Enhancing photoelectrochemical performance and stability of Ti-doped hematite photoanode via pentanuclear Co-based MOF modification

Modifying photoanodes with metal-organic frameworks (MOFs) as oxygen evolution reaction (OER) cocatalysts has emerged as a promising approach to enhance the efficiency of photoelectrochemical (PEC) water oxidation. However, designing OER-active MOFs with both high photo- and electrochemical stability remains a challenge, limiting the advancement of this research. Herein, we present a facile method to fabricate a MOF-modified photoanode by directly loading a pentanuclear Co-based MOF (Co-MOF) onto the surface of a Ti-doped hematite photoanode (Ti:Fe2O3). The resulting Co-MOF/Ti:Fe2O3 modified photoanode exhibits an enhanced photocurrent density of 1.80 mA∙cm−2 at 1.23 V, surpassing those of the Ti:Fe2O3 (1.53 mA∙cm−2) and bare Fe2O3 (0.59 mA∙cm−2) counterparts. Additionally, significant enhancements in charge injection and separation efficiencies, applied bias photon-to-current efficiency (ABPE), incident photon to current conversion efficiency (IPCE), and donor density (Nd) were observed. Notably, a minimal photocurrent decay of only 5% over 10 h demonstrates the extraordinary stability of the Co-MOF/Ti:Fe2O3 photoanode. This work highlights the efficacy of polynuclear Co-based MOFs as OER cocatalysts in designing efficient and stable photoanodes for PEC water splitting applications.


Introduction
Nowadays we are more and more reliable on fossil fuels in our daily life.However, the intensive use of fossil fuel has led to significant environmental issues, including air pollution, global warming, and damage to ecosystems and human health (Hassan et al., 2024).Consequently, there is an urgent need to explore new, less polluting, and more sustainable energy sources.In this context, hydrogen emerges as an ideal candidate to replace fossil fuels due to its clean emission profile, high energy content, sustainability, and diverse applications (Crabtree et al., 2004).Since hydrogen is predominantly available on Earth within water molecules, the most straightforward and environmental method for hydrogen production is to split water molecules into hydrogen and oxygen by electrolysis, particularly using renewable energy sources like solar, hydro, geothermal, and wind (Dincer and Zamfirescu, 2012).Currently, the cost of hydrogen ranges from $2.50 to $6.80 per kilogram.The US Department of Energy (DOE) has set a goal to reduce the cost of hydrogen production to $1 per kilogram by 2030 (Zainal et al., 2024).However, the process of water splitting is energy-intensive, and the sequential conversion of renewable energy into electricity and subsequently into chemical energy (H 2 ) can lead to high production costs (Ahad et al., 2023).This economic obstacle is a major technological challenge that must be addressed to facilitate the broader application of hydrogen as a fuel.
In recent years, metal-organic frameworks (MOFs), particularly Co-based MOFs, have been utilized to enhance water oxidation efficiency by directly synthesizing them in situ on Fe 2 O 3 photoanodes (Zhang et al., 2018;Li et al., 2019;Wu et al., 2020;Ali et al., 2021;Wang et al., 2021;Xiao et al., 2021;Cai et al., 2023).The validity of this strategy stems from the advantages of MOFs, such as their large specific surface areas and adjustable pore structures, which allow easy accessibility of catalytic active sites, as well as smooth transport of reactants and products.However, because hydrothermal self-assembly of MOFs is usually not a clean process, fabrication of MOF modified photoanodes through in situ synthesis would be encountered with difficulties in purity and loading control of the MOF overlayers.Moreover, the majority of MOFs are not so photo-and electro-chemically stable to be used for practical PEC applications due to their inherent nature of coordination bonds between metal ions and organic ligands.Nevertheless, several researches demonstrated that MOF stability could be effectively improved by replacing single metal nodes with polynuclear cluster nodes (Feng and Du, 2016).In this study, we employ a pentanuclear Co-based MOF, formulated as which enhance both the activity and stability of a Ti-doped Fe 2 O 3 photoanode (Ti:Fe 2 O 3 ).This MOF (Co-MOF) features a pentanuclear {Co 5 } cluster node, which is extended into a 3D triangle network through the ligand (L 2− ) (Du et al., 2021).Our results demonstrate that the incorporation of Co-MOF not only improves the PEC activity but most significantly also the overall stability of the modified photoanode Co-MOF/Ti:Fe 2 O 3 .

Experimental section 2.1 Materials
Unless otherwise specified, the reagents used in the experiments were analytical pure and used as received without further purification.Deionized water was used throughout all experiments.Samples of Co-MOF were prepared following the literature method (Du et al., 2021).

Preparation of photoanodes
Fe 2 O 3 materials with or without Ti doping were grown on a fluorine-doped tin oxide (FTO) glass substrates by a modified hydrothermal-annealing method (Bu et al., 2019).Typically, a piece of FTO substrate (30 × 10 × 2.2 mm) was ultrasonically washed with acetone, ethanol and water for 15 min in sequence, followed by drying in an oven at 60 °C.The FTO glass substrate was sealed with high-temperature resistant tape, leaving area of 1 × 1 cm −2 for following reactions.A 100 mL aqueous solution containing 0.15 M FeCl 3 • 6H 2 O and 1 M NaNO 3 was stirred for 1 h and a 15 mL of the mixed solution was transferred to a 25 mL Teflon-lined stainless steel autoclave.A clean FTO glass was immersed into the solution and the autoclave was heated at 95 °C for 4 h.The yellow film obtained was washed repeatedly with water and annealed in muffle oven at 550 °C (5 °C/min) for 2 h and subsequently at 770 °C (10 °C/min) for 15 min in air to obtain a Fe 2 O 3 film.The preparation of Ti-doped Fe 2 O 3 film was exactly the same as above, except that 1 μL TiCl 4 was added to the solution before it was transferred to the autoclave.The as-prepared Fe 2 O 3 and Ti-doped Fe 2 O 3 photoanodes are designated as Fe 2 O 3 and Ti: Fe 2 O 3 , respectively.For Co-MOF modified photoanode, a sample of Co-MOF (2 mg) was dispersed in 1 mL of ethanol and Nafion solution (v/v: 1:100) through sonication.The Ti:Fe 2 O 3 photoanode was immersed in the above suspension for 5 min, then dried at 60 °C for 5 min in oven.The above process was repeated once to obtain the Co-MOF/Ti:Fe 2 O 3 composite photoanode (Figure 1A).

Material characterization
The surface morphology of the samples was examined using a Hitachi SU8000 scanning electron microscope (SEM) and a JEOL 2100 High-resolution transmission electron microscope (HRTEM).The element distribution was analyzed by energy-dispersive X-ray spectroscopy (EDS) and mapping.The X-ray diffraction (XRD) measurements were performed on a Mini FLEX600 spectrometer with Cu-Kα radiation.X-ray photoelectron spectroscopy (XPS) measurements were carried out using an EscaLab 250Xi (Thermo Fisher Scientific, USA) with an achromatic Al Kα source (1486.6 eV).No surface cleaning was employed before the XPS analysis (Bae et al., 2020).Photoluminescence (PL) spectra were taken by using an Edinburgh FLS1000 fluorescence spectrometer.

Photoelectrochemical measurements
The PEC performances were measured by an electrochemical workstation (CHI 660E) using a standard three-electrode configuration with as-prepared photoanodes as the working electrode, a platinum foil (1 × 1 cm) as the counter electrode, a saturated Ag/AgCl electrode as the reference electrode and 1.0 M NaOH (pH = 13.4) as the electrolyte.A 300 W xenon lamp (PLS-FX300HU) coupled with an AM 1.5 G filter was used as the light source, and the light intensity was adjusted to 100 mW • cm 2 .All asprepared photoanodes were illuminated from the back side and the irradiated areas were 1.0 cm 2 .All electrode potentials reported herein were converted to the reversible hydrogen electrode (RHE) using equation E vs RHE = E (Ag/AgCl) + 0.197 + 0.059 × pH.The linear sweep voltammogram (LSV) curves were recorded at a scan rate of 5 mV/s from −0.4-0.6 V.The photoelectrochemical impedance spectroscopy (PEIS) was performed under illumination with a frequency ranging from 0.01 to 100 kHz and the perturbation amplitude is 5 mV.The Motte-Schottky (M-S) plots were evaluated in dark at a frequency of 1 kHz with a perturbation amplitude of 5 mV.Durability test of the photoanodes was conducted under successive illumination for 10 h at 1.23 V.The evolved H 2 and O 2 were collected and tested in a three-electrode system by a gas chromatograph spectrometer (GC9790Ⅱ) with a thermal conductivity detector (TCD).The electrolyte was purged with Ar for 30 min to eliminate any dissolved oxygen before the measurement.

Characterization of the photoanodes
The morphology, crystal structure, and elemental composition of the samples were characterized using scanning electron

microscopy (SEM) and transmission electron microscopy (TEM).
As depicted in Figure 1B and Supplementary Figure S1D, Fe 2 O 3 nanorods, with mean diameters ranging from 70 nm to 100 nm, are uniformly grown perpendicular to the FTO substrate.After Ti doping, the morphology of Ti:Fe 2 O 3 remains largely unchanged, though slight necking and coalescence among adjacent nanorods are observed (Figure 1C).The SEM images show that after loading, Co-MOF particles are evenly dispersed on the surface of the Ti:Fe 2 O 3 nanorod layer (Figure 1D).Corresponding TEM analyses confirm the one-dimensional rod-like morphology of both Fe 2 O 3 and Ti: Fe 2 O 3 , along with a homogeneous distribution of Co-MOF particles (Supplementary Figure S1).HR-TEM images of Co-MOF/Ti:Fe 2 O 3 display a lattice spacing of 0.25 nm, which corresponds to the (110) plane of Fe 2 O 3 (Figure 1E) (Zhang et al., 2019).The elemental composition was further analyzed using EDX (Supplementary Figure S2 and Table S1) and elemental mapping via SEM (Supplementary Figure S3), confirming a uniform distribution of Fe, Co, Ti, O, N, and S over the sample.
To elucidate the surface chemical composition and electronic states of the photoanodes, XPS analyses were conducted (Supplementary Figure S4, S5).The comprehensive XPS survey confirmed the presence of Fe, O, Ti, Co, and S in Co-MOF/Ti: Fe 2 O 3 .In the fine Fe 2p spectrum of Co-MOF/Ti:Fe 2 O 3 , three distinct peaks were observed: Fe 2p 3/2 at 711.8 eV, Fe 2p 1/2 at 724.5 eV, and a satellite peak at 719.1 eV, corroborating the presence of Fe₂O₃ (Peng et al., 2021).The O 1s spectrum of Co-MOF/Ti: Fe 2 O 3 was deconvoluted into three distinct peaks: lattice oxygen at 530.5 eV, surface hydroxyl groups at 532.6 eV, and water molecules at 535.5 eV (Wei et al., 2015).Compared to Fe 2 O 3 and Ti:Fe 2 O 3 , the binding energies of Fe 2p and O 1s in Co-MOF/Ti:Fe₂O₃ exhibited positive shifts of approximately 0.6 and 0.7 eV, respectively.These shifts suggest potential electronic couplings at the interface between Co-MOF and Ti:Fe₂O₃, likely enhancing photogenerated charge transfer (Li et al., 2020).Additionally, a new area at 532.6 eV was noted, attributable to the oxygen species from the organic ligands in Co-MOF.The Ti 2p XPS spectra of Ti:Fe 2 O 3 and Co-MOF/Ti:Fe 2 O 3 revealed two symmetric peaks at around 458.2 eV and 464.0 eV, with a splitting energy of ~5.6 eV, indicative of Ti doping (Peng et al. 2021).The Co 2p spectrum of Co-MOF/Ti:Fe 2 O 3 displayed peaks at 781.1 eV and 795.7 eV, corresponding to Co 2p 3/2 and Co 2p 1/2 , respectively, confirming the presence of Co 2 ⁺ (Du et al., 2021).

Photoelectrochemical properties
The linear sweep voltammetry (LSV) curves of the photoanodes were recorded under AM 1.5 G illumination.Fe 2 O 3 and Ti:Fe 2 O 3 exhibited photocurrent densities of 0.59 mA•cm −2 and 1.53 mA•cm −2 at 1.23 V, respectively.In contrast, the Co-MOF/ Ti:Fe 2 O 3 photoanode demonstrated a superior photocurrent density of 1.80 mA•cm −2 (Figure 3A).Additionally, Co-MOF/Ti:Fe 2 O 3 displayed an onset potential of only 0.86 V, showing a 100 mV cathodic shift relative to Ti:Fe 2 O 3 .The chopped current-time (I−t) curves for these photoanodes, measured at 1.23 V under interrupted illumination, are presented in Figure 3B.All samples showed rapid photocurrent response and achieved a stable current upon light activation, indicating excellent light sensitivity and robust stability.The photocurrent densities are consistent with those observed in the LSV results.The PEC performance of Co-MOF/Ti:Fe 2 O 3 is either better than or comparable to previously reported results (Supplementary Table S2).This enhancement is attributed to the improved conductivity from Ti doping and the enhanced surface oxidation kinetics due to Co-MOF, synergistically advancing the PEC efficiency of the hematite photoanode.
The ABPE of the photoanodes are detailed in Figure 3C.Co-MOF/Ti:Fe 2 O 3 achieved a ABPE value of 0.18% at 1.05 V, which is significantly higher than that of Fe 2 O 3 (0.062% at 1.03 V) and Ti: Fe 2 O 3 (0.11% at 1.09 V), representing increases of 2.9 times and 1.6 times, respectively.Furthermore, the IPCE were measured, with results shown in Figure 3D.All photoanodes demonstrated photoresponses across the wavelength range of 380 nm-650 nm, with IPCE values at 400 nm being 2.6% for Fe 2 O 3 , 20.9% for Ti: Fe 2 O 3 , and 38.1% for Co-MOF/Ti:Fe 2 O 3 .The enhanced ABPE and IPCE for Co-MOF/Ti:Fe 2 O 3 can be attributed to improved charge transfer kinetics on the surface facilitated by the incorporation of Co-MOF.
To clarify the roles of Ti doping and the Co-MOF co-catalyst, the photocurrent densities of photoanodes were evaluated under illumination from an AM 1.5 G light source in a 1 M NaOH electrolyte, both with and without a 0.5 M Na 2 SO 3 hole sacrificial agent, as depicted in Supplementary Figure S6.The efficiencies of charge injection and separation for these photoanodes are presented in Figures 4A, B  XRD patterns of the photoanodes.
Frontiers in Chemistry frontiersin.orgreached 80%, superior to that of Ti:Fe 2 O 3 (71%) and bare Fe 2 O 3 photoanode (59%).The charge separation efficiency for the Co-MOF/Ti:Fe 2 O 3 (21.0%)was comparable with that of Ti:Fe 2 O 3 (21.4%),both of which were much higher than that of bare Fe 2 O 3 photoanode (8.2%).These results demonstrat significant improvements in reducing surface recombination and enhancing water oxidation kinetics through the synergistic effects of Ti doping and Co-MOF loading.Photoelectrochemical impedance spectroscopy (PEIS) measurements provide insights into the charge transfer behavior of photoanodes under illumination.Figure 4C displays the Nyquist plots for the bare Fe 2 O 3 , Ti:Fe 2 O 3 , and Co-MOF/Ti:Fe 2 O 3 photoanodes, which were analyzed using an equivalent resistance-capacitance (RC) circuit model (inset in Figure 4C).Within this model, R s represents the series resistance associated with the FTO substrate and the test system, while R ct denotes the charge transfer resistance at the electrode-electrolyte interface during the water oxidation reaction.The fitted parameters for these resistances for the different photoanodes are detailed in Supplementary Table S3.Notably, all photoanodes display similar external resistances (R s ) ranging from 23.1 Ω to 25.2 Ω, indicating a consistent interface between the semiconductor and the FTO substrate.However, the Co-MOF/Ti:Fe 2 O 3 photoanode demonstrates a significantly lower R ct value of 289 Ω compared to Fe 2 O 3 (1,047 Ω) and Ti:Fe 2 O 3 (310 Ω).These results suggest that Ti doping and the addition of Co-MOF on Fe 2 O 3 can markedly enhance the charge transfer processes at the electrode/electrolyte interface, contributing to improved PEC performance.
Mott-Schottky (M-S) analysis was utilized to further assess the donor concentration (N d ) and the flat band potential (V fb ) of the photoanodes.The positive slopes of the M-S plots confirm the n-type semiconductor characteristics of these hematite photoanodes.A decrease in the slope indicates an increase in carrier density, reflecting enhanced charge carrier concentration in the modified hematite photoanodes (Figure 4D).The N d value for the Co-MOF/Ti:Fe 2 O 3 photoanode was determined to be approximately 9.32 × 10 20 cm −3 .In comparison, the N d values for the Ti:Fe 2 O 3 and bare Fe 2 O 3 photoanodes were 2.63 × 10 20 cm −3 and 1.42 × 10 20 cm −3 , respectively (Supplementary Table S4).This result suggests that Ti doping and Co-MOF loading incrementally enhance the charge transfer capabilities of the photoanodes (Kumar et al., 2022).The flat band potentials (V fb ) were measured as 0.56 V for Co-MOF/Ti:Fe 2 O 3 , 0.61 V for Ti:Fe 2 O 3 , and 0.63 V for bare Fe 2 O 3 , which is consistent with the trend of the onset potentials observed in the LSV curves.
The steady-state open-circuit photovoltage (OCP) was measured both under light illumination and in the dark, as shown in Supplementary Figure S7.For Fe 2 O 3 , the OCP in the dark is recorded at 0.905 V, deviating from the equilibrium value of 1.23 V and indicating the presence of surface states.After Ti doping and Co-MOF loading, the OCP in the dark shifts positively (0.97 V for Ti:Fe 2 O 3 and 1.02 V for Co-MOF/Ti:Fe 2 O 3 ), suggesting partial passivation of these surface states, and hence the enhancement of This reduction in photovoltage for the Ti-doped samples could be due to the increased carrier density, which narrows the space charge layer, a phenomenon also observed in other similar systems (Cai et al., 2023).Photoluminescence (PL) measurements were conducted in air without applied bias to assess the impact of Ti doping and Co-MOF loading on charge separation, as illustrated in Supplementary Figure S8.A notable reduction in PL intensity for Ti: Fe 2 O 3 compared to bare Fe 2 O 3 corresponds with the decreased charge carrier recombination and enhanced electron and hole separation efficiency (Wang et al., 2018b).Conversely, the Co-MOF modification led to an increase in PL intensity, which suggests effective hole storage within the material, enhancing the likelihood of hole transfer to the electrode surface during PEC water oxidation (Zhang et al., 2018;Wu et al., 2020).Stability of the photoanodes was examined through chronoamperometry (Figure 5A).The Co-MOF/Ti:Fe 2 O 3 photoanode demonstrated excellent stability, maintaining 95% of its initial photocurrent density throughout the test duration.In contrast, the bare Fe 2 O 3 and Ti:Fe 2 O 3 photoanodes retained only 80% and 84% of their original activity, respectively.Subsequently, the generation and quantification of H 2 and O 2 by the Co-MOF/Ti: Fe 2 O 3 photoanode were conducted and are presented in Figure 5B.The Co-MOF/Ti:Fe 2 O 3 photoanode achieved PEC H 2 and O 2 evolution rates of 13.0 μmol•cm −2 •h −1 and 6.28 μmol•cm −2 •h −1 , respectively.In comparison, the Ti:Fe 2 O 3 photoanode exhibited lower evolution rates of 7.58 μmol•cm −2 •h −1 for H 2 and 3.92 μmol•cm −2 •h −1 for O 2 .This results indicate the superior PEC stability of the Co-MOF/Ti:Fe 2 O 3 photoanode, highlighting its effectiveness in sustained water splitting under operational conditions.

Conclusions
In conclusion, the performance of a hematite photoanode in PEC water splitting was enhanced through Ti doping and Co-MOF modification.The modified photoanode, Co-MOF/Ti:Fe 2 O 3 , was easily fabricated by directly loading Co-MOF particles onto the surface of the Ti-doped hematite photoanode.LSV measurements under illumination demonstrated that modifying with Co-MOF increased the photocurrent density at 1.23 V from 1.53 mA • cm −2 for Ti:Fe 2 O 3 and 0.59 mA • cm −2 for Fe 2 O 3 to 1.80 mA • cm −2 , while shifting the onset potential by 100 mV compared to Ti:Fe 2 O 3 .Additionally, the modification led to reduced charge transfer resistance at the electrode-electrolyte interface, as well as enhanced charge injection and separation efficiencies, ABPE and IPCE values, and donor density.The improved PEC performance of Co-MOF/Ti: Fe 2 O 3 can be attributed to the Ti doping and loading of Co-MOF cocatalyst, both of which reduce surface charge recombination and improve charge transfer and water oxidation kinetics.Furthermore, the polynuclear cluster nodes in Co-MOF enhance framework connectivity, ensuring MOF integrity during the PEC process.Consequently, the Co-MOF modified photoanode exhibited remarkable stability, with only a 5% decrease in photocurrent after 10 h of PEC, which is superior to most of the recently reported MOF-modified hematite photoanodes.This study provides valuable insights for the development of stable MOF materials for PEC water splitting applications.

FIGURE 4
FIGURE 4 Charge injection efficiency (A), charge separation efficiency (B), Nyquist plots (C), and Mott-Schottky plots (D) of the photoanodes.Inset in (C) is the equivalent electric circuit for impedance fitting.