Structural, Morphological, Optical, And Electrochemical Properties Of Nickel-Doped Tungsten Trioxide (Ni–WO?) Nanomaterials Synthesised By Hydrothermal Method For Supercapacitor Electrode Applications

The global transition towards renewable energy systems and electrified transportation has placed an unprecedented demand on electrochemical energy storage technologies that combine high power density, long cycle life, and reliable operation across wide temperature ranges [1]. Supercapacitors, also known as electrochemical double-layer capacitors or, when pseudocapacitive mechanisms dominate, pseudocapacitors, have emerged as indispensable components in hybrid energy storage architectures that pair their rapid charge–discharge characteristics with the high energy density of batteries [2]. The electrode material is the central determinant of supercapacitor performance, and the development of electrode materials with simultaneously high capacitance, low internal resistance, and long cycle stability remains a primary research objective in the field [3].

Tungsten trioxide (WO₃) is a transition metal oxide of considerable fundamental and applied interest. Its monoclinic crystal structure, which is the thermodynamically stable polymorph at room temperature, features corner-sharing WO₆ octahedra arranged in a three-dimensional framework containing rectangular tunnels that are geometrically well-suited to reversible intercalation of alkali and proton species [1,4]. The multi-step redox chemistry of tungsten, spanning the W⁶⁺/W⁵⁺ and W⁵⁺/W⁴⁺ couples within an accessible electrochemical potential window, provides a rich source of faradaic charge storage. WO₃ has accordingly been investigated as a pseudocapacitive electrode material, a photocatalyst, a photoanode for water splitting, and an electrochromic material, the electrochromic colouration mechanism being mechanistically identical to the electrochemical charge storage process [4,8].

A well-recognised limitation of undoped WO₃ as a supercapacitor electrode material is its relatively modest intrinsic electronic conductivity, which arises from the wide optical band gap (~2.6–2.8 eV) of the monoclinic phase [8]. This imposes a significant series resistance penalty at high charge–discharge rates, degrading power density and rate capability. Substitutional doping with transition metal cations represents a powerful strategy for addressing this limitation, as the introduction of aliovalent or isovalent dopants with accessible redox states simultaneously narrows the optical and electrical band gap through the creation of intra-gap electronic states, creates additional crystallographic defects and grain boundaries that serve as electrochemically active sites, and potentially introduces supplementary redox couples that broaden the faradaic charge storage repertoire of the material [6,9].

Nickel is an attractive dopant choice for WO₃ for several reasons. The Ni²⁺ ion (ionic radius 0.69 Å) is sufficiently close in size to W⁶⁺ (ionic radius 0.60 Å) to permit solid-solution substitution without disrupting the WO₃ framework beyond the solubility limit at moderate concentrations, while being large enough to induce measurable lattice expansion [3,7]. The Ni²⁺/Ni³⁺ redox couple is electrochemically active in neutral aqueous electrolytes, adding a supplementary pseudocapacitive pathway to the host WO₃ redox chemistry. Moreover, nickel is earth-abundant, low-cost, and compatible with scalable hydrothermal synthesis methods widely used for the preparation of nanostructured metal oxide materials [6,9].

Despite the considerable interest in Ni-doped metal oxides for energy storage, systematic studies tracking the evolution of all key structural, morphological, optical, and electrochemical properties of Ni–WO₃ as a function of doping concentration across the full range from undoped to moderately doped (2.5–10%) under identical synthesis and evaluation conditions are limited. Such data are essential for identifying the optimum doping level, understanding the physical mechanisms responsible for performance enhancement, and providing quantitative design guidance for synthesis optimisation. The present study addresses this gap by reporting a comprehensive characterisation of hydrothermally synthesised Ni–WO₃ nanoparticles across doping concentrations of 2.5%, 5%, 7%, and 10%.

2. MATERIALS AND METHODS

2.1 Synthesis of Ni–WO₃ Nanoparticles

All chemicals, including sodium tungstate dihydrate (Na₂WO₄·2H₂O), nickel nitrate hexahydrate (Ni(NO₃)₂·6H₂O), hydrochloric acid, and ethanol (analytical reagent grade, Sigma-Aldrich), were used as received without further purification. Deionised water (resistivity >18.2 MΩ cm, Milli-Q system) was used throughout.

Undoped WO₃ was prepared by dissolving 2.0 mmol of Na₂WO₄·2H₂O in 30 mL deionised water, adjusting the pH to 2 using dilute hydrochloric acid under vigorous magnetic stirring, and transferring the resulting suspension to a 50 mL Teflon-lined stainless steel autoclave. The autoclave was sealed and maintained at 180 °C for 18 hours in an oven, then allowed to cool naturally to room temperature. For Ni-doped samples, the appropriate mole fraction of Ni(NO₃)₂·6H₂O corresponding to 2.5%, 5%, 7%, or 10% Ni substitution for W was dissolved together with Na₂WO₄·2H₂O prior to pH adjustment, ensuring homogeneous distribution of the dopant precursor in the reaction solution. The products were collected by centrifugation at 8000 rpm for 10 minutes, washed three times each with deionised water and ethanol to remove residual ions, and dried at 80 °C under vacuum for 12 hours to obtain the final powder samples.

2.2 Characterisation

X-ray diffraction (XRD) patterns were recorded on a Bruker D8 Advance diffractometer using Cu Kα radiation (λ = 1.5406 Å) over the 2θ range 10–80° at a scan rate of 2° min⁻¹ with a step size of 0.02°. Crystallite sizes were calculated from the full width at half maximum (FWHM) of the most intense diffraction peak using the Scherrer equation [4]. Microstrain was evaluated using the Williamson–Hall plot method [5]. Field-emission scanning electron microscopy (FE-SEM) was performed on a JEOL JSM-7100F instrument at an accelerating voltage of 15 kV. Energy dispersive X-ray (EDAX) spectra were collected using an Oxford Instruments X-Max detector integrated with the FE-SEM. Fourier transform infrared (FTIR) spectra were recorded on a Shimadzu IRTracer-100 spectrometer using KBr pellets over the wavenumber range 400–4000 cm⁻¹ at a resolution of 4 cm⁻¹. UV–Visible diffuse reflectance spectra were recorded on a PerkinElmer Lambda 950 spectrophotometer and converted to absorbance using the Kubelka–Munk function; optical band gaps were determined from Tauc plots assuming direct-allowed transitions [15].

2.3 Electrode Fabrication and Electrochemical Measurements

Working electrodes were prepared by mixing 80 wt% active material, 10 wt% acetylene black, and 10 wt% polyvinylidene fluoride (PVDF) binder in N-methyl-2-pyrrolidone (NMP) to form a homogeneous slurry, which was blade-coated onto nickel foam current collectors (1 cm² geometric area), dried at 80 °C under vacuum for 12 hours, and pressed at 10 MPa. Active material loadings were approximately 2 mg cm⁻². Cyclic voltammetry was performed using a Bio-Logic SP-200 potentiostat in a three-electrode cell comprising the fabricated electrode (working), a platinum wire (counter), and a saturated calomel electrode (SCE, reference) in 1 M Na₂SO₄ aqueous electrolyte at ambient temperature. CV profiles were recorded at scan rates of 2, 5, 10, 20, 50, and 100 mV s⁻¹ over the potential window −0.2 to +0.8 V vs. SCE. Specific capacitance was calculated as Cs = ∫i dV / (2 × m × v × ΔV) [16].

3. RESULTS AND DISCUSSION

3.1 X-ray Diffraction Analysis

The XRD patterns of the undoped WO₃ and the Ni-doped series are presented in Figure 1. The diffraction pattern of undoped WO₃ was fully indexed to the monoclinic gamma-WO₃ phase (JCPDS card no. 43-1035), with characteristic peaks corresponding to the (002), (020), (200), (112), (022), (202), and (122) crystallographic planes [1]. The monoclinic phase represents the thermodynamically stable polymorph of WO₃ at room temperature and the faithful reproduction of its diffraction pattern confirmed phase-pure synthesis of the undoped material. The relatively sharp and intense diffraction peaks of the undoped sample indicated a reasonably high degree of long-range crystalline order in the as-prepared nanoparticles.

Progressive Ni doping from 2.5% to 7% produced a systematic and monotonic shift of all peak positions towards lower 2θ angles, consistent with lattice expansion arising from the partial substitution of the smaller W⁶⁺ ion (ionic radius 0.60 Å) by the larger Ni²⁺ ion (ionic radius 0.69 Å) [3]. The magnitude of this peak shift was smaller than that observed in the analogous MoO₃ system under equivalent doping conditions, which is attributable to the more rigid and three-dimensionally interconnected corner-sharing octahedral network of WO₃ compared to the more compliant layered structure of MoO₃. This structural rigidity means that a given dopant-induced lattice mismatch produces a smaller unit cell expansion per cent in WO₃. No diffraction peaks attributable to secondary phases such as NiO, Ni(OH)₂, or NiWO₄ were detected at any doping level, confirming that Ni²⁺ ions were fully incorporated into the WO₃ host lattice as solid-solution substitutants within the studied concentration range [2].

Scherrer analysis of the FWHM of the most intense (200) reflection yielded average crystallite sizes of approximately 35 nm for undoped WO₃, decreasing progressively to approximately 29 nm at 2.5% Ni, 26 nm at 5% Ni, and reaching a minimum of approximately 22 nm at 7% Ni doping [4]. This monotonic trend of crystallite size reduction with increasing dopant concentration is a well-documented consequence of dopant-induced lattice strain, which introduces local stress fields that impede the coalescence of crystallites during the hydrothermal growth process, effectively capping the maximum grain size achievable at a given dopant level [5,6]. Smaller crystallite sizes translate directly into higher grain boundary area per unit volume, providing a greater density of electrochemically active surface and near-surface sites. At 10% Ni, the crystallite size increased marginally to approximately 24 nm relative to the 7% sample, suggesting that the solubility limit of Ni in the WO₃ lattice has been approached and that a fraction of excess Ni may begin to segregate to grain boundaries or form amorphous clusters, disrupting the regular grain-growth inhibition mechanism.

Williamson–Hall analysis of peak broadening across multiple reflections yielded microstrain values increasing monotonically from 0.08% in undoped WO₃ to 0.27% at 10% Ni doping [5]. The increasing microstrain corroborates the crystallite size data in confirming the progressive accumulation of lattice distortion with dopant concentration. The concurrent reduction of crystallite size and increase of microstrain at the 7% Ni level, combined with the absence of secondary phases, identifies this composition as the effective solid-solution limit for Ni in hydrothermally synthesised WO₃ under the conditions employed.

Figure 1:  XRD patterns of (a) pure WO₃, (b) 2.5% Ni–WO₃, (c) 5% Ni–WO₃, (d) 7% Ni–WO₃, and (e) 10% Ni–WO₃ samples recorded using CuKα radiation (λ = 1.5406 Å) in the 2θ range 10–80°.

3.2 Morphological Analysis (FE-SEM)

The FE-SEM micrographs presented in Figure 2 illustrate the systematic evolution of particle morphology with increasing Ni doping concentration. The undoped WO₃ sample (Figure 2a) exhibited a compact, densely packed microstructure composed of quasi-spherical and irregularly polyhedral particles with diameters in the range of 200–500 nm. The particle surfaces appeared smooth and the inter-particle contacts were intimate, leaving minimal void space accessible to the electrolyte [8]. This dense microstructure, while indicative of well-crystallised material, is unfavourable for supercapacitor applications because it restricts electrolyte penetration to the outer surface layer, thereby limiting the volume of the active material that participates in electrochemical charge storage.

Upon introducing 2.5% Ni (Figure 2b), the primary particle size distribution was narrowed and shifted towards smaller values, with average particle diameters in the range 150–350 nm. The particle surfaces became noticeably rougher and more textured, suggesting that Ni incorporation perturbs the regular growth fronts of the WO₃ crystallites during the hydrothermal process and promotes the nucleation of smaller secondary crystallites on the primary particle surfaces. At 5% Ni (Figure 2c), the morphology evolved further towards a more open, interconnected assembly of smaller particles with visible inter-particle voids, providing channels for electrolyte access into the electrode bulk [9].

The 7% Ni–WO₃ sample (Figure 2d) displayed the most favourable microstructure among the series: a granular, loosely aggregated nanoparticle morphology with particle sizes predominantly in the range 80–200 nm and a well-developed network of inter-particle pores of 10–50 nm diameter. This hierarchical porous architecture simultaneously maximises the effective electrode–electrolyte interfacial area available for double-layer charge storage and surface faradaic reactions, and minimises the mean ion diffusion path length through the electrode bulk, both of which contribute directly to enhanced electrochemical performance. The 10% Ni sample (Figure 2e) showed evidence of incipient agglomeration, with some particles merging into larger secondary clusters with reduced inter-particle porosity, consistent with the non-monotonic crystallite size trend observed in the XRD analysis and suggesting that the 7% level represents the optimum between grain refinement and agglomeration effects [32].

Figure 2:  FE-SEM micrographs of (a) pure WO₃, (b) 2.5% Ni–WO₃, (c) 5% Ni–WO₃, (d) 7% Ni–WO₃, and (e) 10% Ni–WO₃ samples showing the progressive morphological evolution from compact dense particles to open porous nanoparticle assemblies with increasing Ni doping concentration.

3.3 Elemental Analysis (EDAX)

EDAX spectra were collected from multiple regions of each sample to confirm elemental composition and verify the successful incorporation of Ni at the intended concentrations. The EDAX spectrum of undoped WO₃ (Figure 3a) showed only the characteristic W Mα, W Lα, and O Kα emission lines, with the measured W:O atomic ratio of approximately 1:3.0 confirming stoichiometric WO₃ composition and the absence of contaminant elements within the detection limit of the technique (~0.1 at.%) [11].

The EDAX spectra of the Ni-doped samples (Figures 3b–3e) clearly revealed the progressive emergence and growth of the Ni Lα (0.85 keV) and Ni Kα (7.47 keV) emission lines with increasing doping concentration. The quantified Ni atomic percentages measured from the EDAX data were 2.3 ± 0.3%, 4.8 ± 0.4%, 6.9 ± 0.5%, and 9.7 ± 0.6% for the nominally 2.5%, 5%, 7%, and 10% Ni–WO₃ samples, respectively [11]. The close correspondence between measured and nominal values, within experimental analytical uncertainty, validates the efficacy of the hydrothermal synthesis protocol in incorporating the dopant at the designed concentrations. No emission lines attributable to sodium, chlorine, or other process-derived contaminants were detected, confirming thorough washing of the products.

The W:O atomic ratios in the doped samples deviated slightly from the ideal 1:3 stoichiometry, showing values in the range 1:2.85–2.92, which is consistent with the creation of oxygen vacancies as a charge-compensating defect mechanism for the aliovalent Ni²⁺ substitution at W⁶⁺ sites [12]. This charge-compensation mechanism is well established for aliovalent doping in metal oxides and has important implications for the electronic properties of the material: oxygen vacancies introduce localised electron states in the band gap that enhance electronic conductivity and increase the carrier density available for charge transfer at the electrode–electrolyte interface.

Figure 3:  EDAX spectra of (a) pure WO₃, (b) 2.5% Ni–WO₃, (c) 5% Ni–WO₃, (d) 7% Ni–WO₃, and (e) 10% Ni–WO₃ samples confirming the progressive incorporation of Ni into the WO₃ host matrix. Inset values indicate measured Ni atomic percentages.

3.4 FTIR Spectral Analysis

The FTIR spectra of the undoped and Ni-doped WO₃ samples recorded in the range 400–1100 cm⁻¹ are presented in Figure 4. The spectrum of undoped WO₃ exhibited two principal absorption features characteristic of the monoclinic gamma-WO₃ structure: a broad, intense band centred near 800 cm⁻¹ assigned to the doubly bridging W–O–W stretching vibration involving oxygen atoms shared between adjacent WO₆ octahedra, and a sharper band near 950 cm⁻¹ attributed to the terminal W=O stretching mode of non-bridging terminal oxygen atoms [13]. A broad absorption feature below 550 cm⁻¹ arises from W–O bending modes. These assignments are fully consistent with the monoclinic WO₃ structure characterised by corner-sharing WO₆ octahedra with varying degrees of octahedral distortion imposed by the cooperative tilting pattern of the monoclinic phase.

Upon Ni doping at 2.5% (Figure 4b), the principal W–O–W and W=O stretching bands were retained at positions close to those in undoped WO₃, confirming that the fundamental WO₃ octahedral framework was preserved [13,34]. However, a systematic broadening of both the W–O–W bridging band and the terminal W=O band was observed, the degree of broadening increasing with Ni content. Band broadening in doped materials of this type is typically attributed to the distribution of local bond-length environments created by the introduction of dopant atoms at W sites: the varied W–O–W bond angles and W=O bond lengths in the vicinity of Ni²⁺ substitution sites produce a distribution of vibrational frequencies, manifested as inhomogeneous broadening of the infrared absorption band [34].

At 5% and 7% Ni doping (Figures 4c and 4d), a discernible red-shift of the terminal W=O stretching frequency was observed, from approximately 950 cm⁻¹ in undoped WO₃ to approximately 940 cm⁻¹ at 7% Ni. This red-shift indicates a reduction in the W=O bond force constant in the vicinity of Ni-substituted sites, consistent with electron density redistribution arising from the introduction of the less electronegative Ni²⁺ into the W⁶⁺ coordination environment. A weak feature near 460 cm⁻¹, which intensified with increasing Ni content and was most clearly resolved in the 7% and 10% Ni spectra, is tentatively assigned to a Ni–O stretching mode contribution, indicating the formation of a Ni–O local bonding environment within the WO₃ matrix [35].

The 10% Ni sample (Figure 4e) exhibited continued evolution of these features, with the W–O–W band further broadened and slightly red-shifted relative to the 7% sample. The increasing degree of spectral perturbation with Ni content is fully consistent with the increasing lattice strain and crystallite size reduction established by the XRD analysis and corroborates the conclusion that Ni is genuinely incorporated as a substitutional dopant rather than merely adsorbed on the particle surfaces.

Figure 4:  FTIR spectra of (a) pure WO₃, (b) 2.5% Ni–WO₃, (c) 5% Ni–WO₃, (d) 7% Ni–WO₃, and (e) 10% Ni–WO₃ in the wavenumber range 400–1100 cm⁻¹.

3.5 UV–Visible Spectroscopy and Optical Band Gap

UV–Visible absorption spectra and corresponding Tauc plots for the undoped and Ni-doped WO₃ series are presented in Figure 5. The absorption spectrum of undoped WO₃ displayed a well-defined absorption onset in the near-UV region, with the Tauc plot analysis yielding an optical band gap of approximately 2.65 eV, consistent with values reported in the literature for monoclinic gamma-WO₃ and reflecting the indirect semiconductor character of the undoped material [1,15]. This relatively wide band gap implies a low intrinsic charge carrier density and a correspondingly high electronic resistivity in the undoped state.

Progressive Ni doping produced a systematic red-shift of the optical absorption onset towards longer wavelengths, indicating a monotonic narrowing of the optical band gap with increasing Ni content. The Tauc-derived band gap values were approximately 2.57 eV at 2.5% Ni, 2.48 eV at 5% Ni, 2.35 eV at 7% Ni, and 2.22 eV at 10% Ni [2,15]. This systematic band gap narrowing with increasing Ni concentration is attributed to two complementary mechanisms. First, the introduction of Ni²⁺ energy levels within the WO₃ band gap creates intermediate electronic states that enable optical transitions at sub-gap photon energies, effectively reducing the measured optical gap [14]. Second, the increasing density of oxygen vacancies created as charge-compensating defects for the Ni²⁺ substitution contributes additional intra-gap donor states located just below the WO₃ conduction band minimum, further narrowing the onset of strong optical absorption.

The sub-gap absorption region, corresponding to photon energies below the fundamental WO₃ band gap, showed progressively increasing absorption intensity with Ni content, which is directly attributable to the growing density of Ni-related and oxygen vacancy-related intra-gap states. This broadband sub-gap absorption is practically beneficial because it implies a higher density of electronic states accessible at room temperature, translating into improved electronic conductivity and lower charge-transfer resistance at the electrode–electrolyte interface. The reduction of the optical band gap from 2.65 eV (undoped) to 2.35 eV (7% Ni) represents a 11% narrowing, corresponding to a substantial increase in the thermally excited carrier density at room temperature [15].

Figure 5:  UV–Visible absorption spectra and (inset) Tauc plots for (a) pure WO₃, (b) 2.5% Ni–WO₃, (c) 5% Ni–WO₃, (d) 7% Ni–WO₃, and (e) 10% Ni–WO₃ samples

3.6 Electrochemical Analysis: Cyclic Voltammetry

The cyclic voltammograms of undoped and Ni-doped WO₃ electrodes recorded at 5 mV s⁻¹ in 1 M Na₂SO₄ aqueous electrolyte are shown in Figure 6. The CV profile of undoped WO₃ displayed broad, asymmetric redox features superimposed on a quasi-rectangular background, consistent with a combination of electrical double-layer capacitance and pseudocapacitive contributions from the multi-step W⁶⁺/W⁵⁺ and W⁵⁺/W⁴⁺ redox transitions occurring reversibly in the accessible potential window [17,18]. The multi-peak character of the WO₃ CV profile reflects the well-known two-phase intercalation mechanism of WO₃, in which Na⁺ ions insert into the WO₃ tunnel structure in distinct structural transitions corresponding to different degrees of intercalation.

Upon Ni doping at 2.5% (Figure 6b), the enclosed CV area increased perceptibly relative to the undoped sample, indicating a higher specific capacitance, and the redox features became somewhat more symmetric and better-defined. These observations are consistent with the improved electronic conductivity implied by the band gap narrowing and the increased active grain boundary area arising from the crystallite size reduction confirmed by XRD. The 5% Ni–WO₃ electrode (Figure 6c) showed a further increase in CV area and more pronounced redox humps, reflecting the growing contribution of Ni²⁺/Ni³⁺ pseudocapacitive transitions supplementing the intrinsic WO₃ intercalation redox activity [19].

The 7% Ni–WO₃ electrode (Figure 6d) delivered the highest specific capacitance among the series at all measured scan rates. The CV profile at 5 mV s⁻¹ showed a substantially enlarged area relative to the undoped sample, well-defined redox peaks corresponding to the combined W⁶⁺/W⁵⁺ and Ni²⁺/Ni³⁺ faradaic processes, and good profile shape retention with increasing scan rate up to 100 mV s⁻¹. The retention of profile shape at high scan rates indicates that the improved electronic conductivity (from band gap narrowing) and the reduced ion diffusion distances (from smaller particles and higher porosity) together enable the electrode to sustain the required charge and ion transport rates even at high current densities [19,20]. The calculated specific capacitance at 5 mV s⁻¹ followed the order: 7% Ni–WO₃ > 5% Ni–WO₃ > 10% Ni–WO₃ > 2.5% Ni–WO₃ > undoped WO₃.

The 10% Ni–WO₃ electrode (Figure 6e) showed a slightly reduced specific capacitance relative to the 7% sample, consistent with the partial agglomeration observed in the FE-SEM micrographs and the slight crystallite size increase at 10% Ni in the XRD data. Agglomeration reduces the effective electrode–electrolyte contact area and lengthens the mean ion diffusion path length, partially counteracting the benefits of higher dopant concentration. With increasing scan rate from 2 to 100 mV s⁻¹ for all samples, the CV profiles retained their general shape but the redox humps became progressively less resolved, characteristic of diffusion-limited intercalation pseudocapacitance in which the rate of ionic diffusion within the electrode cannot keep pace with the rate of potential sweep [21,22].

Figure 6:  Cyclic voltammograms of (a) pure WO₃, (b) 2.5% Ni–WO₃, (c) 5% Ni–WO₃, (d) 7% Ni–WO₃, and (e) 10% Ni–WO₃ electrodes recorded at 5 mV s⁻¹ in 1 M Na₂SO₄ aqueous electrolyte (potential window: −0.2 to +0.8 V vs. SCE). The progressive increase in enclosed CV area with doping up to 7% indicates enhancement of specific capacitance.