Enhancing the pure water splitting using carbon-iron oxide–carbon nitride (Fe₂O₃-C/CN) heterostructure

2.1 Chemicals and reagents

Ferric tri-chloride hexahydrate (FeCl₃·6H₂O, 99 %; pure), dicyandiamide (DCDA), N, N-dimethylformamide (DMF), terephthalic acid (H₂BDC), and ethanol anhydrous (C₂H₆O) were purchased from Sinopharm Chemical Reagent Co., Ltd. Na₂SO₄ electrolyte, platinum salt for Pt photo-deposition. In all the experiments, DI water was used. All chemical reagents were of analytical grade and used without any purification.

2.2 Synthesis of Fe₂O₃-C catalyst

Initially, 2.0 g of FeCl₃·6H₂O and 100 mg of terephthalic acid (H₂BDC) were added in a solution comprising 20 mL of ethanol and 20 mL of N, N-dimethylformamide (DMF) in an autoclave of total volume 100 mL. A clear solution was obtained after stirring for 20 min, and then sealed in the autoclave and placed in an oven at 120 °C for 6 h. Finally, a dark brown suspension was washed several times with ethanol and DMF to remove the soluble contaminants. Eventually, Fe₂O₃-C was obtained by drying the obtained product at 90 °C for 8 h in a vacuum oven.

2.3 Synthesis of carbon nitride CN and Fe₂O₃-C/CN composite

For the synthesis of CN, typically, DCDA (10.0 g) was dissolved in water at 40 °C, then transferred to an autoclave and put in an oven at 140 °C for 12 h. The obtained sample was washed with DI water many times to eliminate the residuals and heated at 80 °C to dry. Finally, the white powder is placed in a horizontal muffle furnace and heated at 600 °C about 4 h at a heating rate of 5 °C/min and then cooled down naturally to obtain porous carbon nitride (CN).

A simplistic synchronous calcination method was applied to the synthesis of Fe₂O₃-C/CN composite, with MIL-235 acting as the template to synthesize Fe₂O₃ enclosed by an amorphous carbon layer (Fe₂O₃-C). In a typical procedure, 100 mg of synthesized CN and a certain amount of Fe₂O₃-C (1.0, 3.0, 5.0, and 10.0 wt%) were transferred into DI (40. mL) water and sonicated for about half an hour. After that, the solution was filtered and heated at 90 °C to completely evaporate the water. The dried composite was then transferred to a combustion corundum boat with cover and calcinated at 300 °C for about 2 h at a heating rate of 5 °C/min in air.

3.1 Structural and surface area characterizations

Starting from MOF (MIL-235) which serves as an excellent template for the controlled synthesis of Fe₂O₃ nanoparticles. The simulated and synthesized XRD analysis of MIL-235 is shown in Fig. S1. During the synthesis process, the organic ligands of MIL-235 can decompose to form a carbon layer at about 300 °C that coats the Fe₂O₃ nanoparticles. This carbon layer enhances the electrical conductivity and protects the Fe₂O₃ nanoparticles from agglomeration, maintaining their high surface area and active sites. The combination of Fe₂O₃-C nanoparticles and the CN results in a synergistic effect that enhances the photocatalytic performance and improves light absorption, charge transfer, and overall stability of the photocatalytic system. Next, we perform XRD and FTIR analysis of pristine and composite heterostructure.

Fig. 1a shows the X-ray diffraction (XRD) patterns and crystal structure of the obtained pristine Fe₂O₃-C, CN, and Fe₂O₃-C/CN heterostructure photocatalysts. For CN the XRD pattern (red line) shows two characteristic peaks appeared at 12.8° and 27.8° suggesting a pristine nature of CN and corresponding well to the crystal plan of (1 0 0), and (0 0 2) respectively (Sarkar et al., 2022). The crystal face at (1 0 0) represents the repeating unit of tri-s-triazine in its planar structure, whereas the crystal face at (0 0 2) appeared from the interlayer stacking of aromatic segments. All the composite photocatalytic materials showed similar diffraction peaks around 27.8° indexed as the (0 0 2) crystal plane of CN indicated that the crystal phase of CN is well maintained after incorporation of Fe₂O₃-C catalyst. On the other hand, sharp and obvious peaks were detected for the Fe₂O₃-C/CN composite samples consistent with the standard atlas card (JCPDS 24-0072) suggesting a high crystalline nature was achieved using the solvothermal method. Notably, the peak intensity of CN in Fe₂O₃-C /CN showed a slight opposite trend with increasing the amount of Fe₂O₃-C. This phenomenon can be attributed to the effect of Fe₂O₃-C on the inhibition of CN diffraction, indicating that there is a tight interface between Fe₂O₃-C and CN (Qasim et al., 2021).

Close

Fig. 1

A) xrd patterns of fe₂O₃-C, CN, and Fe₂O₃-C/CN photocatalyst composites, b) the corresponding FTIR patterns, and c, d) N₂ adsorption–desorption isotherms of pristine CN, Fe₂O₃-C, and their composite, along with corresponding pore size distribution.

Export to PPT

The FT-IR spectrum was acquired to further investigate the formation of pristine CN, Fe₂O₃-C, and Fe₂O₃-C/CN heterostructure in a range of 450–4000 cm⁻¹ as shown in Fig. 1b. For pristine CN sharp peak is observed around 800–890 cm⁻¹ representing the breathing mode of tri-s-triazine units, and the absorption peaks appeared at 1200–1650 cm⁻¹ indicating the stretching vibration mode of C-N, and C=N heterocycles. Additionally, broad peaks observed between 3000 and 3500 cm⁻¹ suggest the stretching vibration of N–H bonds. The characteristic peaks of pristine CN are also observed in the composite (Fe₂O₃-C/CN) heterostructure representing a well-maintained chemical structure of CN and Fe₂O₃-C. Furthermore, the two intense peaks appeared around 460 and 526 indicating the stretching vibration mode of the Fe-O bond (Qasim et al., 2021).

Next, the nitrogen adsorption–desorption isotherms of the obtained photocatalyst composites were conducted to measure the porosity, diameter, and pore volume (BET-BJH) analysis as shown in Fig. 1c. The composites’ adsorption–desorption isotherms show type IV adsorption–desorption isotherm according to IUPAC type classification, which is correlated with a hysteresis cycle suggested the meso and macroporous material architecture (Qasim et al., 2019). Additionally, the plot of pore volume and pore radius derived using the BJH method (Fig. 1d) revealed the pore size distribution within the range from 20 to 100 Å. The average pore size and surface area of pristine CN, Fe₂O₃-C, and optimized (5 wt% Fe₂O₃-C/CN) composite are 1.72/22.96, 1.90/22.37, and 7.40 nm/44.19 m² g⁻¹, respectively as shown in supporting information Table S1. These results confirm the presence of mesoporous nanotexture in the photocatalytic composites which is consistence with the SEM and TEM morphology characterizations shown above not only enhanced the photocatalytic activity but also enhanced the light absorption and decreased scattering.

3.2 Morphological and compositional analyses

Fig. 2a provided a schematic illustration for the conversion of MOF(MIL-235) into Fe₂O₃-C and formation of (Fe₂O₃-C/CN) heterostructure composites while Fig. 2b shows the microstructure analysis of optimized composite (5 wt% Fe₂O₃-C/CN), where iron oxide (Fe₂O₃-C) nanoparticles (blue circles) are embedded on the surface of pristine carbon nitride nanosheet and free from agglomeration. This understanding can be further verified by magnified HRTEM analysis where the lattice spacing of 0.27 nm can be assigned to the Fe₂O₃ {1 0 4} plane as shown in (Fig. 2c). A closer investigation indicates that the surface of the nanoparticle is covered by a thin carbon layer of thickness about 3–4 nm. Next, (Fig. 2 e-h) shows the energy dispersive X-ray (EDX) mapping analysis of the photocatalyst composite to verify the presence of C, N, Fe, and O elements respectively. The elemental analysis was utilized to determine the composition of C in the Fe₂O₃-C, Fe₂O₃-C/CN heterostructure. Upon meticulous measurement, we have determined that the weight percentage of carbon (C) in the composite material Fe₂O₃-C/CN is greater, specifically 2.15, in comparison to the weight % of carbon (C) in pure Fe₂O₃/CN, which is 1.65. This investigation presents additional proof supporting the presence of a carbon layer on Fe₂O₃. This result further corroborates our prior findings. In addition, (Fig. S2 a, d) shows the SEM images of porous pristine CN structure along with the successful formation of Fe₂O₃-C nanoparticles through the solvothermal route. Similarly, (Fig. S2 b, c) showed that the Fe₂O₃-C nanoparticles are successfully embedded on the surface of CN without any agglomeration. Furthermore, to verify the formation of the carbon layer we perform a mapping analysis of pristine Fe₂O₃-C nanoparticles as shown in Fig. S3.

Close

Fig. 2

A) schematic diagram b) tem, c) hrtem images of 5%fe₂O₃-C/CN composite along with carbon layer, d) STEM and e-h) the corresponding mapping images of C, N, O, and Fe, respectively.

Export to PPT

To determine the atomic valance state of the photocatalyst composites, X-ray photoelectron spectroscopy (XPS) was then performed as depicted in Fig. 3. Fig. 3a, represents the C 1S core XPS spectra of pristine CN, and composite photocatalyst, respectively. It shows three typical binding energy peaks located at 284.86 eV, 286.14 eV, and 288.23 eV, indicated to (C–C), sp²-C connected to –NH₂ (C-NH₂), and sp²-C bounded to N in aromatic ring (N-C=N), respectively (Iqbal et al., 2023; Ma et al., 2023; Palanivel et al., 2023). Similarly, N 1 s XPS spectra split into four peaks situated at 398.60, 399.52, 401.10, and 404.10 eV, respectively for both pristine CN and 5 wt% Fe₂O₃-C/CN composite. These peaks can be indexed to sp²-hybridized nitrogen (C-N = C), tertiary nitrogen (N-3C) groups, amino groups (C-N-H), and π-excitation, respectively (Fig. 3b) (Iqbal et al., 2023; Ma et al., 2023; Yavuz and Erten-Ela, 2022). However, a slight peak shift towards higher binding energies was observed in the composite heterostructure as compared to pristine CN suggesting the successful construction of heterojunction with strong interaction between Fe₂O₃-C and CN substrate which increases the charge transfer process at the interfaces (Danish et al., 2022; Vignesh et al., 2022). Usually, peak shifted to higher binding energy indicates a decrease in electron concentration due to modulated interfacial charge transfer attributed to the formation of heterostructure (Qasim et al., 2021). Furthermore, Fe 2p core spectra of pristine Fe₂O₃-C, and 5 wt.%Fe₂O₃-C/CN, composite exhibit characteristic peaks arising at 710.73 eV, 724.34 eV, 710.83 eV, and 724.45 eV respectively, along with two shake-up satellite peaks. Finally, Fig. 3d presents the high-resolution XPS spectra of O 1 s of pristine Fe₂O₃-C, and composite photocatalyst. The spectra exhibit three characteristic peaks two of them located at around 533.30 eV and 531.81 eV corresponding to C=O and C-O, respectively. In addition, the third peak located at 529.60 eV originated from Fe-O-C bonding (Qasim et al., 2023; Vignesh et al., 2022).

Close

Fig. 3

XPS high-resolution spectra of pristine CN along with 5.0 wt% Fe₂O₃-C/CN samples, a) C 1s, b) N 1s, c) Fe 2p, and d) O core spectra, respectively.

Export to PPT

3.3 UV–vis, bandgap, and band structure analysis

To understand the electronic and band structure of the composites, Fig. 4a shows the collective UV–vis absorption spectra of pristine CN, and Fe₂O₃-C, and their composites are measured in the wavelength range from 300 to 800 nm. The absorption edges of the composite materials exhibited no discernible alterations when contrasted with pristine g-C₃N₄. However, there is a notable enhancement in the absorption intensity of the composite within the visible light spectrum. This enhancement can be attributed to the inclusion of Fe₂O₃, which possesses a broad and intense absorption of visible light. Essentially, the redshift in the absorption region and the augmentation of absorption serves to facilitate the absorption of more solar energy, leading to a greater number of excited electrons. Consequently, this phenomenon enhances the photocatalytic efficiency. Fig. 4(b, c) shows the corresponding Kubelka-Munk plots of the Fe₂O₃-C and CN respectively, where the band gaps of (Fe₂O₃-C and CN can be estimated to be 2.09 eV and 2.52 eV, respectively. To estimate the conduction band (CB) levels under the dark of Fe₂O₃-C and CN sample, Fig. 4(d, e) shows the Mott-Schottky results performed at a frequency of 1.0 kHz in 0.1 M Na₂SO₄ solution. The flat band potential (E_fb) of the semiconductors was represented by the intercept created from an extended line of the Mott-Schottky curve along the x-axis. The results indicated that the E_fb of Fe₂O₃-C and CN samples are −0.30 and −1.65 V (vs. Ag/AgCl), corresponding to 0.31 and −1.04 V (vs. RHE), respectively. Generally, the CB value is supposed to be 0.1 to 0.2 eV more negative than that of the E_fb. Therefore, the actual values of Fe₂O₃ and CN are 0.11 and −1.24 eV vs. RHE, respectively (Dang et al., 2023). However, the reliability of the VBM was also confirmed through experimental validation employing valence band X-ray photoelectron spectroscopy (VB-XPS) on the manufactured samples (Fig. S4 a, b). The VBM values for Fe₂O₃-C and CN were determined to be 2.20 and 1.27 eV, respectively, based on observations of M-S plots. To further the understanding of the energy band structure, UPS analysis was conducted to determine the work functions (Φ) and VBM of the as-fabricated samples using equations (S2-S6).

Close

Fig. 4

A) uv–vis absorption spectra of pristine CN, Fe₂O₃-C, and Fe₂O₃-C/CN composites, b, c) Tauc plots of the UV–vis analysis by using Kubelka-Munk (KM) transformation of the pristine Fe₂O₃-C and CN, d, e) Mott-Schottky plots for Fe₂O₃-C and CN photocatalysts measured at 1.0 kHz in a Na₂SO₄ electrolyte and f) the electronic band structure of Fe₂O₃-C, and CN obtained from a collective analysis of bandgap energy.

Export to PPT

Fig. S4 (c) shows E_cut-off for Fe₂O₃-C and CN were 16.26 and 17.24 eV, respectively. The electron work function (Φ) for Fe₂O₃-C and CN was estimated to be 4.96 and 3.98 eV, respectively using Equation (S2). The Fermi level (E_f), which is the negative value of Φ, was determined to be − 4.96 and −3.98 eV for Fe₂O₃-C and CN, respectively. The valence band top energy (E_VB) relative to the vacuum level was determined by adding the values of VB_UPS (the lower emission onset energy) with Φ, as indicated in Equation (S3). The VB_UPS spectra for Fe₂O₃-C and CN were produced by projecting the left slope to the baseline, resulting in values of 2.02 and 2.13 eV, respectively Fig. S4 d. Thus, the estimated E_VB was −7.054 eV for Fe₂O₃-C and −6.123 eV for CN. The values of E_VB and E_f for Fe₂O_3-C were determined as +2.20 and + 0.18 V (E, vs. NHE, pH = 7) using the conversion formula (S4) and Nernst equation (S5). On the other hand, values of E_VB and E_f for CN were determined as +1.27 and −0.86 V (E, vs. NHE, pH = 7). As a result, the CB potentials (E_CB) for Fe₂O₃-C and CN were determined to be 0.11 and −1.25 V (E, versus NHE, pH = 7) respectively, using Eq. (S7). The comprehensive calculation is provided in the supplementary information. Fig. S4 e) depicts the band structures of Fe₂O₃-C and CN before and after the creation of the connection, as inferred from the observations. The primary factor that drives charge transportation in heterojunction nanocomposites is the fluctuation in Φ or E_f for semiconductor photocatalysts. Fe₂O₃-C exhibits a higher work function (Φ) of 4.96 eV and a lower Fermi energy (E_f) of −4.96 eV, while CN has a lower work function (Φ) of 3.98 eV and a higher E_f of −3.98 eV (Fig. S4 e i). In addition, the formation of heterojunction results in the rapid movement of electrons from Fe₂O₃-C to CN through the conductive carbon layer until the equilibrium E_f is reached (Fig. S4 ii). The photo-generated electrons in the conduction band of Fe₂O₃-C are effectively transferred to the valence band of CN through the conductive carbon layer. They then recombine with the holes generated in the VB of CN (Fig. S4 iii). The charge transfer behavior of this heterojunction corresponds to the XPS data and is in line with the Z-scheme mechanism. As a result, the electrons in the conduction band of CN and the holes produced in the valence band of Fe₂O₃-C are retained. These retained electrons and holes play a role in the photocatalytic redox reactions, as shown in Fig S4 e iii) (Qaraah et al., 2023, 2022).

Based on the above calculation and the band alignment the proposed mechanism can also be drawn and shown in Fig. 4f. Furthermore, additional electrochemical impedance spectroscopy (EIS) measurements at multiple frequencies (i.e. 1000 Hz, 1500 Hz, and 2000 Hz), respectively were conducted as shown in Fig. S5 to obtain a more comprehensive dataset. This will allow us to accurately determine the flat band potential of Fe₂O₃-C and CN. After careful observation, we concluded that the flat band potential values of Fe₂O₃-C and CN are 0.30, and −1.65 eV respectively which are consistent with our previous findings.

3.4 Mechanism analysis

Fig. 5 shows the photocatalytic behavior of pure water splitting of the optimized composite (5 wt% Fe₂O₃-C/CN) composites along with their mass activity under solar irradiation (λ ≥ 420 nm) without using any sacrificial agent. The optimized heterostructure demonstrates an efficient Z-scheme mechanism for pure water splitting, with a conductive carbon layer acting as a crucial mediator between Fe₂O₃ and CN. Upon light irradiation, both Fe₂O₃ and CN absorb visible light, generating electron-hole pairs. The carbon layer facilitates the transfer of electrons from the conduction band (CB) of Fe₂O₃ to the valence band (VB) of CN, effectively reducing the recombination rate of photogenerated charge carriers due to its excellent conductivity. This conductive pathway provided by the carbon layer enhances charge mobility and spatially separates electron-hole pairs, thereby increasing the overall photocatalytic efficiency. Additionally, the carbon layer increases the surface area and offers additional active sites for photocatalytic reactions, further improving efficiency. In this Z-scheme mechanism, electrons in the CB of CN reduce water to produce hydrogen gas (H₂), while holes in the VB of Fe₂O₃ oxidize water to produce oxygen gas (O₂). The integration of Fe₂O₃ and CN through the carbon layer enhances photocatalytic performance by ensuring efficient charge transfer, protecting Fe₂O₃ from photo-corrosion, and synergistically increasing the photocatalytic activity compared to individual components. Thus, the Fe₂O₃-C/CN heterostructure, facilitated by the carbon layer, is a promising candidate for sustainable hydrogen production through solar-driven water splitting. The photocatalytic activity increased significantly at an optimum Fe₂O₃-C concentration (i.e. 5 wt%.) where the rate of hydrogen and oxygen evolution reaches up to 408 and 199 μmol*g^−1*h⁻¹ (i.e. 2:1), respectively. The efficiency of converting solar-to-hydrogen energy reaches 0.29 % when utilizing an AM 1.5 G filter using the S1 equation. Usually, pristine CN and Fe₂O₃ have no potential for pure water splitting because of the high recombination rate whereas in the case of Fe₂O₃ the conduction band is more positive and it is not suitable for pure water splitting. As shown in Fig. S6 and for comparison, the photocatalytic activity of the optimized (5 wt% Fe₂O₃/CN) composite was found to be significantly lower than that of the optimized (5 wt% Fe₂O₃-C/CN) composite. The obtained results were also compared to those of other CN-based heterostructures including Fe₂O₃, as shown in Table S2 of the Supporting Information. From Table S2 it is evident that our Fe₂O₃-C/CN composite demonstrates superior photocatalytic performance in terms of pure water splitting compared to other state-of-the-art photocatalysts. The hydrogen/oxygen evolution rate of 499/199 μmol·g⁻¹·h⁻¹ is among the highest reported, indicating the effectiveness of our material in photocatalytic water splitting.

Close

Fig. 5

A) hydrogen and oxygen evolution rate at 5 wt% of Fe₂O₃-C/CN photocatalyst during pure water splitting under visible light illumination, b) the corresponding mass activity, and c) photocatalytic stability of the 5 wt% of Fe₂O₃-C/CN as tested for 20 h with an evacuation at 5 h intervals.

Export to PPT

The highest activity observed for 5 wt% Fe₂O₃-C/CN, can be attributed to the heterojunction following the Z-scheme mechanism (Fig. 4f) that suppresses the recombination process through the conductive carbon layer and enhances the photocatalytic activity. A Z-scheme band alignment (Fig. 4f) could be formed for the heterostructure, where the CB and VB of CN lie at −1.24 and 1.27 eV, corresponding to Fe₂O₃-C located at 0.11 and 2.20 eV, respectively. It is worth pointing out that the positive CB potential excluded the possibility of hydrogen generation from Fe₂O₃. This idea suggests that hydrogen evolves from the CB of CN. Simply we can say that the photogenerated electrons at the CB from Fe₂O₃ will be consumed by the holes generated at VB of CN through a conductive carbon layer following a Z-scheme mechanism and suppresses the recombination process. On the other hand, electrons in the CB of CN react with water molecules to generate H⁺ which is further converted to H₂, while the holes generated in the VB of Fe₂O₃-C can oxide the water molecules to generate O₂.

To confirm the Z-scheme mechanism of the Fe₂O₃-C/CN heterostructure, we performed Electron Paramagnetic Resonance (EPR) analysis. EPR is a powerful tool for detecting and characterizing paramagnetic species, including unpaired electrons and reactive oxygen species, which are crucial for understanding the charge transfer processes in photocatalytic systems. No signals of DMPO- •OH, and •O₂ − were found in the absence of light as shown in Fig. S7 a, c. The EPR spectra of the optimized composite (5 wt% Fe₂O₃-C/CN) reveal the presence of hydroxyl radicals (•OH) under visible light irradiation as shown in Fig. S7 b. The •OH radicals are detected by their characteristic quartet signal in the presence of the spin-trapping agent DMPO. The formation of •OH radicals indicates that photoexcited holes in the valence band (VB) of Fe₂O₃-C are involved in the oxidation of water molecules, producing •OH radicals. Similarly, the EPR spectra of the Fe₂O₃-C/CN composite show a characteristic signal corresponding to superoxide radicals (•O₂⁻) under visible light irradiation as shown in Fig. S7 d. This indicates the generation of •O₂⁻ radicals in the composite, suggesting efficient charge separation and transfer. The EPR analysis provides direct evidence for forming both superoxide (•O₂⁻) and hydroxyl (•OH) radicals, confirming the Z-scheme charge transfer mechanism in the Fe₂O₃-C/CN heterostructure. The observed EPR signals under visible light irradiation demonstrate that the photoexcited electrons in the CB of CN effectively participate in the reduction process, while holes in the VB of Fe₂O₃-C actively participate in the oxidation process. This charge transfer pathway is consistent with the Z-scheme mechanism, where the spatial separation of charge carriers reduces recombination and enhances photocatalytic efficiency (Xu et al., 2024; Zhang et al., 2024b).

Platinum (Pt) is widely recognized as an effective cocatalyst in photocatalysis due to its exceptional ability to facilitate charge separation and provide active sites for hydrogen evolution reactions (HER). Optimizing the Pt content is crucial for maximizing the photocatalytic activity of the samples, as it directly influences the efficiency of electron capture, hydrogen production, and overall stability of the photocatalyst. Pt provides numerous active sites for the HER, facilitating the reduction of protons to hydrogen gas. The high catalytic activity of Pt for HER is due to its optimal hydrogen adsorption energy, which ensures rapid and efficient hydrogen evolution (Tsao et al., 2021). To further confirm the suggested mechanism, here we also conducted an additional experiment involving the photo-deposition of Pt onto the surface of the optimized composite photocatalyst (5 wt% Fe₂O₃-C/CN). In this scenario, Pt nanoparticles will deposit in situ at sites where their high-valence counterpart is reduced. As shown in Fig. 6, the STEM mapping image of the Fe₂O₃-C/CN heterostructure revealed the deposition of Pt deposit generally located at the surface of CN, verifying the reduction place over the Fe₂O₃-C/CN heterostructure. The direct Z-scheme proposed mechanism of the Fe₂O₃-C/CN heterojunction photocatalyst is demonstrated in Fig. 4f. Despite the absence of a carbon layer, the redox potentials of the composite photocatalyst have been effectively maintained through the utilization of this Z-scheme pathway, the relatively slow charge transfer at the interface of Fe₂O₃ and CN still leads to severe recombination of the charge carriers at each component. As a result, photocatalytic pure water splitting over 5 wt% Fe₂O₃-C/CN was promoted at a limited level. Due to this reason, we believe that the formation of the conductive carbon layer on the surface of Fe₂O₃ plays a critical role in the acceleration of interfacial charge transfer, suppresses the recombination process, and enhances the activity of pure water splitting.

Close

Fig. 6

A) stem image of fe₂O₃-C/CN heterostructure with deposited Pt nanoparticles, and b-f) the corresponding elemental mapping analysis of C, N, O, Fe, and Pt deposition on the composite, respectively.

Export to PPT

The photocatalyst stability is crucial for practical applications, therefore, the optimized photocatalyst composition of 5 wt% of Fe₂O₃-C/CN was tested for pure water splitting for a period of 20 h with an evacuation at 5 h intervals as shown in Fig. 5c. The hydrogen and oxygen evolution rates were retained which shows excellent stability. After the photocatalytic reaction, the composite photocatalyst was characterized by XRD, FTIR, and TEM analyses (see Figs. S8). These outcomes also determine the excellent stability of the composite photocatalyst (Fe₂O₃-C/CN). Finally, the activity of the remaining composite photocatalysts is shown in Fig. S9.

3.5 Optical analysis

To study the charge transfer, separation, superiority of conductive carbon layer, and carrier efficiency mechanism of the photo-generated e^-/h⁺ pairs in pristine CN and optimized composite a photo-luminescence (PL) analysis was carried out at room temperature as shown in Fig. 7a. The PL spectra of pristine CN, and the 5 wt% Fe₂O₃-C/CN composite was obtained at an excitation wavelength of 337 nm. As depicted in Fig. 7a, the peak intensity of the optimized composite (5 wt. %Fe₂O₃-C/CN) quenched as compared to the pristine CN, representing the supportive role of the conductive carbon layer (C) in suppressing the recombination process and enhancing the lifetime of photoexcited charge carriers. Notably the optimized composite of 5 wt% Fe₂O₃-C/CN exhibits the weakest peak intensity, being consistent with the pure water splitting results. It also implies intimate interfacial contact between Fe₂O₃-C and CN, which forms the basis for achieving rapid interfacial charge transfer. Furthermore, an electrochemical impedance spectroscope (EIS) was carried out to examine the behavior of charge transfer and resistance in the composite during interfacial charge transfer as shown in Fig. 7b. In comparison to pristine CN and Fe₂O₃-C, the optimized composite (5 wt% Fe₂O₃-C/CN) exhibits a substantially smaller arc radius in EIS Nyquist plot, indicating significantly lower charge transfer resistance. Thus, it is hypothesized that the conductive carbon layer enhances the charge transfer behavior at the interface between CN and Fe₂O₃-C photocatalytic materials. This understanding was further confirmed via Transient photocurrent response and linear sweep voltage (LSV) measurements of the pristine CN, Fe₂O₃-C, and 5 wt% Fe₂O₃-C/CN photoelectrode in Na₂SO₄ electrolyte solution as shown in Fig. 8. Fig. 8a illustrates the photocurrent response curves of the pure CN, Fe₂O₃-C, and optimized composite photocatalysts when exposed to visible light in multiple cycles of intermittent illumination. Upon exposure to visible light, the photocatalyst experiences a progressive increase in photocurrent until it reaches a steady state. The experiment shows that the charge carriers generated in the photocatalyst were initially trapped by the surface trap state and then transferred to the deep trap state. When the light source is switched off, it is not possible to measure any photocurrent. The optimized composite photocatalyst demonstrates exceptionally elevated photoelectrochemical activity. This photocurrent increase confirms the conductive carbon layer enhances the interface connection and then promotes the electron transfer behavior which is also consistent with the order of pure water splitting activity. Finally, linear sweep voltage (LSV) analysis provides that optimized composite 5 wt% Fe₂O₃-C/CN is another evidence that shows a much-reduced overpotential and a significantly enhanced cathodic current density, suggesting a faster movement of electrons between CN and Fe₂O₃-C as shown in Fig. 8b. All these results demonstrate the vital role of the conductive carbon layer in facilitating a reliable path for fast charge transfer and reducing the electron-hole (e^-/h⁺) pair recombination effect.

Close

Fig. 7

A) pl analysis plots for cn and optimized composite fe₂O₃-C/CN and b) the corresponding Nyquist plots of electrochemical impedance spectroscopy in the dark obtained at a frequency ranging from 1 Hz to 10 kHz.

Export to PPT

Close

Fig. 8

A) transient photocurrent responses of fe₂O₃-C, CN, and an optimized composite of the as-prepared photocatalysts at a potential of 0.6 V (vs. Ag/AgCl) respectively, and b) LSV curves of pristine Fe₂O₃-C, CN, and optimized composite respectively.

Export to PPT