Investigation of Substrate and Deposition Temperature on Mo–Ni–Cr Thin Films for Alkaline Hydrogen Evolution Reaction

Abstract

In this work, ternary Mo–Ni–X (X = Al, Co, Cr, Cu, Fe, W) thin films with nominal composition Mo₈₀Ni₁₀X₁₀ (at. %) were prepared by magnetron sputtering and evaluated as electrocatalysts for the hydrogen evolution reaction (HER) in alkaline media. The influence of alloy composition, substrate type, and deposition temperature on catalytic performance was systematically investigated. Electrochemical screening revealed a strong dependence of HER activity on both substrate conductivity and ternary alloying, with Al-, Cr-, and W-containing systems showing the best performance on glassy carbon substrates. This highlights the importance of interfacial charge-transfer efficiency in determining catalytic behavior. The Mo₈₀Ni₁₀Cr₁₀/GC system was selected for detailed analysis. Deposition temperatures ≥ 500 °C resulted in enhanced HER activity, reaching an overpotential of η₁₀ = −222 mV at j = −10 mA cm⁻². The improved performance is attributed to temperature-induced microstructural optimization and electrochemically driven surface reconstruction, leading to the formation of a Ni-enriched active interface. AFM analysis confirmed surface restructuring during operation, with roughness increasing from ~1 to ~3 nm, indicating the formation of additional electrochemically accessible active sites. XPS results suggest partial depletion of Mo during cycling, while Cr mainly contributes to structural stabilization of the evolving thin film. Overall, the results demonstrate that HER performance is governed by the coupled effects of alloy composition, substrate-dependent charge transport, and in situ surface reconstruction. This work highlights magnetron sputtering as a scalable approach for designing homogeneous noble-metal-free thin-film electrocatalysts with tunable activity.

1. Introduction

In the context of the ongoing global energy transition, the extensive reliance on non-renewable fossil fuels has led to severe environmental challenges, including greenhouse gas emissions and accelerating climate change. Consequently, the development of sustainable and carbon-neutral energy carriers has become a critical scientific and technological priority [1].

Among various emerging solutions, hydrogen production via water electrolysis powered by renewable energy sources represents one of the most promising pathways toward a clean energy economy. In this process, the hydrogen evolution reaction (HER) is a key cathodic half-reaction, in which water molecules or protons are reduced to molecular hydrogen, directly determining the overall efficiency of electrolytic systems [1].

Despite the maturity of alkaline water electrolyzers, their performance is still strongly governed by the catalytic materials used at the cathode. State-of-the-art systems rely on Pt-based catalysts, which exhibit outstanding HER activity but are hindered by high cost and scarcity, limiting large-scale deployment [2]. Therefore, the development of earth-abundant, non-noble metal electrocatalysts remains a central research focus [3,4].

Among transition-metal-based systems, Ni-based catalysts have attracted sustained attention due to their intrinsic activity in alkaline media, stability, and tunable electronic structure. Numerous studies have demonstrated that Ni-based materials can achieve high HER activity through compositional and morphological engineering [5,6,7,8,9,10,11,12,13,14], focusing on mechanistic understanding in alkaline media together with morphology and compositional optimization.

In particular, Ni–Mo systems have emerged as a benchmark class of non-precious electrocatalysts, often exhibiting markedly enhanced performance compared to monometallic counterparts. Recent experimental and theoretical studies have shown that the high activity of NiMo alloys originates from a complex interplay between water dissociation and hydrogen adsorption energetics. For instance, it has been demonstrated that oxidized or reconstructed Ni₃Mo-type surfaces can facilitate the Volmer step via Mo-assisted water dissociation, while Ni sites optimize hydrogen binding energy, resulting in a cooperative dual-site mechanism rather than a single active phase interpretation. This insight has helped clarify long-standing controversies regarding the origin of activity in Ni–Mo systems [15,16,17,18,19]. Representative nanoporous Ni–Mo architectures have been reported to achieve ultralow overpotentials (as low as 10–15 mV at 10 mA cm⁻²), highlighting the exceptional potential of this materials class [20]. Recent theoretical and experimental studies indicate that such performance enhancements originate from electronic structure modulation at the atomic scale rather than from a generic synergistic effect alone [21]. Charge redistribution between neighboring metal atoms, defect-induced electronic perturbations, and optimized adsorption energetics of reaction intermediates have been identified as decisive factors governing HER activity. These findings emphasize that rational catalyst design requires consideration of both composition and local electronic environment [21].

Beyond binary systems, molybdenum-containing catalysts offer additional advantages due to their flexible electronic structure, corrosion resistance, and strong affinity for hydrogen intermediates [22,23,24]. Building upon these properties, the incorporation of a third metallic element (X = Al, Co, Cr, Cu, Fe, W) into Mo–Ni systems has emerged as an effective strategy to further tune electronic structure, optimize adsorption energetics, and stabilize surface intermediates. The choice of Al, Co, Cr, Cu, Fe, and W was primarily guided by available phase-diagram data and alloy-forming tendencies with Ni and Mo. The selected elements enable the formation of stable or metastable alloy structures suitable for thin-film deposition, while simultaneously providing a broad range of chemical and structural effects that may influence HER performance [15,16,22,23].

Such ternary and multicomponent alloying strategies are closely related to recent developments in high-entropy alloys (HEAs), single-atom catalysts (SACs), and multi-site catalytic systems, where synergistic interactions between different metal species enable enhanced control over catalytic pathways [25,26,27,28].

In parallel, recent in situ and operando studies have revealed that HER catalysts often undergo dynamic surface reconstruction under reaction conditions, emphasizing that the true active phase may differ significantly from the as-prepared material. These findings highlight the importance of surface-sensitive characterization and mechanistic understanding at the atomic level [27,28].

Despite these advances, the fundamental role of the third metal in Mo–Ni–X systems remains insufficiently understood, particularly in terms of electronic redistribution, strain effects, and site-specific reaction energetics [16,29].

Many studies [30,31,32,33,34] have focused on powder catalysts through conventional wet-chemical synthesis routes, leading to compositional inhomogeneity and limited structural control.

In contrast, physical vapor deposition (PVD) techniques, particularly plasma-assisted magnetron sputtering, enable the fabrication of homogeneous thin films with precisely controlled composition, thickness, and microstructure [35]. However, systematic investigations of Mo–Ni–X thin films prepared via PVD remain relatively scarce.

Equally, the influence of substrate properties on catalytic performance is often overlooked. Substrate surface roughness and electronic interactions can significantly affect film adhesion, active site distribution, and charge-transfer kinetics [36].

In this work, we employ magnetron sputtering to fabricate a series of Mo–Ni–X (X = Al, Co, Cr, Cu, Fe, W) ternary thin-film catalysts. Special attention is given to the Mo₈₀Ni₁₀Cr₁₀ system, which remains relatively unexplored. The films (~30 nm) are deposited on different substrates, including glassy carbon (GC), nickel foil, and graphite, to systematically evaluate substrate–catalyst interactions. Electrochemical performance is investigated in 1 M NaOH, focusing on HER activity, charge-transfer behavior, and interfacial effects. Furthermore, the Mo₈₀Ni₁₀Cr₁₀ system is studied in detail to elucidate the influence of deposition temperature on morphology evolution, surface reconstruction, catalytic activity, and long-term electrochemical stability, aiming to establish atomic-level structure–property relationships in multicomponent alkaline HER electrocatalysts.

2. Results and Discussion

2.1. Initial Electrochemical Screening for Optimization of Mo₈₀Ni₁₀X₁₀ Thin Film Preparation and Testing Conditions

In the initial screening (Figure 1), the hydrogen evolution reaction (HER) activity of Mo₈₀Ni₁₀X₁₀ thin films (X = Al, Co, Cr, Cu, Fe, W) (confirmed by EDX) was evaluated on three substrates (glassy carbon (GC), nickel foil (Ni- foil), and graphite (C)) and at two deposition temperatures (20 °C and 800 °C). The obtained overpotential data (η₁₀) demonstrate that HER performance is governed by three experimentally independent variables: deposition temperature, substrate type, and alloy composition [35,36,37,38,39].

Figure 1. Overpotential (η₁₀) of Mo₈₀Ni₁₀X₁₀ thin films deposited on GC, Ni foil, and graphite at 20 °C and 800 °C. (a) MoNiAl, (b) MoNiCo, (c) MoNiCr, (d) MoNiCu, (e) MoNiFe, (f) MoNiW. Bar plots with error bars.

Across all investigated compositions and substrates, films deposited at 800 °C consistently exhibit lower overpotentials compared to those prepared at 20 °C (Figure 1). This trend is observed for all Mo–Ni–X systems, indicating that deposition temperature is a universal controlling parameter of catalytic performance. Quantitatively, the improvement is reflected in a systematic shift in η₁₀ toward lower values for high-temperature-deposited films, which is further supported by reduced charge-transfer resistance in subsequent impedance analysis (Figure 2, Figure 3 and Figure 4).

Figure 2. Electrochemical characterization of Mo₈₀Ni₁₀X₁₀ thin films deposited on GC at 800 °C: (a) polarization curves, (b) Tafel plots, (c) Nyquist plots, (d) η₁₀ comparison with error bars, (e) correlation between η₁₀ and Rct, and (f) dependence of Rct on η₁₀.

Figure 3. Electrochemical characterization of Mo₈₀Ni₁₀X₁₀ thin films deposited on Ni foil at 800 °C: (a) polarization curves, (b) Tafel plots, (c) Nyquist plots, (d) η₁₀ comparison with error bars, (e) correlation between η₁₀ and Rct, and (f) dependence of Rct on η₁₀.

Figure 4. Electrochemical characterization of Mo₈₀Ni₁₀X₁₀ thin films deposited on graphite at 800 °C: (a) polarization curves, (b) Tafel plots, (c) Nyquist plots, (d) overpotential (η₁₀) comparison with error bars, (e) correlation between overpotential and charge transfer resistance (Rct), and (f) dependence of charge transfer resistance on overpotential.

A clear substrate dependence is observed for all compositions (Figure 1, Figure 2, Figure 3 and Figure 4). Among the investigated supports, GC consistently yields the lowest overpotentials and the most favorable electrochemical response. For GC-supported films, η₁₀ values fall in the range −192 to −241 mV (Figure 2a,d), whereas Ni foil-supported samples show significantly higher overpotentials, with most systems exceeding −260 mV (Figure 3d). Graphite exhibits intermediate behavior with η₁₀ values around −203 to −205 mV for the most active compositions (Figure 4a). This trend is further confirmed by electrochemical impedance spectroscopy, where GC-supported catalysts exhibit the lowest charge-transfer resistance (Rct ≈ 14–15 Ω for the most active systems), while Ni foil shows significantly higher values (>300 Ω), indicating strongly hindered interfacial charge transport. The Rct values were extracted from EIS spectra by fitting with Randles and Henderson equivalent circuit models, as shown in Figure 5.

Figure 5. Randles (a) and Henderson–Armstrong (b) equivalent circuit models used for fitting EIS data and extraction of Rct for Mo₈₀Ni₁₀X₁₀ thin films [15].

The alloying element X also plays a decisive role in determining HER activity. Among all investigated compositions, Mo₈₀Ni₁₀Al₁₀, Mo₈₀Ni₁₀Cr₁₀, and Mo₈₀Ni₁₀W₁₀ consistently show the lowest overpotentials across all substrates. On GC at 800 °C, these systems reach η₁₀ values of −192 mV, −227 mV, and −217 mV, respectively, while Mo₈₀Ni₁₀Fe₁₀ exhibits systematically inferior performance. Tafel analysis (63–78 mV dec⁻¹ for GC-supported systems) indicates a Volmer–Heyrovsky mechanism, with Mo₈₀Ni₁₀Cr₁₀ exhibiting the lowest slope (63 mV dec⁻¹), suggesting faster charge-transfer kinetics.

A direct correlation between η₁₀ and Rct is observed for all systems (Figure 2f, Figure 3f and Figure 4f), confirming that HER activity is primarily governed by interfacial electron-transfer kinetics rather than solely by surface area effects. Ni foil-supported catalysts show systematically suppressed activity despite identical composition and deposition conditions, indicating that the substrate imposes a dominant kinetic limitation on charge transfer. In contrast, graphite provides intermediate behavior, partially preserving intrinsic catalytic trends while still limiting full performance compared to GC.

Overall, the screening results (Table 1 and Table 2) demonstrate that HER activity in Mo₈₀Ni₁₀X₁₀ thin films is governed by a hierarchical structure–property relationship: deposition temperature controls structural activation, substrate determines charge-transfer efficiency, and alloying element tunes intrinsic catalytic behavior. The combined analysis confirms that the observed catalytic performance arises from the interplay between intrinsic alloy properties and extrinsic interfacial transport limitations.

Table 1. Overpotential (η₁₀), Tafel slope (b), and exchange current density (j₀) of Mo₈₀Ni₁₀X₁₀ thin films on glassy carbon, Ni foil, and graphite at 800 °C.

Alloy (at 800 °C)	η10 (mV)	b (mV dec−1)	j0 (mA cm−2)	η10 (mV)	b (mV dec−1)	j0 (mA cm−2)	η10 (mV)	b (mV dec−1)	j0 (mA cm−2)
	Sample on GC			Sample on Ni foil			Sample on C
Mo80Ni10Al10	−192	78	3.45 × 10−2	−278	72	1.37 × 10−3	−217	68	6.67 × 10−3
Mo80Ni10Co10	−241	70	3.43 × 10−3	−270	108	3.16 × 10−2	−239	125	1.31 × 10−1
Mo80Ni10Cr10	−227	63	2.32 × 10−3	−257	89	1.26 × 10−2	−204	115	1.68 × 10−1
Mo80Ni10Cu10	−247	74	4.73 × 10−3	−248	56	4.21 × 10−4	−207	92	6.06 × 10−2
Mo80Ni10Fe10	−277	69	9.75 × 10−4	−287	72	1.06 × 10−3	−272	154	1.49 × 10−1
Mo80Ni10W10	−217	78	1.38 × 10−2	−278	106	2.49 × 10−2	−203	107	1.29 × 10−1

Table 2. EIS-derived parameters (Rs, Rct, Rp) and equivalent circuit models (Henderson–Armstrong and Randles, see Figure 5) for Mo₈₀Ni₁₀X₁₀ thin films on different substrates.

Alloy (at 800 °C)	Rs (Ω)	Rct (Ω)	Rp (Ω)	Electrical Circuit	Rs (Ω)	Rct (Ω)	Rp (Ω)	Electrical Circuit	Rs (Ω)	Rct (Ω)	Rp (Ω)	Electrical Circuit
	Sample on GC				Sample on Ni foil				Sample on C
Mo80Ni10Al10	15.08	33.50	49.92	(H-A)	40.25	433.5	-	(R)	11.07	55.51	-	(R)
Mo80Ni10Co10	14.97	34.47	175.5	(H-A)	14.84	430.6	-	(R)	17.09	86.84	-	(R)
Mo80Ni10Cr10	14.85	34.26	70.86	(H-A)	44.12	339.6	-	(R)	8.301	27.98	46.63	(H-A)
Mo80Ni10Cu10	21.55	34.81	165.1	(H-A)	34.87	306.2	-	(R)	9.491	56.94	-	(R)
Mo80Ni10Fe10	31.06	55.24	362.1	(H-A)	22.63	857.0	-	(R)	11.32	447.6	-	(R)
Mo80Ni10W10	14.49	34.80	70.33	(H-A)	18.57	433.3	-	(R)	14.77	15.68	45.37	(H-A)

2.2. Standardized Evaluation of HER Activity Based on ECSA

For a more reliable comparison of catalytic performance, the HER activity of Mo₈₀Ni₁₀X₁₀ thin films was additionally evaluated after normalization to the electrochemically active surface area (ECSA). This approach allows separation of intrinsic catalytic trends from differences arising from surface area variations among substrates [22].

To determine ECSA, a 20 nm Pt thin film was deposited on GC, Ni foil, and C. The electrochemical response was recorded in the potential range from −0.2 to 0.8 V vs. Ag/AgCl (Figure 6a–c). The resulting cyclic voltammograms show that carbon-based substrates (GC and C) exhibit higher capacitive currents compared to Ni foil, indicating a larger electrochemically accessible surface area. In contrast, Ni foil shows a significantly lower capacitive response, suggesting reduced electrochemically active surface development under identical conditions.

Figure 6. CV-derived ECSA evaluation using Pt reference films (a–c) and ECSA-normalized HER polarization curves of Mo₈₀Ni₁₀X₁₀ thin films on different substrates (d–f).

After normalization of HER polarization curves to ECSA (Equation (2), Figure 6d–f), distinct differences between substrates become evident. On GC, the relative activity trends between individual Mo–Ni–X compositions are preserved, indicating that intrinsic differences between alloys remain detectable after surface-area correction. In contrast, on Ni foil, these differences are strongly attenuated, and all investigated systems converge toward similar apparent activity. This behavior indicates that Ni foil imposes a limitation on the effective utilization of the catalytic surface, which affects the apparent electrochemical response.

C exhibits intermediate behavior, where partial differentiation between alloy compositions is still observable after normalization; however, the separation of catalytic trends remains less pronounced compared to GC.

A direct comparison of overpotentials (η₁₀) after ECSA normalization (Figure 7) confirms the trends observed in the raw polarization data (Figure 2, Figure 3 and Figure 4). GC consistently provides the lowest η₁₀ values across most compositions, while Ni foil shows the highest values. C remains in an intermediate regime between these two extremes.

Figure 7. Substrate-dependent evaluation of HER activity for Mo₈₀Ni₁₀X₁₀ thin films prepared at 800 °C: overpotential (η₁₀) values for different alloy compositions with error bars.

Importantly, Mo₈₀Ni₁₀Al₁₀, Mo₈₀Ni₁₀Cr₁₀, and Mo₈₀Ni₁₀W₁₀ maintain the highest relative activity after ECSA normalization, confirming that their enhanced performance is not solely related to geometric surface effects. The corresponding η₁₀ values are summarized in Table 3.

Table 3. Summary of overpotential (η₁₀) values at j = −10 mA cm⁻² (ECSA-normalized current density) for Mo₈₀Ni₁₀X₁₀ thin films (800 °C) deposited on different substrates in 1 M NaOH for the hydrogen evolution reaction (HER).

Alloy (at 800 °C)	η10 (mV) at jECSA = −10 mA cm−2	η10 (mV) at jECSA = −10 mA cm−2	η10 (mV) at jECSA = −10 mA cm−2
	Sample on GC	Sample on Ni foil	Sample on C
Mo80Ni10Al10	−203	−275	−251
Mo80Ni10Co10	−249	−264	−350
Mo80Ni10Cr10	−237	−254	−276
Mo80Ni10Cu10	−258	−245	−271
Mo80Ni10Fe10	−286	−283	−370
Mo80Ni10W10	−230	−271	−276

Overall, the ECSA-normalized analysis confirms that the substrate plays a decisive role not only in determining the apparent catalytic performance, but also in controlling the extent to which intrinsic differences between Mo–Ni–X compositions are expressed in electrochemical measurements.

2.3. Investigation of Mo–Ni–Cr-Based Alloys for Alkaline Water Electrolysis

The next part of this study focused on the detailed investigation of Mo–Ni–Cr-based alloys for alkaline HER. The selection of the Mo–Ni–Cr system was based on a combination of electrochemical performance considerations and practical material constraints. Although the Mo₈₀Ni₁₀W₁₀/GC system exhibits a slightly lower overpotential (η₁₀ = −217 mV) compared to Mo₈₀Ni₁₀Cr₁₀/GC (η₁₀ = −227 mV) at 800 °C, this difference lies within the typical experimental variability observed for sputtered thin-film electrocatalysts. Therefore, both systems can be considered comparable in terms of catalytic activity.

From a materials perspective, chromium was selected as a more practically relevant alloying element due to its lower cost compared to tungsten and its well-established contribution to corrosion resistance in alkaline environments. In addition, Mo–Ni–Cr thin-film systems have been significantly less explored in the context of magnetron-sputtered HER catalysts, which provides additional motivation for a detailed investigation of this composition.

A series of Mo₈₀Ni₁₀Cr₁₀ thin films was therefore prepared on GC substrates at different deposition temperatures (20, 100, 300, 500, and 800 °C), and their electrochemical performance is summarized in Figure 8.

Figure 8. Electrochemical characterization of Mo₈₀Ni₁₀Cr₁₀ thin films deposited on GC at substrate temperatures ranging from 20 °C to 800 °C: (a) polarization curves, (b) Tafel plots, (c) Nyquist plots, (d) η₁₀ on deposition temperature, (e) Tafel slope values for individual samples, and (f) correlation between η₁₀ and Rct with error bars.

The polarization curves (Figure 8a) demonstrate a clear dependence of HER activity on deposition temperature. A systematic shift toward lower overpotentials is observed with increasing temperature above 300 °C, with η₁₀ values improving from −251 mV (low-temperature samples) to −222 mV (500 °C). The best performance is achieved for the sample prepared at 500 °C, which shows the lowest overpotential among all investigated conditions. The corresponding trend is summarized in Figure 8d, confirming the reproducibility of the temperature-dependent behavior.

Tafel analysis (Figure 8b,e and Table 4) further reveals that all samples follow a Volmer–Heyrovsky mechanism, with Tafel slopes ranging from 69 to 124 mV dec⁻¹. The lowest value (69 mV dec⁻¹), observed for the 500 °C sample, indicates the most favorable reaction kinetics among the investigated conditions.

Table 4. Summary of electrochemical parameters for Mo₈₀Ni₁₀Cr₁₀ thin films deposited on GC at different substrate temperatures, including η₁₀, b, j₀, Rs, and Rct.

	η10 (mV)	b (mV dec−1)	j0 (mA cm−2)	Rs (Ω)	Rct (Ω)	n	CPE
Mo80Ni10Cr10
at 20 °C	−306	124	1.04 × 10−1	25.58 (3.14%)	119.8 (3.76%)	7.93 × 10−5	0.83494
						(5.02%)	(1.38%)
at 100 °C	−284	102	3.38 × 10−2	27.78 (1.38%)	79.60 (1.64%)	19.4 × 10−5	0.77015
						(2.89%)	(0.67%)
at 300 °C	−251	99	3.80 × 10−3	22.31 (1.12%)	48.34 (1.08%)	17.5 × 10−5	0.77362
						(8.02%)	(0.43%)
at 500 °C	−222	69	2.34 × 10−3	27.34 (1.50%)	44.43 (1.85%)	12.9 × 10−5	0.82939
						(3.32%)	(0.64%)
at 800 °C	−237	83	2.39 × 10−3	31.73 (1.09%)	44.62 (1.10%)	0.0005	0.08530
						(4.26%)	(0.84%)

EIS (Figure 8c) shows a systematic decrease in Rct with increasing deposition temperature. For samples prepared at ≥300 °C, significantly smaller Nyquist semicircles are observed, corresponding to reduced Rct values. The lowest values are obtained for samples prepared at 300 °C (48.34 Ω), 500 °C (44.43 Ω), and 800 °C (44.62 Ω). This trend is consistent with the polarization and Tafel results, indicating improved interfacial charge-transfer kinetics at elevated deposition temperatures.

The overall electrochemical trends observed across polarization, Tafel, and impedance analyses consistently indicate that deposition temperature is a key parameter governing the kinetic performance of the Mo₈₀Ni₁₀Cr₁₀ system in alkaline HER conditions.

2.4. Surface Chemical Analysis of the Mo₈₀Ni₁₀Cr₁₀ Catalyst

High-resolution X-ray photoelectron spectroscopy (XPS) was employed to investigate the surface chemical states of the Mo₈₀Ni₁₀Cr₁₀ catalyst prepared at 500 °C and to monitor changes induced by electrochemical cycling under alkaline HER conditions. Figure 9 presents the high-resolution Mo 3d, Ni 2p, and Cr 2p spectra, while Figure 10 shows the corresponding survey spectra before and after electrochemical activation.

Figure 9. High-resolution XPS spectra of Mo 3d (a) and Ni 2p (b) core levels for Mo₈₀Ni₁₀Cr₁₀ thin film.

Figure 10. XPS survey spectra of Mo₈₀Ni₁₀Cr₁₀ thin film before and after electrochemical cycling.

The Mo 3d spectra exhibit three distinct chemical states of molybdenum, with each Mo 3d₅/₂ component accompanied by a corresponding 3d₃/₂ peak, separated by a spin–orbit splitting of approximately 3.1 eV [40,41,42,43]. In the as-prepared state, three components are observed at 228.34, 229.70, and 232.40 eV, assigned to Mo⁰, Mo⁴⁺, and Mo⁶⁺, respectively. The presence of metallic Mo indicates a partially reduced surface, while oxidized species originate from surface passivation under ambient conditions. After electrochemical cycling, the oxidation states remain unchanged; however, a pronounced decrease in overall Mo signal intensity is observed in both high-resolution and survey spectra. This confirms a reduction in surface Mo concentration during HER operation, consistent with partial leaching or redistribution of Mo species under alkaline conditions. Importantly, the preservation of binding energies suggests that the remaining Mo species maintain their chemical environment, although at reduced surface coverage.

The Ni 2p spectrum of the as-prepared catalyst consists of contributions from Ni⁰ (852.5 eV), Ni²⁺ (NiO, 853.5 eV), and Ni(OH)₂ (856.21 eV) [44,45,46], together with characteristic satellite features confirming the presence of Ni²⁺ species. After electrochemical cycling, a clear redistribution of surface Ni species is observed. The relative contribution of Ni(OH)₂ increases, while metallic Ni and NiO components are partially suppressed. A weak high-binding-energy feature (~857.2 eV) is also detected, which may be tentatively associated with NiOOH-like species; however, its presence is more likely related to partial surface oxidation occurring during post-mortem exposure to air, as NiOOH is not thermodynamically stable under cathodic HER conditions [47]. Overall, the Ni 2p results indicate the formation of a hydroxide-enriched Ni surface layer while preserving a fraction of metallic Ni, suggesting the coexistence of metallic and hydroxylated Ni species after activation.

Due to the low signal intensity and strong multiplet splitting of Cr(III) species, reliable quantitative deconvolution of the Cr 2p spectra was not possible. The overlapping nature of Cr 2p₃/₂ and Cr 2p₁/₂ signals limits a precise assignment of individual chromium chemical states. Nevertheless, the presence of Cr in oxidized form is confirmed in both survey and high-resolution spectra. These limitations are typical for Cr-containing thin films and have been reported in similar systems [44,48,49], preventing further quantitative interpretation of Cr speciation in the present study.

For selected Mo₈₀Ni₁₀Cr₁₀ systems deposited on a GC substrate, stability tests including chronoamperometry, chronopotentiometry, and multicycling were performed, as shown in Figure 8. All measurements were carried out in an alkaline electrolyte.

Survey spectra confirm the presence of Mo, Ni, Cr, O, and C on the catalyst surface in the as-prepared state. The O 1s and C 1s signals are attributed to surface oxidation and adventitious carbon contamination. After HER cycling, a clear change in surface composition is observed, characterized by a decrease in Mo signal intensity and a relative increase in Ni contribution. This trend indicates preferential surface depletion or redistribution of Mo species under alkaline electrochemical conditions, accompanied by the formation of a Ni-enriched surface region [50]. Such behavior is commonly observed in Mo–Ni-based systems exposed to alkaline HER environments and is associated with dynamic surface reconstruction processes.

Overall, the combined XPS results demonstrate that the Mo₈₀Ni₁₀Cr₁₀ catalyst undergoes significant surface reconstruction during HER operation. The post-activation surface consists of a mixed Ni⁰/Ni(OH)₂ matrix with reduced but still detectable Mo content and oxidized Cr species. Rather than forming a completely Ni-dominated surface, the data suggest a dynamically evolving multicomponent interface, where Ni provides the primary hydrogen evolution sites, while residual Mo and Cr likely contribute to electronic and structural modulation of the surface. This dynamic restructuring consists of the observed electrochemical activation behavior and supports the role of surface evolution in governing catalytic performance.

2.5. Durability Test of Mo₈₀Ni₁₀Cr₁₀/GC

For the selected Mo₈₀Ni₁₀Cr₁₀ catalyst deposited on GC substrate, long-term stability was evaluated using chronoamperometric measurements in alkaline electrolyte (Figure 11). The current response was monitored over 12,000 s at a constant potential of −1.3 V vs. Ag/AgCl.

Figure 11. Electrochemical stability evaluation of Mo₈₀Ni₁₀Cr₁₀/GC catalysts, including (a) chronoamperometric response, (b) polarization curves before and after durability testing, (c) Nyquist plots before and after testing, and (d) comparison of key electrochemical parameters.

The chronoamperometric curve (Figure 11a) shows an initial decrease in current density during the early stage of the measurement. This behavior is commonly attributed to surface stabilization processes, including restructuring of the catalyst surface and the formation of steady-state active sites under cathodic polarization [51]. After this initial activation period, the current stabilizes and remains nearly constant throughout the remainder of the experiment, confirming stable catalytic operation under prolonged HER conditions.

The initial 3 h stability window, used as a comparative benchmark for rapid durability screening, revealed a relatively low degradation of jGEO, with a decrease of approximately 11%. This value is lower than that observed for the Mo–Ni–Al system (≈17%), while significantly outperforming a reference Ni-based catalyst, which exhibited a substantial decline of up to ~55% under identical conditions. These results confirm the improved intrinsic stability of the Mo–Ni–Cr composition compared to both benchmark ternary alloys and monometallic Ni systems.

To further evaluate structural stability, post-stability electrochemical characterization was performed (Figure 11b). A comparison of polarization curves before and after the chronoamperometric test shows only a minor shift in overpotential, from −222 mV to −241 mV at j = −10 mA cm⁻². The relatively small change (19 mV) indicates good preservation of catalytic activity after prolonged electrochemical operation. This conclusion is further supported by electrochemical impedance spectroscopy, where only a slight increase in charge-transfer resistance is observed (from ~44 Ω to ~48 Ω), suggesting that the interfacial electron-transfer properties remain largely intact after durability testing.

Overall, these results demonstrate that the Mo₈₀Ni₁₀Cr₁₀/GC catalyst exhibits a favorable combination of high HER activity and good electrochemical stability in alkaline media.

To correlate electrochemical stability with compositional evolution, post-mortem EDX analysis was performed (Table 5). After durability testing, measurable changes in the relative elemental composition of Mo, Ni, and Cr were observed. The most pronounced variation was detected for molybdenum, indicating partial redistribution or leaching of Mo species during prolonged electrochemical operation.

Table 5. Energy-dispersive X-ray spectroscopy (EDX) results for as-prepared Mo_xNi_γCr_z thin films and after electrochemical stability testing.

	Mo (at. %)	Ni (at. %)	Cr (at. %)
As-prepared	78.82	10.74	10.44
	After chronoamperometry
at 500 °C	72.93	9.84	17.23

This behavior suggests that the stability of the multicomponent film is governed not only by its initial stoichiometry but also by the evolution of its microstructure under electrochemical conditions. In particular, higher deposition temperatures appear to promote improved incorporation and retention of Mo within the film matrix, thereby mitigating compositional drift during HER operation [16].

These findings are consistent with the surface reconstruction phenomena identified by XPS (Section 2.4), confirming that the Mo–Ni–Cr catalyst undergoes dynamic but stable restructuring under alkaline HER conditions, while maintaining its overall catalytic performance.

2.6. Morphological and Structural Characterization

The surface morphology of the prepared Mo₈₀Ni₁₀Cr₁₀ catalysts was investigated using atomic force microscopy (AFM) (Figure 12). The upper row (Figure 12a–e) shows as-prepared films deposited at different substrate temperatures (20–800 °C), while the lower row (Figure 12f–j) presents the corresponding samples after electrochemical testing.

Figure 12. AFM topography of Mo₈₀Ni₁₀Cr₁₀ thin films deposited at different substrate temperatures (20–800 °C), shown before (a–e) and after cyclic voltammetry (f–j) testing.

The AFM images of the as-prepared films (Figure 12a–e) reveal a relatively smooth and homogeneous surface at lower deposition temperatures. With increasing deposition temperature, particularly at 800 °C (Figure 12e), the formation of larger grains and island-like surface features becomes evident, indicating a transition in the thin-film growth mechanism toward a more pronounced three-dimensional growth mode. This evolution reflects enhanced adatom mobility and increased atomic rearrangement during high-temperature deposition.

After electrochemical cycling, all samples exhibit pronounced morphological changes (Figure 12f–j). The surface becomes significantly rougher, accompanied by the formation of nanoscale defects, local restructuring, and surface fragmentation. These changes are especially evident for samples deposited at higher temperatures [52,53]. Importantly, rather than indicating degradation, these transformations are consistent with electrochemically induced surface reconstruction, which is commonly associated with the formation of newly exposed active sites during HER [53].

The quantitative evolution of surface roughness (Rq) as a function of deposition temperature is shown in Figure 13. In the as-prepared state, all samples exhibit low roughness values (0.96–2.2 nm), which increase systematically with deposition temperature, reaching the maximum for the 800 °C sample. After electrochemical activation, a substantial increase in roughness is observed for all samples, with values approaching ~3 nm. The increase is most pronounced for films deposited at ≥500 °C, which correlates well with their superior electrocatalytic performance (Figure 8), suggesting a direct relationship between thermally induced microstructure and electrochemically activated surface evolution.

Figure 13. Dependence of Rq of Mo₈₀Ni₁₀Cr₁₀ thin films on deposition temperature before and after HER, with error bars (a). The corresponding Rq values are shown in a bar graph (b) with error bars.

Overall, the AFM results demonstrate that the catalyst surface undergoes dynamic restructuring under HER conditions. This process can be interpreted as surface activation rather than degradation and is consistent with the reconstruction phenomena identified by XPS (Section 2.4), where preferential Mo depletion and Ni enrichment were observed. The formation of a roughened, defect-rich surface is therefore considered beneficial for enhancing the density of catalytically active sites.

Transmission electron microscopy (TEM) analysis of the Mo₈₀Ni₁₀Cr₁₀ sample deposited at 500 °C further reveals the internal microstructure of the catalyst (Figure 14). A representative high-resolution image (Figure 12a) shows isolated nanocrystalline domains with an average size of approximately 10 nm embedded in a disordered matrix. Low-magnification TEM (Figure 14b) confirms a uniform distribution of these nanostructures within a continuous background.

Figure 14. TEM and SAED analysis of Mo₈₀Ni₁₀Cr₁₀ (500 °C): (a) isolated nanoparticle (10 nm); (b) low-magnification TEM image showing uniform nanoparticle dispersion (yellow square indicates the region shown in (a)); (c) SAED pattern confirming a nanocrystalline bcc phase embedded in an amorphous matrix. Red markings indicate reflections from crystallographic planes.

Selected area electron diffraction (SAED) (Figure 14c) exhibits a combination of diffuse halo features and distinct Debye–Scherrer rings. The diffuse halo indicates the presence of an amorphous phase, while the indexed diffraction rings correspond to a body-centered cubic (bcc) structure with reflections assigned to the (101), (200), and (211) planes. The calculated lattice parameter is a = 0.31 ± 0.01 nm. These results confirm a dual-phase nanostructure consisting of nanocrystalline bcc grains embedded in an amorphous matrix, which is known to facilitate a high density of structural defects and potentially beneficial electronic heterogeneity for electrocatalysis.

2.7. Comparison with Reported Results in Literature

Due to the limited availability of directly comparable studies on ternary Mo–Ni–Cr thin films prepared by magnetron sputtering, the electrocatalytic performance of the Mo₈₀Ni₁₀Cr₁₀/GC system was benchmarked against representative Ni–Mo-based and multicomponent HER electrocatalysts operating in alkaline media (Table 6).

Table 6. Comparison of the electrochemical HER performance of the Mo₈₀Ni₁₀Cr₁₀/GC thin film prepared by magnetron sputtering with selected Ni–Mo- and ternary Ni-based electrocatalysts reported in the literature.

Catalyst	Class	Preparation Method	Electrolyte	η10 (mV)	b (mV dec−1)	Mechanistic Insight	Reference
Mo/Ru–Mo2C/MoO2	Multi-site catalyst	Interface engineering	1 M KOH	19 @ 10 mA cm−2	~30–60	Spatially separated Volmer/Tafel sites	[28]
Ni–Mo	Binary alloy	Electrodeposition	1 M KOH	250–300	70–120	Ni/Mo synergistic sites (Volmer–Heyrovsky)	[54]
Ni–Fe–Cr	Multicomponent alloy	Sintering	30 wt.% KOH	260–320	n.a.	Composition-induced electronic tuning	[55]
Ni–Mo–Fe	Ternary alloy	Electrodeposition	1 M KOH	240–290	n.a.	Multimetal electronic modulation	[56]
O-covered Ni3Mo(111)	Reconstructed alloy surface	Theoretical/operando	Alkaline	~100–150 (effective)	~70–120	Dual-site (Mo: water dissociation, Ni: H adsorption)	[57]
Mo80Ni10Cr10/GC (this work)	Ternary thin film	Magnetron sputtering	1 M NaOH	222	69	Alloying + substrate-driven charge transfer + surface reconstruction	This work

The obtained overpotential of η₁₀ = −222 mV at j = −10 mA cm⁻² places the prepared catalyst within the typical performance range of conventional Ni–Mo-derived systems. For example, Ni–Mo catalysts exhibit overpotentials of approximately −250 to −300 mV, attributed to synergistic interactions between Ni and Mo active sites [54]. Similarly, Ni–Fe–Cr and Ni–Mo–Fe systems generally operate in the range of −240 to −320 mV, confirming the performance window of multicomponent Ni-based alloys [55,56].

More recent studies, however, indicate that HER activity is governed not only by macroscopic alloy synergy but also by dynamic surface reconstruction and the formation of catalytically relevant interfacial structures. In Ni–Mo systems, for instance, reconstructed alloy–oxide interfaces have been identified as the true active phase, where Mo facilitates water dissociation, and Ni optimizes hydrogen adsorption energetics, forming a cooperative dual-site mechanism [41,57]. These findings highlight the importance of considering operando surface evolution rather than static bulk composition.

Beyond conventional alloy systems, advanced catalyst architectures such as multi-site and interface-engineered materials demonstrate significantly enhanced HER kinetics by spatially separating elementary reaction steps. A representative example is a Mo/Ru-based catalyst containing Ru/Mo₂C/MoO₂ interfacial sites, which achieves overpotentials as low as 19 mV at 10 mA cm⁻² by accelerating both Volmer and Tafel steps through distinct active centers [28]. This illustrates the upper performance limit of deliberately engineered heterointerface catalysts.

Similarly, single-atom and defect-engineered systems further confirm that catalytic activity is highly sensitive to local coordination environments and can evolve under operating conditions. Operando studies show that dynamic reconstruction of metal–oxygen coordination motifs plays a key role in defining the active phase and reaction kinetics [27,58].

In this context, the Mo₈₀Ni₁₀Cr₁₀/GC system (η₁₀ = −222 mV, b = 69 mV dec⁻¹) demonstrates competitive performance among Ni–Mo–X catalysts, particularly considering its compact thin-film morphology without porous architectures or high-surface-area scaffolds. The results suggest that its activity arises from a combination of alloy-induced electronic modulation, efficient substrate-mediated charge transport, and electrochemically induced surface reconstruction.

This is further supported by the experimental observations in this work, including (i) substrate-dependent charge-transfer behavior (Section 2.1 and Section 2.2), (ii) surface reconstruction evidenced by AFM and XPS (Section 2.4, Section 2.5 and Section 2.6), and (iii) stable long-term operation under alkaline HER conditions (Section 2.5). Together, these findings indicate that catalytic performance is governed by a dynamic equilibrium between alloy composition, interfacial electron transfer, and operando surface restructuring.

3. Experimental Section

3.1. Preparation of the Thin Films

Thin films were prepared by magnetron sputtering using an Orion 8 deposition system (AJA International, Scituate, MA, USA). Alloys with the nominal composition Mo₈₀Ni₁₀X₁₀ (X = Al, Co, Cr, Cu, Fe, and W) were deposited by co-sputtering from metallic targets: Mo (99.99%), Ni (99.99%), and the corresponding alloying element X (99.99%).

The sputtering process (Figure 15) was carried out in an argon atmosphere at a working pressure of 3 mTorr. High-purity argon (99.999%, Messer, Bad Soden am Taunus, Germany) was supplied at a flow rate of 18 cm³ min⁻¹. The substrate temperature was maintained at either 20 °C or 800 °C during deposition. The deposition time and sputtering power were adjusted to obtain a film thickness of approximately 30 nm. The films were deposited onto three different substrates: glassy carbon (Sigradur G, HTW Hochtemperatur-Werkstoffe GmbH, Thierhaupten, BA, Germany), nickel foil (Alfa Aesar, Haverhill, MA, USA), and graphite plates (Fuel Cell Store, College Station, TX, USA). Prior to deposition, all substrates were ultrasonically cleaned in acetone and ethanol and subsequently dried in air.

Figure 15. Schematic illustration of the preparation of Mo–Ni–X thin films by magnetron co-sputtering.

3.2. Materials Characterization

The elemental composition of the deposited films was determined using energy-dispersive X-ray spectroscopy (EDX) attached to a TESCAN VEGA scanning electron microscope (SEM, VEGA3, TESCAN, Brno, Czech Republic) operating at an accelerating voltage of 30 kV.

Surface topography and roughness were analyzed by atomic force microscopy (AFM, Dimension Icon, Bruker, Billerica, MA, USA) operating in tapping mode. The root-mean-square roughness (Rq) values were calculated from the acquired AFM images.

The microstructure of selected samples was investigated using transmission electron microscopy (TEM, JEM-2100F UHR, JEOL Ltd., Tokyo, Japan) operated at 200 kV. Crystallographic information was obtained from selected-area electron diffraction (SAED) patterns.

X-ray photoelectron spectroscopy (XPS) measurements were performed using a PHOIBOS 100 hemispherical analyzer (SPECS GmbH, Berlin, Germany) equipped with a Mg Kα X-ray source operated at 200 W. The base pressure during analysis was maintained at approximately 1 × 10⁻⁸ mbar. High-resolution spectra were recorded for the Mo 3d, Ni 2p, Cr 2p, O 1s, and C 1s regions.

3.3. Electrochemical Measurement

Electrochemical measurements were performed using a three-electrode cell connected to a potentiostat/galvanostat (PGSTAT302N, Metrohm Autolab B.V., Utrecht, The Netherlands).

The sputtered thin film deposited on the corresponding substrate served as the working electrode. A gold wire was used as the counter electrode, while an Ag/AgCl electrode (3 M KCl) served as the reference electrode. All measurements were carried out in 1 M NaOH prepared from analytical-grade sodium hydroxide and deionized water.

Prior to each experiment, the electrolyte was purged with argon (99.999%, Messer, Bad Soden am Taunus, Germany) for 30 min.

HER activity was evaluated by cyclic voltammetry in the potential range from −0.9 to −1.7 V vs. Ag/AgCl. All measured potentials were converted to the reversible hydrogen electrode (RHE) scale according to:

$${\mathrm{E}}_{\mathrm{R}\mathrm{H}\mathrm{E}}={\mathrm{E}}_{\mathrm{A}\mathrm{g}/\mathrm{A}\mathrm{g}\mathrm{C}\mathrm{l}}+0.222+0.059\times \mathrm{p}\mathrm{H}$$

(1)

where 0.222 V is the standard potential of the Ag/AgCl reference electrode.

Electrochemical impedance spectroscopy (EIS) measurements were performed under HER conditions. The impedance spectra were fitted using equivalent electrical circuits based on the Randles and Henderson–Armstrong models to determine the solution resistance (R_s), charge-transfer resistance (Rct), and polarization resistance (Rp) [16].

In alkaline media, the HER is generally described by the Volmer–Heyrovsky mechanism involving electrochemical water dissociation (Volmer step) followed by electrochemical hydrogen desorption (Heyrovsky step) [15,16].

The electrochemically active surface area (ECSA) was determined using Pt reference films (20 nm) deposited on identical substrates. Cyclic voltammograms were recorded in 0.1 M HClO₄ within the potential range −0.2 to 0.8 V vs. Ag/AgCl. ECSA values were calculated according to

$$\mathrm{E}\mathrm{C}\mathrm{S}\mathrm{A}={\displaystyle \frac{{\mathrm{Q}}_{\mathrm{H}}}{210\times {10}^{-4}\times {\mathrm{m}}_{\mathrm{P}\mathrm{t}}}}$$

(2)

where Q_H is the charge associated with hydrogen adsorption, 210 × 10⁻⁴ C m⁻² is the charge required for monolayer hydrogen adsorption on Pt, and m_Ptis the mass of deposited Pt [22].

4. Conclusions

In this work, ternary Mo–Ni–X (X = Al, Co, Cr, Cu, Fe, and W) thin films were prepared by magnetron sputtering and evaluated as electrocatalysts for the hydrogen evolution reaction in alkaline media. The results demonstrate that both deposition temperature and substrate type strongly affect the morphology, electrochemical properties, and catalytic performance. Among the investigated systems, alloys containing Al, Cr, and W showed the highest activity, with the best performance obtained for films deposited on glassy carbon. Detailed analysis of a novel Mo₈₀Ni₁₀Cr₁₀ system revealed that deposition temperatures of 500 °C and above lead to enhanced HER activity and excellent long-term stability, which is associated with electrochemically induced surface reconstruction and an increased density of active sites. These findings highlight the importance of microstructural control and substrate selection in the design of efficient HER catalysts and identify Mo–Ni–Cr thin films as promising candidates for alkaline water electrolysis.