Oxygen vacancy engineering in MXenes for sustainable electrochemical energy conversion and storage applications

Abstract

Ever-increasing global energy requirements and environmental pollution have directed major research focus on developing sustainable energy conversion and energy storage technologies. Ti₃C₂T_X MXenes are widely considered a potential electrode material for electrochemical hydrogen energy generation through water splitting and electrochemical energy storage supercapacitor applications. Herein, multifunctional Ti₃C₂T_X MXene-based nanocomposites with varying Eu₂O₃ concentrations were synthesized and systematically investigated for electro/photocatalytic water splitting and two-electrode supercapacitor properties. Ti₃C₂T_X MXenes exhibit high electrical conductivity, tunable surface functionalities and multivalent Ti oxidation states. Meanwhile, the incorporation of Eu₂O₃ supplements the electrochemical performance by altering the physio-chemical structure and overcoming the intersheet restacking and oxidative degradation issues of pristine Ti₃C₂T_X MXenes. More importantly, the interfacial charge transfer synergism between Ti₃C₂T_X MXenes and Eu₂O₃ creates oxygen vacancies that modulate the electronic structure of nanocomposites, aiding in the formation of abundant hydrogen evolution reaction (HER) and oxygen evolution reaction (OER) sites, photo-generated charge trapping centres and pseudocapacitive sites. The key findings of the present study showed that the Eu₂O₃/MXene nanocomposite with an optimum oxygen vacancy content exhibited excellent performance with a small overpotential of 63 mV and 169 mV and a high faradaic efficiency of 96.2% and 95.23% to drive the HER and OER, respectively. Additionally, upon combination with CdS as the photoabsorber, the optimized nanocomposite achieved a high photocurrent density of 4.86 mA cm⁻², leading to a H₂ evolution rate of 56.67 μL min⁻¹. Considering supercapacitor characteristics, the optimized nanocomposite exhibited a high specific capacitance of 374.98 F g⁻¹ with an energy density and power density of 13.02 Wh kg⁻¹ and 300 W kg⁻¹, respectively. Thus, the results of the present study establish a facile approach to develop high-performance multifunctional electrodes for advancing electrochemical energy conversion and storage technologies.

---

1. Introduction

Rapid population growth and industrial expansion have placed immense pressure on the overuse of fossil fuels as a primary source of energy production.¹ However, their exhausting reserves and subsequent environmental concerns related to the emission of harmful greenhouse gases pose significant threats to climate change and the health of living beings. These adverse effects, coupled with the severe energy shortage crisis, have intensified the exploration of alternative clean and sustainable sources for energy conversion and storage.²

In this context, electrochemical hydrogen (H₂) energy generation water splitting devices and electrochemical energy storage supercapacitors stand out as the pioneering pillars of sustainable energy future.³ In the context of electrochemical H₂ energy generation, the approach that has gained a rapid upsurge involves the splitting of water into H₂ and O₂ either via direct electric current (which is termed electrocatalysis) or using sunlight energy (which is termed photocatalysis).^4,5 It involves two half-cell reactions, viz. the hydrogen evolution reaction (HER) at the cathode and the oxygen evolution reaction (OER) at the anode, respectively, which control the reaction kinetics and H₂ production rate.⁶ However, practical implementation of this process is greatly hindered by the higher overpotential of existing electrocatalytic materials. Furthermore, limited light absorption, slow charge transport properties and easy electron–hole recombination hamper the photocatalytic water splitting process. In this context, researchers are actively investigating different potential materials and strategies by focusing on these rate-controlling parameters of electro/photocatalytic water splitting. Furthermore, on account of electrochemical energy storage devices, supercapacitors (SCs) have gained significant attention in recent few years due to their intriguing features of quick charge–discharge time profile, high power density, long cyclic stability, safe operation, and compact size.^7–9 In SCs, pseudocapacitive materials store nearly 100 times greater charge than electrochemical double layer capacitors (EDLCs) due to the fast redox reaction of the electrolyte with the bulk of the electrode, whereas EDLCs interacts only at the outer interface.^10,11 Typically, pseudocapacitive materials alleviate the energy density of SCs due to their wider potential window, but at the cost of sacrificing the power density and cyclability.^12,13 However, the widened potential window negatively impacts the overall specific capacitance due to the ongoing HER and OER water-splitting reactions during the charging–discharging processes in SCs.¹⁴ Therefore, as the core part of the two electrochemical technologies mentioned above, the material selection is crucial to support the potential window for practicing electro/photocatalytic water splitting and SC applications.

In this direction, researchers have focused on constructing efficient materials with multifunctional catalytic performance.^15,16 The conventionally used noble metal-based materials are limited in commercial electrochemical applications due to their high cost and long-term stability issues.^17,18 Considering this, MXenes, especially Ti₃C₂T_X MXenes with their notable properties of high electrical conductivity, strong hydrophilic character, high specific surface area, high electrocatalytic activity, and other properties, have become a widely investigated and leading two-dimensional electrode material in the field of electrochemical devices involving catalysis and storage technologies.^19,20 It exhibits a rich density of redox sites originating from the transitions of the Ti oxidation states upon reaction with the electrolyte and swift redox activity of the surface-attached functional groups.^21,22 Moreover, the high conductivity derived from its carbide core promotes the swift charge transfer kinetics. Nevertheless, the theoretically expected electrochemical parameters, i.e., overpotential of water splitting reactions and specific capacitance of SC have been rarely achieved in experimental analysis mainly due to the oxidative degradation and easy restacking issues of the Ti₃C₂T_X MXene in ambient environment conditions.^23,24 The self-restacking of the Ti₃C₂T_X MXene layers due to the strong van der Waals forces reduces the effective surface area of the Ti₃C₂T_X MXene, which suppresses the exposed redox active sites and retards the ion intercalation between the nanosheets.²⁵ This limits the potential window of the pristine Ti₃C₂T_X MXene-based electrodes, and thus degrades their catalytic performance and pseudo capacitance.^26–29 Therefore, to derive the high electrochemical performance from the Ti₃C₂T_X MXene, it has been widely proposed to hybridize Ti₃C₂T_X MXene with a promising pseudocapacitive material such as metal oxides, metal hydroxide, and conductive polymers, which modify the physio-chemical structure and induce the efficient interfacial charge transfer properties of Ti₃C₂T_X MXene for higher adsorption characteristics, swift electron transfer and separation properties.^30,31

Considering this, metal oxides have attracted much research interest as perspective dopants in Ti₃C₂T_X MXene for their potential use in water splitting and supercapacitors in response to their low cost, easy synthesis routes, environmental friendliness, corrosion resistance, enlarged active surface area and high theoretical capacitance.^24,32–34 Aside from the advantages of metal oxides, their intrinsically poor electrical conductivity results in sluggish ion transfer and hampers the power density and cycle life.^35–37 In this context, researchers have adopted different strategies to boost the conductivity, such as elemental doping, defect engineering, morphology modification, and heterostructure forming. Among them, the creation of oxygen vacancy (O_V) defects is a satisfactory strategy for enhancing the electrical conductivity.^38,39 In addition to improving the electrical conductivity, O_V offers other advantages in the following ways: (i) O_V promotes the adsorption of OH⁻ ions, boosting the otherwise sluggish OER and specific capacitance;^40,41 (ii) the delocalized electrons of O_V modulate the electronic environment of adjacent Ti active sites by reducing the prevalent Ti⁴⁺ in Ti₃C₂T_X MXene to Ti²⁺ for swift HER kinetics; (iii) O_V creates additional charge states below the CB, which help to narrow the optical band gap and trap photogenerated electrons, resulting in a significant effect on the photoconversion efficiency;^42,43 (iv) the positively charged core of O_V surrounded by negatively charged delocalized electrons creates an electric field, which accelerates the interfacial charge transfer characteristics and thus faradaic reactions.⁴⁴ However, excess O_V has a negative impact as recombination centres, hindering electron transfer processes and light absorption. The excess of O_V reduces the capacitance by decreasing the available surface area.^45,46 Thus, the precise control of O_V is crucial for achieving high performance in water splitting and supercapacitor applications.

Towards the controllable incorporation of O_V, the investigation of rare-earth metal oxides as an electrode material has gained much attention due to the rapid redox transitions between its mixed-valent oxidation states upon contact with a reducing surface, leading to the formation of O_V to maintain charge neutrality.⁴⁷ Also, these materials exhibit high packing density, which allows for their high volumetric capacitance. Among them, Eu₂O₃ stands out due to its unique shielded 4f electronic configuration, where the Eu³⁺/Eu²⁺ redox couple transitions promote the formation of active reaction sites. Furthermore, the compositing of Eu₂O₃ with the Ti₃C₂T_X MXene matrix would result in O_V formation, which further leads to superior electrochemical performance for electro/photocatalytic water splitting and supercapacitor devices.

Considering the above discussion, in the present work, Ti₃C₂T_X MXene/Eu₂O₃ nanocomposites were synthesised via a simple sonication method, and the concentration of O_V was controlled by varying the weight ratios of Eu₂O₃ (3%, 5%, and 7 wt%) in Ti₃C₂T_X MXene. The prepared nanocomposites were investigated for HER, OER, overall water splitting H₂ energy generation and supercapacitor energy storage device applications. The photocatalytic water splitting properties of these nanocomposites were realized using CdS as the photo absorber material, which demonstrated the efficient impact of the interfacial charge transfer mechanism between MXene and Eu₂O₃ on the electron/hole pair recombination properties, and thus the photoinduced H₂ production efficiency.

2. Experimental

2.1. Materials

Titanium aluminium carbide (Ti₃AlC₂) MAX powder was purchased from Nanoshel, UK. Europium oxide (Eu₂O₃) powder, hydrofluoric acid (HF, 40% concentration), sulphuric acid (H₂SO₄, 98% concentration), Nafion 117 solution, and ethanol were all obtained from Merck, India.

2.2. Synthesis of the MXene/Eu₂O₃ nanocomposites

Ti₃C₂T_X MXene was prepared using the top-down liquid exfoliation process, as described in our previous reports.⁴⁸ Eu₂O₃/Ti₃C₂T_X MXene nanocomposites were prepared using the following procedure: 100 mg of Ti₃C₂T_X MXene powder was dispersed in 20 mL of ethanol and sonicated for 30 min. Afterwards, 3 mg of Eu₂O₃ powder was added to the above dispersion, and the mixture was stirred for 30 min and later sonicated for another 30 min. After 1 h of reaction, the mixture was left to dry at 50 °C. The dried powder was collected and labelled as 3-EuM. The two other Eu₂O₃/Ti₃C₂T_X MXene nanocomposites were prepared with 5 mg and 7 mg of Eu₂O₃, and labelled as 5-EuM and 7-EuM, respectively.

2.3. Characterizations

X-ray diffraction (XRD) measurements were conducted on D8 FOCUS, Bruker Ettlingen with Cu-K_α radiation (λ = 1.54 Å) to collect structural information. The surface morphological visualization of the samples was obtained via field emission scanning electron microscopy (FESEM), Carl Zeiss, Supra 55, and transmission electron microscopy (TEM, JEOL, JEM-2100Plus). X-ray photoelectron spectroscopy data were collected to investigate the chemical composition via Mac-2 electron analyzer using the Mg K_α radiation source.

2.4. Electrocatalytic measurements

The electrocatalytic activity of the prepared samples was analyzed in a standard three-electrode configuration on an AUTOLAB Metrohm PGSTAT302 electrochemical workstation. Pt wire was used as a counter electrode, Ag/AgCl as the reference electrode, and a prepared sample slurry in ethanol deposited on glassy carbon (GC) electrode with an active area of 0.07 cm² was used as the working electrode. 1 M H₂SO₄ was used as an electrolyte solution throughout the electrochemical measurements. Cyclic voltammetry (CV) scans were conducted at a scan rate of 10 mV s⁻¹ before performing linear sweep voltammetry (LSV). All measured potentials in this work were converted to reversible hydrogen electrode (RHE), in accordance with the following Nernst eqn (1):

E_RHE (V) = E_Ag/AgCl (V) + 0.197 + 0.059pH (1)

LSV measurements were recorded from (0 to −0.4 V vs. RHE) and (1.7 to 1.0 V vs. RHE) for HER and OER, respectively, at a scan rate of 5 mV s⁻¹ to obtain the polarization curves. Electrochemical impedance spectroscopy (EIS) results were obtained in a frequency range of 0.1 Hz to 10⁵ Hz. CV studies were carried out at different scan rates (10 to 80 mV s⁻¹) to obtain data for the double layer capacitance (C_dl) calculation. The electrochemical surface area (ECSA) was derived from C_dl using the following relation (2):⁴⁹

ECSA (cm²) = C_dl/C_s (2)

where C_s refers to the electrode's specific capacitance, which is considered 60 μF cm⁻² in the present work.

Furthermore, the roughness factor (RF) was calculated using eqn (3) to estimate the catalytic active sites exposed on the surface.

The number of active sites (N) was calculated according to the following relation (4):

N (mol) = Q/2F (4)

where Q represents the charge in the electrochemical reaction and F refers to Faraday's constant.⁵⁰

2.5. Photocatalytic measurements

The photocatalytic experiment involved a similar three-electrode configuration in 0.1 M Na₂SO₃ and 0.1 M Na₂S electrolyte. An equal quantity of CdS powder was added to the prepared samples to prepare the photocatalysts. 10 mg of photocatalyst powder was added into a mixture of 15 μL ethanol and 7 μL Nafion solution. The solution was then sonicated for 30 min to form a homogenous slurry, which was drop-casted on a pre-cleaned FTO substrate of 0.5 cm² active area. The OAI, TriSOL solar simulator with an intensity of 100 mW cm⁻² was used as a light source to illuminate the samples and examine their H₂ production efficiency under light irradiation. The photocurrent vs. voltage LSV curves were obtained at a scan rate of 5 mV s⁻¹ to determine the photocurrent densities. The photoconversion efficiency (η) was calculated from the LSV curve data using the following relationship (5):

η (%) = I(1.23 − E_RHE)/I₀ (5)

where I is the photocurrent density, E_RHE is the measured potential, and I₀ is the incident light intensity at 100 mW cm⁻².⁵¹ Mott–Schottky plots were derived at a frequency of 100 Hz to obtain the band edge positions of the photocatalysts.

2.6. Supercapacitor measurements

Electrochemical measurements for supercapacitor device applications were performed using the two-electrode configuration on the AUTOLAB Metrohm PGSTAT302 electrochemical workstation. At first, a thick slurry was prepared by mixing the electroactive material and carbon black in an 85 : 15 ratio, using absolute ethanol as the solvent, to fabricate the working electrodes. Thereafter, the slurry was doctor bladed onto the aluminum substrates (3 cm × 1.5 cm) and dried for 2 h in an oven. The aluminium substrates were weighed before and after loading the electroactive material to measure the mass loading, which was ∼6 mg. Finally, the two as-prepared identical working electrodes were sandwiched separated with a piece of cellulose filter paper dipped in 1 M H₂SO₄ (acting as a separator) to obtain the symmetric supercapacitor. Furthermore, the fabricated symmetric supercapacitor was tested using electrochemical techniques, such as cyclic voltammetry (CV), galvanostatic charge–discharge (GCD) and electrochemical impedance spectroscopy (EIS). The specific capacitance of the electrode (C_s,e) and cell (C_s,cell) from CV and GCD curves were evaluated using the following eqn (6)–(9):

Furthermore, the specific energy (E_cell) and specific power (P_cell) of the cell were evaluated using the calculated value of specific capacitance obtained from the GCD curves, using eqn (10) and (11):

where, A_s is the area under the CV curve, K is the scan-rate (mV s⁻¹), m_t is the total mass loading of the cell (g), ΔV is the potential window (V), I_m is the current-density (A g⁻¹), and Δt is the discharging time (s).⁵²

The coulombic efficiency (η%) was calculated using the following relation:⁵³

where, t_D is the discharging time and t_C is the charging time in seconds.

2.7. DFT studies

First-principles DFT calculations were performed using the Quantum Espresso package.^54,55 The Perdew–Burke–Ernzerhof (PBE) approach within the generalized-gradient approximation (GGA) framework was used to describe the exchange–correlation function.⁵⁶ Spin-restricted plane-wave self-consistent field (PWscf) calculations were performed using ultra-soft pseudopotentials. The cutoff values for the energy and charge density were set equal to 35 Ry and 350 Ry, respectively, and a 10⁻⁶ Ry convergence threshold was used.⁵⁷ To study the MXene system, a slab model of size 3 × 3 × 1 was constructed, which has 63 atoms, and a vacuum of 15 Å was added along the Z direction to avoid the interaction between the adjacent MXene layers.⁵⁸ Also, to construct the Eu₂O₃/MXene system, a molecule of Eu₂O₃ was added above the MXene layer, which is 3 Å away from the layer. The van der Waals force of interaction between MXene and Eu₂O₃ was taken into account with the help of the Grimme vdW correction (DFT+D3) scheme.⁵⁹ A Monkhorst–Pack k-point grid of size 3 × 3 × 1 was employed to integrate the Brillouin zone.⁶⁰ The structural optimizations were carried out by utilizing the Broyden–Fletcher–Goldfarb–Shanno (BFGS) method, and the structure was relaxed to obtain a force (acting on each atom) below the value of 10⁻³ Ry Bohr⁻¹.^61,62 Furthermore, to realize the Eu₂O₃/MXene system with an oxygen vacancy, one of the oxygen atoms was removed from the MXene layer, and the optimization was then performed.

To study the hydrogen adsorption, an H atom was added to the systems, namely MXene, and Eu₂O₃/MXene with an oxygen vacancy (Eu₂O₃/MXene-O_V), and then DFT calculations (structural optimizations) were performed. Also, the Grimme vdW correction (DFT+D3) scheme was used to consider the van der Waals interaction between H and the adsorbent system. The hydrogen adsorption energy (E_ads) of the adsorbent system has been calculated using the formula:

where E_{Adsorbent with H adsorbed} is the total energy of the system with hydrogen adsorbed on it, E_Adsorbent is the total energy of the system, and E_H2 is the total energy of a hydrogen gas molecule.^58,63

3. Results and discussion

The X-ray diffraction (XRD) results of Ti₃C₂T_X MXene, 3-EuM, 5-EuM, and 7-EuM are depicted in Fig. 1a. The XRD spectrum of Ti₃C₂T_X MXene shows the typical diffraction peaks located at 2θ = 9.09°, 18.46°, 28.5°, 36.2°, 41.87°, 60.52°, and 72.36°, which are well indexed to the (002), (004), (006), (101), (105), (110), and (311) crystallographic planes of Ti₃C₂T_X MXene, respectively (JCPDS 32-1383).^64,65 Furthermore, the diffractograms of all EuM nanocomposites exhibited identical Ti₃C₂T_X MXene diffraction peaks, along with the characteristic XRD peaks of Eu₂O₃ positioned at 2θ = 18.46°, 28.5°, 32.7°, and 46.5° assigned to the (211), (222), (400), and (440) planes (JCPDS 00-043-1008), respectively.⁶⁶ Notably, the (211) and (222) XRD planes of Eu₂O₃ coincide with the (004) and (006) planes of Ti₃C₂T_X MXene. In addition, the intensity of the Ti₃C₂T_X MXene diffraction peaks was found to decrease with a rise in the Eu₂O₃ doping concentration, which indicates the degeneration of the crystallinity.⁶⁷ Moreover, it was noticed that the (002) and (004) XRD peaks of Ti₃C₂T_X in the composite samples were shifted to lower 2θ angles at 8.98° and 18.28° in 3-EuM and 8.89° and 18.18° in 5-EuM, suggesting an interplanar lattice expansion of 0.01 nm due to the compositing of Eu₂O₃ with Ti₃C₂T_X MXene.⁴⁸ However, in 7-EuM, a slight rise in 2θ was observed for the (002) and (004) XRD peaks to 9.02 and 18.35, respectively, due to the agglomeration effect of Eu₂O₃ upon excessive doping between the MXene layers.

Fig. 1 Structural and morphological analysis of EuM nanocomposites: (a) comparative XRD patterns and (b–d) FESEM images of 3-EuM, 5-EuM, and 7-EuM; (e) TEM image, (f) HRTEM image with inset showing corresponding FFT patterns, and (g) SAED pattern for the 5-EuM sample.

FESEM images were obtained to visualize the morphological features of the EuM nanocomposites, and the results are presented in Fig. 1b–d. The 3-EuM sample (Fig. 1b) exhibits a few Eu₂O₃ nanoparticles sparsely dispersed on the lamellar Ti₃C₂T_X MXene sheet structure. With an increase in the Eu₂O₃ doping content in 5-EuM (Fig. 1c), more nanoparticles are seen growing over the Ti₃C₂T_X MXene surface, which effectively prevents the restacking of the Ti₃C₂T_X MXene sheets for the higher electrochemically accessible specific surface area and electron diffusion efficiency. Importantly, the Ti₃C₂T_X MXene sheets prevented the aggregation of the Eu₂O₃ nanoparticles due to their layered structure, thus maximizing the nanoparticle potential merits. However, with the further increase in the doping content of Eu₂O₃ in 7-EuM (Fig. 1d), the lamellar structure of Ti₃C₂T_X MXene gets distorted, leading to the formation of a thicker MXene sheet structure with the collapse of the effective pore channels, causing the severe blockage of the potential redox active sites. It reduces the specific surface area, which is considered to negatively impact the electrochemical application prospects.

Furthermore, to examine the elemental composition and uniform distribution of elements throughout the materials, energy-dispersive X-ray spectroscopy (EDS) measurements were conducted and elemental mapping was obtained. The EDS spectra (Fig. S1a–c†) indicate the presence of Ti, O, C, F, and Eu elements in the EuM nanocomposites, where the content of the Eu element was found to increase in the 3-EuM, 5-EuM and 7-EuM samples due to the increasing doping concentration of Eu₂O₃. This is evident from their respective elemental distribution table, which demonstrates the wt% and atomic% concentration of each element in the samples. Furthermore, the EDS elemental mapping shows the distribution pattern of elements in the nanocomposites, which reveal the uniform dispersion of the Eu₂O₃ nanoparticles on the MXene surface, thus confirming the formation of EuM nanocomposites. A more detailed microstructural observation of optimized 5-EuM was noted using TEM investigations. Consistent with the FESEM results, the TEM image (Fig. 1e) shows Eu₂O₃ nanoparticle growth on the Ti₃C₂T_X MXene sheets morphology. The corresponding high-resolution TEM (HRTEM) micrograph (Fig. 1f) demonstrates the construction of numerous heterointerfaces between Ti₃C₂T_X MXene and Eu₂O₃, which depicts the distinctly visible parallel ridges. Fast Fourier Transform (FFT) studies carried out on the ridges revealed an interlayer spacing of 0.31 nm and 0.28 nm (insets: Fig. 1f), matching well with the (222) and (101) planes of Eu₂O₃ and Ti₃C₂T_X MXene, respectively.⁶⁸ The slightly enlarged lattice spacing of Ti₃C₂T_X MXene in the nanocomposite, relative to the pristine Ti₃C₂T_X MXene of 0.27 nm, is attributed to the increased intersheet distance upon insertion of the Eu₂O₃ nanoparticles.⁶⁹ Furthermore, the selected area diffraction (SAED) pattern of the 5-EuM nanocomposite (Fig. 1g) represents a combination of bright diffraction spots and circular rings assigned to the single crystalline Ti₃C₂T_X MXene structure and polycrystalline Eu₂O₃ nanoparticles, respectively.^48,70

XPS was carried out to determine the oxidation states of the nanocomposites upon interfacial interactions between the doped Eu₂O₃ and Ti₃C₂T_X MXene surface. The XPS survey spectra (Fig. S2†) of the EuM composites depict the presence of Ti, O, C, F, and Eu with increasing intensity of Eu 4d from the 3-EuM sample to the 7-EuM sample, confirming the formation of the composite with no detectable impurity peaks. The Ti 2p core-level XPS spectra (Fig. 2a–c) display two spin–orbital doublets, viz., Ti 2p_3/2 and Ti 2p_1/2 at 459.1 eV and 464.9 eV, respectively. Further deconvolution of the Ti 2p spectra provides detailed oxidation states, i.e., Ti–C (454.8 eV), Ti²⁺ (456.5 eV), Ti³⁺ (458.5 eV), and Ti⁴⁺ (459.1 eV).⁷¹ The atomic (at) ratios of these species in the nanocomposites have been calculated from their relative peak area and are represented in Table 1. It is seen that with an increase in concentration from 3-EuM to 5-EuM, Ti²⁺/Ti⁴⁺ increases from 0.9 to 4.9, suggesting that the effective reduction of the Ti species of Ti₃C₂T_X MXene is related to the O_V creation. Specifically, the formation of O_V creates two free electrons, which reduces the Ti⁴⁺ states to Ti²⁺ states.⁷² However, with the further increase in concentration in 7-EuM, Ti²⁺/Ti⁴⁺ exhibits a decrease, which could be attributed to the agglomeration impact of the excessive concentration, resulting in the interfacial charge transfer bottleneck.⁷³ In addition, a similar trend was noticed upon analyzing the detailed Eu 4d spectra (Fig. 2d–f), in which peaks at B.E. of 128.82 eV and 137.21 eV could be assigned to Eu²⁺ and Eu³⁺.⁷⁴ A comparison of the Eu 4d spectra revealed that the Eu²⁺/Eu³⁺ content increased from 0.08 in the 3-EuM sample to 0.14 in the 5-EuM sample, indicating that a portion of Eu³⁺ was transformed into Eu²⁺ states, and consequently created O_V to maintain the charge neutrality. Furthermore, the high-resolution O 1s XPS spectra (Fig. 2g–i) were probed to estimate the O_V contents. It depicts three distinct peaks positioned at 530.3 eV, 531.8 eV, and 533.2 eV associated with lattice oxygen (O_L), surface oxygen vacancies (O_V), and surface adsorbed water H₂O_ads, respectively.⁷⁵ As tabulated in Table 1, the O_V concentration increased to 2.61 in the 5-EuM sample, which is indicative of the increased O_V content, while a sharp rise was seen in the 7-EuM sample, which is ascribed to the excessive generation of oxygen vacancies upon higher doping levels.

Fig. 2 High-resolution XPS spectra of (a–c) Ti 2p, (d–f) Eu 4d, and (g–i) O 1s for the 3-EuM, 5-EuM, and 7-EuM nanocomposites.

Table 1 Atomic percentages (at%) of the valence states for the 3-EuM, 5-EuM, and 7-EuM nanocomposites

Sample	Ti^2+/Ti^4+	Eu^2+/Eu^3+	OV (%)	H2Oads
3-EuM	0.91	0.08	27.90	25.83
5-EuM	4.91	0.14	41.42	42.69
7-EuM	1.8	0.11	42.89	30.60

---

Thus, it is inferred that the precursor concentration greatly influences the structural, morphological, and composition of the synthesized samples. The structural and morphological analyses definitively confirmed the anchoring of nanoparticles on the Ti₃C₂T_X MXene sheets during ultrasonic treatment. The XPS results confirmed the Eu₂O₃ loading and revealed the at% of the Ti, Eu species and oxygen moieties. The overall analyses show that among the prepared samples, 5-EuM exhibits greater potential in electrochemical energy conversion and storage applications, owing to its structural suitability, larger hydrophilicity, large density of Ti²⁺ and Eu²⁺ as the potential HER active sites, optimal O_V content which would ease otherwise sluggish OER pathway, and multivalent oxidation states with greater OH⁻ ion adsorption ability favoring charge storage ability.

Considering the above discussion, the electrocatalytic HER performance of EuM nanocomposites was first investigated from the cathodic polarization curves (Fig. 3a) obtained by conducting linear sweep voltammetry (LSV) measurements in 1 M H₂SO₄ electrolyte. For comparison, a state-of-the-art HER catalyst, Pt/C, was also tested among the prepared catalysts. In the HER LSV polarization study, Pt/C clearly exhibited excellent performance in HER activity, requiring a minimal overpotential (η) of 47 mV to attain 10 mA cm⁻². Among the prepared synthesized materials, it has been observed that the 5-EuM sample exhibits near Pt activity, presenting a lower overpotential of 63 mV, and was found to be catalytically more active than the 3-EuM and 7-EuM samples (Table 2). The excessive generation of O_V in 7-EuM negatively impacts the electroactivity by reducing the electrical conductivity and imposing structural instability.⁷⁶ This is because the higher O_V content creates a larger defect that localizes a high electron density around the defect site, thereby hindering the material's charge transfer ability and increasing the overpotentials.⁷⁷ Removing the excessive lattice oxygen atoms distorts the crystal lattice, which affects the active site and adsorption/desorption process. More importantly, the highly defective structure is prone to corrosion under electrochemical testing conditions, thereby accelerating the degradation process.⁷⁵ Therefore, the performance of the 7-EuM sample with the highest oxygen vacancy concentration was lower as compared to that for 5-EuM. The superior electrochemical performance of 5-EuM suggests that the controlled generation of O_V played a vital role in enhancing the HER catalytic activity. To further obtain insight into the reaction kinetics and understand the underlying reaction mechanism on the catalyst surface, the Tafel plot was extracted from the corresponding LSV curves. The reaction catalyzed by Pt/C, 3-EuM, 5-EuM, and 7-EuM showed Tafel slopes (Fig. 3b) of 45.0 mV dec⁻¹, 134.1 mV dec⁻¹, 65.4 mV dec⁻¹, and 125.1 mV dec⁻¹, respectively. Aside from the superior catalytic activity, the stability of an electrocatalyst in its electrolyte environment is also crucial to ensure its long-term operation in water electrolysis for commercial-scale H₂ production. Thus, to probe the stability of the 5-EuM electrocatalyst, chronoamperometric (CP) electrochemical testing (Fig. 3c) was conducted at an overpotential of 63 mV for a continuous 50 h. It revealed that 5-EuM retained a nearly steady current, revealing its outstanding durability. A slight catalytic decrement with time can be ascribed to the bubble accumulation and release from the electrode surface, which obstructs the effective electrolyte interaction with the electrode surface. This electrochemical stability is further confirmed from the negligible change in the overpotential observed in the LSV measurement performed before and after conducting 100 CV cycles (inset: Fig. 3c), signifying no obvious catalyst degradation.

Fig. 3 Comparison of the electrochemical performance of the as-synthesized EuM nanocomposites for the HER and OER in 1.0 M H₂SO₄: (a) LSV polarization curves, (b) corresponding Tafel slopes for Pt/C, 3-EuM, 5-EuM, and 7-EuM; (c) chronoamperometry (i–t curve) stability test for 5-EuM for 50 h; (d) LSV polarization curves and (e) corresponding Tafel slopes of RuO₂, 3-EuM, 5-EuM, and 7-EuM; (f) chronoamperometry (i–t curve) stability test for 5-EuM for 50 h.

Table 2 Electrochemical HER and OER parameters for the 3-EuM, 5-EuM, and 7-EuM nanocomposites

Electrocatalyst	HER		OER
	η (mV)	Tafel slope (mV dec^−1)	η (mV)	Tafel slope (mV dec^−1)
3-EuM	205	134.1	237	184.0
5-EuM	63	65.4	169	92.0
7-EuM	155	125.1	173	140.0

---

It is well known that electrocatalysis is highly pH-dependent.⁷⁸ Thus, to compare the electrocatalytic water splitting kinetics of the 5-EuM sample in acidic, alkaline, and neutral solutions, LSV measurements were conducted in 1 M H₂SO₄ acidic, 1 M KOH alkaline, and 1 M PBS neutral media (Fig. S3†). In comparison to acidic electrocatalysis, relatively deteriorated HER performance is seen in alkaline and neutral solutions with higher overpotentials of 81.2 mV and 141.9 mV, respectively, indicating higher HER kinetics in acidic media. This is primarily attributed to the higher H⁺ ion concentration in acidic media (H₂SO₄ and H₂O), which are involved in the electro-reduction processes towards H₂ generation, whereas the limited H⁺ availability in alkaline and neutral electrolytes (derived from H₂O only) restricts the electrocatalysis kinetics.⁷⁹

The OER activity of the prepared samples was also evaluated in 1 M H₂SO₄. RuO₂, a benchmarking OER catalyst, was also investigated for its water oxidation ability. The OER LSV polarization curves in Fig. 3d show that 5-EuM requires an η of 169 mV, which is even lower than that required for the RuO₂ catalyst (η = 220 mV), demonstrating the superior activity of the EuM composites, especially 5-EuM in the oxidation reaction. The derived Tafel slopes (Fig. 3e) for RuO₂, 3-EuM, 5-EuM, and 7-EuM are found to be 184.0 mV dec⁻¹, 92.0 mV dec⁻¹, and 140.0 mV dec⁻¹, respectively. The smaller η and Tafel slope values of 5-EuM as the OER catalyst are in accordance with the HER results, which signify the strong redox properties of the 5-EuM sample towards both H₂ and O₂ evolutions. This remarkable performance can be ascribed to the presence of the optimal O_V content in the 5-EuM sample calculated from the XPS results, which significantly promotes the rate of charge transfer even at lower applied η values. Likewise, CP tests conducted at 169 mV OER overpotential (Fig. 3f) and long-term cycling stability tests (inset: Fig. 3f) further demonstrated the stable OER kinetics over a long duration of 50 h.

In addition to the long-term stability measurements, post-catalytic structural and morphological characterizations were obtained for the optimized 5-EuM sample to validate the electrode durability. The post-stability PXRD pattern of the 5-EuM sample (Fig. S4a†) indicated no significant structural alterations, demonstrating a diffraction pattern that was similar to that obtained before electrochemical testing. Likewise, the SEM image (Fig. S4b†) indicated the dispersion of the Eu₂O₃ nanoparticles while preserving the layered morphology of MXene, further revealing the high stability of the sample after the continuous stability test. This could be attributed to the high corrosion resistance of the dispersed Eu₂O₃ rare-earth oxide, which encapsulated the MXene surface in aqueous solution and protected it from oxidative degradation during acidic electrolysis.

Electrochemical double layer capacitance (C_dl) and electrochemical surface area (ECSA) were obtained to further analyze the intrinsic electrochemical behaviour of the electrocatalysts. ECSA is directly related to C_dl, which is measured by conducting CV in a non-faradaic potential range at different scan rates from 10 to 80 mV s⁻¹, as shown in Fig. 4a–c. The linear fitting of the scan rate vs. current density plot (Fig. 4d) gives the C_dl values (Table 3). The C_dl of the 5-EuM sample (11.35 mF) is much larger than that of the other tested electrodes, indicating the presence of a rich density of active sites at the electrode/electrolyte interface, which is responsible for ion adsorption and interaction. Consequently, the ECSA calculated from the corresponding C_dl values gives the largest electrochemically accessible surface area for the 5-EuM sample (189.1 cm²), validating a greater number of adsorbed species on the catalyst surface. The number of active sites (N) were calculated from the C_dl values to estimate the density of the surface exposed redox active sites. Clearly, 5-EuM gives the highest concentration of catalytically active sites (11.7 × 10⁻⁸ mol), evidencing its high HER performance. The roughness factor (RF) was evaluated further to compare the HER performance of the prepared sample. The 5-EuM sample revealed the highest RF (2701.4), again confirming the presence of abundant actives sites, which is beneficial for the higher H₂ evolution.

Fig. 4 (a–c) CV curves at different scan rates (10–80 mV s⁻¹), (d) C_dl plots, and (e) Nyquist plots fitted using the equivalent circuit for the 3-EuM, 5-EuM, and 7-EuM nanocomposites. (f) LSV curve of the overall electrochemical water splitting for 5-EuM before and after the cycling test for a two-electrode setup.

Table 3 C _dl, ECSA, RF, N and R_ct values for 3-EuM, 5-EuM, and 7-EuM nanocomposites

Electrocatalyst	C dl (mF)	ECSA (cm^2)	RF	N (10^−8 mol)	R ct (Ω)
3-EuM	3.78	63.0	900	3.91	45.91
5-EuM	11.35	189.1	2701.4	11.7	7.13
7-EuM	7.66	127.6	1822.85	7.93	27.81

---

The conductivity of an electrocatalyst is an important parameter to evaluate the electroactivity. Thus, EIS measurements were carried out to determine the resistance offered to the flow of charges during an electrochemical reaction. The Nyquist plots (Fig. 4e) obtained after fitting the EIS data to an equivalent circuit (inset: Fig. 4e) present a semicircle with a diameter corresponding to the internal charge transfer resistance (R_ct), revealing that 5-EuM delivers an obvious smallest R_ct of 7.13 Ω in comparison to the 3-EuM and 7-EuM samples. This signifies the improved scope of the electrode/electrolyte interfacial contact for 5-EuM, owing to the synergistic effects of the electron-rich O_V, Eu₂O₃ nanospacers, and highly conductive Ti₃C₂T_X MXene support, which are responsible for its enhanced performance.

Finally, 5-EuM was assembled as an anode and cathode in an overall water-splitting electrolyzer setup. As shown in the polarization curves (Fig. 4f), the 5-EuM catalyst delivered a current density of 10 mA cm⁻² at a small cell voltage of 1.48 V, revealing the remarkable bifunctionality of the prepared catalyst. This could be ascribed to the presence of O_V in a controlled manner, which not only modulates the electronic environment of the neighbouring Ti valence states of Ti₃C₂T_X MXene for generating highly active Ti²⁺ HER sites, but also that they act as adoption centers for OER intermediates and enhance the charge transfer kinetics. The water-splitting polarization curves before and after the repeated CV scans have been recorded (Fig. 4f). The results indicate the outstanding cycling stability, as observed from the negligible cell voltage drift recorded after continuous 100 CV cycles.

Finally, the faradaic efficiencies for HER and OER of 5-EuM were evaluated using the experimentally quantified volumes of H₂ and O₂ obtained at the cathode and anode, respectively, in a typical water-gas displacement process conducted at a constant overpotential in chronoamperometric measurement and theoretically calculated volumes. After 3600 s, a total of 8.96 C of the charge passing through the electrolyzer cell generated 2 mL and 0.99 mL of H₂ and O₂, which agrees well with the theoretically estimated volume of 2.079 μL and 1.040 μL for H₂ and O₂, respectively. This gives the faradaic efficiency of 96.2% and 95.23% for HER and OER, respectively.

Furthermore, the photocatalytic water-splitting performance of the synthesized composites was investigated by combining the nanocomposites with a potential photoabsorber, CdS. LSV was first carried out under artificial sunlight illumination. Then, the current density vs. voltage (I–V) curves were obtained to compare the current density change with the voltage upon light illumination. Fig. 5a demonstrates that the photocurrent density is altered significantly upon varying the doping concentration. The maximum photocurrent density for 3-EuM/CdS, 5-EuM/CdS, and 7-EuM/CdS at 1.0 V vs. RHE reached 0.3 mA cm⁻², 4.86 mA cm⁻², and 2.07 mA cm⁻², respectively. The remarkable increment of the photo response in 5-EuM/CdS is attributable to the promoted charge separation efficiency, which generated a greater number of e⁻/h⁺ pairs with high charge transfer ability. Beyond this doping concentration in 7-EuM/CdS, the decrease in the current density indicates the diminished charge carrier density, which could be directly correlated with a greater e⁻/h⁺ pair annihilation rate over charge separation. This is because the higher O_V concentration in 7-EuM/CdS acts as photogenerated charge recombination centers and reduces the charge mobility, which negatively effects the photocatalytic performance. Furthermore, the photocurrent efficiency was calculated from the corresponding photocurrent spectra to quantitatively estimate the photocatalytic H₂ generation efficiency of the nanocomposites. Fig. 5b shows that 5-EuM/CdS exhibits η% of 0.034%, which is higher than that for 3-EuM/CdS (0.002%) and 7-EuM/CdS (0.02%). Thus, consistent with the LSV results, the η% analysis also reveals that the 5-EuM sample should be considered optimal for achieving high photoelectrochemical activity, which is due to the desired physio/chemical structure modulation induced by O_V formation. The higher concentration of Eu₂O₃ reduced the photoactivity, which could be related to the creation of excessive O_V that act as charge recombination centers. Thus, it is concluded that when the O_V concentration is too low, it becomes impossible for electrons to reach the electrode surface and participate in surface chemical reactions, whereas an excessive O_V concentration increases the charge recombination and thus degrades the photocatalytic activity. Therefore, the optimal O_V concentration is crucial for promoting the photoexcited charge carrier separation and transfer properties.

Fig. 5 (a) LSV polarization curves, (b) photoconversion efficiency, and (c) Nyquist plots. (d) UV-vis absorption spectra, (e) Tauc plots, and (f) PL spectra for the 3-EuM/CdS, 5-EuM/CdS, and 7-EuM/CdS photocatalysts.

Additionally, EIS was conducted to analyze the charge transfer kinetics at the photoanode/electrolyte interface and the corresponding Nyquist plots are shown in Fig. 5c. Among the prepared photocatalysts, 5-EuM/CdS exhibited the smallest semi-circular arc radius of 16.37 Ω when compared to 3-EuM/CdS (41.89 Ω) and 7-EuM/CdS (21.87 Ω). This signifies that the incorporation of Eu₂O₃ in the controlled content and conductive Ti₃C₂T_X MXene support greatly diminished the resistance towards the charge transport.

Importantly, the use of Pt wire as the counter electrode is very challenging, especially for HER activity since Pt itself is a catalyst. Therefore, to explore the efficiency of the prepared catalyst towards electrochemical and photoelectrochemical testing, the measurements were collected using a graphite rod as the counter electrode under similar experimental conditions.

The HER polarization curves for electrocatalytic water splitting (Fig. S5a†) and photocatalytic water splitting (Fig. S5b†) demonstrated no significant shift in the overpotential in electrocatalytic water splitting. A nearly similar current density in photocatalytic water splitting was attained while using a graphite rod compared to the Pt counter electrode, indicating that the HER contribution is mainly derived from the intrinsic catalytic activity of the prepared Eu₂O₃/MXene sample. This could be attributed to the high conductivity of the 5-EuM sample derived from the electron-rich oxygen vacancies, compensating for the otherwise deteriorated conductivity upon using a graphite rod instead of the Pt counter electrode.

To explain the significantly enhanced photocatalytic properties of the 5-EuM/CdS sample, the light absorption behaviour and charge separation efficiency of all samples were comparatively investigated using UV-vis reflectance spectroscopy and PL studies, respectively. Fig. 5d presents the UV-vis absorption spectra of the 3-EuM/CdS, 5-EuM/CdS, and 7-EuM/CdS photocatalysts. Notably, the absorption band edge of 5-EuM/CdS is greatly shifted to the higher wavelength region, which suggests the enlarged light absorption capability. The increase of the visible range absorbance of 5-EuM/CdS is ascribed to the formation of the intra-band states of the O_V defects below the conduction band, which alter the band structure to expand the light absorption response. Also, the presence of delocalised free electrons in O_V creates electron charge clouds, which absorb incident light energy and undergo intra-band excitation processes. Nevertheless, the excess O_V concentration disrupts the band structure, where the excessive electrons accumulated in the forbidden band hinder the light absorption. Considering this, the band gaps of the nanocomposite samples were estimated from the Tauc plots (Fig. 5e) obtained by using Kubelka–Munk theory on the corresponding DRS spectra. Upon extrapolation of the linear portion of Tauc plot to the horizontal axis, the band gap values are noted as follows, i.e., 2.30 eV, 2.05 eV, and 2.10 eV for 3-EuM/CdS, 5-EuM/CdS and 7-EuM/CdS, respectively. Furthermore, the PL spectra of the doped samples were obtained at 520 nm to study the recombination rate of the photo-generated e⁻/h⁺ pairs and to better understand the role of the underlying O_V formation on the photocatalytic properties of the samples. As illustrated in Fig. 5f, all of the samples exhibited an emission peak around 687 nm, and the PL intensity was found to decrease from the 3-EuM sample to the 5-EuM sample, revealing the greatly diminished charge recombination. This indicates the positive aspect of O_V in the effective charge separation and transfer characteristics of the conductive Ti₃C₂T_X MXene support, which elongates the charge carrier lifetime. With the further rise in the Eu₂O₃ concentration in the 7-EuM sample, the PL intensity greatly increased due to the swift recombination rate of the e⁻/h⁺ pairs, showing the negative impact of excessive O_V, which behave as recombination centers. The combined UV and PL analysis confirms that the creation of O_V in a controlled concentration does not significantly increase the light harvesting efficiency, but suppresses the charge carrier recombination by trapping photogenerated e⁻/h⁺ pairs, improving the interfacial charge transfer properties, and thus enhancing the photocatalytic efficiency.

Furthermore, the photocatalytic H₂ evolution plots (Fig. 6) were obtained to confirm the superiority of 5-EuM/CdS. Fig. 6a and b shows the maximal photocatalytic efficiency of 3.4 mL in 60 min over the 5-EuM/CdS photocatalyst, which shows a linear trend over the time course. As a result, 5-EuM/CdS exhibited the highest average H₂ evolution rate of 56.67 μL min⁻¹, which presents remarkably improved photocatalytic performance compared to 3-EuM/CdS (38.34 μL min⁻¹) and 7-EuM/CdS (45 μL min⁻¹) shown in Fig. 6c. Thus, it is evidenced that the improvement in the photocatalytic water splitting for 5-EuM/CdS primarily originates from the optimal O_V concentration by the controlled Eu₂O₃ doping content, which efficiently alleviated the charge recombination processes. In addition, the photocatalytic durability of 5-EuM/CdS was assessed under similar reaction conditions to explore the viability for long-term photocatalysis application. The results from Fig. 6d showed that 5-EuM/CdS, when recycled three times over 60 min, exhibited a nearly stable H₂ generation rate, indicating its high operational stability.

Fig. 6 (a) Photocatalytic H₂ evolution curves over a time course, (b) comparison of the photocatalytic activity, and (c) average H₂ evolution per min for 3-EuM/CdS, 5-EuM/CdS, and 7-EuM/CdS. (d) Stability test for 5-EuM/CdS cycled three times.

Based on the above discussed results, the plausible charge transfer and redox reaction mechanism for the enhanced photocatalytic performance can be proposed. Firstly, the band gap energies (E_g) for CdS and the Eu₂O₃ components are estimated to be 2.36 eV and 4.23 eV, respectively, using UV-vis (Fig. S6a and b†) and the corresponding Tauc plots (inset: Fig. S6a and b†). Furthermore, Mott–Schottky plots were obtained to determine the energy band structure of the semiconductor photocatalyst. The positive slope of the Mott–Schottky plots (Fig. S6c and d†) confirms that both CdS and Eu₂O₃ are n-type semiconductors. The extrapolation of the slope of Mott–Schottky plots to the x-axis gives a flat band potential (V_fb) of −0.29 and −0.19 V vs. RHE for CdS and Eu₂O₃, respectively. E_RHE can be further converted into a vacuum energy level (E_vac) using eqn (14):

E_vac = −4.5 − E_RHE (14)

Generally, in n-type semiconductors, the conduction band (CB) lies at 0.2 eV above V_fb. Meanwhile, in p-type semiconductors, the valence band (VB) is positioned at 0.2 eV below the V_fb. Thus, considering the band gap values, the CB/VB positions for CdS and Eu₂O₃ are −4.01/−6.42 eV and −4.11/−8.41 eV, respectively. Considering these derived band edge positions of CdS and Eu₂O₃, the energy level structure with Ti₃C₂T_X MXene as a co-catalyst is shown in the schematic (Fig. 7). Upon light irradiation, both CdS and Eu₂O₃ components are simultaneously excited by absorbing light of energy greater than their respective band gap values. In CdS, the electrons are excited from their valence bands and migrate quickly to fill the CB, leaving behind holes in its VB and creating an electron/hole pair. Meanwhile, as confirmed from XPS analysis, the presence of O_V in Eu₂O₃ forms a defect energy level just below its CB minima edge, lowering the band gap for broader light absorption capability.⁸⁰ Moreover, these O_V act as trapping centers for photoexcited electrons from CB and prevent them from recombining with holes in VB. This process greatly minimizes the electron/hole recombination and facilitates their effective separation in bulk Eu₂O₃.⁴² Furthermore, the close intimate contact between the two semiconductors causes the electrons in the CB of Eu₂O₃ to recombine with the holes present in the VB of CdS via a Z-scheme charge transfer route. Consequently, the photogenerated electrons in CB of CdS and photogenerated holes in VB of Eu₂O₃ are spatially separated on their respective band energy levels with strong reduction and oxidation ability, respectively. Typically, the local electric field generated across the positive core and negatively charged delocalized electron cloud of O_V accelerates the swift charge transfer for effective interfacial charge transfer. Furthermore, when CdS and Ti₃C₂T_X MXenes are in contact with each other, the difference in their E_f position causes the electrons to flow from the CB of CdS to Ti₃C₂T_X MXenes until an equilibrium state of E_f of both CdS and Ti₃C₂T_X MXenes is reached. This raises the E_f of Ti₃C₂T_X MXene to a suitable reduction potential, i.e., more negative than 0 V vs. NHE. After E_f alignment, a space charge layer is created on the CdS side, which causes the upward band bending of the CdS energy levels. This leads to the creation of a Schottky barrier at the metal–semiconductor interface region, which prevents the reflux of electrons, thus promoting the unidirectional electron flow to MXene. Thus, Ti₃C₂T_X MXene now acts as an electron sink, where these electrons effectively reduce the Ti valence states for higher H⁺ ion adsorption and release of molecular H₂. Meanwhile, holes accumulated on VB of Eu₂O₃ exhibits high oxidisability due to its VB position at the more positive potential than the standard water oxidation potential, i.e., 1.23 V vs. NHE. Thus, the holes effectively oxidise the water molecules to generate O₂. It is revealed that interactions among CdS, Eu₂O₃, and Ti₃C₂T_X MXene could effectively promote the interfacial charge separation and boost the charge transfer efficiency of the photogenerated electron/hole pairs.

Fig. 7 Schematic illustrating the photocatalytic water splitting mechanism.

To study the energy storage supercapacitor device applications, electrochemical analysis including CV, GCD, and EIS measurements were carried out in a two-electrode sandwiched cell configuration with 1.0 M H₂SO₄ aqueous electrolyte. Initially, the CV response of Ti₃C₂T_X MXene and EuM nanocomposites were compared at 10 mV s⁻¹ (Fig. 8a). It was found that all of the nanocomposites achieved a wider operating voltage window (0 V to 1 V) in comparison to Ti₃C₂T_X MXene (0 V to 0.6 V) under the same scan rate, which holds a direct correlation with the energy density. For Ti₃C₂T_X MXene, the CV curve depicts neither an exact rectangular profile (EDLC characteristic) nor distinct redox peaks. The plausible reason for this behaviour may be ascribed to the interlayer restacking issues of Ti₃C₂T_X MXene, which first inhibit the intercalation/deintercalation movement of the electrolyte ions, obstructing the charge storage via EDLC mechanism. Secondly, the interlayer collapse reduces the specific surface area and thus masks the redox active sites, inhibiting the charge storage through reversible surface redox reactions. The slight humps are only observed due to the interaction of hydronium ions with the –O functional entity on the Ti₃C₂T_X MXene surface, which alters the Ti oxidation state in an acidic medium and thus imparts a slight pseudo influence. Thus, the shape of the CV curve of Ti₃C₂T_X MXene suggests the quasi-pseudo behaviour. Meanwhile, on compositing with Eu₂O₃, the CV curves of the EuM nanocomposites display a relatively larger CV peak area and prominent pairs of redox peaks, which reflects the higher specific capacitance (C_s) compared to that for the pure Ti₃C₂T_X MXene, evidencing the pseudo capacitance behaviour. The EDLC contribution in capacitance could also be seen from the nearly flat middle section of the CV curve. Thus, the combination of pseudo-capacitance and EDLC behaviour contributes to the nanocomposites' overall electrochemical energy storage capacity. This could be ascribed to the in situ generation of O_V upon incorporation of Eu₂O₃ in Ti₃C₂T_X MXene. The positive core of O_V acts as an active site that favors adsorption and accumulation of OH⁻ on the electrode surface. It modulates the local electronic structure of the underlying Ti₃C₂T_X MXene Ti sites towards the active redox ability for pseudo-capacitance. Simultaneously, O_V boosts the electrochemical charge transfer and expansion of the interlayer diffusion channels of Ti₃C₂T_X MXene with Eu₂O₃. Considering the vital role of O_V, the 5-EuM sample exhibits the highest current density and largest CV integrated area, signifying the relatively higher charge storage capacity. Moreover, the 5-EuM CV curve exhibits more prominent redox peaks relative to other nanocomposites, which is directly correlated with multiple redox reactions attributable to the greater number of active sites on the electrode surface, along with swift redox kinetics due to the optimal O_V content and reversible transitions between Eu²⁺/Eu³⁺. The smaller O_V level offers less redox active sites; meanwhile, an excessive O_V concentration imposes a negative impact by reducing the conductivity.^81,82 Thus, the C_s calculated from the CV plots using eqn (6) and (7) were 135.46 F g⁻¹, 201.95 F g⁻¹, 364.35 F g⁻¹, and 245.98 F g⁻¹ for Ti₃C₂T_X MXene, 3-EuM, 5-EuM, and 7-EuM, respectively.

Fig. 8 (a) CV curves and (b) GCD curves comparison for Ti₃C₂T_X MXene, 3-EuM, 5-EuM, and 7-EuM.

Furthermore, GCD measurements were employed to confirm the enhanced electrochemical performance of the 5-EuM sample. Fig. 8b shows the symmetric GCD profiles of the prepared electrodes, which were collected at 0.6 A g⁻¹ and compared. 5-EuM revealed greatly improved capacitance with the C_s reaching as high as 374.98 F g⁻¹, higher than Ti₃C₂T_X MXene (139.88 F g⁻¹), 3-EuM (206.28 F g⁻¹), and 7-EuM (246.98 F g⁻¹). Moreover, a comparison of the GCD curves demonstrates a much longer discharging time of 156 s for 5-EuM. These results indicate that a suitable amount of Eu₂O₃ increases the charge storage ability of Ti₃C₂T_X MXene by contributing to the pseudo capacitance and increasing the interlayer spacing between the Ti₃C₂T_X MXene layers.

Furthermore, the CVs and GCDs of 3-EuM, 5-EuM and 7-EuM were recorded at different scan rates (10 to 80 mV s⁻¹) and current densities (0.6 to 1.4 A g⁻¹), as shown in Fig. 9. It can be seen from the CV studies that with increasing scan rate, the current response increases, while C_s decreases (Table 4). With the increasing scan rate, the electrolyte ions do not have enough time to interact with the electrode's surface. Thus, the specific capacitance value is reduced at higher scan rates. The shape of the GCD studies at higher current density represents an exact triangular shape. Meanwhile, as the current density decreases, GCD demonstrates a slow and sloppy discharge, elongating the discharge time and thus resulting in an increase in C_s (Table 5). This may be due to the deep penetration of ions at the slower current density, which resulted in faradaic reactions at the inner surface and additional charge storage. Slower discharge arises from the formation of soluble redox shuttles upon decomposition of the loosely bonded surface functional groups at lower current density. The highest current response, largest specific capacitance, and longest discharging time altogether for the 5-EuM sample confirm its suitability for charge storage in supercapacitor applications.

Fig. 9 (a, c, and e) CV curves at different scan rates and (b, d, and f) GCD curves at various current densities for the 3-EuM, 5-EuM, and 7-EuM samples.

Table 4 Comparison of the cell and electrode specific capacitance values at different scan rates for 3-EuM, 5-EuM, and 7-EuM obtained from CV studies

Scan rate (mV s^−1)	Cell specific capacitance from CV studies (F g^−1)			Electrode specific capacitance from CV studies (F g^−1)
	3-EuM	5-EuM	7-EuM	3-EuM	5-EuM	7-EuM
10	50.49	91.09	61.49	201.95	364.35	245.98
20	34.41	75.26	55.24	137.65	301.05	220.95
40	22.51	51.99	38.69	90.04	207.98	154.78
60	20.16	43.63	32.95	80.42	174.52	131.82
80	17.99	36.67	29.05	71.99	146.68	116.18

---

Table 5 Comparison of the cell and electrode specific capacitance values at different scan rates for 3-EuM, 5-EuM, and 7-EuM obtained from GCD studies

Current density (A g^−1)	Cell specific capacitance from GCD studies (F g^−1)			Electrode specific capacitance from GCD studies (F g^−1)
	3-EuM	5-EuM	7-EuM	3-EuM	5-EuM	7-EuM
0.6	51.57	93.74	61.75	206.28	374.98	246.98
0.8	37.01	75.14	59.13	148.03	300.54	236.54
1.0	26.51	53.32	41.61	106.04	213.28	166.44
1.2	22.63	46.03	36.85	90.53	184.13	147.41
1.4	20.41	36.69	30.40	81.65	146.78	121.58

---

Furthermore, the energy densities and power densities were calculated from the GCD parameters of 3-EuM, 5-EuM, and 7-EuM, and compared in Ragone plots (Fig. 10a). At every power density value, 5-EuM depicts the highest energy density of 13.02 Wh kg⁻¹ at 0.6 A g⁻¹ current density among the other tested samples, which is attributable to the increased density of exposed redox active species upon creation of O_V at a controlled concentration. Another observation suggests that the energy density is increased upon doping without sacrificing the power density due to the increased electrical conductivity and ion diffusion rate associated with the delocalized free electrons of O_V, leading to high power density. Furthermore, EIS revealed that 5-EuM exhibited superior interfacial charge transfer and capacitive behaviour compared to the bare Ti₃C₂T_X MXene and other nanocomposites. The Nyquist plots (Fig. 10b) derived from EIS measurements could be divided into three major components: the internal resistance (R_s) of the electrode material, electrolyte ionic movement and metal electrode are obtained from the intersection of the plot with the X-axis in the high-frequency region; the diameter of the semicircular part gives the charge transfer resistance (R_ct) at the electrode/electrolyte interface; and the linear part in the low-frequency region represents the Warburg resistance in which the slope of line corresponds to the diffusion/transportation of electrolytic ions. Table 6 lists the R_s and R_ct values calculated for different electrodes. The larger the R_s and R_ct, the greater the charge transfer kinetics. Furthermore, the steep line indicates the facile electrolyte intercalation, and thus higher capacitive performance. Among the prepared samples, 5-EuM displayed the smallest R_s and R_ct of 1.58 Ω and 3.86 Ω, respectively, and a nearly vertical line in the low-frequency region, which projects its fast kinetic merits. Finally, a long-term cycling stability test (Fig. 10c) was also conducted for 5-EuM by repeated GCD tests at 1.0 A g⁻¹ current density. It maintained 93.39% of specific capacitance even after 10 000 continuous cycles, which demonstrates the durability of the electrode material. The coulombic efficiency of the 5-EuM sample at a current density of 1 A g⁻¹ has been calculated from the GCD curves in order to understand the material's performance during the long-term charge–discharge cycles. The coulombic efficiency of the 5-EuM sample decayed from 91.69% at the 1st cycle to 90.31% at the 10 000th cycle after continuous measurement (inset: Fig. 10c). This suggests the high stability of the MXene electrode materials, which is attributed to the high corrosion resistance properties upon composite formation of Eu₂O₃.

Fig. 10 (a) Ragone plot, (b) Nyquist plot; inset shows the magnified high-frequency region for the 3-EuM, 5-EuM, and 7-EuM electrodes, and (c) cycling stability of 5-EuM at 1.0 A g⁻¹.

Table 6 EIS parameters for Ti₃C₂T_X MXene, 3-EuM, 5-EuM, and 7-EuM

Samples	R s (Ω)	R ct (Ω)
Ti3C2TX MXene	3.604	13.34
3-EuM	2.07	5.97
5-EuM	1.58	3.86
7-EuM	1.68	4.24

---

A comparison of the water splitting and supercapacitor performance of Ti₃C₂T_X MXene/Eu₂O₃ in this present work and the previously reported MXene-based electrochemical devices is tabulated in Table S1.† It is easy to see that the present device exhibits lower overpotential values in HER and OER and higher specific capacitance due to the optimal generation of O_V, which aids in swift charge transfer, modulates an electronic structure that creates abundant redox active sites, maximizes the light absorption, and favors greater OH⁻ ion adsorption for higher charge storage capacity.

The electronic structure of a material plays a vital role in the HER process. Thus, DFT calculations were carried out to calculate the electronic structure and hydrogen (H) adsorption energy of MXene and the Eu₂O₃/MXene heterostructure with an oxygen vacancy. The optimized structure of the MXene system is shown in Fig. 11a. The optimized structure of MXene with an H adsorbed on the O-site is shown in Fig. 11b, and the calculated value of H adsorption energy is equal to −0.48 eV. A high value of H adsorption energy (−0.48 eV) on the O-site suggests a strong interaction (coupling) between the H and O atoms and subsequent strong adsorption. However, this strong adsorption tendency reflects the difficulty in the desorption of H from the surface of the adsorbent (MXene), consequently hindering the HER process. These findings are also supported by the DFT studies from other researchers.^58,63,83,84

Fig. 11 (a) Top and front views of the optimised structure of MXenes. (b) Top and front views of H adsorbed on the O site of MXenes. (c) TDOS and PDOS plots of MXenes.

Furthermore, O_V are formed in the Eu₂O₃/MXene system, and play a vital role in the electrocatalytic process.⁶³ Therefore, the adsorption/desorption of hydrogen and the electronic structure of Eu₂O₃/MXene-O_V were also investigated. The structure of the Eu₂O₃/MXene-O_V system before and after optimization is shown in Fig. 12a. To realize the Eu₂O₃/MXene-O_V system, O_V has been created by removing an O atom from MXene. However, after structural optimization, the O_V shifted to the Eu₂O₃ group, as shown in Fig. 12a. The H adsorption on the Eu₂O₃/MXene-O_V system (Fig. 12b) is calculated to −0.084 eV, which is favorable for both the adsorption and desorption of H, and thus for the HER process.^58,63,84 Hence, DFT studies confirm that the presence of O_V in the Eu₂O₃/MXene system helps to improve the HER performance.⁶³

Fig. 12 (a) Top and front views of the structure of the Eu₂O₃/MXene system with an O_V before and after optimisation. (b) Top and front views of H adsorbed on the O site of the Eu₂O₃ molecule of the Eu₂O₃/MXene-O_V system. (c) TDOS and PDOS plots of Eu₂O₃/MXene-O_V.

Furthermore, the electronic structure (density of states) helps to study the charge carrier transfer rate.⁸⁵ The total density of states (TDOS) plot of MXene and the partial density of states (PDOS) plot of the constituent atoms (Ti, C, O) are shown in Fig. 11c. The top of the valence band is formed from the hybridization between the orbitals of C, Ti, and O. In contrast, the bottom of the conduction band is mainly made up of Ti orbitals, with some contribution from the orbitals of C and O.⁸⁶ It should be noted that while plotting the DOS plots and band structures, the Fermi energy level was set at 0 eV. The DOS and PDOS plots of Eu₂O₃/MXene-O_V are shown in Fig. 12c. Apart from the contribution of Ti, C, and O of MXene, the orbitals of Eu mainly contribute to the conduction band, whereas O (of the Eu₂O₃ group) primarily contributes to the valence states particularly close to the Fermi level. From the TDOS plots, we can see that there are a greater number of electronic states around the Fermi level in the case of Eu₂O₃/MXene-O_V as compared to Eu₂O₃/MXene, which indicates that the electron transfer rate is higher in Eu₂O₃/MXene-O_V. In other words, the larger the number (density) of electronic states near the Fermi level, the higher the concentration of electrons around the Fermi level, which leads to better electron mobility and hence higher electron transfer rate.⁸⁵ For a better comparison of the density of states of the MXene and Eu₂O₃/MXene-O_V near the Fermi level, see Fig. S7.†

Thus, the DFT study reveals that Eu₂O₃/MXene-O_V demonstrates better adsorption/desorption energetics, greater electron transfer rate, and reaction kinetics, making it a suitable contender for electrochemical energy conversion and electrochemical energy storage applications.

4. Conclusion

In summary, Eu₂O₃/Ti₃C₂T_X MXene electrodes with tunable oxygen vacancies (O_V) were prepared and tested for electro/photocatalytic water splitting and supercapacitor device applications. FESEM, TEM and XPS studies revealed the physio/chemical alterations in the Ti₃C₂T_X MXene structure with O_V variations upon changing the concentration of Eu₂O₃. O_V modulated the electronic structure of Ti₃C₂T_X MXene by providing highly active Ti²⁺ sites for effective H⁺ reduction, and eased the sluggish charge dissociation process in OER. Among the prepared samples, the 5 wt% Eu₂O₃-doped Ti₃C₂T_X MXene (5-EuM) with optimal O_V content demonstrated superior performance in electrocatalytic water splitting, requiring smaller 63 mV and 169 mV overpotential to drive HER and OER, respectively, and also exhibited long-term stable operation in an acidic medium for 15 h. Furthermore, 5-EuM assembled in a two-electrode configuration required a small cell voltage of 1.48 V to achieve 10 mA cm⁻² current density for the overall water splitting reaction. Upon combination with the CdS photoabsorber, the 5-EuM/CdS sample exhibited superior performance in photocatalysis with a higher photocurrent density of 4.86 mA cm⁻² among the other photocatalysts, and a high photoconversion efficiency of 1.86%. This enhanced photocatalytic response could be ascribed to the maximized light absorption and an effective charge separation mediated by the controlled O_V content in 5-EuM/CdS and suitable redox potentials of the adjoining photocatalysts, leading to an efficient Z-scheme interfacial charge transfer. When used as a supercapacitor electrode, 5-EuM delivered a high specific capacitance of 374.98 F g⁻¹ and a long-discharging time of 156 s derived from the combined effect of the redox-active Eu₂O₃ and highly conductive underlying Ti₃C₂T_X MXene matrix. Thus, the present work paves the way for tuning the electrochemical catalytic activity and charge storage by utilizing the synergistic effects of the O_V-enriched Eu₂O₃/Ti₃C₂T_X MXene framework and constructed multifunctional electrodes, exhibiting superior electrochemical energy generation water splitting and electrochemical energy storage supercapacitor applications.

Data availability

The data supporting this article have been included as part of the ESI.†

Author contributions

Vaishali Sharma: conceptualization (equal); data curation (lead); methodology (lead); writing – original draft (lead). Jasvir Singh: investigation (equal); resources (equal); Rajnish Dhiman: investigation (equal); Aman Mahajan: supervision (lead); conceptualization (equal); writing – review & editing (equal).

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors are thankful to UGC-DAE CSR, Indore, for providing financial support and experimental facility through CRS project no. CRS/2021–22/01/386.

References

1. P. Wang, S. Zhang, Z. Wang, Y. Mo, X. Luo, F. Yang, M. Lv, Z. Li and X. Liu, J. Mater. Chem. A, 2023, 11, 5476–5494.
2. S. A. Thomas and J. Cherusseri, J. Mater. Chem. A, 2023, 11, 23148–23187.
3. G. M. Tomboc, J. Kim, Y. Wang, Y. Son, J. Li, J. Y. Kim and K. Lee, J. Mater. Chem. A, 2021, 9, 4528–4557.
4. H. Lee, J. Park, J.-M. Myoung and B. Han, J. Mater. Chem. A, 2025,(13), 9244–9251.
5. S. Balachandran, K. Jeeva Jothi, K. Selvakumar, D. K. Bhat, K. Sathiyanarayanan and M. Swaminathan, Mater. Chem. Phys., 2020, 256, 123624.
6. A. De, M. S. Kim, A. Adhikari, R. Patel and S. Kundu, J. Mater. Chem. A, 2024, 12, 19720–19756.
7. J. Libich, J. Máca, J. Vondrák, O. Čech and M. Sedlaříková, J. Energy Storage, 2018, 17, 224–227.
8. A. Shiralizadeh Dezfuli, E. Kohan, H. R. Naderi and E. Salehi, New J. Chem., 2019, 43, 9260–9264.
9. W. Sun and Z. Mo, Mater. Sci. Eng., B, 2013, 178, 527–532.
10. R. Zou, H. Quan, M. Pan, S. Zhou, D. Chen and X. Luo, Electrochim. Acta, 2018, 292, 31–38.
11. M.-K. Song, S. Cheng, H. Chen, W. Qin, K.-W. Nam, S. Xu, X.-Q. Yang, A. Bongiorno, J. Lee, J. Bai, T. A. Tyson, J. Cho and M. Liu, Nano Lett., 2012, 12, 3483–3490.
12. W. Chen, R. B. Rakhi, L. Hu, X. Xie, Y. Cui and H. N. Alshareef, Nano Lett., 2011, 11, 5165–5172.
13. R. Liu, L. Ma, G. Niu, X. Li, E. Li, Y. Bai and G. Yuan, Adv. Funct. Mater., 2017, 27, 1701635.
14. Y. Zhu, X. Liu, S. Jin, H. Chen, W. Lee, M. Liu and Y. Chen, J. Mater. Chem. A, 2019, 7, 5875–5897.
15. D. Gogoi, R. S. Karmur, M. R. Das and N. N. Ghosh, J. Mater. Chem. A, 2023, 11, 23867–23880.
16. C. Yang, Y. Lu, W. Duan, Z. Kong, Z. Huang, T. Yang, Y. Zou, R. Chen and S. Wang, Energy Fuels, 2021, 35, 14283–14303.
17. A. Bashir, A. Inayat, R. Bashir, S. Jamil, S. M. Abbas, M. Sultan, A. Iqbal and Z. Akhter, New J. Chem., 2023, 47, 3560–3571.
18. Q. Yue, J. Sun, S. Chen, Y. Zhou, H. Li, Y. Chen, R. Zhang, G. Wei and Y. Kang, ACS Appl. Mater. Interfaces, 2020, 12, 18570–18577.
19. D. Ontiveros, F. Viñes and C. Sousa, J. Mater. Chem. A, 2023, 11, 13754–13764.
20. S. Ponnada, M. S. Kiai, O. Eroglu, C.-O. Kim, R. K. Sharma and G. S. Martynkova, Energy Fuels, 2024, 38, 18242–18264.
21. A. Raveendran, M. Chandran, M. R. Siddiqui, S. M. Wabaidur, M. Eswaran and R. Dhanusuraman, ACS Omega, 2023, 8, 34768–34786.
22. K. Yang, H. Cheng, B. Wang, Y. Tan, T. Ye, Y. Yang and C. Wang, Adv. Mater. Technol., 2022, 7, 2100675.
23. S. Venkateshalu, G. M. Tomboc, S. P. Nagalingam, J. Kim, T. Sawaira, K. Sehar, B. G. Pollet, J. Y. Kim, A. Nirmala Grace and K. Lee, J. Mater. Chem. A, 2023, 11(27), 14469–14488.
24. S. N. Ansari, M. Saraf, Z. Abbas and S. M. Mobin, Nanoscale, 2023, 15, 13546–13560.
25. X. Liu, M. Wang, B. Qin, Y. Zhang, Z. Liu and H. Fan, Chem. Eng. J., 2022, 431, 133796.
26. J. B. Lee, G. H. Choi and P. J. Yoo, J. Alloys Compd., 2021, 887, 161304.
27. Q. Zhang, X. Wang, Z. Pan, J. Sun, J. Zhao, J. Zhang, C. Zhang, L. Tang, J. Luo, B. Song, Z. Zhang, W. Lu, Q. Li, Y. Zhang and Y. Yao, Nano Lett., 2017, 17, 2719–2726.
28. M. Majumder, R. B. Choudhary, A. K. Thakur and U. Kumar, J. Electroanal. Chem., 2017, 804, 42–52.
29. R. Tong, Z. Sun, X. Wang, S. Wang and H. Pan, ChemElectroChem, 2019, 6, 1338–1343.
30. Q. Ke, C. Guan, X. Zhang, M. Zheng, Y.-W. Zhang, Y. Cai, H. Zhang and J. Wang, Adv. Mater., 2017, 29, 1604164.
31. A. Padhy, S. Lakshmy, B. Chakraborty and J. N. Behera, Sustainable Energy Fuels, 2023, 7, 5271–5282.
32. S. Yang, Y. Liu, Y. Hao, X. Yang, W. A. Goddard III, X. L. Zhang and B. Cao, Adv. Sci., 2018, 5, 1700659.
33. M. Z. Iqbal, M. Shaheen, U. Aftab, Z. Ahmad, M. Y. Solangi, M. I. Abro and S. M. Wabaidur, Energy Fuels, 2024, 38, 2416–2425.
34. M. M. Ibrahim, A. Mezni, M. Alsawat, T. Kumeria, M. R. Das, S. Alzahly, A. Aldalbahi, K. Gornicka, J. Ryl, M. A. Amin and T. Altalhi, Int. J. Energy Res., 2019, 43, 5367–5383.
35. M. Zhi, C. Xiang, J. Li, M. Li and N. Wu, Nanoscale, 2013, 5, 72–88.
36. C. An, Y. Zhang, H. Guo and Y. Wang, Nanoscale Adv., 2019, 1, 4644–4658.
37. Y. Ma, X. Xie, W. Yang, Z. Yu, X. Sun, Y. Zhang, X. Yang, H. Kimura, C. Hou, Z. Guo and W. Du, Adv. Compos. Hybrid Mater., 2021, 4, 906–924.
38. J. Liu, W. Chen, Q. Sun, Y. Zhang, X. Li, J. Wang, C. Wang, Y. Yu, L. Wang and X. Yu, ACS Appl. Energy Mater., 2021, 4, 2864–2872.
39. Y. He, W. Zhou, D. Li, Y. Liang, S. Chao, X. Zhao, M. Zhang and J. Xu, Small, 2023, 19, 2206956.
40. D. Yan, W. Wang, X. Luo, C. Chen, Y. Zeng and Z. Zhu, Chem. Eng. J., 2018, 334, 864–872.
41. J. Hao, S. Peng, H. Li, S. Dang, T. Qin, Y. Wen, J. Huang, F. Ma, D. Gao, F. Li and G. Cao, J. Mater. Chem. A, 2018, 6, 16094–16100.
42. X. Zhao, Y. Jing, Z. Dai, Y. Chu, Z. Liu, Y. Cong and J. Song, ACS Omega, 2024, 9, 16868–16875.
43. F. Lei, Y. Sun, K. Liu, S. Gao, L. Liang, B. Pan and Y. Xie, J. Am. Chem. Soc., 2014, 136, 6826–6829.
44. A. Zhang, R. Gao, L. Hu, X. Zang, R. Yang, S. Wang, S. Yao, Z. Yang, H. Hao and Y.-M. Yan, Chem. Eng. J., 2021, 417, 129186.
45. M. Sun, R.-T. Gao, J. He, X. Liu, T. Nakajima, X. Zhang and L. Wang, Angew. Chem., Int. Ed., 2021, 60, 17601–17607.
46. Q. Yang, Y. Wang, Q. Tian, X. Li, A. Pan, M. Zhao, Y. Zhu, T. Wu and G. Fang, J. Mater. Chem. A, 2024, 12, 7207–7214.
47. P. Aryanrad, H. R. Naderi, E. Kohan, M. R. Ganjali, M. Baghernejad and A. Shiralizadeh Dezfuli, RSC Adv., 2020, 10, 17543–17551.
48. V. Sharma, R. Dhiman and A. Mahajan, Appl. Surf. Sci., 2023, 612, 155883.
49. J. D. Benck, Z. Chen, L. Y. Kuritzky, A. J. Forman and T. F. Jaramillo, ACS Catal., 2012, 2, 1916–1923.
50. M. H. Suliman, A. Adam, M. N. Siddiqui, Z. H. Yamani and M. Qamar, Carbon, 2019, 144, 764–771.
51. T. Li and D. Ding, Materials, 2019, 12(24), 4102.
52. S. Sardana, P. Mahajan, A. Mishra, A. K. Chawla and A. Mahajan, Adv. Mater. Technol., 2024, 2400829.
53. A. Padhy, R. Samal, C. S. Rout and J. N. Behera, Sustainable Energy Fuels, 2022, 6, 2010–2019.
54. P. Hohenberg and W. Kohn, Phys. Rev. [Sect.] B, 1964, 136, B864–B871.
55. P. Giannozzi, S. Baroni, N. Bonini, M. Calandra, R. Car, C. Cavazzoni, D. Ceresoli, G. L. Chiarotti, M. Cococcioni, I. Dabo, A. Dal Corso, S. de Gironcoli, S. Fabris, G. Fratesi, R. Gebauer, U. Gerstmann, C. Gougoussis, A. Kokalj, M. Lazzeri, L. Martin-Samos, N. Marzari, F. Mauri, R. Mazzarello, S. Paolini, A. Pasquarello, L. Paulatto, C. Sbraccia, S. Scandolo, G. Sclauzero, A. P. Seitsonen, A. Smogunov, P. Umari and R. M. Wentzcovitch, J. Phys.: Condens. Matter, 2009, 21, 395502.
56. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868.
57. J. Singh, G. P. Singh, K. J. Singh, N. Virdi, D. A. Tonpe and R. C. Singh, Desalin. Water Treat., 2025, 322, 101155.
58. D. Cao, H. Liu, Q. Sun, X. Wang, J. Wang, C. Liu, C. Liu, Q. Hao and Y. Li, Appl. Surf. Sci., 2021, 563, 150210.
59. S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem. Phys., 2010, 132, 154104.
60. H. J. Monkhorst and J. D. Pack, Phys. Rev. B: Solid State, 1976, 13, 5188–5192.
61. C. G. Broyden, IMA J. Appl. Math., 1970, 6, 222–231.
62. R. Fletcher, Comput. J., 1970, 13, 317–322.
63. Z. Wang, H. Di, R. Sun, Y. Zhu, L. Yin, Z. Zhang and C. Wang, ChemPhysMater, 2022, 1, 321–329.
64. S. Gasso, M. K. Sohal and A. Mahajan, Sens. Actuators, B, 2022, 357, 131427.
65. H. Wang, R. Zhao, J. Qin, H. Hu, X. Fan, X. Cao and D. Wang, ACS Appl. Mater. Interfaces, 2019, 11, 44249–44262.
66. T. R. Poorani, C. Ramya and M. Ramya, J. Inorg. Organomet. Polym. Mater., 2023, 33, 3413–3440.
67. B. Sun, Y. Qian, Z. Liang, Y. Guo, Y. Xue, J. Tian and H. Cui, Sol. Energy Mater. Sol. Cells, 2019, 195, 309–317.
68. C.-H. Zeng, K. Zheng, K.-L. Lou, X.-T. Meng, Z.-Q. Yan, Z.-N. Ye, R.-R. Su and S. Zhong, Electrochim. Acta, 2015, 165, 396–401.
69. K. L. Firestein, J. E. von Treifeldt, D. G. Kvashnin, J. F. S. Fernando, C. Zhang, A. G. Kvashnin, E. V Podryabinkin, A. V Shapeev, D. P. Siriwardena, P. B. Sorokin and D. Golberg, Nano Lett., 2020, 20, 5900–5908.
70. J. Yang, Z. Quan, D. Kong, X. Liu and J. Lin, Cryst. Growth Des., 2007, 7, 730–735.
71. V. Natu, M. Benchakar, C. Canaff, A. Habrioux, S. Célérier and M. W. Barsoum, Matter, 2021, 4, 1224–1251.
72. A. Karmakar, K. Karthick, S. S. Sankar, S. Kumaravel, R. Madhu, K. Bera, H. N. Dhandapani, S. Nagappan, P. Murugan and S. Kundu, J. Mater. Chem. A, 2022, 10, 3618–3632.
73. Y. Zare, Composites, Part A, 2016, 84, 158–164.
74. S. Kumar, R. Prakash, R. J. Choudhary and D. M. Phase, Mater. Res. Bull., 2015, 70, 392–396.
75. T. Zhang, M.-Y. Wu, D.-Y. Yan, J. Mao, H. Liu, W.-B. Hu, X.-W. Du, T. Ling and S.-Z. Qiao, Nano Energy, 2018, 43, 103–109.
76. Y. Guo, Y. Tong, P. Chen, K. Xu, J. Zhao, Y. Lin, W. Chu, Z. Peng, C. Wu and Y. Xie, Adv. Mater., 2015, 27, 5989–5994.
77. G. Li, R. Huang, C. Zhu, G. Jia, S. Zhang and Q. Zhong, Colloids Surf., A, 2021, 614, 126156.
78. S. P. Narayanan, P. Thakur, A. P. Balan, A. A. Abraham, F. Mathew, M. Yeddala, T. Subair, C. Tiwary, S. Thomas, T. N. Narayanan, P. M. Ajayan and M. M. R. Anantharaman, Phys. Status Solidi RRL, 2019, 13, 1900025.
79. S. A. Ali and T. Ahmad, Int. J. Hydrogen Energy, 2023, 48, 22044–22059.
80. A. Ahmed, M. Naseem Siddique, U. Alam, T. Ali and P. Tripathi, Appl. Surf. Sci., 2019, 463, 976–985.
81. X. Zhang, X. Liu, Y. Zeng, Y. Tong and X. Lu, Small Methods, 2020, 4, 1900823.
82. G. Wang and Z. Jin, J. Mater. Chem. C, 2021, 9, 620–632.
83. C. Kang, C. Li, J. Zhang, J. Ren and M. Chen, Int. J. Hydrogen Energy, 2024, 78, 503–512.
84. Y. Su, M. Song, X. Wang, J. Jiang, X. Si, T. Zhao and P. Qian, Nanomaterials, 2021, 11(10), 2497.
85. I. Ullah, A. Irfan, M. Li, S. Saddique, T. Yang, H. Zhao, N. Irshad, K. Yang, C. Wang and P. K. Chu, J. Power Sources, 2025, 643, 237046.
86. M. Das and S. Ghosh, J. Electrochem. Soc., 2022, 169, 90525.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5ta01195g

---