Short-Term Annealing Effect on Hydrogen Evolution Activity of Amorphous Al₈₇Y₄Gd₁Ni₄Fe₄

Highlights

What are the main findings?

• Heat treatment at 647±2 K forms Al(Gd,Ni,Y,Fe), AlFe2Ni and GdFe2 phases.
• Short annealing boosts HER via catalytic AlFe2Ni and GdFe2 nanophases.
• Surface reconstruction and Ni enrichment increase H* recombination rate.

What are the implications of the main findings?

• Thermal treatment enables controlled formation of active catalytic phases.
• Short annealing is an effective route to enhance HER performance.
• Ni-rich reconstructed surfaces promote faster hydrogen evolution.

Abstract

This work investigates the structural evolution and electrocatalytic activity of the amorphous metal alloy Al₈₇Y₄Gd₁Ni₄Fe₄ during short-term annealing and its effect on the kinetics of the hydrogen evolution reaction (HER) in 1 M KOH. It is shown that a 5 min heat treatment at 647 ± 2 K initiates controlled nanocrystallisation with the formation of AlFe₂Ni, GdFe₂ and Al(X) (X = Gd, Ni, Y, Fe) phases, which are uniformly dispersed in the amorphous matrix. According to XRD, DSC and HRTEM data, it was established that the formation of intermetallic nanodomains leads to a decrease in charge transfer energy barriers and the appearance of additional active centres of H* adsorption. Electrochemical studies have shown an increase in cathode current density, an increase in i₀ by 2–3 orders of magnitude, and a decrease in R_ct after annealing, confirming the improvement in HER kinetics. Potentiostatic tests showed an increase in the volumetric hydrogen evolution rate from 35.1 to 106.0 mL/(g·min) during the first immersion and up to 217.9 mL/(g·min) during reuse. SEM/EDS analysis revealed surface reconstruction and Ni enrichment after HER, which contributes to the acceleration of the H* recombination stage. The synergy of the amorphous matrix and nanophases ensures high electrocatalytic activity and stability of the system, making annealed AMA a promising low-cost catalyst for alkaline hydrogen evolution.

1. Introduction

Hydrogen energy is considered a key area for the development of green technologies, and efficient hydrogen evolution reaction (HER) remains one of the critical stages in hydrogen technologies [1,2,3,4,5]. Among the various hydrogen production technologies, electrochemical water splitting is considered the simplest and most environmentally friendly way to produce hydrogen on a large scale [3]. However, the overall efficiency of this process depends significantly on the kinetics of the hydrogen evolution reaction (HER), which requires efficient and stable electrocatalysts to minimise overvoltage and increase the rate of hydrogen generation [4,5,6].

It is known that platinum-based materials demonstrate the best catalytic activity for HER due to their nearly optimal hydrogen adsorption energy, but their high cost and limited availability complicate large-scale application [7]. Therefore, the development of inexpensive catalysts from common metals with comparable activity and long-term stability remains an important research task [8]. Over the past decade, significant progress has been made through the development of transition metal compounds such as sulphides, phosphides, carbides and MOF-based hybrids [9,10]. Recently, a separate class of electrocatalysts has emerged: amorphous metal alloys, also known as metal glasses (MGs), due to their unique atomic and electronic structures [11,12,13].

Hydrogen generation processes through the interaction of aluminium and its alloys with water in alkaline environments are considered a promising direction in alternative energy; however, their efficiency is largely limited by the passivation of the aluminium surface with an oxide film [14]. Aluminium in Al-5Co and Al-10Ni alloys corrodes under the action of NaOH, releasing heat and OH⁻ ions, which promotes further corrosion; Ni and Co form microgalvanic cells with Al, accelerating hydrogen evolution; and active Co catalyzes the decomposition of water, generating ⋅OH, H*, and O radicals. The authors of the article [15] considered overcoming passivation of aluminium alloying with transition metals, in particular Ni and Co, which are capable of forming cathodically active intermetallic phases and disrupting the integrity of the oxide layer. Microstructural analysis indicates a uniform distribution of intermetallic inclusions Al₃Ni and Al₉Co₂ in the aluminium matrix, which creates the prerequisites for the formation of microgalvanic pairs [16]. Article [16] discusses in detail the effect of alloying elements, such as Ni, Co, In, Zn, and Sn, on the kinetics of anodic dissolution of aluminium and the stability of the passive film. In article [17] the authors showed that the efficiency of aluminium anodes in alkaline electrolytes is determined by a complex synergistic interaction between the internal microstructure of the material (distribution of intermetallic phases, grain size, segregation of alloying elements), the chemical composition and pH of the electrolyte, the concentration of hydroxide ions, the presence of inhibitors, as well as the conditions of electrochemical loading (current density, temperature, polarization duration). Doping with transition metals increases the catalytic activity of cathodic reactions, while elements with a low melting point contribute to the destruction of the oxide film and the transition of the anode to the active state [18]. The synergistic effect of multicomponent alloying is manifested in reducing the anodic overvoltage and increasing the operating voltage of electrochemical systems [19].

AMAs are characterized by the absence of long-range order, uniform distribution of atoms and isotropic surface properties [20]. Such structural irregularity forms a large number of coordination-unloaded centres and atomic-scale heterogeneities, which act as active centres for hydrogen adsorption and evolution [21]. Unlike crystalline materials, where catalytic activity is determined by grain boundaries and lattice defects, amorphous alloys exhibit a high density of catalytically active centres uniformly distributed over the surface [12]. The metallic type of bonds provides excellent electrical conductivity, while the homogeneous amorphous matrix increases corrosion resistance in acidic and alkaline electrolytes [22]. Various amorphous systems have demonstrated excellent HER performance. For example, Ni–P, Co–P, and Fe–B metallic glasses have shown low overpotential (η10 < 100 mV) and high stability during prolonged electrolytic treatment [23]. Similarly, amorphous Ni–Mo–S and Co–W–P catalysts demonstrated improved hydrogen evolution kinetics compared to their crystalline counterparts [24]. In aluminium amorphous alloys, partial replacement of Al with transition and rare earth elements leads to significant structural and electronic changes, which provides increased catalytic activity and improved chemical stability [25,26]. The synergy between the Al matrix and the doping components promotes electron transfer and improves the adsorption of intermediate hydrogen, which accelerates the Volmer–Heyrovsky HER stages [27].

Heat treatment of AMAs additionally allows their catalytic properties to be adjusted. Short-term annealing near the onset of crystallisation causes the formation of nanocrystalline domains in the amorphous matrix, creating heterogeneous interfaces that increase charge transport and surface reactivity [28,29]. Such ‘controlled nanocrystallisation’ increases the availability of catalytic centres while maintaining the stability of the amorphous structure. Studies have shown that annealed amorphous Ni–P and Fe–B alloys exhibit increased exchange current density and reduced Tafel slopes after partial crystallisation [30,31]. This approach allows the structural and electronic states of amorphous alloys to be tuned without compromising corrosion resistance. Compared to crystalline catalysts, amorphous alloys are characterised by greater structural flexibility, mechanical stability and corrosion resistance, which is critical for long-term operation in alkaline environments [32]. Their synthesis by rapid cooling or electro-reduction is relatively simple, economical, and scalable [33]. The absence of grain boundaries minimises localised corrosion and ensures uniform activity across the entire electrode surface. These properties make AMAs one of the most promising non-precious catalysts for efficient hydrogen evolution.

In this context, amorphous alloys based on aluminium modified with transition and rare earth elements are of particular scientific interest. The combination of high specific electrical conductivity of the aluminium matrix with the catalytic activity of RE and TM components creates the conditions for a synergistic effect during the hydrogen evolution reaction. Transition metals (Ni, Fe) promote the formation of electron-active centres and facilitate the recombination stage of atomic hydrogen, while rare earth elements (Y, Gd) stabilise the amorphous structure and increase the corrosion resistance of the alloy in an alkaline environment [34,35,36,37].

The additional effect of short-term thermal annealing, which leads to partial nanocrystallisation, can significantly change the local electronic structure and distribution of active centres on the surface. Therefore, the aim of this work is to investigate how short-term annealing at the temperature of the second stage of nanocrystallisation affects the corrosion resistance and activity in the hydrogen evolution reaction of the amorphous alloy Al₈₇Y₄Gd₁Ni₄Fe₄ in 1 M KOH aqueous solution.

2. Materials

Amorphous tapes composed of Al₈₇Y₄Gd₁Ni₄Fe₄ were obtained by melt spinning at the G.V. Kurdyumov Institute of Metal Physics of the National Academy of Sciences of Ukraine. The starting metals (purity ≥ 99.9%) were pre-melted in high-purity argon in an induction melting furnace. The melt was ejected onto the surface of a copper wheel rotating at a linear speed of approximately 40 m s⁻¹, providing a cooling rate higher than 10⁵ K s⁻¹. The obtained ribbons had a typical thickness of 25–35 μm and a width of 2–3 cm. The amorphous structure was confirmed by X-ray diffraction (XRD).

3. Methods

3.1. Thermal Treatment

To initiate partial nanocrystallisation, the tape was subjected to short-term annealing in a furnace at 647 ± 2 K for 5 min in an air atmosphere. The annealing temperature was determined based on the results of DSC analysis, which showed that the specified range corresponds to the second stage of exothermic crystallisation. After annealing, the samples were cooled in air.

3.2. Structural Analysis

The thermal stability of the alloy was determined by differential scanning calorimetry (DSC) at heating rates of 10, 15, and 20 K/min (Perkin Elmer Pyris 1, 710 Bridgeport Ave Shelton, CT 06484, USA). Phase analysis was performed by X-ray diffraction (XRD) using a PANalytical Empyrean Diffractometer with Cu-Kα radiation (λ Kα1 = 1.5418Å) and the PIXcell detector, Almelo, The Netherlands). The XRD parameters were as follows: 2θ range from 20 to 120°, step size 0.03°, scan speed 1◦ per minute. Phase analysis was done basing on the HighScore Plus PANalytical software integrated with the ICDD PDF4 + 2025 crystallographic database.

3.3. Surface Morphology and Elemental Analysis

The surface morphology was studied using scanning electron microscopy (SEM) (Tescan Brno s.r.o. Tescan Vega 3 LMU, 623 00 Brno, Czech Republic). The chemical composition and distribution of elements were determined by energy dispersive spectroscopy (EDS), including mapping of the elements O, Al, Fe, Ni, Y, and Gd. The analysis was performed for the initial and annealed samples before and after electrochemical testing.

3.4. Electrochemical Analysis

Electrochemical experiments were carried out in 1 M KOH at 298 K using a standard three-electrode cell and a PARstat 2273 (Revision 2273) potentiostat. The amorphous alloy electrode with a geometric surface area of 1 cm² served as the working electrode, while a platinum plate (surface area of 2.5 cm²) and an Ag/AgCl electrode (3 M KCl) were used as the counter and reference electrodes, respectively. All measured potentials were recalculated with respect to the hydrogen electrode (HE). The active surface of the tape was cleaned with distilled water and isopropanol.

Potentiodynamic polarisation curves were recorded at a rate of 50 mV/s. The Tafel equation was used to determine the parameters E_corr, i_corr and the Tafel slopes by analysing the corresponding regions of the curves. The catalytic parameters of HER (i₀, E_HER, b) were obtained by approximating the cathodic polarisation dependences in the range of 10 mA·cm⁻². The measured potentials of the reversible hydrogen electrode (RHE) were reported according to the Nernst equation: E_RHE = E (Ag/AgCl) + 0.210 + 0.059 pH [37]. Electrochemical impedance spectroscopy was performed in the frequency range of 100 kHz–10 mHz at an amplitude of 5 mV. The resulting EIS data were fitted using an equivalent circuit, as described in Section 4.2, which consisted of the solution resistance (R_s), film resistance (R_f), charge-transfer resistance (R_ct), and the corresponding constant phase elements (CPE).

3.5. Potentiostatic Hydrogen Evolution and Quantification

Potentiostatic experiments were performed at two potentials, (i) the corrosion potential (E_corr) and (ii) the hydrogen evolution potential (E_HER), which were determined from the voltammetric curves presented in Section 4.2. The time-dependent current density was recorded to assess the stability and kinetics of gas evolution. The total charge (Q) was obtained by integrating the current over time (Q = ∫i dt), and the amount of hydrogen evolved was calculated using Faraday’s law. The time-dependent hydrogen evolution current density was measured in potentiostatic mode for both the initial and 5 min-annealed Al₈₇Y₄Gd₁Ni₄Fe₄ amorphous metallic alloys in 1 M KOH during the first and repeated use of the electrode. The volumetric hydrogen evolution rate (V, mL g⁻¹ min⁻¹) was determined from the cathodic current and normalized by the sample mass and reaction time [3].

4. Results and Discussion

4.1. Structural Characterization and Thermal Analysis

Analysis of the diffractograms of the amorphous alloy Al₈₇Y₄Gd₁Ni₄Fe₄ (Figure 1) shows the presence of a broad diffuse maximum, a typical feature of the amorphous state.

Figure 1. (a) Profiles of the main maxima of amorphous alloys Al₈₇Y₄Gd₁Ni₄Fe₄ (black line—experiment data, red line—fitted, green line—separation of three phases by the Lorentz approximation method); (b) HREM images of the initial amorphous alloy Al₈₇Y₄Gd₁Ni₄Fe₄.

Figure 1. (a) Profiles of the main maxima of amorphous alloys Al₈₇Y₄Gd₁Ni₄Fe₄ (black line—experiment data, red line—fitted, green line—separation of three phases by the Lorentz approximation method); (b) HREM images of the initial amorphous alloy Al₈₇Y₄Gd₁Ni₄Fe₄.

Analysis of the diffraction curve showed that this amorphous alloy is also characterised by a micro-heterogeneous atomic structure, as seen in the HREM image (Figure 2b).

In the small scattering angle region (2θ≈20°), an additional maximum is clearly visible, indicating the existence of intermediate (average) atomic ordering in the amorphous alloy alongside the nearest-neighbour ordering. It can be assumed that the existence of this maximum confirms the cluster structure of the alloy and is the result of intercluster interference of X-rays. The interatomic distance determined by the position of the maximum (R₁ = 0.566 nm) can be considered as the average distance between the centres of clusters formed from rare earth elements (Y, Gd) and surrounded by Al atoms.

The separation of maximum (27–52°) into two components (AP₂, AP₃) made it possible to identify two local regions of short-range order with different structural organization parameters (Table 1). The main component AP₂ (2θ = 38.3°, S₁ = 26.7 nm⁻¹) accounts for approximately 78% of the alloy volume and is characterized by a relatively larger interatomic distance (R₁ = 0.289 nm) and a short coherent scattering radius (L ≈ 1.4 nm). These parameters indicate the predominance of loosely packed Al-rich clusters, which determine the chemical activity of the amorphous matrix. The second maximum AP₃ (2θ = 43.3.3°, s¹ = 30.0 nm⁻¹) has a lower intensity (32.2 of the volume) and a reduced average interatomic distance (R₁ = 0.258 nm), which corresponds to locally denser regions enriched with transition metals Fe and Ni and partially with rare earth atoms Gd and Y. The slightly larger size of coherent regions (L ≈ 1.5 nm) in this fraction indicates a higher degree of short-range order, which may facilitate the nucleation of nanocrystals upon heating. Thus, the combination of two types of local structures—(i) loosened Al-dominant areas (AP₂) and (ii) dense Fe/Ni/Gd-enriched domains (AP₃)—forms the microstructural heterogeneity characteristic of multicomponent amorphous alloys [36,37]. This heterogeneity determines the characteristics of nanocrystallisation and subsequent electrochemical behavior. In further experiments, it is precisely the AP₂ regions that function as potential nuclei for intermetallic phases, while the AP₃ regions serve as an amorphous matrix that ensures chemical reactivity and ease of dissolution of Al during corrosion-electrochemical processes. This complementarity of structural domains is a prerequisite for the combined hydrogen evolution mechanism, which combines chemical decomposition of water and electrochemical recombination of protons on the Fe/Ni-enriched surface.

Table 1. Structural parameters of amorphous alloys.

Amorphous Phase	2θ°	S, nm−1	I/Io	FWHM	R, nm	L, nm
(AP)1	19.3	13.65	61.8	6.70	0.566	0.9
(AP)2	38.3	26.70	100.0	4.60	0.289	1.4
(AP)3	43.3	30.00	32.2	4.25	0.258	1.5

2θ—the angular position of the maxima; S—the modulus of the wave vector corresponding to the maximum scattering intensity; I/I₀—the relative scattering intensity; FWHM—the angular half-width of the maximum; L—the size of the coherent scattering regions; R—the average interatomic distance; X—the volume fraction of the phase.

Table 1. Structural parameters of amorphous alloys.

Amorphous Phase	2θ°	S, nm−1	I/Io	FWHM	R, nm	L, nm
(AP)1	19.3	13.65	61.8	6.70	0.566	0.9
(AP)2	38.3	26.70	100.0	4.60	0.289	1.4
(AP)3	43.3	30.00	32.2	4.25	0.258	1.5

2θ—the angular position of the maxima; S—the modulus of the wave vector corresponding to the maximum scattering intensity; I/I₀—the relative scattering intensity; FWHM—the angular half-width of the maximum; L—the size of the coherent scattering regions; R—the average interatomic distance; X—the volume fraction of the phase.

DSC curves obtained at heating rates of 10–20 K/min (Figure 2) show five consecutive exothermic peaks corresponding to five-stage crystallisation (1 pre peak, 2–3 peaks and 4–5 post peaks). The first thermal effect (≈450–550 K) reflects the nucleation of fcc-Al nanocrystals [32,38], while the second (≈640–660 K) reflects the formation of solid solution phases and Al(X) (X = Gd, Ni, Y, Fe).

So, the selected temperature for short-term annealing at T₃ (647 ± 2 K) corresponds to the completion of the second stage of nanocrystallisation, during which the optimal number of active centres is formed without significant loss of the amorphous component.

Figure 2. (a,b) DSC-curves (β = 10, 15, 20 K/min) of AMA Al₈₇Y₄Gd₁Ni₄Fe₄ (a) in the temperature range 323–900 K and (b) in the temperature range 625–650 K.

Figure 2. (a,b) DSC-curves (β = 10, 15, 20 K/min) of AMA Al₈₇Y₄Gd₁Ni₄Fe₄ (a) in the temperature range 323–900 K and (b) in the temperature range 625–650 K.

From the DSC curves, the phase transition temperatures were determined and are given in Table 2.

Table 2. Characteristic temperatures (T ± 2 K) of the second maxima of the maxima on the DSC-curves of AMAs based on aluminium at different heating rates (β).

AMA	β	T1 ± 2, K	T2 ± 2, K	T3 ± 2, K
Al87Y4Gd1Ni4Fe4	10	625	636	642
	15	630	639	644
	20	633	642	647

Table 2. Characteristic temperatures (T ± 2 K) of the second maxima of the maxima on the DSC-curves of AMAs based on aluminium at different heating rates (β).

AMA	β	T1 ± 2, K	T2 ± 2, K	T3 ± 2, K
Al87Y4Gd1Ni4Fe4	10	625	636	642
	15	630	639	644
	20	633	642	647

After short-term annealing (5 min at 647 ± 2 K), peaks appear on the diffractogram (Figure 3), corresponding to the nucleation of Al(X) (X = Gd, Ni, Y, Fe) nanocrystal-line phases, as well as intermetallic inclusions of the AlFe₂Ni and GdFe₂ type. The formation of these nanocrystals does not destroy the amorphous matrix but instead forms a two-phase structure, an amorphous–crystalline composite. This structure is thermodynamically more stable and provides efficient electron transport between phases during electrochemical reactions. It is assumed that Al(X) nanocrystals act as electron donors with high s-electron mobility, while the AlFe₂Ni and GdFe₂ phases containing d-elements form local zones with high density of states near the Fermi level.

Figure 3. Diffractogram of Al₈₇Y₄Gd₁Ni₄Fe₄ of annealed 5 min at T₃ = 647 ± 2 K, (CuKα).

Figure 3. Diffractogram of Al₈₇Y₄Gd₁Ni₄Fe₄ of annealed 5 min at T₃ = 647 ± 2 K, (CuKα).

Table 3 shows the parameters of the crystal lattices of the phases formed during the AMA crystallisation process of AMA Al₈₇Y₄Gd₁Ni₄Fe₄.

Table 3. The unit cell parameters and size of crystallites and lattice strains obtained for the nano-and microstructural main phases identified in AMA Al₈₇Y₄Gd₁Ni₄Fe₄.

	Al(X), X = Gd, Y, Ni, Fe	GdFe2
Theoretical data: ICDD PDF4 + card number Space group Lattice parameters	01-089-2837 Fm-3 m, cubic a = 4.0592(1) Å	04-004-0708 Fd-3 m, cubic a = 7.3940(3) Å
	a = 4.0732(1) Å D = 370 Å η = 0.12%	a = 7.4967(9) Å D = 300 Å η = 0.37%

Table 3. The unit cell parameters and size of crystallites and lattice strains obtained for the nano-and microstructural main phases identified in AMA Al₈₇Y₄Gd₁Ni₄Fe₄.

	Al(X), X = Gd, Y, Ni, Fe	GdFe2
Theoretical data: ICDD PDF4 + card number Space group Lattice parameters	01-089-2837 Fm-3 m, cubic a = 4.0592(1) Å	04-004-0708 Fd-3 m, cubic a = 7.3940(3) Å
	a = 4.0732(1) Å D = 370 Å η = 0.12%	a = 7.4967(9) Å D = 300 Å η = 0.37%

The image in Figure 4a shown a homogeneous amorphous contrast without signs of periodic lattice stripes, which is typical for a completely amorphous state of the alloy. This is consistent with the results of X-ray diffraction (Figure 1), where only a broad diffuse maximum without crystal phase peaks was detected. According to the structural analysis (Table 1), the amorphous matrix contains two local regions of short-range order: Al-rich regions (AP₂) with a larger interatomic distance and Fe/Ni-enriched domains (AP₃) with a higher degree of short-range order. It is the AP₂ type regions that act as potential nucleation centres for nanocrystallisation during heating.

After short-term annealing at 647 ± 2 K for 5 min (Figure 4b), nanoscale crystalline inclusions with clearly defined lattice stripes appear in the structure, indicating the onset of nanocrystallisation. These changes are consistent with the DSC data (Figure 2), where the first exothermic effect (≈450–550 K) corresponds to the nucleation of fcc-Al nanocrystals [32], and the second (≈640–660 K) corresponds to the formation of intermetallic phases. The annealing temperature used (647 ± 1 K) coincides with the second stage of the exothermic process, ensuring controlled devitrification.

The HRTEM images obtained confirm the data of X-ray structural analysis after annealing (Figure 3), where diffraction peaks appear, belonging to the Al(X) (X = Gd, Ni, Y, Fe), AlFe₂Ni and GdFe₂ phases.

Figure 4. High-resolution TEM images of amorphous metal alloys Al₈₇Y₄Gd₁Ni₄Fe₄ of the initial (a) and annealed at 5 min (b) at T₃ = 647 ± 2 K.

Figure 4. High-resolution TEM images of amorphous metal alloys Al₈₇Y₄Gd₁Ni₄Fe₄ of the initial (a) and annealed at 5 min (b) at T₃ = 647 ± 2 K.

Therefore, short-term annealing leads to the formation of a two-phase amorphous–nanocrystalline structure in which finely dispersed nanocrystals are evenly distributed in an amorphous matrix (Figure 4b). This microstructure contributes to improved electron transport and the appearance of additional active centres, which further ensures an increase in the electrocatalytic activity of the alloy during the hydrogen evolution reaction.

4.2. Electrochemical Behavior

Voltammetric studies of amorphous Al₈₇Y₄Gd₁Ni₄Fe₄ alloys in 1 M KOH (Figure 5, Table 4) were conducted to: (i) determine the corrosion potential (E_corr) before and after annealing and (ii) evaluate the hydrogen evolution potentials (E_HER) at a current density of 10 mA·cm⁻². For the initial alloy, E_corr is −0.21 V, and after annealing it shifts to −0.27 V, indicating an increase in anodic activity. Extended polarization times (cycles 3–5, corresponding to voltammetric curves 3–5) of the AMA result in an anodic shift in the E_corr. The E_HER potential shifts to more negative values after annealing, reflecting a decrease in the reaction overpotential due to the appearance of additional electroactive sites. The i_corr values increase from 0.74 to ≈3 mA·cm⁻², consistent with nanocrystalline structure development and an enlarged in the effective surface area. Changes in anodic Tafel slopes (b = 50–120 mV/dec) indicate a transition from activation-controlled to diffusion-controlled electrode behavior. In annealed samples, a mixed mechanism occurs, involving local activation at the Al/AlFe₂Ni phase boundaries and partial passivation due to Al₂O₃· xH₂O films formation (Table 4, Figure 5).

Figure 5. Voltammetric curves of the initial (a) and annealed for 5 min. (b) AMAs Al₈₇Y₄Gd₁Ni₄Fe₄ in 1.0 M KOH solution. Cycle numbers are indicated by colors: black—1, red—2, green—3, blue—4, cyan—5. Panel (c) shows Tafel slopes and exchange current density measurements. Curves 1, 3, 5, 7 correspond to initial sample; 2, 4, 6, 8 correspond to the annealed for 5 min. sample.

Figure 5. Voltammetric curves of the initial (a) and annealed for 5 min. (b) AMAs Al₈₇Y₄Gd₁Ni₄Fe₄ in 1.0 M KOH solution. Cycle numbers are indicated by colors: black—1, red—2, green—3, blue—4, cyan—5. Panel (c) shows Tafel slopes and exchange current density measurements. Curves 1, 3, 5, 7 correspond to initial sample; 2, 4, 6, 8 correspond to the annealed for 5 min. sample.

Table 4. Corrosion characteristics of initial and 5 min-annealed AMAs Al₈₇Y₄Gd₁Ni₄Fe_4. in 1 M KOH.

Initial AMA
Cycles	Parameters
	−Ecorr, V	icorr, mA·cm−2	b, mV/dec
1	0.21	0.47	70
2	0.14	0.42	58
3	0.12	0.48	50
4	0.13	0.50	60
5	0.12	0.50	50
Annealed 5 min. at T3 = 647 ± 2 K
Cycles	Parameters
	−Ecorr, V	icorr, mA·cm−2	b, mV/dec
1	0.27	0.58	60
2	0.19	0.65	70
3	0.19	0.58	100
4	0.17	0.58	110
5	0.17	0.60	120

Table 4. Corrosion characteristics of initial and 5 min-annealed AMAs Al₈₇Y₄Gd₁Ni₄Fe_4. in 1 M KOH.

Initial AMA
Cycles	Parameters
	−Ecorr, V	icorr, mA·cm−2	b, mV/dec
1	0.21	0.47	70
2	0.14	0.42	58
3	0.12	0.48	50
4	0.13	0.50	60
5	0.12	0.50	50
Annealed 5 min. at T3 = 647 ± 2 K
Cycles	Parameters
	−Ecorr, V	icorr, mA·cm−2	b, mV/dec
1	0.27	0.58	60
2	0.19	0.65	70
3	0.19	0.58	100
4	0.17	0.58	110
5	0.17	0.60	120

HER parameters (Table 5) show that annealing increases the exchange current density i₀ by 100–1000 times (up to 10⁻³–10⁻² A·cm⁻²), while cathodic Tafel slopes decrease. This indicates a reduction in the energy barrier for H₂ formation and improved HER kinetics.

Annealed samples exhibit higher cathode currents at identical overvoltages, confirming enhanced electron transfer and surface activity.

Table 5. Electrochemical parameters of hydrogen evolution reaction on Al₈₇Y₄Gd₁Ni₄Fe₄ AMAs in 1 M KOH.

Initial AMA
Cycles	Parameters
	−b, mV/dec	io·10−4, mA·cm2
1	28	0.03
2	41	5.62
3	40	6.31
4	45	2.51
5	45	1.20
Annealed 5 min. at T3 = 647 ± 2 K
Cycles	Parameters
	−b, mV/dec	io·10−3, mA·cm2
1	23	0.0008
2	41	0.01
3	39	0.78
4	50	7.76
5	55	5.25

Table 5. Electrochemical parameters of hydrogen evolution reaction on Al₈₇Y₄Gd₁Ni₄Fe₄ AMAs in 1 M KOH.

Initial AMA
Cycles	Parameters
	−b, mV/dec	io·10−4, mA·cm2
1	28	0.03
2	41	5.62
3	40	6.31
4	45	2.51
5	45	1.20
Annealed 5 min. at T3 = 647 ± 2 K
Cycles	Parameters
	−b, mV/dec	io·10−3, mA·cm2
1	23	0.0008
2	41	0.01
3	39	0.78
4	50	7.76
5	55	5.25

EIS measurements at hydrogen evolution potential show (Figure 6 and Figure 7, Table 6) a significant decrease in the total charge transfer resistance (R_s + R_f + R_ct) and an increase in the double layer capacitance (CPE_dl) after annealing. This indicates an intensification of electron transport and an increase in the activity of surface H⁺ adsorption centres. The n_dl parameter increases to 0.99, indicating increased surface irregularity and nanophases formation.

Figure 6. Equivalent electrical circuit used for fitting the EIS data of the amorphous Al₈₇Y₄Gd₁Ni₄Fe₄ alloy in 1 M KOH solution.

Figure 6. Equivalent electrical circuit used for fitting the EIS data of the amorphous Al₈₇Y₄Gd₁Ni₄Fe₄ alloy in 1 M KOH solution.

Figure 7. Nyquist plots of the amorphous Al₈₇Y₄Gd₁Ni₄Fe₄ alloy in 1 M KOH (a) initial and (b) after 5 min of annealing. Experimental points (appropriate shapes) and fits (lines) using the equivalent circuit from Figure 6 are shown.

Figure 7. Nyquist plots of the amorphous Al₈₇Y₄Gd₁Ni₄Fe₄ alloy in 1 M KOH (a) initial and (b) after 5 min of annealing. Experimental points (appropriate shapes) and fits (lines) using the equivalent circuit from Figure 6 are shown.

Table 6. EIS parameters AMAs Al₈₇Y₄Gd₁Ni₄Fe₄ alloy in 1 M KOH (a) initial and (b) after 5 min of annealing at hydrogen evolution potential.

Time of Annealing, min	Rs, Ω·cm2	Rf, Ω·cm2	CPEf 10−2, Ω−1cm−2	nf	Rct Ω·cm2	CPEdl 10−5, Ω−1cm−2	ndl
as cast	1.83	1.56	2.55	0.82	0.39	3.52	0.93
5	3.12	1.32	9.97	0.67	0.26	4.27	0.99

Table 6. EIS parameters AMAs Al₈₇Y₄Gd₁Ni₄Fe₄ alloy in 1 M KOH (a) initial and (b) after 5 min of annealing at hydrogen evolution potential.

Time of Annealing, min	Rs, Ω·cm2	Rf, Ω·cm2	CPEf 10−2, Ω−1cm−2	nf	Rct Ω·cm2	CPEdl 10−5, Ω−1cm−2	ndl
as cast	1.83	1.56	2.55	0.82	0.39	3.52	0.93
5	3.12	1.32	9.97	0.67	0.26	4.27	0.99

Overall, short-term annealing induced structural and surface modifications that enhance both corrosion behavior and HER kinetics. A combination of voltammetric and impedance analysis confirms that annealing increases the number of active sites, reduces energy barriers for HER, and improves the electrochemical performance of Al₈₇Y₄Gd₁Ni₄Fe₄ AMA in alkaline environments.

4.3. Hydrogen Evolution Performance

The HER performance of the as-quenched and annealed Al₈₇Y₄Gd₁Ni₄Fe₄ AMAs was evaluated in 1 M KOH for 60 min under potentiostatic conditions. Time-dependent current density (i–t) curves are shown in Figure 8. After the initial immersion (Figure 8a), the electrodes were removed, rinsed with distilled water, dried, and re-immersed in 1 M KOH to assess their stability and reusability (Figure 8b).

The time dependencies of the current density (Figure 8a,b) show significant differences between the initial and annealed AMAs at fixed potentials E_corr and E_HER (Figure 5). In the i–t curves, cathodic currents (i < 0, curves 1–2) correspond to HER, while anodic currents (i > 0, curves 3–4) reflect the partial corrosion dissolution of Al at fixed potentiostatic Ecorr. This distinction arises from the intrinsic polarity of the electrode reactions: HER proceeds as a cathodic process, whereas slight deviations from equilibrium potential induce anodic Al dissolution (Al → Al³⁺ + 3e⁻).

The stability of the current density during repeated electrode use indicates preservation of the amorphous–nanocrystalline microstructure and sustained electrocatalytic activity. The volumetric hydrogen evolution rates during the first (V^I) and second (V^II) immersions are summarized in Table 7. The annealed sample demonstrates a marked increase in HER activity compared to the initial AMA. At E_HER, the volumetric hydrogen formation rate increases from 35.1 to 106.0 mL·g⁻¹·min⁻¹ upon annealing, and further to 217.9 mL·g⁻¹·min⁻¹ upon repeated use.

Figure 8. Time dependencies of hydrogen evolution current density in potentiostatic mode of initial (black line—1, green line—3) and annealed (red line—2, blue line—4) Al₈₇Y₄Gd₁Ni₄Fe₄ AMAs in 1 M KOH during the first (a) and repeated (b) use of the electrode at the hydrogen evolution (black line—1, red line—2) and corrosion (green line—3, 4 blue line—4) potentials.

Figure 8. Time dependencies of hydrogen evolution current density in potentiostatic mode of initial (black line—1, green line—3) and annealed (red line—2, blue line—4) Al₈₇Y₄Gd₁Ni₄Fe₄ AMAs in 1 M KOH during the first (a) and repeated (b) use of the electrode at the hydrogen evolution (black line—1, red line—2) and corrosion (green line—3, 4 blue line—4) potentials.

The data in Table 7 consistently confirm the increase in the activity of the annealed alloy at both potentials. The volumetric hydrogen formation rate increases from 35.1 to 106 mL/(g·min) at E_HER, indicating the effective participation of the formed AlFe₂Ni and GdFe₂ nanophases in the stages of adsorbed hydrogen recombination and reduction in kinetic limitations at the amorphous matrix–nanocrystal boundaries.

The obtained values correlate well with (i) an increase in HER cathode currents (Figure 5), (ii) a decrease in R_ct according to EIS data (Table 6), (iii) nanocrystallisation confirmed by XRD and TEM (Figure 3 and Figure 4), and (iv) Ni-/Fe-enrichment of the surface after HER (Figure 9 and Figure 10, Table 8).

Table 7. Volumetric hydrogen evolution rate on AMAs electrodes in 1 M KOH at the hydrogen evolution potential during the first (V^I) and repeated (V^II) immersions.

AMAs	VI, mL/(g·min)	VII, mL/(g·min)
initial	35.1	83.2
5 min	106.0	217.9

Table 7. Volumetric hydrogen evolution rate on AMAs electrodes in 1 M KOH at the hydrogen evolution potential during the first (V^I) and repeated (V^II) immersions.

AMAs	VI, mL/(g·min)	VII, mL/(g·min)
initial	35.1	83.2
5 min	106.0	217.9

AlFe₂Ni and GdFe₂ nanophases are assumed to facilitate H* recombination due to the presence of Fe and Ni d-electron states, which is consistent with the observed increase in exchange current density and surface Ni enrichment. For comparison, the literature data for Al–10Ni and Al–5Co show the following stable volumetric hydrogen evolution rates: 40–60 mL g⁻¹ min⁻¹ for Al–10Ni and 117.7 mL g−¹ min⁻¹ for Al–5Co [30,31,32]. Short-term annealing of Al₈₇Y₄Gd₁Ni₄Fe₄ provides a significant increase in HER in 1 M KOH, characterised by a high volumetric hydrogen evolution rate, structural stability and preservation of catalytic activity upon reuse. Compared to classical Al–Ni and Al–Co alloys [31,32], the amorphous–nanocrystalline Al–Y–Gd–Ni–Fe system demonstrates superiority due to the effect of nanocrystalline intermetallic phases and an amorphous matrix.

4.4. Surface Morphology and Composition

SEM images (Figure 9) show a significant change in morphology after annealing: the surface becomes more granular with nanoscale protrusions corresponding to newly formed phases. After electrochemical tests (conditions in Figure 8b), micropores and channels characteristic of areas of intense gas evolution are observed.

SEM images (Figure 9) show a clear evolution of surface morphology as a result of heat treatment and subsequent electrochemical loading. The initial AMAs is characterised by a smooth amorphous surface (a), which after corrosion (b) and HER (c) is covered with micropores and local channels associated with hydrogen bubble detachment and uneven Al dissolution.

Figure 9. SEM images (left—SE-secondary electrons, right—SE-backscattered electrons) of AMAs: (a)—initial, (b)—initial after corrosion, (c)—initial after hydrogen evolution, (d)—annealed, (e)—annealed after corrosion, (f)—annealed after hydrogen evolution.

Figure 9. SEM images (left—SE-secondary electrons, right—SE-backscattered electrons) of AMAs: (a)—initial, (b)—initial after corrosion, (c)—initial after hydrogen evolution, (d)—annealed, (e)—annealed after corrosion, (f)—annealed after hydrogen evolution.

After annealing (d), a grainy structure is formed, corresponding to the formation of nanocrystals of intermetallic phases. Further loading at E_corr (e) and E_HER (f) causes the development of cracks, pores and pitted areas. The presence of cracks is typical for intensive gas evolution: local accumulation of H₂ under hydroxide–oxide films leads to their mechanical rupture. Such morphological degradation is not critical; on the contrary, it increases the actual activated surface area and provides access to internal catalytic centres, which is characteristic of efficiently functioning HER electrodes.

Elemental distribution maps (Figure 10) confirm a significant reformatting of the surface after potentiostatic modes. The annealed alloy (a) shows a uniform distribution of Al, Fe, Ni, Y, and Gd with a relatively low oxygen content, which is consistent with a weakly oxidized surface in its initial state. After exposure to E_corr (b), the surface is strongly enriched with oxygen (up to ~67 at.%), which corresponds to the intensive formation of Al(OH)₃/AlOOH layers, while the Al content on the surface drops sharply. The distribution of Fe and Ni remains uniform, indicating their lower susceptibility to dissolution in an alkaline environment. Hydroxide layers stabilize the surface and promote reactivation during cyclic use.

Fixing the potential in the HER region (c) leads to Ni enrichment of the surface (up to ~15 at.%), which is a key factor in increased catalytic activity. The simultaneous decrease in O content compared to the corrosion mode indicates partial dihydroxylation due to bubble destruction of surface films. Such surface reconstruction confirms that Fe/Ni nanophases remain electrochemically stable and provide the main contribution to the HER process.

Figure 10. Elemental maps of annealed (a) AMAs and annealed AMAs after potentiostatic modes at corrosion potential (b) and hydrogen evolution potential (c).

Figure 10. Elemental maps of annealed (a) AMAs and annealed AMAs after potentiostatic modes at corrosion potential (b) and hydrogen evolution potential (c).

Table 8. Elemental composition (at. ± 0.1%) of the Al₈₇Y₄Gd₁Ni₄Fe₄ alloy in different states: initial, after corrosion, after 5 min of annealing, and after reuse in hydrogen evolution experiments (Figure 9b).

Elements at. %	Initial	Annealed 5 min. at T = 647 ± 2, K	Initial After Corrosion Process	Annealed 5 min After Corrosion Process	Initial After HER	Annealed 5 min. at T = 647 ± 2, K After HER
Al	86.9	87.1	54.9	30.5	86.6	52.7
Fe	4.2	3.9	13.4	10.3	4.5	9.3
Ni	4.0	4.2	13.3	30.5	4.1	21.0
Y	3.9	3.7	15.1	23.5	3.8	13.1
Gd	1.0	1.1	3.3	5.2	1.0	3.9

Table 8. Elemental composition (at. ± 0.1%) of the Al₈₇Y₄Gd₁Ni₄Fe₄ alloy in different states: initial, after corrosion, after 5 min of annealing, and after reuse in hydrogen evolution experiments (Figure 9b).

Elements at. %	Initial	Annealed 5 min. at T = 647 ± 2, K	Initial After Corrosion Process	Annealed 5 min After Corrosion Process	Initial After HER	Annealed 5 min. at T = 647 ± 2, K After HER
Al	86.9	87.1	54.9	30.5	86.6	52.7
Fe	4.2	3.9	13.4	10.3	4.5	9.3
Ni	4.0	4.2	13.3	30.5	4.1	21.0
Y	3.9	3.7	15.1	23.5	3.8	13.1
Gd	1.0	1.1	3.3	5.2	1.0	3.9

From the point of view of electronic structure, Ni and Fe, due to their unfilled 3d orbitals, are capable of forming semi-covalent bonds with adsorbed H atoms, reducing the recombination barrier. Gd and Y, on the contrary, are more electropositive and contribute to the local polarization of the Al matrix, which facilitates the breaking of Al–H bonds. Thus, after a short annealing, the surface acquires a heterogeneous composition with functionally different centres, where chemical and electrochemical hydrogen evolution proceed synergistically.

4.5. Correlation Between Structure and Activity

The behaviour of the amorphous alloy Al₈₇Y₄Gd₁Ni₄Fe₄ in 1 M KOH is determined by the interaction of three main processes: aluminium leaching, hydride formation of rare earth elements (Y, Gd), and the catalytic activity of Ni–Fe nanoclusters formed as a result of short-term annealing.

The combination of these phenomena significantly affects the mechanism of the hydrogen evolution reaction (HER) and ensures the stabilisation of the electrochemical behaviour of the alloy. In the initial polarisation cycles (Figure 5, Table 4), intense anodic dissolution of aluminium is observed. The value of the cathodic Tafel slope b ≈ 100 mV/dec indicates a Volmer-controlled hydrogen adsorption mechanism, which corresponds to electron transfer in the reaction:

$${\mathrm{H}}_2\mathrm{O} + e^{-}\to {\mathrm{H}}_{\mathrm{ads}} + {\mathrm{OH}}^{-}.$$

In this mode, hydrogen evolution is predominantly corrosive and is accompanied by degradation of the natural passive film Al₂O₃·xH₂O. EDX analysis after corrosion shows a sharp decrease in the Al content from 84.87% to 24.91% and an increase in the oxygen content from 2.41% to 54.56%, confirming the intensive leaching of aluminium and the formation of an oxide layer. After partial leaching, the surface is enriched with Ni, Fe, Y, and Gd—more stable elements that form a porous reaction-active structure (Figure 9 and Figure 10), which further contributes to the activation of HER. Short-term annealing at T = 647 ± 2 K (5 min) promotes partial nanocrystallisation, which creates conditions for the hydride formation of rare earth elements. Due to the high affinity of Y and Gd to hydrogen, stable phases such as YH₃ and GdH_2-3 can form in the subsurface zone. This is reflected in a significant decrease in corrosion activity, with the b value decreasing to 60–70 mV/dec. These values correspond to the transition to the Heyrovsky-controlled or mixed Volmer–Heyrovsky stage, when the rate is determined by the desorption of adsorbed hydrogen:

$${\mathrm{H}}_{\mathrm{ads}} + {\mathrm{H}}_2\mathrm{O} + e^{-}\to {\mathrm{H}}_2 + {\mathrm{OH}}^{-}$$

The formed hydrides Y and Gd act as temporary hydrogen buffers, binding part of H_ads, suppressing local dissolution of Al and at the same time maintaining electronic conductivity through the defective areas of the hydride phase. This stabilises the potential and promotes the formation of a denser passive film of Al₂O₃–Y₂O₃–Gd₂O₃ composition, as seen in the EDX spectra (increase in O to 68.56% after annealing, Table 4).

4.6. Catalytic Activity of Ni–Fe Nanoclusters

After structural relaxation due to annealing, small areas of coherent scattering rich in Ni and Fe are formed in the alloy matrix (Figure 3 and Figure 4, Table 2). These nanoclusters act as active catalysts for the water reduction reaction, where HER occurs via a mixed mechanism:

$$\mathrm{Volmer}: {\mathrm{H}}_2\mathrm{O} + e^{-}\to {\mathrm{H}}_{\mathrm{ads}} + {\mathrm{OH}}^{-},$$

$$\mathrm{Heyrovsky}: \mathrm{Hads}+{\mathrm{H}}_2\mathrm{O}+e^{-} \to {\mathrm{H}}_2 + {\mathrm{OH}}^{-}$$

Metastable NiHx hydrides formed during polarisation increase the rate of H₂ desorption. Electrochemical impedance (Figure 7) shows an increase in R_ct after annealing, indicating an increase in charge transfer resistance due to a more stable film and the activity of Ni–Fe centres. The time dependence of the current (Figure 8) shows the stability of the cathode current during repeated use of the electrode, in contrast to the initial state, where its decline is observed.

Thus, after short-term annealing, the hydrogen evolution rate increases significantly: at HER potential—from 35.1 to 106.0 mL/(g·min), and at reuse—from 83.2 to 217.9 mL/(g·min). Such changes indicate a shift in the predominant mechanism from corrosive to electrochemical and confirm the role of Ni–Fe nanophases as the main centres of HER. The comprehensive analysis of the results shows that short-term annealing of AMAs Al₈₇Y₄Gd₁Ni₄Fe₄ initiates controlled nanocrystallisation with the formation of AlFe₂Ni, GdFe₂ and Al(X) (X = Gd, Ni, Y, Fe) phases, which significantly change the electronic structure of the surface. The formation of intermetallic nanodomains creates heterogeneous boundaries with an amorphous matrix, which serve as effective areas for charge transfer and H^* adsorption. The decrease in R_ct and increase in CPE_dl after annealing, combined with an increase in cathode currents in voltammetric studies, indicates an improvement in HER kinetics due to the increased density of active d-metals and the facilitation of the Volmer stage. The presence of a chemically active Al matrix provides an additional contribution to hydrogen formation through hydrolysis, which is particularly evident in E_corr, while Fe/Ni-containing nanophases catalyse the electrochemical pathway of H₂ evolution in E_HER. SEM analysis after potentiostatic loading demonstrates the development of a porous morphology and cracks, which is associated with intense gas evolution and contributes to an increase in surface area. Elemental maps reveal Ni enrichment of the surface under HER conditions, which is consistent with the increase in hydrogen evolution rate (Table 7). In general, the synergistic effect of the amorphous matrix and nanophases provides a significant enhancement of HER activity and stability, making annealed AMAs a promising inexpensive electrocatalyst for alkaline hydrogen evolution. Thus, the results obtained demonstrate a close relationship between nanostructural evolution and electrocatalytic activity. Short-term annealing creates a two-phase structure of amorphous matrix + nanocrystalline phases, where

• The amorphous component provides chemical activation and a source of H₂;
• The AlFe₂Ni and GdFe₂ nanophases catalyze the electrochemical recombination stage;
• Al nanocrystals and rare earth elements improve electron transport and reduce reaction overvoltage.

As a result, the complex structure after 5 min of annealing provides increased speed and stability of hydrogen evolution in an alkaline environment.

5. Conclusions

Short-term annealing of the Al₈₇Y₄Gd₁Ni₄Fe₄ amorphous alloy at 647 ± 2 K induces controlled nanocrystallisation with the formation of Al(X) (X = Gd, Ni, Y, Fe), AlFe₂Ni and GdFe₂ nanophases embedded in the amorphous matrix. This structural transformation significantly modifies the local electronic environment and enhances charge-transfer efficiency at the amorphous–crystalline interfaces.

Electrochemical measurements demonstrate a strong improvement of HER kinetics, expressed by a 2–3 order increase in the exchange current density, reduced Tafel slopes and lower charge-transfer resistance. Potentiostatic experiments confirm that the volumetric hydrogen evolution rate increases from 35.1 to 106.0 mL·g⁻¹·min⁻¹ during the first run and exceeds 200 mL·g⁻¹·min⁻¹ upon reuse. SEM/EDS mapping shows surface reconstruction and Ni enrichment under HER conditions, which correlates with accelerated H* recombination.

The synergistic contribution of the amorphous matrix (chemical H₂ formation via Al hydrolysis) and intermetallic nanophases (electrocatalytic recombination on Ni/Fe sites) provides high activity and stability in alkaline media. These results identify annealed Al-based AMAs as promising low-cost electrocatalysts for hydrogen evolution.