Ni₃S₄/NiS/rGO as a promising electrocatalyst for methanol and ethanol electro-oxidation

All chemical precursors, such as deionized water, pristine graphite, potassium permanganate, hydrochloric acid, sulfuric acid, hydrogen peroxide, phosphoric acid, nickel nitrate hexahydrate (Ni(NO₃)₂.6H₂O), polyvinylidene fluoride, N-methyl pyrrolidone, and thiourea (CH₄N₂S), were purchased with high purification from MERK Company. The crystalline structure of the prepared samples was analyzed using an x-ray diffraction (XRD, PHILIPS PW1730) model with monochromatized Cu-Kα radiation (λ = 1.541874 Å) at room temperature within the range of two thetas of 20°–70°. Raman spectra were analyzed with a micro Raman spectrometer at 532 nm wavelengths (Takram N1-541). The morphology and structure of the synthesized samples were characterized by field emission scanning electron microscope (Mira3 XMU from TESCAN Company), transmission electron microscope (TEM, JEOL 2010), and an energy-dispersive x-ray analyzer.

The synthesis of NN was performed using the hydrothermal method. To prepare NN, 5.4 mmol of Ni(NO₃)₂.6H₂O was dissolved in a 35 ml mix of ethanol and distilled water. Afterwards, 2.4 mmol of CH₄N₂S was added dropwise into the solution until the final color turned green and then stirred for 30 min. Simultaneously, a 1 × 1 cm² piece of cleaned nickel foam (NF) was immersed in the prepared solution. Finally, suspensions were sealed into a Teflon-lined autoclave and maintained at 180 °C for 21 h. Subsequently, the autoclave was cooled to room temperature naturally. The resulting samples were washed several times with distilled water and finally with ethanol to remove the impurities and dried at 50 °C in a vacuum oven. In this research, four samples were synthesized, including NN (S1) (sample without rGO) and three samples of NNR with 0.03 (S2), 0.05 (S3), and 0.07 (S4) grams of GO added to the precursors.

The synthesis method of NNR with the optimal amount of GO was almost the same as that of NN, with the difference that we added 0.05 grams of GO to the precursors at the beginning of the synthesis [34]. It should be mentioned that the optimal sample had the best electrochemical results in the process of methanol and ethanol oxidation.

A three-electrode system was used for electrochemical analysis. A platinum wire with a diameter of 0.5 mm was used as an auxiliary electrode and the Ag/AgCl electrode was our reference electrode. The modified NN and NNR on NF (1 × 1 cm) was our working electrode to investigate the capability of NN and NNR in the methanol and ethanol oxidation process that takes place in the anode of the fuel cell.

For this purpose, first, 1 M KOH solution was prepared, and cyclic voltammetry (CV) analysis of the modified electrodes was performed. As figure 5(a) shows, the addition of rGO to the NN structure increased the current density of the hybrid catalyst. rGO as a conductive material with a high active surface area has a positive effect on the electrochemical process. The CV analysis of the NN catalyst shows a redox peak that is due to the conversion of Ni⁺³ $\leftrightarrow {\text{ }}$Ni⁺² [35]. We can see that adding rGO to the catalyst increases the peak intensity in the CV curve and improves the pseudocapacitive behavior of the catalyst in alkaline media.

Zoom In Zoom Out Reset image size

The electrochemical properties of the catalysts were also investigated through electrochemical impedance spectroscopy (EIS) analysis. Measurements were made in the frequency range from 0.01 Hz to 100 kHz. Figure 5(b) compares the EIS spectra for NNR and NN. As can be seen, all the Nyquist plots of the EIS spectra consist of a rough semicircle in the high-frequency region followed by a linear component in the low-frequency region. The diameter of the semicircle indicates the surface charge transfer resistance (R_ct) caused by Faraday reactions, which is lower in NNR than NN, at 12 and 16 ohms, respectively. According to the slope of the Warburg line, it can be concluded that the ion diffusion rate in NNR is higher than in NN, indicating higher electrochemically active sites in the catalyst.

According to figures 5(a) and (b), we can conclude that the NNR catalyst, having lower R_ct and higher current density, is a better performing catalyst than NN. However, the potential of both catalysts in the oxidation process of methanol and ethanol alcohols was evaluated.

To investigate the behavior of NN and NNR catalysts in the methanol oxidation process, a solution containing 1 M KOH and 0.1 M methanol was prepared. According to figure 6, both NN and NNR catalysts have relatively good potential in the MOR process. The behavioral comparison of these two catalysts indicates the apparent superiority of NNR in the MOR process. This advantage can be inferred from comparing the overvoltage and the maximum current density oxidation peak of the two catalysts. The maximum current density for NN is 8 and for NNR it is 15 mA cm⁻², and the overvoltage for these catalysts is 400 and 420 mV, respectively. In the following, we intend to obtain the optimal concentration of methanol and the optimal scan rate in the MOR process for the two catalysts. However, a solution containing 1 M KOH and different concentrations of methanol (0.1, 0.3, 0.5, 0.7, 0.9, and 1 molar of methanol) was prepared.

Zoom In Zoom Out Reset image size

Figures 7(a) and (b) show the behavior of these two catalysts at different concentrations of methanol at a scan rate of 10 mV s⁻¹. In the MOR process with NN, the current density of methanol oxidation increases in the concentration of 0.5 M methanol, and from this concentration onwards, the current density decreases. Also, the current density variation trend for NNR increases up to a concentration of 0.7 M, and from this concentration onwards it decreases. This behavior is probably due to blocking the penetration of methanol into the catalyst core and preventing the MOR process by-products [36]. It can be concluded that rGO increases the active surface area of NNR and simultaneously causes delays in the saturation of the NNR surface by methanol oxidation by-products.

Zoom In Zoom Out Reset image size

To obtain the optimal scan rate, the CV analysis of NN and NNR catalysts was performed at different scan rates and given concentrations. According to figure 7(c), by increasing the scan rate from 10 to 60 mV s⁻¹, the current density in the MOR process increases, but from 60 mV s⁻¹ onwards, we see a decreasing trend in the current density. For NNR (figure 7(d)), we see an increase in the oxidation peak current density up to 80 mV s⁻¹, followed by a decrease in the 100 and 120 mV s⁻¹ range. It seems that at higher scan rates than the optimal scan rate, the electrolyte and methanol do not have sufficient contact with the catalyst and do not penetrate deep into the catalyst core. The optimal scan rate in the methanol oxidation process for NNR is 80 mV/s, which is higher than NN (60 mV/s). rGO provides a more active surface area in the face of the electrolyte and methanol by enhancing more active electrochemical sites in the catalyst.

In many studies, the mechanism of methanol oxidation has been mentioned as a six-electron mechanism, and based on this, the MOR mechanism in the research is proposed as follows [37]:

$$\begin{equation*}{\text{Catalyst}} + {\text{C}}{{\text{H}}_3}{\text{OH}} \to {\text{Catalyst}} - {\text{C}}{{\text{H}}_3}{\text{O}}{{\text{H}}_{{\text{ads}}}}\end{equation*}$$

$$\begin{equation*}{\text{Catalyst}} - {\text{C}}{{\text{H}}_3}{\text{O}}{{\text{H}}_{{\text{ads}}}} + 4{\text{O}}{{\text{H}}^ - } \to {\text{Catalyst}} - {\left( {{\text{CO}}} \right)_{{\text{ads}}}} + 4{{\text{H}}_2}{\text{O}} + 4{{\text{e}}^ - }\end{equation*}$$

$$\begin{equation*}{\text{Catalyst}}\, + \,{\text{O}}{{\text{H}}^ - }\, \to \,{\text{Catalyst}}\, - \,{\text{O}}{{\text{H}}_{{\text{ads}}}} + \,{{\text{e}}^ - }\end{equation*}$$

$$\begin{equation*}{\text{Catalyst}} - {\text{C}}{{\text{O}}_{{\text{ads}}}} + {\text{Catalyst}} - {\text{O}}{{\text{H}}_{{\text{ads }}}} + {\text{O}}{{\text{H}}^ - } \to {\text{Catalyst}} + {\text{C}}{{\text{O}}_2} + {{\text{H}}_2}{\text{O}} + {{\text{e}}^ - }.\end{equation*}$$

To investigate the effect of temperature on the MOR process, CV analysis was performed at the optimal concentration and optimal scan rate at different temperatures (ambient temperature of 50 °C). As shown in figures 8(a) and (b), the oxidation peak for both catalysts increases with the increasing temperature. Methanol oxidation and the absorption of OH⁻ seem to be facilitated by the increasing temperature.

Zoom In Zoom Out Reset image size

The stability of the catalysts in the MOR process was evaluated by performing 1000 CV scans at the optimal concentration and optimal scan rate. As shown in figure 8(c), NN has a good cyclic stability of 92%, and this stability for NNR is 97% (figure 8(d)); both stability values are relatively acceptable. rGO increases the electrical conductivity and provides a high surface area for NNR, thus facilitating the MOR process at NNR in comparison with NN. In addition, in order to check the stability of the NNR catalyst in the methanol oxidation process, chronoamperometry analysis was performed at peak potential. As shown in the inset of figure 8(d), the catalyst maintains about 70% of the initial current density after 2000 s, which is a relatively acceptable amount.

The ethanol oxidation capability of NN and NNR catalysts was evaluated by performing a CV test in an alkaline medium (figure 9). For this purpose, a solution consisting of 1 M KOH and 0.1 M ethanol was prepared. As shown in figure 8, the two catalysts are relatively capable of ethanol oxidation. Due to the stronger carbon bonds of ethanol than methanol, the oxidation of this alcohol by catalysts naturallyoccurs at higher overvoltages and lower current densities [38]. A comparison of NN and NNR behavior at different concentrations of ethanol was evaluated to obtain the optimal concentration by performing a CV test.

Zoom In Zoom Out Reset image size

As shown in figures 10(a) and (b), the oxidation current density in NN and NNR catalysts increases to 0.3 and 0.5 methanol concentrations, respectively, and after these critical concentrations, we see a decrease in the peak current density with the saturation of the surface of the electrodes by the by-products of ethanol oxidation. In addition, the presence of rGO in the catalyst structure increases the surface area of the catalyst and delays the creation of contamination-affected by-products [39].

Zoom In Zoom Out Reset image size

A CV test was performed using the optimal concentrations of ethanol at different scan rates, to obtain the optimal scan rate in the process of ethanol oxidation by two catalysts.

Figures 10(c) and (d) show that by changing the scan rate to 40 and 60 mV s⁻¹, the peak oxidation current density has an upward trend, and from this scan rate onwards, it was decreased in NN and NNR, respectively.

In general, at scan rates higher than the optimal scan rate, the electrolyte and alcohol do not have enough opportunity to contact and penetrate deep into the catalyst, so the peak current density will be decreased.

The mechanism of the ethanol oxidation process by catalysts can be presented as follows [40]:

$$\begin{equation*}{\text{Catalyst}} + {\text{O}}{{\text{H}}^ - } \to {\text{Catalyst}} - {\text{O}}{{\text{H}}_{{\text{ads}}}} + {{\text{e}}^ - }\end{equation*}$$

$$\begin{equation*}{\text{Catalyst}} + {\text{C}}{{\text{H}}_3}{\text{C}}{{\text{H}}_2}{\text{OH}} \to {\text{Catalyst}} - {({\text{C}}{{\text{H}}_3}{\text{C}}{{\text{H}}_2}{\text{OH}})_{{\text{ads}}}}\end{equation*}$$

${\text{Catalyst}} - {({\text{C}}{{\text{H}}_3}{\text{C}}{{\text{H}}_2}{\text{OH}})_{{\text{ads}}}} + 3{\text{O}}{{\text{H}}^ - } \to {\text{Catalyst}} - {({\text{C}}{{\text{H}}_3}{\text{CO}})_{{\text{ads}}{} }} + 3{{\text{H}}_2}{\text{O}} + 3{{\text{e}}^ - }$ ${\text{Catalyst}} - {({\text{C}}{{\text{H}}_3}{\text{CO}})_{{\text{ads}}}} + {\text{Catalyst}} - {\text{O}}{{\text{H}}_{{\text{ads}}}} \to {\text{Catalyst}} - {({\text{C}}{{\text{H}}_3}{\text{COOH}})_{{\text{ads }}}} + {\text{Catalyst}}$

$$\begin{equation*}{\text{Catalyst}} - {({\text{C}}{{\text{H}}_3}{\text{C}}OOH)_{{\text{ads}}}} + {\text{O}}{{\text{H}}^ - } \to {\text{Catalyst}} + {\text{C}}{{\text{H}}_3}{\text{CO}}{{\text{O}}^ - } + { }{{\text{H}}_2}{\text{O}}.\end{equation*}$$

In the EOR process, we also see an increase in the current density with the increasing temperature. Figures 11(a) and (b) show that by increasing the temperature in the range of environments to 50 °C, the current density for both NN and NNR catalysts increases, and the ethanol oxidation process is facilitated by increasing the temperature.

Zoom In Zoom Out Reset image size

The stability measurements of the two catalysts were also investigated by performing 1000 consecutive CV cycles at the optimal scan rate and concentration for NN and NNR in the EOR process.

As figure 11(c) indicates, after this number of consecutive cycles, NN still has a relatively good stability of 90% and NNR shows a stability of 94% (figure 11(d)) due to the role of rGO in its structure as a superior component for the creation of higher electrical conductivity and active surface area than NN. The stability of the NNR catalyst in the EOR process was determined by chronoamperometry analysis at the oxidation peak potential. As shown in the inset of figure 11(d), the catalyst maintains about 65% of the initial current density after 2000 s.

Table 1 compares the performance of the NNR catalyst in the methanol and ethanol oxidation process with other research.