Enhanced Oxygen Reduction with Ethanol Tolerant Ni-Te Nanoparticles on Carbon Support Through Vapour-Solid Preparation

Abstract

Recent endeavours to promote the widespread use of renewable and sustainable energy technologies depend heavily on the development and design of new catalytic materials. In this context, intermetallic compounds have come into the spotlight of recent research as a promising material class to tune the catalytic properties and stability for various uses. In this work, vapour–solid synthesis is highlighted as an outstanding method for its control over the composition and crystal structure of prepared intermetallic nanoparticles. Carbon black-supported nickel-telluride nanoparticles of different compositions and crystallographic structures have been synthesised and investigated regarding their oxygen reduction reaction performance in alkaline media. The relation between catalytic activity and ethanol tolerance depending on the various intermetallic phases has been investigated. The addition of tellurium into nickel-based nanoparticles allowed a two-fold increase of the mass activity from 43.6 A g_Ni⁻¹ for Ni/C to 88.5 A g_Ni⁻¹ for Ni-Te/C. Onset and half-wave potentials were comparable to commercial Pt/C benchmark catalyst. Furthermore, chronoamperometric testing showed that the ethanol-tolerant Ni-Te/C catalysts were stable under electrocatalytic conditions during in alkaline media. The trend in catalytic activity of the Ni-Te phases was followed the order: Ni₃Te₂ > NiTe > NiTe_2−x > Ni.

1. Introduction

The global increase in greenhouse gas emissions and the associated impact on the environment caused by the extensive burning of fossil fuels necessitate the development of alternative, cost-effective, renewable, and sustainable energy sources [1]. In this context, alkaline direct ethanol fuel cells (ADEFCs) have gained increasing attention in the recent years [2,3]. In ADEFCs, chemically stored energy is converted into electrical energy via oxidation of ethanol at the anode side of the cell at low temperatures. The hydroxide ions required to enable the current flow within the cell are formed by the oxygen reduction reaction (ORR) at the cathode side. Figure 1 shows an illustration of the process as well as the reactions involved in an ADEFC. Contrary to operation in acidic media, ADEFCs offer the advantages of improved reaction kinetics, reduced ethanol crossover rates due to ion counter flows, and most importantly, the potential use of non-noble metal catalysts, such as nickel-, iron-, or cobalt-based materials [4,5].

Ethanol, as a liquid energy carrier, has a high energy density, is environmentally friendly, facile to implement into existing infrastructure, and nontoxic, and can be produced from renewable feedstocks. However, major challenges for the wide application of ADEFCs include partial fuel oxidation on the anode side, high cost of the state-of-the-art noble metal catalyst materials, and fuel crossover towards the cathode side. Ethanol crossover leads to side reactions at the cathode, accompanied with mixed potentials of the cell, ultimately reducing the overall performance of the ADEFC [6]. Hence, the development of highly active, cost-efficient, and ethanol-tolerant ORR catalysts for the cathode is of critical importance.

To date, platinum (Pt) was used as the benchmark electrocatalyst for ORR, but its use has severe drawbacks including limited availability, high cost, poor long-term durability, and distinct sensitivity towards ethanol crossover. As a result, recent research has focused on the implementation of carbon-supported, non-noble metals as ORR catalysts [7]. Nickel (Ni) is especially interesting as catalyst material, as it has a comparatively high intrinsic activity for ORR in alkaline media. In contrast to precious metals, Ni is less expensive and exhibits high stability under alkaline operating conditions. In addition, Ni shows a remarkable affinity for oxygen species, which favours efficient oxygen adsorption and reduction [5]. Despite these advantages, pure Ni has limitations, particularly regarding the kinetics of ORR and the tendency to be oxidised under operating conditions. To overcome these challenges, nickel-based intermetallic compounds are of particular interest [8,9]. Intermetallic compounds consist of two or more metals arranged in crystal structures differing from their constituting metals. In intermetallic compounds, a combination of metallic, covalent, and ionic bond contents leads to their unique physical, chemical, and mechanical properties [10,11]. Through targeted combination of various metals, the materials characteristics as well as interactions with reactants can be selectively tuned which enables improved catalyst performance [12,13]. Such intermetallic systems offer the possibility of optimising both the activity and the long-term stability of the catalysts, making them promising candidates for future applications [14]. To study the performance of the various intermetallic phases and find the most promising material, a very precise synthesis method is of utmost importance [15]. Conventional synthesis routes like thermal annealing or wet-chemistry approaches suffer from the lack of compositional control, lead to particle aggregation and sintering, and require individual development for each system limiting product comparability. Recently, the direct vapour–solid (VS) synthesis method (Figure 2) has been shown to overcome those challenges and allow for a facile route to produce a variety of supported, intermetallic nanoparticle systems with unprecedented compositional control and product quality [12,16]. In the VS approach, stoichiometric amounts of the constituting metals are sealed in an evacuated quartz vessel, separated in space and equilibrated within a temperature gradient. The temperature gradients ensures that deposition of the more volatile compound only takes place on the coolest point of the vessel, and that the reaction takes place solely by direct vapour–solid interaction. The system pressure and respective activities are controlled via the selected temperatures. The use of simple, commercially available educts of high purity, the absence of additives, and the good compositional control enable the synthesis and study of various comparable, intermetallic samples within an intermetallic system.

Although various elements could theoretically be used for the VS preparation of intermetallic nanoparticles, Te is particularly promising due to its low oxidation tendency, making it well-suited for forming stable compounds in oxygen-rich environments like the ORR.

Table 1. Crystallographic and phase diagram information for intermetallic Ni-Te system.

Name	Formula	Structure Type	Space Group	Homogeneity Range [17] [at% Te]	Lattice Parameters [nm]	Reference
(Ni)	Ni	Cu	Fm-3m	-	a = 3.5181	P. Juhás et al. [18]
β2	Ni3+xTe2	own type	P121/m1	41–42	a = 0.3376 b = 0.3800 c = 0.6100 β = 91.25°	J. Barstard et al. [19]
γ1	NiTe0.775	own type	Pmc21	-	a = 0.3916 b = 0.6860 c = 1.232	K. Klepp et al. [20]
δ	NiTe2−x	NiAs	P63/mmc	51–67	a = 0.3956 c = 0.5370	J. Barstard et al. [19]

summarises the crystallographic and phase diagram data of the intermetallic system nickel–tellurium obtained from bulk studies. The listed NiTe_2−x phase ranges from the structures NiTe to NiTe₂ depending on the intermetallic composition.

In this work we aim to present the study of carbon black-supported single- and multiphase nickel-telluride nanoparticles (Ni-Te/C) as ORR catalysts, as well as highlight the potential of the VS synthesis method for future catalyst development.

2. Results and Discussion

2.1. X-Ray Fluorescence Spectroscopy

The elemental composition of the samples obtained after VS synthesis was analysed using total X-ray fluorescence spectroscopy (TXRF). Observed Ni concentrations for Ni standard solution were within a 95% confidence interval of the nominal concentration and the results for all samples were used without correction. For the Te concentrations, a correction factor of 0.45 was applied based on the findings of recovery rate experiments with a 10 mg L⁻¹ Te standard solution. The product compositions are close to the targeted concentrations and only deviate in the range of a few percents. A systematic minor shift towards higher Ni and Te concentrations than expected was observed. TXRF results are illustrated in Figure 3, and composition values are presented in

Table 2. TXRF results of investigated catalysts.

Sample	ωTXRF (Ni) [wt%]	ωtarget (Ni) [wt%]	ωTXRF (Te) [wt%]	ωtarget (Te) [wt%]
Ni/C	16.2	-	-	-
Ni67-Te33/C	9.6	6.7	9.4	7.2
Ni60-Te40/C	8.6	6.5	10.6	9.5
Ni49-Te51/C	15.6	11.8	28.7	26.9
Ni39-Te61/C	12.3	10.5	39.8	35.0

based on the total powder mass. The good agreement of targeted and observed Te concentrations hints full conversion during the VS synthesis step.

2.2. X-Ray Diffraction and Rietveld Refinement

The pattern of the carbon black-supported material revealed mainly amorphous behaviour with a few minor reflections. The material appeared to be stable during the annealing and reduction step. Since the pattern could not solely be described by the graphite structure (graphite-type, P6₃/mmc) [21], the 14 residual signals were indicated by empirically added peaks. This augmented pattern description is referred to as “simulated support”. The individually refined reflection positions, profile function, and scaling factors yielded a satisfactory fit to the observed pattern of the carbon black material. Figure 4a shows measured PXRD patterns of the catalyst materials and Figure 4b presents a close-up of the refined phases of the representative Ni₆₇-Te₃₃/C sample. Additional refined phase patterns of all samples are presented in Figure S1.

The expected crystal structures include Ni (Cu-type, Fm-3m) [18], NiO (NaCl-type, Fm-3m) [22], Ni_3+xTe₂ (Ni₃Te₂-type, P12₁/m1) [23], NiTe_0.775 (Pmc2₁) [23], and NiTe_2−x (NiAs-type, P6₃/mmc) [20], according to the known phase diagram data (see Table 1). Except for the stoichiometric compound NiTe_0.775, all phases were successfully verified in the various samples. The individual crystallographic parameters for the different phases obtained from Rietveld refinement are listed in

Table 3. Lattice parameters and calculated volume weighted mean column height (LVol-FWHM, Lorentzian Volume Full Width at Half Maximum) obtained from Rietveld refinements of Ni/C and Ni-Te/C samples. * LVol-FWHM value not observed due to overlapping composition sets of the same phase.

Sample	Phase	Lattice Parameters [Å]	LVol-FWHM [nm]
Ni/C	Ni NiO	a = 3.529(9) a = 4.177(0)	27.(9) 1.(2)
Ni67-Te33/C	Ni NiO NiTe2−x Ni3Te2	a = 3.531(3) a = 4.177(0) a = 3.977(4), c = 5.368(2) a = 7.592(2), b = 3.777(1), c = 6.142(5)	54.(8) 1.(3) 36.(1) 26.(2)
Ni60-Te40/C	Ni NiO NiTe2−x Ni3Te2	a = 3.530(5) a = 4.177(0) a = 3.948(8)–3.975(5), c = 5.369(2)–5.371(2) a = 7.586(3), b = 3.770(3), c = 6.136(3)	47.(2) 1.(2) * 22.(9)
Ni49-Te51/C	NiO NiTe2−x	a = 4.177(0) a = 3.902(7)–3.953(0), c = 5.356(1)–5.369(6)	1.(3) *
Ni39-Te61/C	NiO NiTe2−x	a = 4.177(0) a = 3.889(7), c = 5.340(1)	1.(2) 89.(0)

.

Interestingly, NiTe_2−x was present in all intermetallic samples. Signals of the NiTe_2−x reflexes for samples where the targeted composition was not aimed at the NiTe_2−x homogeneity range showed a broadening of the diffraction peak width. This peak broadening is caused by compositional variations and reflected through slightly differing sets of lattice parameters, indicating that those samples did not reach thermodynamic equilibrium during the synthesis procedure. Due to the peak broadening an explicit evaluation of the crystal domain size of the respective NiTe_2−x in the sample is not achievable. The NiTe_2−x phase with NiAs structure-type has a broad, homogeneity range and would allow for intermetallic Ni-Te compositions to extend from Ni₂Te to NiTe to NiTe₂, depending on the occupation of vacancies and interstitial sites [24]. With increasing Te intake of NiTe_2−x, the lattice parameter a decreases [17]. In nanoparticulate state, the stability of intermetallic phases is greatly dependent on the size and morphology of the particles. Phases with a homogeneity range can be beneficial for the stability due to their flexibility regarding composition and activity, contrary to stoichiometric line compounds (like NiTe_0.775). Also, the simpler and comparatively smaller structures are more facile to form in nanoparticles in contrast to complex structures with high unit cell volume. Our observations indicate that NiTe_2−x is the most favourable structure formed from our starting materials. Vacancy formation on the Ni sites in samples with high Te concentration were confirmed by Rietveld refinement. Ni doping on interstitial sites on the other hand could not be seen. Therefore, the phase-pure catalysts composition ranges from NiTe_1.8 to NiTe_1.9. For the samples which were not in thermodynamic equilibrium, NiTe was the dominant composition. A slight shift towards higher Te concentrations is caused by partial oxidation of Ni, reflected by the observation of rather amorphous NiO in the patterns.

2.3. Transmission Electron Microscopy and Energy Dispersive X Ray Spectroscopy

Images of the initial Ni/C starting material before VS synthesis and the resulting intermetallic catalysts were obtained by transmission electron microscopy (TEM). Figure 5a presents representative images of Ni₃₉-Te₆₁/C. The carbon black surface was covered with randomly oriented, well-dispersed, edgy nanoparticles. For particle size distribution analysis, the diameters in both the horizontal and vertical directions of about 400 particles per sample were measured. Comparison of various powder particles allowed the chance of outliers to be minimized. The particle size distributions of Ni/C and Ni₃₉ Te₆₁/C are compared in Figure 5b. Ni/C displayed a bimodal particle size distribution, particles of 11 nm in size being the dominant species accompanied with bigger 45 nm particles. The intermetallic Ni-Te/C powders, on the other hand, showed a more uniform particle size distribution pattern which was shifted towards an average particle size of about 30 nm. The increase in particle size fulfils the expectations of particle growth due to the addition of Te into the nanoparticles. The minor presence of several rather big particles up to 225 nm in the intermetallic products indicates some partial agglomeration and sintering of particles during the VS synthesis, but this effect remains limited. Energy dispersive X-ray spectroscopy (EDS) confirmed the expected nanoparticle compositions. For Ni/C, superficial oxidation in contact to air was observed, confirming the necessity of pre-reduction and handling under inert atmosphere for the vapour–solid reactions. Furthermore, EDS mapping for the elements Ni and Te (Figure 5a,b) confirmed selective Te intake into the Ni nanoparticles during VS synthesis.

2.4. Physisorption

Nitrogen adsorption–desorption isotherms at −196 °C show type IV isotherms typical for mesoporous carbon materials. Figure 6 presents the resulting physisorption isotherms and QSDFT [25] pore size distributions in staggered form. The broad hysteresis loops in the region from 0.45 up to 0.95 P/P₀ can be assigned to a H4-type according to IUPAC standards [26]. Those materials are usually associated with non-rigid aggregates of plate-like particles leading to slit-shaped pores, which could include micropores [27]. At around 0.40–0.45 P/P₀, the cavitation effect leads to the closing of the hysteresis, which does not reflect the true pore size but rather an artefact in the desorption branch [28]. Hence, data evaluation was based on the adsorption branch of the isotherms. The different carbon support of the Pt/C standard material and the in-house prepared Ni/C and Ni-Te/C samples is reflected in the high relative pressure regime. Since the metal loading of the catalysts only slightly affects the porosity, it is concluded that the (inter-)metallic nanoparticles are mostly loaded at the surface of the carbon black support.

For QSDFT calculations performed using the adsorption branch of the isotherms, the kernel for carbon with slit/cylindrical pore geometry was selected. Calculated data including the total pore volume, specific surface area, according to QSDFT and BET, and mean pore size are presented in

Table 4. BET specific surface areas, QSDFT calculated specific surface areas, total pore volume, and pore width.

Sample	SBET [m2 g−1]	SDFT [m2 g−1]	Vtot [cm3 g−1]	wpore [nm]
Pt/C	1198	992	0.84	3.32
Ni/C	662	622	0.60	3.25
Ni67-Te33/C	788	746	0.72	3.39
Ni60-Te40/C	779	749	0.70	3.32
Ni49-Te51/C	411	380	0.38	3.17
Ni39-Te61/C	415	384	0.37	3.25

. QSDFT and BET lead to comparable specific surface area values and an expected trend of decreasing surface areas with increasing metal loading. For the samples Ni₆₇ Te₃₃/C and Ni₆₀ Te₄₀/C, higher surface areas are attributed and align to the lower initial metal loading of the respective educt batch. Furthermore, the calculations revealed the presence of mesopores in all samples. The relatively high adsorption volumes measured at low relative pressure (<0.02) also suggest the presence of micropores in the samples, which could be expected for the carbon-based support.

2.5. Electrocatalytic ORR Investigations

2.5.1. Catalytic Activity

First, rotating disc electrode cyclic voltammetry (CV) measurements at 1600 rpm rotation rate in the 0.1–1.0 V vs. RHE potential range were carried out in argon-purged 1 M KOH electrolyte solution and 1 M KOH + 1 M ethanol electrolyte mixture to investigate ethanol tolerance (Figure S2). All species showed a minor oxidation–reduction couple around 0.8 V vs. RHE, which is attributed to the reduction of residual O₂ quantities in the electrolyte. For Ni-Te materials a small, irreversible peak in the anodic branch of the CV curve occurred at around 0.6 V vs. RHE, which was assigned to oxidation of Te. While for the Pt/C standard catalyst, major ethanol oxidation peaks were observed with the respective electrolyte mixture, the addition of ethanol did not lead to significant effects on the CV runs with the Ni-based materials.

The ORR performance in alkaline media of the catalysts was investigated by linear sweep voltammetry (LSV) investigations in the 1.0–0.1 V vs. RHE potential range in cathodic scan direction at different rotation speeds between 400 to 2000 rpm. LSV measurements were conducted in O₂ saturated 1 M KOH at 30 °C. A commercial Pt/C (10 wt% Pt) standard catalyst is analysed for comparison. The LSV reveal typical behaviour for ORR catalysts with a kinetically controlled region above the reaction onset potential (E_os) of ~0.8 V vs. RHE, followed by a mixed controlled region down to about 0.7 V vs. RHE, where the diffusion-controlled region leads to the respective limitation of the current (Figure 7a) [29]. Onset potentials (E_os), half wave potentials (E_1/2), and diffusion limited current densities (j_D) act as significant characteristics to compare the ORR catalysts performance; the precise values are listed in

Table 5. Results of electrochemical characterization. E_OS = onset potential; E_1/2 = half-wave potential; j_D = limiting current density at 0.4 V vs. RHE; j_Ni = mass activity based on Ni content at 0.4 V vs. RHE; n = electron transfer number.

Catalyst	EOS [V vs. RHE]	E1/2 [V vs. RHE]	jD [mA cm−2]	jNi [A g−1]	n []
Pt/C	0.89	0.79	−2.51	-	3.85
Ni/C	0.86	0.77	−1.72	−43.6	2.31
Ni67-Te33/C	0.84	0.76	−1.81	−76.8	2.30
Ni60-Te40/C	0.84	0.75	−1.86	−88.5	2.30
Ni49-Te51/C	0.85	0.76	−1.74	−45.4	2.30
Ni39-Te61/C	0.86	0.74	−1.63	−54.2	2.25

. The LSV curves of all ORR catalysts at 1600 rpm are presented in Figure 7a. Generally, a positive effect on the ORR activity is observed by the addition of Te. Solely the richest sample in Te, Ni₃₉-Te₆₁/C, did not show a performance gain compared to the Ni/C starting material when comparing the geometric current densities. The onset and half-wave potentials of the Ni-based catalysts reveal similar values as the Pt/C standard catalyst of about 0.85 V vs. RHE and 0.75 V vs. RHE, respectively. However, these observations do not consider that the amount of ‘active’ material significantly decreases with the addition of the comparably heavy Te. Hence, to understand the activity changes due to Te addition by VS synthesis, the Ni-based mass activity (j_Ni) of the catalysts is the characteristic of interest (Figure 7b). In the diffusion-limited region at 0.4 V vs. RHE, the Ni/C catalyst achieved an apparent mass activity of −43.6 A g_Ni⁻¹, by addition of Te the activity increases not in a direct relation with the Te concentration but more dependent on the present crystal structure of the intermetallic Ni-Te compound (Table 3). The highest mass activity of −88.5 A g_Ni⁻¹ was achieved for Ni₆₀-Te₄₀/C and represents a two-fold increase the activity compared to Ni/C. Active intermetallic phases of this catalyst include Ni₃Te₂ and NiTe_2−x with high Ni occupation leading to the NiTe composition. The lower Ni occupation in the NiTe_2−x phase of sample Ni₆₇-Te₃₃/C leads to slightly lower activities of −76.8 A g_Ni⁻¹. The samples containing only NiTe_2−x led to lower performance gains. We concluded that the mass activity of the Ni-based catalysts follows the particular order: Ni₃Te₂ > NiTe > NiTe_2−x > Ni.

Oxygen reduction in alkaline media can rely on various complex mechanisms; the most prominent ones proceed either via a direct four-electron pathway (Equation (1)) or the less efficient two electron hydrogen peroxide pathway (Equation (2)), where HO₂⁻ is reduced in a subsequent step (Equation (3)) [30].

O₂ + 2H₂O + 4e⁻ ↔ 4OH⁻

(1)

O₂ + H₂O + 2e⁻ ↔ HO₂⁻ + OH⁻

(2)

HO₂⁻ + H₂O + 2e⁻ ↔ 3OH⁻

(3)

To determine the influence of different Ni-Te phases present in the catalysts on the reaction pathway of the ORR, the electron transfer numbers (n) were determined by Koutecký–Levich (KL) analysis (Figure 8). A KL plot according to Equation (4) by plotting the reciprocal of the diffusion limiting current over the reciprocal of the square root of the rotation rate was prepared (Figure 8b). From the slope of the respective curve the electron transfer number (n) can be calculated with the linear regression given (Equation (5)).

$${\displaystyle \frac{1}{\mathrm{i}}}={\displaystyle \frac{1}{{\mathrm{i}}_{\mathrm{k}}}}+{\displaystyle \frac{1}{{\mathrm{i}}_{\mathrm{D}}}}={\displaystyle \frac{1}{\mathrm{n}\mathrm{F}\mathrm{A}{\mathrm{k}}_{\mathrm{h}}{\mathrm{C}}_{\mathrm{r}}}}+{\displaystyle \frac{1}{0.62\mathrm{n} \mathrm{F} \mathrm{A} {\mathrm{D}}_{\mathrm{r}}^{2/3}{\mathsf{\nu }}^{-1/6}{\mathrm{C}}_{\mathrm{r}}{\mathsf{\omega }}^{1/2}}}$$

(4)

$$\mathrm{s}\mathrm{l}\mathrm{o}\mathrm{p}\mathrm{e}=0.62\mathrm{n}\mathrm{F}\mathrm{A}{\mathrm{D}}_{\mathrm{r}}^{2/3}{\mathsf{\nu }}^{-1/6}{\mathrm{C}}_{\mathrm{r}}$$

(5)

where i, i_k, and i_D represent the measured, kinetic, and diffusion limiting current, respectively (A), F is the Faraday constant (96,485 C mol⁻¹), A is the geometric electrode area (0.196 cm⁻²), D_r is the diffusion coefficient of O₂ in solution (1.8 × 10⁻⁵ cm² s⁻¹), ν is the kinematic viscosity (0.01 cm² s⁻¹) in 1 M KOH, C_r is the bulk O₂ concentration in solution (7.8 × 10⁷ mol cm⁻³), and ω is the rotation rate (rad s⁻¹) [31]. The resulting electron transfer numbers are presented in Table 5. Overall, the different intermetallic phases appear not to influence the tendency of Ni-based catalysts to operate the ORR via the two-electron pathway with n about 2.3. As expected, the compared Pt/C benchmark catalyst favoured the direct four-electron ORR mechanism.

2.5.2. Ethanol Tolerance and Chronoamperometric Stability

Although the fuel crossover in ADEFCs is comparably small, it cannot be neglected. Ethanol tolerance of the cathode catalysts is therefore targeted [32]. To investigate undesired side reactions in case of ethanol crossover, LSV measurements were repeated in a 1 M KOH + 1 M EtOH electrolyte mixture. The comparison to the mass activity without ethanol present is shown in Figure 9a. No ethanol oxidation was observed during CV and LSV investigations for Ni-based catalysts in contrast to Pt/C which leads to severe oxidation peaks in both the anodic and cathodic CV branches (Figure S2f). Also, the activity trend of the Ni-based catalysts did not change due to ethanol in the solution (Figure 9). Nevertheless, the samples Ni₆₀-Te₄₀/C and Ni₃₉-Te₆₁/C suffer from a significantly stronger decrease in mass activity compared to all other Ni-based catalysts. This effect is attributed to the presence of NiTe in respective samples which appears to induce stronger surface interactions with ethanol, lowering the ORR activity. Additionally, chronoamperometric (CA) measurements were done to screen the catalyst stability during operation as well as simulate ethanol crossover by instantaneously inserting 2 mL ethanol to the electrolyte after 20 min of operation. Immediately after operation starts, an activation of the electrode is observed in CA, which is typically caused by surface reorganisation of catalysts [33]. Ni and Ni-Te catalysts maintain their active state and reach a stable state after a few minutes of operation. The Pt/C benchmark catalyst on the other hand continuously loses activity and only reaches a quasi-stable state after 10 min of operation. When 2 mL of ethanol was added after 20 min of CA operation, a sudden drop of performance to about 50% of the initial activity was observed for Pt/C due to the ethanol oxidation reaction, while the Ni-based catalysts showed remarkable ethanol tolerance and did not show any performance drop. The Ni/C and Ni-Te/C catalysts are not only very stable under operation conditions but also act highly selective towards ORR in alkaline media.

3. Experimental

3.1. Chemicals

Nickel(II) acetylacetonate, for synthesis (Merck, Darmstadt, Germany); carbon powder, steam activated and acid washed, Norit GSX (Alfa Aesar, Karlsruhe, Germany); platinum on carbon, 10 wt% Pt on matrix activated carbon support (Sigma-Aldrich, Darmstadt, Germany); ethanol, 96% (Brenntag, Guntramsdorf, Austria); Nafion^®-117, around 5% in a mixture of water and lower aliphatic alcohols (Sigma-Aldrich, Darmstadt, Germany); KOH pellets, 85% (VWR Chemicals, Darmstadt, Germany); 2-propanol ≥99.5% (Sigma-Aldrich, Darmstadt, Germany); Al₂O₃ MicroPolish suspension, 0.05 µm and 1 µm (Buehler Ltd., Leinfelden-Echterdingen, Germany); Tellurium, 99.99+ % (ASARCO; Tucson, AZ, USA); Tellurium AAS-standard solution, 1000 mg L⁻¹ Te in 2% HNO₃ (Carl Roth, Karlsruhe, Germany); nitric acid, 69% HNO₃ suprapur (Merck, Darmstadt, Germany); Inoxline H5, 95 vol% Ar/5 vol % H₂ (Messer, Gumpoldskirchen, Austria); Hydrogen 5.0, 99.999% H₂ (Messer, Gumpoldskirchen, Austria); Argon 5.0, 99.999% Ar (Messer, Gumpoldskirchen, Austria).

All chemicals were used as purchased. Water was purified and deionized (18.2 MΩcm) in-house by a Milli-Q^® (Merck, Darmstadt, Germany) system.

3.2. Preparation of Ni/C

The synthesis of carbon black-supported nickel was carried out similar to our previous work by a facile wet impregnation–thermal decomposition procedure [16]. In a 250 mL round-bottomed flask, the nickel(II) acetylacetonate was diluted in about 50 mL of ethanol. The stoichiometric amount of carbon black was added and dispersed by 15 min of sonification. Constant stirring at 40 °C and reduced pressure were applied to remove the solvents. Subsequently, the powder was dried at 95 °C under ambient atmosphere for 15 min. The loaded carbon powder was calcinated at 350 °C for 10 h. Reduction of the as-prepared materials was carried out in a tube furnace (Nabertherm, Lilienthal, Germany) at 350 °C in an Inoxline (Ar/H₂ = 95/5 = v/v) gas stream of about 30 mL_N min⁻¹ for 2 h. Adsorbed gases were removed under reduced pressure at 350 °C for 2 h. The produced Ni/C powder were handled and stored under argon atmosphere.

3.3. Vapour Solid Synthesis of Ni Te/C

In a custom-built quartz glass tube, pure tellurium (Te) was weighed to an accuracy of ±0.1 mg. The tellurium was evaporated and condensed as a thin layer on the top side of the quartz tube by heating with an H₂/O₂ torch under static, reduced pressure of about ≤2 × 10⁻² mbar. The respective amount of as prepared Ni/C powder were transferred into the tube under argon atmosphere, to avoid direct contact between the reactants a long funnel was used. The quartz glass tube, with the Ni/C powder on the bottom and the tellurium layer on the top side, was mounted to the vacuum line and sealed under reduced pressure of ≤2 × 10⁻² mbar.

For the vapour–solid reaction, the filled quartz glass tube was aligned in a horizontal, 2-zone furnace (HTM Reetz, Berlin, Germany) in a temperature gradient of 422 °C at the Te side to 448 °C at the Ni/C reaction side. The deposition of Te on the coolest point of the final experimental setup, followed by fading and ultimately disappearance was observed and used to monitor the reaction progress. Full conversion was observed at the earliest after 1 day of reaction. An isothermal post-reaction step of 1 to 3 days at 400 °C was applied for better homogenisation of the samples. The final Ni Te/C catalysts were handled and stored under argon atmosphere.

3.4. Materials Characterization

The quantitative analysis of the materials was conducted using total X-ray fluorescence spectroscopy with a Picofox S2 (Bruker, Berlin, Germany) setup. The catalyst powders, approximately 30 mg each, were dispersed in 20 mL of 2 mol L⁻¹ HNO₃ and stirred continuously at 105 °C for one hour. Afterward, the dispersions were filtered, and the resulting solutions were diluted with ultra-pure water to a total volume of 100 mL. A volume of 500 μL from each solution was mixed in 1.5 mL Eppendorf tubes with 500 μL of a 30 mg L⁻¹ Ga standard solution and 100 μL of a 0.3 g L⁻¹ polyvinyl alcohol solution. For analysis, 5 μL of these sample solutions were transferred onto quartz plates, where they were dried under infrared irradiation and reduced pressure for 45 min. The dried plates were then mounted in the Picofox S2 apparatus and measured for 400 s per sample. Gain correction was performed using an arsenic standard placed on a quartz sample holder. Additionally, for quality control, standard solutions of 10 mg L⁻¹ Ni and 10 mg L⁻¹ Te were measured with 30 mg L⁻¹ Ga as the internal standard.

The phase composition and structural information of the catalysts were analysed through powder X-ray diffraction using a D8 Advance (Bruker, Germany) diffractometer. Measurements were taken in Bragg–Brentano pseudo-focusing mode with theta/theta geometry, and structure refinement was carried out with the Topas 7.13 software. The samples were placed on zero-background silicon single-crystal plates, which were covered with a thin layer of grease to ensure powder fixation. A one-dimensional silicon strip detector, Lynxeye (Bruker, Germany), was installed in the diffractometer. Diffraction patterns were recorded over a 3 h period within a 30° to 90° 2-theta range, using an acceleration voltage of 40 kV and a beam current of 40 mA.

Electron microscopy was used to verify the size of the intermetallic nanoparticles and to perform elemental mapping. Scanning and high-resolution images were captured with a Talos F200i (Thermo Scientific™, Dreieich, Germany) transmission electron microscope (TEM) operating at 200 kV. The microscope was equipped with a field electron gun (FEG) and a 4k × 4k Ceta 16M camera. Data evaluation was done using Velox software (version 3.15, 2024).

Nitrogen adsorption–desorption isotherms were recorded at −196 °C using a QuantaTech Inc. iQ3 (Anton Paar, Graz, Austria) physisorption setup. Prior to measurement, the samples were outgassed at 80 °C for 12 h. The acquired data were evaluated using the Anton Paar QuantaTech Inc. ASiQWin 5.2 software. The Brunauer–Emmett–Teller (BET) equation was applied to determine the specific surface area (SSA) in the relative pressure range of 0.05–0.20 P/P₀. For comparison, SSA and pore size distribution were also calculated using the quenched solid density functional theory (QSDFT) approach, applied to the adsorption branch of the isotherms, which has been reported to yield more accurate values for carbon materials [25,28,34,35] The QSDFT calculations assumed slit/cylindrical pore geometries of carbon. Following the Gurvich rule, the total pore volume was assessed at the plateau region of the isotherm at P/P₀ = 0.95.

3.5. Electrochemical Investigation

Electrochemical ex-situ characterization of the ORR catalysts was performed in a quartz glass cell using a PGSTAT302N Autolab electrochemical workstation (Metrohm, Filderstadt, Germany). The cell contained 200 mL of 1.0 M KOH (VWR pellets dissolved in ultra-pure water) and was maintained at a constant temperature of 30 °C via an external thermostat. Before the measurement, the electrolyte solution was purged with argon gas for 30 min. A typical three-electrode setup was used, consisting of a graphite rod as the counter electrode, a HydroFlex reversible hydrogen electrode (Gaskatel, Kassel, Germany) as the reference electrode, and a glassy carbon rotating disc electrode (GC-RDE) with a disc area of 0.196 cm² (Metrohm, Germany) modified with the catalyst material as the working electrode.

To prepare the catalyst suspension, 4.8 mg of the catalyst powder was dispersed in a mixture of 700 μL ultra-pure water, 250 μL 2-propanol, and 50 μL Nafion^® solution, followed by sonication for 15 min. The GC RDE was polished with 1 μm and 0.05 μm Al₂O₃ suspension on a polish cloth to achieve a mirror finish and then sonicated in ultra-pure water for 5 min. A 10 μL aliquot of the freshly prepared catalyst suspension was drop-cast onto the polished electrode disc and dried under light irradiation for 30 min. This procedure resulted in a final catalyst loading of 0.24 mg cm⁻² on the working electrode.

For the cyclic voltammetry (CV) and linear sweep voltammetry (LSV) measurements, 90% of the iR-drop was corrected using the Nova 2.1.5 software (Metrohm, Germany). To ensure reproducibility, the measurements were performed in triplicates. Prior to catalyst screening, the open circuit potential (OCP) was recorded for at least six minutes. The electrode was considered stable if potential changes of ≤10 mV s⁻¹ were observed during the final minute of OCP measurement. Activation of the working electrode was carried out through a series of CV measurements while continuously purging the electrolyte with argon. This procedure involved alternating cycles: one cycle in the potential range of 0.1 to 1.0 V vs. RHE at a scan rate of 10 mV s⁻¹, followed by 25 cycles at 100 mV s⁻¹. This sequence was repeated until a total of 53 activation cycles had been completed.

To investigate the activity towards ORR, LSV was carried out in O₂ saturated 1 M KOH electrolyte solution during constant oxygen bubbling at different rotation rates (ω) of 400, 800, 1000, 1200, 1600, and 2000 rpm with a scan rate of 10 mV s⁻¹ in cathodic direction in the potential range of 1.0 V vs. RHE to 0.1 V vs. RHE. To determine the ethanol tolerance of the materials, the CV and LSV measurements were repeated in an electrolyte mixture of 1 M KOH and 1 M ethanol with otherwise unchanged conditions.

Catalyst stability was investigated by chronoamperometric (CA) measurements in O₂ saturated 1 M KOH electrolyte solution in the diffusion limited current region at 0.6 V vs. RHE and a rotation rate of 1000 rpm for 30 min. To screen the impact of ethanol on the ongoing ORR, after 20 min of CA, an aliquot of 2 mL ethanol was inserted instantaneously to the cell. For data evaluation the LSV and CA curves were smoothed with the moving averages method in windows of 20 data points.

4. Conclusions

In this study, carbon black-supported nickel-telluride nanoparticles of various elemental ratios and phase compositions were prepared. The vapour–solid (VS) synthesis approach is capable of remarkable compositional control for the intermetallic catalysts, as shown by materials characterization. The impact of different Ni-Te phases on activity and stability of carbon black-supported nickel tellurides towards oxygen reduction reaction (ORR) in alkaline media has been presented for the first time. All Ni-Te phases are tolerant against ethanol and the ORR activity follows the order: Ni₃Te₂ > NiTe > NiTe_2−x > Ni. Alloying Ni with Te enabled doubling the intrinsic mass activity up from −43.6 A g_Ni⁻¹ for Ni/C to −88.5 A g_Ni⁻¹ for Ni-Te/C. The ability to benchmark and study whole intermetallic systems under similar reaction conditions has been demonstrated as beneficial for future catalyst development. For further investigations, a variety of systems could be investigated to find the most active phase of the system in catalytic reactions. As a future perspective, noble metal-based materials could be made more affordable and efficient by increasing the mass activities through formation of intermetallic compounds via the herein demonstrated VS method.