Investigation of stability of gold nanoparticles modified zinc–cobalt coating in an alkaline sodium borohydride solution

Chemicals

CoSO₄, ZnO, HAuCl₄ and NaBH₄ were purchased from Sigma-Aldrich Supply. H₂SO₄ and NaOH were purchased from Chempur Company. Titanium (Ti) foil of 99.7% purity and 0.127 mm thickness (Sigma-Aldrich) was used as a substrate. All chemicals were of analytical grade. Deionized water with a resistivity of 18.2 MΩ cm^–1 was used to prepare all the solutions.

Fabrication of catalysts

In this study, the ZnCo/Ti and AuZnCo/Ti catalysts were prepared in the same manner as described in Refs. [43–45]. Briefly, prior to the electrodeposition of ZnCo alloy coating on the Ti, the Ti plates with a geometric area of 2 cm² were degreased with acetone, pre-treated with SiC emery paper (grade 2500) and MgO powder and rinsed with deionized water. Then, the samples were etched in a diluted H₂SO₄ (1:1 vol) solution at 90°C temperature for 10 s and finally rinsed with deionized water. After that, the ZnCo coatings with a thickness of ~5 μm were electroplated on the Ti surface using an electrolyte as described in Refs. [43, 44]. Then, Au nanoparticles were deposited on the as-prepared ZnCo/Ti surface via the galvanic displacement technique [45] by dipping of the latter catalyst into a solution containing 1 mM HAuCl₄ + 0.1 M HCl at 25°C temperature for 0.5, 1 and 5 min, respectively. The surface-to-volume ratio was 1.3 dm² l^–1. After plating, the samples were taken out, thoroughly rinsed with deionized water and air dried at room temperature. Then, the prepared catalysts were used for measurements of the BOR as well as for the hydrolysis reaction of NaBH₄ in an alkaline medium without any further treatment.

Physical characterization of catalysts

The morphology and composition of the fabricated catalysts were characterized using a SEM-focused ion beam facility (Helios Nanolab 650) equipped with an EDX spectrometer (INCA Energy 350 X-Max 20).

Au metal loadings in the catalysts were determined by ICP-OES measurements by recording the ICP optical emission spectra with an Optima 7000DV (Perkin Elmer) spectrometer at wavelengths of λ = 267.595 and 242.795 nm. For calibration, the standard solutions of 1.0 and 10.0 mg l^–1 were used (Perkin Elmer).

Electrochemical measurements

A conventional three-electrode electrochemical cell was used for electrochemical measurements for the BOR. The bare Au, AuZnCo/Ti and ZnCo/Ti catalysts were employed as working electrodes, a Pt sheet was used as a counter electrode and an Ag/AgCl/KCl (3 M KCl) electrode was used as reference. An Au-sputtered quartz crystal with a geometric area of 0.636 cm² was used as a bare gold electrode. All electrochemical measurements were performed with a Metrohm Autolab potentiostat (PGSTAT100) using the Electrochemical Software (Nova 1.6.013). Continuous cyclic voltammograms (CVs) were recorded in a 0.05 M NaBH₄ + 1 M NaOH solution at a scan rate of 10 mV s^–1 at 25°C temperature. The presented current densities were normalized with respect to the geometric area of catalysts. All solutions were deaerated by argon for 30 min prior to the measurements.

Chronoamperometric measurements were carried out on the investigated catalysts in a 0.05 M NaBH₄ + 1 M NaOH solution at a constant potential value of –0.2 V for 2 min. The potential was initially held at open circuit for 10 s, then set to –0.2 V for 2 min.

Hydrogen generation measurements

The amount of generated hydrogen (H₂) was measured using the classic water-displacement method with the aim to determine the catalytic efficiency of the AuZnCo/Ti catalysts for the hydrolysis reaction of NaBH₄. Typically, a reaction solution containing 0.05 M NaBH₄ and 1 M NaOH was placed into a thermostatically controlled airtight flask fitted with an outlet for collection of the evolved H₂ gas. Then, the AuZnCo/Ti catalysts were dipped into the solution at the designated temperature in order to initiate the hydrolysis of NaBH₄. The water displaced from the calibrated cylinder connected to the reaction flask was continuously monitored during the performance of the reaction. In order to determine the activation energy, the rate of H₂ generation was measured at different solution temperatures (30–50°C).

Material characterization

In our previous work [45] we have investigated the electrochemical behaviour of ZnCo/Ti and AuZnCo/Ti catalysts in NaBH₄ and hydrazine (N₂H₄) alkaline solutions and determined that both catalysts are active toward the oxidation reactions of NaBH₄ and N₂H₄. However, the stability and durability of the mentioned catalysts have not been thoroughly analysed in an alkaline medium and their activity toward the hydrolysis reaction of NaBH₄ remained short in detail. Therefore, following the work carried out previously, in this study we continue the investigation regarding the above-mentioned matters and present a more detail analysis for the BOR in an alkaline medium. Since the anodic oxidation of NaBH₄ is a heterogeneous reaction, the changes in the structure and morphology of the surface of the anode material during the oxidation process is of basic importance. Therefore, it is meaningful and very reasonable to compare the physical properties of the as-prepared catalysts and those after some treatment in NaBH₄ solutions. For the sake of simplicity, we briefly overview the SEM images of the as-prepared ZnCo/Ti and AuZnCo/Ti catalysts as was described in detail in Ref. [45] and present them in Fig. 1 under different magnifications. It was found that the ZnCo coating fully covers the Ti surface and produces a compact structure with a spherical cluster surface that consists of a large number of smaller nodular-shaped grains ranging from ca. 0.5 to 2.5 μm in size (Fig. 1a). Dipping of the as-prepared ZnCo/Ti electrodes into the 1 mM HAuCl₄ + 0.1 M HCl solution for various time periods resulted in the deposition of Au nanoparticles that appear as bright crystallites and are homogeneously dispersed on the ZnCo surface (Fig. 1b–d). Moreover, Au nanoparticles sized ca. 16 to 50 nm were deposited on the ZnCo/Ti surface after dipping of ZnCo/Ti electrodes into the 1 mM HAuCl₄ + 0.1 M HCl solution for 0.5 and 1 min (Fig. 1b, c), whereas dipping of the ZnCo/Ti electrode into the 1 mM HAuCl₄ + 0.1 M HCl solution for 5 min resulted in the development of larger Au crystallites on the ZnCo coating surface sized in a range of ca. 50–200 nm (Fig. 1d). Although the Au nanoparticles are hardly discernible in Fig. 1b, c, their residence on the surface of the coating was confirmed by EDX analysis data and are summarized in Table 1. A significant amount of Zn and much lower amounts of Au and Co were determined on the surfaces of ZnCo/Ti and AuZnCo/Ti catalysts. As follows from the data in Table 1, the increase in the modification time of the ZnCo coating by Au nanoparticles leads to a higher amount of Au on the surface of the catalyst. The amounts of Au loadings on the fabricated catalysts were estimated to be 31, 63 and 306 µg cm^–2 corresponding to dipping times of 0.5, 1 and 5 min, respectively.

Table 1. Surface atomic composition of the as-prepared ZnCo/Ti and AuZnCo/Ti catalysts determined by EDX

Electrochemical activity of catalysts for the BOR

In order to determine the electrochemical behaviour of the as-prepared ZnCo/Ti and AuZnCo/Ti catalysts for the BOR, cyclic voltammetry and chronoamperometry measurements were performed in a NaBH₄ alkaline solution. The obtained results of the BOR on the ZnCo/Ti and AuZnCo/Ti catalysts were compared with each other and with those recorded on the bare Au electrode. Figure 2 presents typical CV results of the bare Au electrode recorded in a 0.05 M NaBH₄ + 1 M NaOH solution at a scan rate of 10 mV s^–1. The bare Au electrode is characterized by the onset potential E_onset of ca. –0.8 V and by a typical current peak (denoted as A) at a potential of ca. –0.4 V. Anodic peak A is attributed to the direct and complete eight-electron oxidation of BH₄^– ions (Eq. (1)), as discussed in Refs. [10, 46–49]:

The production of H₂ and BH₃OH^– due to the direct heterogeneous hydrolysis presents a peak in the potential region of the first branch of the BOR wave [50, 51]. In Ref. [52] it is stated that the BOR on Au proceeds through the electrooxidation of BH_3ads species that originates from a partial dissociation of BH₄^– ions, during which the self-dehydrogenation of BH₄^– generates H₂. The origin of H₂ production coming essentially from the dehydrogenation of BH_3ads was monitored and confirmed by differential electrochemical mass spectrometry (DEMS) [53]. Following peak A, a wide-extended oxidation wave from ca. –0.4 up to 0.4 V on the bare Au catalyst appears and corresponds to the oxidation reaction of intermediates on a partially oxidized Au surface [10]. On the reverse potential scan, a sharp peak (denoted as B) is observed at ca. 0.2 V followed by a plateau in the potential region from 0.0 to –0.6 V. This peak corresponds to the oxidation of the adsorbed species such as BH₃OH^– or other borohydrides formed as intermediates during the oxidation of NaBH₄ in the forward scan [10, 51]. A significant decrease in current is observed in the subsequent scan cycles in the NaBH₄ solution (Fig. 2) that can be explained by the formation of intermediates that adsorb on the electrode and simultaneously poison the surface resulting in a diminished electrocatalytic activity of the bare gold electrode [10]. Figure 3 shows the electrochemical behaviour of the AuZnCo/Ti catalysts that have Au loadings of 31 (a), 63 (b) and 306 (c) μg_Au cm^–2 in a 0.05 M NaBH₄ + 1 M NaOH solution at a scan rate of 10 mV s^–1. The first potential cycles of NaBH₄ oxidation onto the AuZnCo/Ti catalysts are shown in Fig. 3a–c (lines 1). The dotted lines represent the CV results of the above-mentioned catalysts recorded in the background solution of 1 M NaOH (Fig. 3a–c, lines 2). Consecutive anodic-going potential scans starting from the 2nd one for the same catalysts with the same loadings are presented in Fig. 3a’–c’. The data obtained in Fig. 3a–c suggest that the oxidation of NaBH₄ at all the AuZnCo/Ti catalysts that have different Au loadings starts at a significantly more negative electrode potential E_onset of ca. –1.2 V as compared to that of bare Au (Fig. 2). The difference in E_onset equals to ca. 0.4 V and points to the presence of better fuel cells operation conditions at the AuZnCo/Ti surface. Unlike on Au, in the CVs recorded in a NaBH₄ solution on the different Au loading containing AuZnCo/Ti catalysts, a set of well-expressed anodic peaks is observed (Fig. 3). A well-defined anodic peak A0 at low potential values and another broad one that collapse into a strongly resolved doublet each of them labelled as A and A’, respectively, can be observed in the CVs (Fig. 3a–c). The nature of anodic peak A0 recorded at AuZnCo/Ti could be ascribed to the oxidation of H₂ generated by the catalytic hydrolysis of NaBH₄ as well as to the oxidation of NaBH₄, similarly as at the ZnCo/Ti catalyst [44]. It should be noted that anodic peak A0 at the Zn/Ti catalyst occurs at ca. –1.27 V and is attributed to the electrooxidation of BH₄^– ions [54]. Meanwhile, anodic peak A0 for the Co/Ti catalyst was earlier attributed only to the oxidation of H₂ generated by the catalytic hydrolysis of NaBH₄ [6, 55]. The same peak A0 for metallic Co was assigned to the direct oxidation of BH₄^– ions [56]. Furthermore, since the oxidation of NaBH₄ proceeds in a strongly alkaline solution, so anodic peak A0 may be also attributed to the formation of cobalt oxide and hydroxide according to the following reactions (Eqs. (2, 3)) [57, 58]:

In order to better understand the low-potential processes occurring at the AuZnCo/Ti catalyst we recorded CVs in a supporting 1 M NaOH solution and superimposed the first scans of CVs in 1 M NaOH (dotted lines 2) with the first scans in a 0.05 M NaBH₄ +1 M NaOH solution (lines 1) in Fig. 3a–c. It is clearly seen that anodic peaks A0 recorded in an alkaline NaBH₄ solution at AuZnCo/Ti with different Au loadings are markedly higher as compared to those recorded using the same catalyst in a NaOH solution. These data indicate that the major part of the charge passed in an alkaline NaBH₄ solution at low potentials may be attributed to the oxidation utilization of fuel rather than the catalyst itself. At more positive potentials peak A emerges. It occurs almost in the same potential region as on the bare Au electrode [10, 46–49] and is reasonably attributed to the direct oxidation of NaBH₄ at the AuZnCo/Ti catalyst. On the backward-going cathodic scan, some oxidation peaks are also observed. They could be related to the oxidation of adsorbed intermediate species that are formed during the BOR in the forward scan as well as to the reactivation of the oxidized Co and Zn surfaces during the previous anodic scan.

Nevertheless, the CVs recorded in a NaBH₄ solution obtained at the AuZnCo/Ti catalyst containing different Au loadings are similar in shape (Fig. 3a–c); however, the electrochemical behaviour of the aforementioned catalysts for the BOR during the long-term cycling (Fig. 3a’–c’) differs significantly. Current density values for anodic peaks A0 and A at the AuZnCo/Ti catalyst that has the Au loading of 31 μg_Au cm^–2 (Fig. 3a’) appreciably increase after the first cycle applied; however, they fall down with continuous potential scans till they receive some stable state. Almost analogous behaviour of the ZnCo/Ti catalyst towards the BOR was determined in Ref. [44], except that the decrease in current during a long-term cycling was already observed from the first scan. One reason of such a decrease in current densities can be explained by the occurrence of NaBH₄ oxidation intermediate species that block the catalyst surface. On the other hand, the degradation of the catalyst should be taken into account. The observed feature points to the fact that the mentioned catalysts are not sufficiently stable and are inclined to degrade during the BOR.

A completely different tendency in current density changes under the same peaks during cycling is obtained on the catalysts that have higher Au loadings of 63 μg_Au cm^–2 (Fig. 3b’) and 306 μg_Au cm^–2 (Fig. 3c’). The values of current densities under anodic peak A associated with the direct oxidation of NaBH₄ for both catalysts increase and finally achieve almost stable ones after a few potential cycles applied, indicating a higher activity of the latter catalysts. Meanwhile, the current densities under anodic peak A0 after the second scan decrease.

To get some further information on the changes in catalyst morphology and elemental composition during the cycling in the NaBH₄ solution, additional FESEM and EDX analysis was performed. Figure 4a–d presents the FESEM images of the same as-prepared ZnCo/Ti and AuZnCo/Ti catalysts as shown in Fig. 1a–d, but after 5 cycles treatment in a 0.05 M NaBH₄ + 1 M NaOH solution. The data in Fig. 1a and Fig. 4a evidence that the view changes drastically after the ZnCo coating undergoes a treatment in a 0.05 M NaBH₄ + 1 M NaOH solution by applying 5 cycles in the potential region from –1.2 to +0.6 V at a potential scan rate of 10 mV s^–1 at 25°C temperature. Continuous cycling in the NaBH₄ solution results in the degradation of a compact coating and leads to the development of an uneven ZnCo layer with larger spaces between agglomerated granules of ca. ~ several hundred nm in size directed both vertically and horizontally. Figure 4a clearly evidences the appearance of a highly porous structure leading to significant changes in the chemical composition of the coating. The porous structure could provide the multi-channel diffusion pathways and enhance the interaction between the electrode and the electrolyte. On the other hand, although the thick volumic active layers contain numerous sites for the reaction to proceed, these sites remain insufficiently fed by a ‘fresh’ NaBH₄ solution, since the layer feeding operates mainly only from the top of the surface [60]. However, the structure/thickness of the active layer as well as the nature and stability of the electrocatalyst is important in order to reach a high BOR efficiency.

The FESEM images of the same-prepared AuZnCo/Ti catalysts, as shown in Fig. 1b–d, after 5 cycles treatment in a 0.05 M NaBH₄ + 1 M NaOH solution are shown in Fig. 4b–d. They clearly indicate that the modification of the ZnCo layer by Au nanoparticles has a great influence on the morphology of the AuZnCo/Ti catalyst when compared to that of ZnCo/Ti surface analogously treated in the tested solution (Fig. 4a). It results in both decrease of larger spaces between agglomerated granules and the number of agglomerated ones. Increasing the modification time of the ZnCo/Ti electrode by Au nanoparticles almost completely prevents the degradation of the modified catalyst during continuous cycling in a 0.05 M NaBH₄ + 1 M NaOH solution. Nevertheless, some separate features of porosity under the semblance of light-coloured spots are observed on the surface of the AuZnCo/Ti catalyst obtained by applying the shortest modification time periods (Fig. 4b), but they rather quickly disappear by prolonging the periods to 5 min (Fig. 4d). The presence of Au nanoparticles in the content of catalyst stabilizes the ZnCo/Ti catalyst and enables one to receive almost unchanged and stable AuZnCo/Ti catalysts, thus enhancing the utilization efficiency in order to provide considerably more stable catalytically active sites.

The data of the above FESEM images are supported by the EDX analysis and are summarized in Table 2. The presence of Zn and Co on the Ti surface was confirmed by energy dispersive X-ray analysis, irrespective of whether the ZnCo/Ti coating was exposed to cycling or not (compare Tables 1 and 2). The amount of Zn significally decreased from 81.32 to 14.7 at.%, meanwhile, the content of Co and Ti increased from 18.48 to 80.40 at.% and from 0.22 to 4.90 at.%, respectively, if 5 cycles treatment in a 0.05 M NaBH₄ + 1 M NaOH solution was applied. The increase in Ti content to 4.90 at.% indicates that the dissolution of Zn occurs not only from the top of the ZnCo coating, but also from all the volume of the layer. Such a phenomenon points to the fact that Zn dissolves from the layer during the cycling and leads to a significant degradation of the ZnCo/Ti coating, declaring that the ZnCo layer is not stable. These data are in line with our previous results [44], where current density in long-term cyclic voltammograms recorded at ZnCo/Ti were shown finally to decrease, which indicates that the ZnCo/Ti catalyst is not sufficiently stable during long-term cycling. Degradation of the catalyst under continuous cycling in a 0.05 M NaBH₄ + 1 M NaOH solution at a scan rate of 10 mV s^–1 was supposed to be related to the simultaneously occurring dissolution of Zn at higher potential values and the changes in the surface structure of cobalt after disappearing of zinc.

Table 2. Surface atomic compositions of the ZnCo/Ti and AuZnCo/Ti catalysts after 5 cycles treatment in a 0.05 M NaBH₄ + 1 M NaOH solution by EDX. The ZnCo/Ti and AuZnCo/Ti catalysts are the same as in Fig. 4

Comparison of the elemental content of the as-prepared AuZnCo/Ti catalysts modified at different time periods to those obtained after 5 cycles treatment in the testing solution has shown that the cycling leads to a lower amount of Zn dissolved from the AuZnCo/Ti catalyst when modification lasts for longer time periods. Finally, when the modification time of the catalyst in the 1 mM HAuCl₄ + 0.1 M HCl solution was extended to 5 min, the content of Zn in the AuZnCo/Ti catalyst remained unchanged whether a cycling was applied or not. It should be also noted that the cycling does not influence the amount of Au in the coating.

Relaying on the aforementioned FESEM and EDX data, we can conclude that the stabilization in current density under anodic peak A (Fig. 3b’, c’) correlates with the reciprocal relation in the changes between the amount of Au loading and the amount of Zn in the catalyst. The higher the loading of Au in the AuZnCo/Ti catalyst layer is, the greater amount of Zn remains in the catalyst after a continuous cycling in the NaBH₄ solution. It remains practically unchanged during the cycling when the Au loading equals 306 μg_Au cm^–2. The presence of sufficient quantities of Au particles in the catalyst layer prevents Zn escape from the surface and from the changes in surface morphology (that are observed in Fig. 4a after the Zn removal) simultaneously leading to the higher stability of the AuZnCo/Ti catalyst. It is evident that the doping of ZnCo layer by small amounts of more noble metal like Au determines the superior quality of the catalyst that highly depends on the amount of the Au loading used.

The comparison data of the 1st and 5th positive-potential going voltammograms recorded on the bare Au (line 1), ZnCo/Ti (a, line 2) and AuZnCo/Ti catalysts that have different Au loadings are shown in Fig. 5. It is clearly seen that the modification of the ZnCo/Ti catalyst with small amounts of Au nanoparticles results both in the potential shift to more negative potential values under anodic peak A with pronounced current peak splitting and to a significant increase in current density values under the both anodic peaks A0 and A.

The electrochemical investigations revealed that the AuZnCo/Ti catalysts show the higher activity towards the processes occurring in the entire potential region that are typically attributed to the hydrolysis reaction of NaBH₄ and the oxidation of NaBH₄ as compared to those at the ZnCo/Ti catalyst and yet more on bare Au. About 1.5 and 2.6 times higher anodic current densities at the potential under peak A after the 1st cycle were observed at the AuZnCo/Ti catalyst that has the Au loading of 31 μg_Au cm^–2 as compared with those at the ZnCo/Ti and bare Au catalysts, respectively. The AuZnCo/Ti catalyst containing Au loadings of 63 and 306 μg_Au cm^–2 gives ca. 1.5 and 1.7 and ca. 2.6 and 3.0 times greater current densities than those at the ZnCo/Ti and the bare Au electrode, respectively.

The aforementioned increase in current under peak A develops yet more during the cycling since anodic current values at the ZnCo/Ti and bare Au catalysts during the cycling fall down, while on the AuZnCo/Ti catalysts that have Au loadings of 63 and 306 μg_Au cm^–2 grow up except the one with the Au loading of 31 μg_Au cm^–2. The increase in current for the aforementioned catalysts after the 5th cycle grows more than 2 times for both loadings (equals ca. 2.2 and 2.0 times higher, respectively) as compared to that of the ZnCo/Ti catalyst, and ca. 12.19 and 11.13 times as compared to that of the bare Au electrode in the potential region under peak A. Meanwhile, for the case when the Au loading was 31 μg_Au cm^–2 the increase in current as compared to that of the Au catalyst equals ca. 6.48 times, but almost no increase in current was observed as compared to that of the ZnCo/Ti catalyst. The enhancement of the catalytic abilities of the catalyst can be attributed to the higher Au loading and to the higher number of active sites on the catalyst surface because of Au–ZnCo interaction leading to higher activities for the BOR.

In order to compare the electrocatalytic activity of the investigated catalysts, the current density values were normalized in reference to Au loadings for each catalyst (Fig. 6). As evident, the NaBH₄ oxidation current density values normalized by Au loadings show that the highest Au mass activity for the BOR was obtained on the AuZnCo/Ti catalyst with the lowest Au loading of 31 μg_Au cm^–2.

To assess the catalytic characteristics of activity and stability of the AuZnCo/Ti catalysts that have different Au loadings chronoamperometric measurements were performed in a NaBH₄ solution at a constant potential value of –0.2 V. The obtained data are shown in Fig. 7a. Figure 7b presents the chronoamperometric curves, which were normalized by Au loadings for each catalyst. The initial current drop-off is observed, possibly due to the formation of intermediate species such as BH₃OH^– or other borohydrides [10, 46–49] during the borohydride oxidation reaction. From Fig. 7 it follows that the current density values of the investigated catalysts stabilize with time and remain constant at the end period of time (t = 120 s). The measured current densities for all the AuZnCo/Ti catalysts are higher than those for bare Au, indicating that the former nanostructured catalysts have a higher catalytic activity for the BOR. The normalized chronoamperograms by Au loadings for each catalyst suggest that the catalyst with the lowest Au loading is the most active even though the current slightly increases at t = 45 s after the potential was stepped to –0.2 V (Fig. 7b, line 1). Possibly, this is the close case as in the chronoamperogram for ZnCo/Ti (Fig. 7a, dashed line), where two consecutive electron transfer steps for the oxidation of two oxidizable species take place [6, 7]. An enhanced electrocatalytic activity of the as-prepared AuZnCo/Ti catalysts can be ascribed to the intrinsic electrocatalytic activity of the Au particles deposited on the ZnCo/Ti surface [34].

To get some information on the nature of anodic peak A0 and the activity of the AuZnCo/Ti catalysts in the potential domain of peak A0, the activity of the AuZnCo/Ti catalysts that have Au loadings of 63 and 306 μg_Au cm^–2 was investigated towards the hydrolysis of NaBH₄. Figure 8 shows the dependence of volume of hydrogen generated during the hydrolysis reaction of the NaBH₄ solution in an alkaline medium at those catalysts on the reaction temperature. The summarized data are given in Table 3. As evidenced from the data in Fig. 8, the rate of catalytic hydrolysis of NaBH₄ in alkaline solutions increases exponentially with the increase of the reaction temperature, and the maximum values of 200 and 150 L g^–1 min^–1 are obtained at the aforementioned catalysts at 50°C temperature. The temperature dependence on the hydrogen generation rate is expressed by the Arrhenius equation

where Ea is the activation energy (J), A is the frequency factor, and R is the general gas constant (8.314 J mol^–1 K^–1). In order to find the activation energy and frequency factor, the Arrhenius plots of ln(k) vs 1/T were constructed from the data presented in Fig. 8a, b and are given in Fig. 8c.

Table 3. Dependence of the hydrogen generation rate on temperature obtained at AuZnCo/Ti in a 0.05 M NaBH₄ + 1 M NaOH solution

The activation energy values calculated on the basis of the Arrhenius plot on the AuZnCo/Ti catalysts that have Au loadings of 63 (a) and 306 μg_Au cm^–2 (b) are 41.5 and 44 kJ mol^–1, respectively (Table 3). The comparison of the mentioned activation energies with those obtained on various catalysts reported in the literature is summarized in Table 4. It should be noted that the E_a values obtained on both AuZnCo/Ti catalysts with Au loadings of 63 and 306 μg_Au cm^–2 in a highly stabilized 0.05 M NaBH₄ + 1 M NaOH solution are higher than those obtained on the Co–B/TiO₂ framework (30.93 kJ mol^–1) [62], Au₅₀Ni₅₀ BNPs (30.3 kJ mol^–1) [65] and Au(Co)/Ti (37.9 kJ mol^–1) [17] (but stability of the former catalyst was noted to be satisfactory), whereas they are almost equal to those determined on ZnCo/Ti (40 kJ mol^–1) [44], Co powder (41.9 kJ mol^–1) [61] or Co–B/Cu (43.3 kJ mol^–1) [64]. Moreover, they are lower than those on Co–B/TiO₂ (51.6 kJ mol^–1) [63], Raney Co (53.7 kJ mol^–1) [61], Co/Ti (54 kJ mol^–1) [44] or Au₂₀Ni₈₀ BNPs (60.0 kJ mol^–1) [65]. The obtained data confirm that the AuZnCo/Ti catalysts catalyze the hydrolysis reaction of NaBH₄ in alkaline solutions.

Table 4. Comparison of hydrogen generation rates (HG) and activation energies (E_a) on various catalysts