE (1.35 eV), indicating significant band-tailing effects due to crystal disorder, further affecting charge extraction. Similarly, the Cr₂O₃-based PSC exhibited moderate performance, with a PCE of 0.63% and a V_OC of 329.1 mV, which is higher than α-Fe₂O₃ but lower than ZnO. Cr₂O₃ also possesses a relatively high UE (1.28 eV), suggesting the presence of structural defects. However, its short-circuit current density (J_SC, 6.84 mA/cm²) was the highest among the three samples, possibly due to better absorption properties in the visible range, as inferred from the UV-Vis analysis in Figure 5(a). Despite this, the lower carrier mobility of Cr₂O₃ compared to ZnO likely limits its overall solar cell performance.

Figure 7.

J-V curve of the perovskite solar cell devices using ETL made from the α-Fe₂O₃, ZnO, and Cr₂O₃. At right-hand side panel shows the photograph of α-Fe₂O₃, ZnO, and Cr₂O₃ ETL layers on conductive FTO substrate, respectively. ETL: electron transport layer; FTO: fluorine-doped tin oxide.

Table 3.

The performance (PCE, FF, V_OC, and J_SC) of reverse bias perovskite solar cell devices using ETLs made from α-Fe₂O₃, ZnO, and Cr₂O₃.

ETL	PCE (%)	Average PCE (%)	FF (%)	V_OC (mV)	J_SC (mA/cm²)	R_S (Ohm)	R_Sh (Ohm)
α-Fe₂O₃	0.43	0.31	27.5	277.2	5.67	170.6	378.4
ZnO	1.07	0.72	33.5	498.1	6.43	182.8	1119.4
Cr₂O₃	0.63	0.38	28.2	329.1	6.84	157.1	422.6

ETL: electron transport layer; PCE: power conversion efficiency.

Moreover, the values of series (R_S) and shunt (R_Sh) resistances also impact the device's performance. R_S is an internal resistance within the device, while R_Sh is a parallel resistance that leads to leakage paths within the structure.³⁹ It means that a lower R_S indicates lower internal resistance, which allows more current to flow without loss, while a higher R_Sh indicates less leakage current. Cr₂O₃-based PSC has the lowest R_S; however, its R_Sh remains lower than that of ZnO-based PSC, which outperforms the other ETLs because it combines low R_S and a high R_Sh value. Overall, these results highlight that ZnO remains a superior ETL candidate for perovskite solar cells due to its better charge transport properties, while α-Fe₂O₃ and Cr₂O₃ exhibit limitations due to higher defect densities and lower carrier mobilities, affecting their photovoltaic performance.

It is interesting to note that the Cr₂O₃-based PSC exhibits a higher J_SC compared to devices using α-Fe₂O₃ ETL and ZnO as ETL. Additionally, the Cr₂O₃-based device demonstrates overall better performance than the α-Fe₂O₃-based device. We speculate that the wider bandgap of Cr₂O₃, compared to α-Fe₂O_3, contributes to its improved solar cell performance. The bandgap of the ETL plays a crucial role in achieving high-performance solar cells. A wide-bandgap ETL could enhance light absorption of the active layer and increase photocurrent. Enhancing the performance of α-Fe₂O₃, ZnO, and Cr₂O₃ as ETLs could be achieved through various modifications, such as structural engineering, doping, composite, defect engineering, and surface engineering. Interestingly, our synthesized α-Fe₂O₃ sample exhibits a smaller lattice constant than its bulk counterpart, which unintentionally results in a higher bandgap. Defect density of the synthesized Cr₂O₃ was also found to be relatively high, which further induces a relatively high bandgap of Cr₂O₃. With proper modifications, we believe that α-Fe₂O₃, ZnO, and Cr₂O₃ can be further optimized as effective ETLs to obtain high-performance PSCs. Fabrication methods also hold crucial roles in obtaining high-performance PSC devices using binary 3d TMO as an ETL. Here, we found that ZnO shows a greater tendency to dissolve in ethanol compared to α-Fe₂O₃ and Cr₂O₃. The enhanced solubility of ZnO in ethanol results in better wetting on the FTO substrate, forming a smoother layer surface compared to α-Fe₂O₃ and Cr₂O₃. The low solubility of the TMO hinders the formation of a uniform and high-quality ETL layer on the FTO substrate, which then further induces bad interfaces with the perovskite layer. The strategies to address this issue, such as exploring more soluble binary 3d TMO and optimizing deposition conditions, are very necessary.

Conclusions

Perovskite solar cell devices based on MAPbI₃ using various ETLs made from α-Fe₂O₃, ZnO, and Cr₂O₃ have been successfully fabricated. It shows that these binary 3d TMO could be used as ETL, though the generated PCE is very small. Proper optoelectronic properties such as bandgap and carrier mobility of these binary 3d TMO, support them as ETL in perovskite solar cells to generate photovoltaic. The perovskite solar cell device using ZnO ETL exhibits more excellent performance compared to using α-Fe₂O₃ and Cr₂O₃ ETL due to its highest carrier mobility. The device using Cr₂O₃ ETL also shows better photovoltaic performances compared to the device using α-Fe₂O₃ ETL due to its wider bandgap compared to α-Fe₂O₃. By using proper modifications, we believe that α-Fe₂O₃, ZnO, and Cr₂O₃ as excellent ETLs to obtain high-performance perovskite solar cells.

Footnotes

ORCID iDs

Yuliar Firdaus

Tahta Amrillah

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Universitas Airlangga through IRCON 2024 Unair grant number 190/UN3.LPPM/PT.01.03/2024 and RIIM-LPDP grant Number 82/II.7/HK/2022.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

Data available on request/reasonable request.

References

Sharma

Agarwal

, et al. Stability and efficiency issues, solutions and advancements in perovskite solar cells: a review. Sol Energy 2022; 244: 516–535.

Dahal

. Configuration of methylammonium lead iodide perovskite solar cell and its effect on the device’s performance: a review. Adv Materials Inter 2022; 9: 2200042.

Aitola

Gava Sonai

Markkanen

, et al. Encapsulation of commercial and emerging solar cells with focus on perovskite solar cells. Sol Energy 2022; 237: 264–283.

Han

Kwon

Kim

, et al. Interfacial engineering of a ZnO electron transporting layer using self-assembled monolayers for high performance and stable perovskite solar cells. J Mater Chem A 2020; 8: 2105–2113.

Gantumur

Hossain

Shahiduzzaman

, et al. Tungsten-doped ZnO as an electron transport layer for perovskite solar cells: enhancing efficiency and stability. ACS Appl Mater Interfaces 2024; 16: 36255–36271.

Zhang

Qin

, et al. Zno electron transporting layer engineering realized over 20% efficiency and over 1.28V open-circuit voltage in all-inorganic perovskite solar cells. EcoMat 2022; 4: e12192.

Balberg

Pinch

. The optical absorption of iron oxides. J Magn Magn Mater 1978; 7: 12–15.

Shinde

Bansode

Bhosale

, et al. Physical properties of hematite α-Fe₂O₃ thin films: Application to photoelectrochemical solar cells. J Semicond 2011; 32: 013001.

Sivula

Le Formal

Grätzel

. Solar water splitting: Progress using hematite (α-Fe₂O₃) photoelectrodes. ChemSusChem 2011; 4: 432–449.

10.

Abdel-Galil

Hussien

MSA

Balboul

. Optimal thickness and annealing temperature for enhancement of structural, optical, and photocatalytic properties of ZnO thin films. J Aust Ceram Soc 2022; 58: 1667–1683.

11.

Marimuthu

Anandhan

. Growth and characterization of ZnO nanostructure on TiO2-ZnO films as a light scattering layer for dye sensitized solar cells. Mater Res Bull 2017; 95: 616–624.

12.

Maragliano

Lilliu

Dahlem

, et al. Quantifying charge carrier concentration in ZnO thin films by scanning Kelvin probe microscopy. Sci Rep 2014; 4: 4203.

13.

Abdullah

Rajab

Al-Abbas

. Structural and optical characterization of Cr2O3 nanostructures: evaluation of its dielectric properties. AIP Adv 2014; 4: 027121.

14.

Singh

Verma

Kumar

, et al. Effect of structural and thermal disorder on the optical band gap energy of Cr₂ O₃ nanoparticles. Mater Res Express 2019; 6: 085039.

15.

Erdoğan

İY

Güllü

. Silicon MIS diodes with Cr2O3 nanofilm: optical, morphological/structural and electronic transport properties. Appl Surf Sci 2010; 256: 4185–4191.

16.

Mehdi

SMZ

Liu

Abbas

, et al. Chromium oxide decorated on carbon materials to tune the electron transport layer (ETL) of perovskite solar cells and X-ray detectors. Appl Surf Sci 2023; 636: 157834.

17.

Prathiksha

Kekuda

. Performance evaluation of p-type Cr₂ O₃ thin films grown by reactive DC magnetron sputtering for Schottky diode applications. Mater Res Express 2024; 11: 105901.

18.

Mote

Purushotham

Dole

. Williamson-Hall analysis in estimation of lattice strain in nanometer-sized ZnO particles. J Theor Appl Phys 2012; 6: 6.

19.

Amrillah

Quynh

Taufiq

, et al. Finding appropriate magnetic properties of BiFeO3-CoFe2O4 vertically aligned nanocomposite by modulating the structure of BiFeO3 matrix and composition ratio of CoFe2O4 nanopillars for memory device applications. Ceram Int 2023; 49: 19615–19623.

20.

Amrillah

Quynh

Rachman

, et al. Thermodynamic control the self-assembled formation of vertically aligned nanocomposite thin film. Nanoscale Microscale Thermophys Eng 2024; 28: 100–113.

21.

Kamble

Naikwade

Garadkar

, et al. Ionic liquid assisted synthesis of chromium oxide (Cr2O3) nanoparticles and their application in glucose sensing. J Mater Sci: Mater Electron 2019; 30: 13984–13993.

22.

Sobhanardakani

Jafari

Zandipak

, et al. Removal of heavy metal (Hg(II) and Cr(VI)) ions from aqueous solutions using Fe2O3@SiO2 thin films as a novel adsorbent. Process Saf Environ Prot 2018; 120: 348–357.

23.

Khan

Abid

Jamil

, et al. Layer by layer assembly of zinc oxide nanotubes and nanoflowers as catalyst for separate and simultaneous catalytic degradation of dyes and fuel additive. ChemistrySelect 2019; 4: 5548–5559.

24.

Ahmed

Usman

, et al. Sustainable fabrication of hematite (α-Fe2O3) nanoparticles using biomolecules of Punica granatum seed extract for unconventional solar-light-driven photocatalytic remediation of organic dyes. J Mol Liq 2021; 339: 116729.

25.

Liang

Y-C

Hsu

Y-W

. Enhanced sensing ability of brush-like Fe2O3-ZnO nanostructures towards NO2 gas via manipulating material synergistic effect. IJMS 2021; 22: 6884.

26.

Mizuno

Yao

. On the electronic transitions of α-Fe2O3 hematite nanoparticles with different size and morphology: analysis by simultaneous deconvolution of UV–vis absorption and MCD spectra. J Magn Magn Mater 2021; 517: 167389.

27.

Cai

Pan

, et al. . c-In2O3/α-Fe2O3 heterojunction photoanodes for water oxidation. J Mater Sci 2016; 51: 8148–8155.

28.

Harshiny

Iswarya

Matheswaran

. Biogenic synthesis of iron nanoparticles using Amaranthus dubius leaf extract as a reducing agent. Powder Technol 2015; 286: 744–749.

29.

Naz

Islam

Tabassum

, et al. Green synthesis of hematite (α-Fe2O3) nanoparticles using Rhus punjabensis extract and their biomedical prospect in pathogenic diseases and cancer. J Mol Struct 2019; 1185: 1–7.

30.

Mashford

Baldauf

Nguyen

T-L

, et al. Synthesis of quantum dot doped chalcogenide glasses via sol-gel processing. J Appl Phys 2011; 109: 094305.

31.

Ghosh

Choudhary

RNP

. Optical emission and absorption spectra of Zn–ZnO core-shell nanostructures. J Exp Nanosci 2010; 5: 134–142.

32.

Khalaji

. Structural, optical and magnetic studies of Cr2O3 nanoparticles prepared by microwave-assisted. Nanochemistry Research 2021; 6: 18–24.

33.

Vayssieres

Manthiram

. 2-D esoparticulate arrays of α-Cr₂ O₃. J Phys Chem B 2003; 107: 2623–2625.

34.

Zhang

Cai

, et al. Characterizations of optical absorption in porous Al2O3 Cr2O3 nanocomposites. Nanostruct Mater 1997; 9: 563–566.

35.

Sen Gupta

Farhad

SFU

Habib

, et al. Characterizations of extrinsically doped CZTS thin films for solar cell absorbers fabricated by sol-gel spin coating method. Appl Surf Sci Adv 2023; 13: 100352.

36.

Najm

Al-Ghamdi

Amin

, et al. Towards a promising systematic approach to the synthesis of CZTS solar cells. Sci Rep 2023; 13: 15418.

37.

Jean

Mahony

Bozyigit

, et al. Radiative efficiency limit with band tailing exceeds 30% for quantum dot solar cells. ACS Energy Lett 2017; 2: 2616–2624.

38.

Lin

Wei

, et al. All-perovskite tandem solar cells with improved grain surface passivation. Nature 2022; 603: 73–78.

39.

Sardar

Bera

Chattopadhyay

, et al. The impact of series (R_s) and shunt resistances (R_sh) on solar cell parameters to enhance the photovoltaic performance of f-PSCs. Opt Mater 2024; 155: 115818.