An Electrocatalytic Activity of AuCeO2/Carbon Catalyst in Fuel Cell Reactions: Oxidation of Borohydride and Reduction of Oxygen

This paper describes the investigation of electrocatalytic activity of the AuCeO2/C catalyst, prepared using the microwave irradiation method, towards the oxidation of sodium borohydride and oxygen reduction reactions in an alkaline medium. It was found that the obtained AuCeO2/C catalyst with Au loading and electrochemically active surface area of Au nanoparticles (AuNPs) equal to 71 μg cm−2 and 0.05 cm2, respectively, showed an enhanced electrocatalytic activity towards investigated reactions, compared with the Au/C catalyst with an Au loading and electrochemically active surface area of AuNPs equal to 78 μg cm−2 and 0.19 cm2, respectively. The AuCeO2/C catalyst demonstrated ca. 4.5 times higher current density values for the oxidation of sodium borohydride compared with those of the bare Au/C catalyst. Moreover, the onset potential of the oxygen reduction reaction (0.96 V) on the AuCeO2/C catalyst was similar to the commercial Pt/C (0.98 V).


Introduction
Conventional combustion-based technologies with high emission rates pose a significant threat regarding air pollution, health, and the climate. Their operation requires enormous natural resources that are not eternal. Therefore, the demand for alternative energy sources has mobilized scientists around the world to search for and research such energy sources. One of them is fuel cells, which can operate at higher efficiencies than combustion engines and can convert the chemical energy in the fuel to electrical energy with efficiencies of up to 60%; it is notable that fuel cells have lower emissions than combustion engines [1]. Different kinds of fuel cells such as direct alcohol fuel cells (DAFC), alkali fuel cells (AFC), direct borohydride fuel cells (DBFC), Proton Exchange Membrane fuel cells (PEMFC), Solid Oxide fuel cells (SOFC), and others have been investigated intensely within the past few decades [2]. All of them have their advantages and disadvantages, however the principle of their operation is similar-fuel oxidation on the anode side and oxygen reduction on the cathode side [3]. Lately, the most attention has attracted DAFC, DBFC, PEMFC because the materials used in these fuel cells can be more easily produced, stored, and transported compared to the other fuel cells, making them more advantageous. DBFC has attracted the attention of researchers because of sodium borohydride (sodium borohydride anion BH 4 − ) potential to generate extremely pure hydrogen on demand or just be directly oxidized in a DBFC. Moreover, the BH 4 − has higher volumetric (7314 Whdm −3 ) and gravimetric (7100 Whkg −1 ) energy density than methanol (4800 Whdm −3 ) and (6000 Whkg −1 ), respectively [4][5][6][7]. BH 4 − is also more stable in an

Results and Discussion
In our previous study, AuCeO 2 /C prepared by microwave irradiation method has been found as an efficient catalyst toward ethanol, methanol, ethylene glycol, and glycerol oxidation reaction [31]. Therefore, we decided to investigate this catalyst more widely. This paper presents an investigation of the electrocatalytic activity of as-prepared AuCeO 2 /C and Au/C catalysts toward sodium borohydride oxidation and oxygen reduction reactions. For comparison, CeO 2 /C substrate (called as CeO 2 /C catalyst) was also investigated for BOR and ORR. The AuCeO 2 /C, Au/C, and CeO 2 /C catalysts were characterized using various methods of materials characterization. The data obtained were described in detail in Ref. [31]. The XRD data confirmed the crystallographic planes of cubic Au and CeO 2 phases for the investigated CeO 2 /C, AuCeO 2 /C, and Au/C catalysts. The diffraction peaks at the 2θ = 38.18, 44.39, and 64.58 • were assigned to the 111, 200, and 220 crystallographic planes of the cubic Au phase for the AuCeO 2 /C and Au/C catalysts. The diffraction peaks at the 2θ = 28.58, 33.12, 47.54, and 56.41 • corresponding to the (111), (200) (220), and (311) crystallographic planes of the cubic CeO 2 phase were established for the AuCeO 2 /C and CeO 2 /C catalysts [31]. The changes in the Au and CeO 2 lattice parameters were not detected, indicating that no metal alloy or metal solid solution was formed in the AuCeO 2 /C catalyst. The estimated size of Au crystallites was equal to 9.3 ± 0.6 nm and 3.6 ± 0.4 for the Au/C and AuCeO 2 /C catalysts, respectively.
The TEM examination showed spherical AuNPs with a diameter less than 20 nm for the Au/C (Figure 1a) and AuCeO 2 /C (Figure 1b) catalysts. The high-resolution TEM (HRTEM) images [31] revealed the differences between the size of Au crystallites and AuNPs. It has been found that AuNP was composed of several Au crystallites of different orientations. Additionally, the AuNPs presented a polycrystalline structure with the characteristic lattice spacing of~0.23 nm, corresponding to the (111) plane of Au [31,32]. (HRTEM) images [31] revealed the differences between the size of Au crystallites and AuNPs. It has been found that AuNP was composed of several Au crystallites of different orientations. Additionally, the AuNPs presented a polycrystalline structure with the characteristic lattice spacing of ~0.23 nm, corresponding to the (111) plane of Au [31,32]. It has been determined that Au in a metallic phase for both Au/C and AuCeO2/C catalysts exists [31,33,34]. The investigation of the O 1s XPS spectra elucidated the lattice oxygen species for the CeO2/C and AuCeO2/C catalysts [31,34]. A Ce was found to correspond to Ce 4+ for both CeO2/C and AuCeO2/C catalysts [31,35,36]. These XPS results show that CeO2 in CeO2/C and AuCeO2/C exists in oxide form. Obviously, C 1s XPS spectra confirmed the presence of carbon for all catalysts studied [31,37,38].
To provide comprehensive information about the structure and dispersion of active components on the surface of the prepared catalysts, the Au/C and AuCeO2/C catalysts were additionally investigated by SEM. Figure 2 shows surface topography images of the synthesized AuCeO2/C ( Figure 2a) and Au/C (Figure 2b) catalysts and total spectra of Au, Ce, O, and C elements (Figure 2(a2,b2)). From the enlarged pictures of the EDS area it can be seen that Au is evenly distributed on the surface of the substrate (Figure 2(a1,b1)). Mass spectra reveal the elemental composition of investigated catalysts (Figure 2(a2,b2)). It has been determined that Au in a metallic phase for both Au/C and AuCeO 2 /C catalysts exists [31,33,34]. The investigation of the O 1s XPS spectra elucidated the lattice oxygen species for the CeO 2 /C and AuCeO 2 /C catalysts [31,34]. A Ce was found to correspond to Ce 4+ for both CeO 2 /C and AuCeO 2 /C catalysts [31,35,36]. These XPS results show that CeO 2 in CeO 2 /C and AuCeO 2 /C exists in oxide form. Obviously, C 1s XPS spectra confirmed the presence of carbon for all catalysts studied [31,37,38].
To provide comprehensive information about the structure and dispersion of active components on the surface of the prepared catalysts, the Au/C and AuCeO 2 /C catalysts were additionally investigated by SEM. Figure 2 shows surface topography images of the synthesized AuCeO 2 /C ( Figure 2a) and Au/C (Figure 2b) catalysts and total spectra of Au, Ce, O, and C elements (Figure 2(a2,b2)). From the enlarged pictures of the EDS area it can be seen that Au is evenly distributed on the surface of the substrate (Figure 2(a1,b1)). Mass spectra reveal the elemental composition of investigated catalysts (Figure 2(a2,b2)).
The electrochemically active surface areas (ESA) of AuNPs were estimated from the integrated charge of the Au oxide reduction region (Q AuO ) in the cyclic voltammograms [31,39] and calculated according to Equation (1): where 400 µC cm −2 is the charge required to reduce a monolayer of Au oxide. The obtained ESA values were 0.05 and 0.19 cm 2 for AuCeO 2 /C and Au/C, respectively [31]. The Au loading in the prepared catalysts determined using ICP-OES was equal to 71 and 78 µg cm −2 for AuCeO 2 /C and Au/C, respectively [31].   Furthermore, Figure 3 presents the cyclic voltammograms of the investigated catalysts recorded in 1 M NaOH and 0.05 M BH 4 − + 1 M NaOH solutions at a sweep rate of 50 mV s −1 . The obtained data correspond to the theoretical ones. Firstly, it can be clearly seen that there were no evident peaks in a 1 M NaOH solution for all investigated catalysts. Meanwhile, the CVs of Au/C and AuCeO 2 /C recorded in a 0.05 M BH 4 − + 1 M NaOH solution showed three typical peaks of the BH 4 − oxidation. As is evident, on the forward scan the obtained peak A1 at approximately −0.26 V was attributed to the direct potentially eight-electron oxidation of BH 4 − [9,13]. However, there was no response to the direct oxidation of BH 4 − associated with the peak A1 in the CV of CeO 2 /C. The obtained peak A2 at approximately 0.26 V was related to the oxidized surface of the catalysts followed by an immediate decrease in the activity of the electrode, which reveals that the oxidized surface is relatively inactive towards the BH 4 − oxidation reaction [9]. In addition, the decrease in current density at 0.3-0.4 V may be attributed to the passivation of the electrode surface because of the adsorption of intermediates of BH 4 − oxidation. On the reverse scan, a sharp peak B that appeared at approximately 0.2 V was attributed to the reactivation of the electrode surface [9,13]. As seen in Figure 3, the AuCeO 2 /C catalyst outperformed Au/C for the oxidation of BH 4 − .
ESA (cm 2 ) = QAuO (µC)/400 (µC cm -2 ), where 400 µC cm -2 is the charge required to reduce a monolayer of Au oxide. The obtained ESA values were 0.05 and 0.19 cm 2 for AuCeO2/C and Au/C, respectively [31]. The Au loading in the prepared catalysts determined using ICP-OES was equal to 71 and 78 µg cm −2 for AuCeO2/C and Au/C, respectively [31]. Furthermore, Figure 3 presents the cyclic voltammograms of the investigated catalysts recorded in 1 M NaOH and 0.05 M BH4ˉ + 1 M NaOH solutions at a sweep rate of 50 mV s -1 . The obtained data correspond to the theoretical ones. Firstly, it can be clearly seen that there were no evident peaks in a 1 M NaOH solution for all investigated catalysts. Meanwhile, the CVs of Au/C and AuCeO2/C recorded in a 0.05 M BH4ˉ + 1 M NaOH solution showed three typical peaks of the BH4ˉ oxidation. As is evident, on the forward scan the obtained peak A1 at approximately −0.26 V was attributed to the direct potentially eight-electron oxidation of BH4ˉ [9,13]. However, there was no response to the direct oxidation of BH4ˉ associated with the peak A1 in the CV of CeO2/C. The obtained peak A2 at approximately 0.26 V was related to the oxidized surface of the catalysts followed by an immediate decrease in the activity of the electrode, which reveals that the oxidized surface is relatively inactive towards the BH4ˉ oxidation reaction [9]. In addition, the decrease in current density at 0.3-0.4 V may be attributed to the passivation of the electrode surface because of the adsorption of intermediates of BH4ˉ oxidation. On the reverse scan, a sharp peak B that appeared at approximately 0.2 V was attributed to the reactivation of the electrode surface [9,13]. As seen in Figure 3, the AuCeO2/C catalyst outperformed Au/C for the oxidation of BH4ˉ. Comparison of current density, mass, and specific activities for the Au/C and AuCeO2/C catalysts is illustrated in Figure 4. It is clear that the AuCeO2/C catalyst exhibited more negative onset potential. Current density values that correspond to the direct BH4ˉ oxidation were approximately 4.5 times higher at the AuCeO2/C compared to the Au/C (cf. Figure 4b,c). The BH4ˉ oxidation current densities normalized by the Au loading (mA mgAu −1 ) and electrochemically active surface area of AuNPs (mA cm −2 ) represent the mass and specific activity of the AuCeO2/C and Au/C catalysts, respectively (Figure 4b,c). As could be expected due to similar Au loading in the catalysts, the mass activity for BH4ˉ oxidation was approximately five times higher at the AuCeO2/C compared with that for Comparison of current density, mass, and specific activities for the Au/C and AuCeO 2 /C catalysts is illustrated in Figure 4. It is clear that the AuCeO 2 /C catalyst exhibited more negative onset potential. Current density values that correspond to the direct BH 4 − oxidation were approximately 4.5 times higher at the AuCeO 2 /C compared to the Au/C (cf. Figure 4b,c). The BH 4 − oxidation current densities normalized by the Au loading (mA mg Au −1 ) and electrochemically active surface area of AuNPs (mA cm −2 ) represent the mass and specific activity of the AuCeO 2 /C and Au/C catalysts, respectively (Figure 4b,c). As could be expected due to similar Au loading in the catalysts, the mass activity for BH 4 − oxidation was approximately five times higher at the AuCeO 2 /C compared with that for the Au/C catalyst ( Figure 4b). The values of specific activities for BH 4 − oxidation were approximately 17 times higher at the AuCeO 2 /C catalyst than those at the Au/C catalyst ( Figure 4c). The electrocatalytic enhancement of the CeO 2 supported catalyst was attributed to the CeO 2 synergistic electronic effect with Au. the Au/C catalyst ( Figure 4b). The values of specific activities for BH4ˉ oxidation were approximately 17 times higher at the AuCeO2/C catalyst than those at the Au/C catalyst (Figure 4c). The electrocatalytic enhancement of the CeO2 supported catalyst was attributed to the CeO2 synergistic electronic effect with Au.  Figure 5 presents the electrochemical durability of the AuCeO2/C, Au/C, and CeO2/C catalysts investigated using chronoamperometry. The chronoamperometric measurement (CA) curves were recorded at constant voltage equal to −0.26 V. As seen in Figure 5a, current density slightly decreases over time (t < 200 s) and gradually reaches the steadystate of the oxidation of BH4ˉ reaction on the investigated catalysts. The current density remains constant throughout the experiment. Moreover, at the end of the experimental period, the BH4ˉ oxidation current density is ca. 5.5 times larger on the AuCeO2/C catalyst than that on the Au/C catalyst. The higher CA current density indicates better electrocatalytic activity. Additionally, the current density values at the end of the experimental period were normalized by the Au loading and ESA of AuNPs for AuCeO2/C and Au/C catalysts. The results obtained showed the same tendency the AuCeO2/C catalyst exhibited: approximately 6-and 21-times higher mass and specific activity values compared with those of the Au/C catalyst.   Figure 5 presents the electrochemical durability of the AuCeO 2 /C, Au/C, and CeO 2 /C catalysts investigated using chronoamperometry. The chronoamperometric measurement (CA) curves were recorded at constant voltage equal to −0.26 V. As seen in Figure 5a, current density slightly decreases over time (t < 200 s) and gradually reaches the steady-state of the oxidation of BH 4 − reaction on the investigated catalysts. The current density remains constant throughout the experiment. Moreover, at the end of the experimental period, the BH 4 − oxidation current density is ca. 5.5 times larger on the AuCeO 2 /C catalyst than that on the Au/C catalyst. The higher CA current density indicates better electrocatalytic activity. Additionally, the current density values at the end of the experimental period were normalized by the Au loading and ESA of AuNPs for AuCeO 2 /C and Au/C catalysts. The results obtained showed the same tendency the AuCeO 2 /C catalyst exhibited: approximately 6-and 21-times higher mass and specific activity values compared with those of the Au/C catalyst. the Au/C catalyst (Figure 4b). The values of specific activities for BH4ˉ oxidation were approximately 17 times higher at the AuCeO2/C catalyst than those at the Au/C catalyst (Figure 4c). The electrocatalytic enhancement of the CeO2 supported catalyst was attributed to the CeO2 synergistic electronic effect with Au.  Figure 5 presents the electrochemical durability of the AuCeO2/C, Au/C, and CeO2/C catalysts investigated using chronoamperometry. The chronoamperometric measurement (CA) curves were recorded at constant voltage equal to −0.26 V. As seen in Figure 5a, current density slightly decreases over time (t < 200 s) and gradually reaches the steadystate of the oxidation of BH4ˉ reaction on the investigated catalysts. The current density remains constant throughout the experiment. Moreover, at the end of the experimental period, the BH4ˉ oxidation current density is ca. 5.5 times larger on the AuCeO2/C catalyst than that on the Au/C catalyst. The higher CA current density indicates better electrocatalytic activity. Additionally, the current density values at the end of the experimental period were normalized by the Au loading and ESA of AuNPs for AuCeO2/C and Au/C catalysts. The results obtained showed the same tendency the AuCeO2/C catalyst exhibited: approximately 6-and 21-times higher mass and specific activity values compared with those of the Au/C catalyst.  Hereinafter, the electrocatalytic activity of the prepared AuCeO 2 /C, Au/C, and CeO 2 /C catalysts toward ORR are presented. Firstly, the cyclic voltammograms were recorded in an Ar or O 2 saturated 0.1 M NaOH solution by comparing the ORR performance of the catalysts ( Figure 6). As seen in Figure 6a-c, all investigated catalysts showed an evident reduction peak in the O 2 saturated solution, however, different oxygen reduc- tion potential and current density values. The most positive oxygen reduction potential together with a higher peak current density was observed for the AuCeO 2 /C catalyst if compared with that for Au/C and CeO 2 /C (Figure 6d). Hereinafter, the electrocatalytic activity of the prepared AuCeO2/C, Au/C, and CeO2/C catalysts toward ORR are presented. Firstly, the cyclic voltammograms were recorded in an Ar or O2 saturated 0.1 M NaOH solution by comparing the ORR performance of the catalysts ( Figure 6). As seen in Figure 6a-c, all investigated catalysts showed an evident reduction peak in the O2 saturated solution, however, different oxygen reduction potential and current density values. The most positive oxygen reduction potential together with a higher peak current density was observed for the AuCeO2/C catalyst if compared with that for Au/C and CeO2/C (Figure 6d). Then, liner sweep voltammograms were scanned in the O2 -saturated 0.1 M NaOH solution from 1.0 V in the cathodic direction to 0.2 V vs. the reversible hydrogen electrode (RHE) at a scan rate of 5 mV s −1 . The rotation speed was varied from 400 to 2000 rpm (Figure 7). The obtained ORR response coincides with that reported in the literature of 0.7 and 0.9 V, indicating that the ORR response was dominated by the kinetics of electrocatalysts (Figure 7a-c) [29,30,40]. Furthermore, the ORR polarization curves were used to obtain the Koutecky-Levich (K−L) plots for each catalyst (Figure 7d-f). In general, a good linear fitting was seen between j −1 and w −1/2 , indicating the first-order dependence of the ORR kinetics at different potentials (0.4-0.7 V). The calculated electron transfer number (n) for the CeO2/C and Au/C catalysts was approximately two, indicating that these catalysts favor a two-electron transfer reaction path and O2 is reduced to HO2ˉ and OH − [41,42]. Meanwhile, the calculated n for the  Figure 7). The obtained ORR response coincides with that reported in the literature of 0.7 and 0.9 V, indicating that the ORR response was dominated by the kinetics of electrocatalysts (Figure 7a-c) [29,30,40]. Furthermore, the ORR polarization curves were used to obtain the Koutecky-Levich (K−L) plots for each catalyst (Figure 7d-f). In general, a good linear fitting was seen between j −1 and w −1/2 , indicating the first-order dependence of the ORR kinetics at different potentials (0.4-0.7 V). The calculated electron transfer number (n) for the CeO 2 /C and Au/C catalysts was approximately two, indicating that these catalysts favor a two-electron transfer reaction path and O 2 is reduced to HO2 − and OH − [41,42]. Meanwhile, the calculated n for the AuCeO 2 /C catalysts was approximately 3.8, indicating the 4 e − transfer reaction and direct reduction of O 2 to H 2 O on the surface of this catalyst [43,44]. AuCeO2/C catalysts was approximately 3.8, indicating the 4 e − transfer reaction and direct reduction of O2 to H2O on the surface of this catalyst [43,44].   (Figure 8a,b), and chronoamperometric data of ORR (Figure 8c,d). Additionally, the LSV scan of the commercial Pt/C catalyst that has the Pt loading of 0.071 mg Pt cm -2 was recorded in the same manner. From the scans in Figure 8a, it can be clearly seen that the ORR onset potential at 0.96 V for AuCeO2/C was similar to the ORR onset potential of 0.98 V for the commercial Pt/C. Meanwhile, the Au/C and CeO2/C catalysts demonstrated the ORR onset potential at 0.88 and 0.75 V, respectively. Noteworthily, AuCeO2/C and Au/C prepared by the microwave irradiation method exhibited more positive potential com-   (Figure 8a,b), and chronoamperometric data of ORR (Figure 8c,d). Additionally, the LSV scan of the commercial Pt/C catalyst that has the Pt loading of 0.071 mg Pt cm −2 was recorded in the same manner. From the scans in Figure 8a, it can be clearly seen that the ORR onset potential at 0.96 V for AuCeO 2 /C was similar to the ORR onset potential of 0.98 V for the commercial Pt/C. Meanwhile, the Au/C and CeO 2 /C catalysts demonstrated the ORR onset potential at 0.88 and 0.75 V, respectively. Noteworthily, AuCeO 2 /C and Au/C prepared by the microwave irradiation method exhibited more positive potential compared with our earlier reported AuCeO 2 /C (0.80 V) [30] and AuCo 3 O 4 /C (0.88 V) [29] catalysts. Figure 8b reveals the comparison of mass and specific activities for AuCeO 2 /C and Au/C calculated from LSV at a potential value of 0.8 V at a rotation speed of 1600 rpm. As seen from the data obtained, the mass and specific activity are approximately 3.5 and Catalysts 2021, 11, 342 9 of 15 12 times higher, respectively, at the AuCeO 2 /C than that obtained at Au/C. From the data obtained can be assumed that when Au and CeO 2 are together in the composition the electrocatalytic activity of the AuCeO 2 /C catalyst towards the oxygen reduction reaction could be significantly enhanced. This enhancement could be attributed to the synergistic effect of present metal and metal oxide in the catalyst composition. The durability of the prepared CeO 2 /C, Au/C, and AuCeO 2 /C catalysts was evaluated by CA measurements at a potential value of 0.55 V for 1900 s. (Figure 8c). Initially, in the first 100-200 s, a decrease in current density in the ORR was observed for all investigated catalysts. However, after approximately 200 s, the current density settled down and remained constant throughout the experiment (Figure 8c). Figure 8d presents the normalized ORR current densities (percentage of current density retained vs. operation time) for each catalyst after 200 s. It was calculated that investigated catalysts can maintain 94.0, 92.0, 87.0, and 80% in the order Pt/C, AuCeO 2 /C, Au/C, and CeO 2 /C, respectively, of the initial current density under continuous operation for 1900 s under a constant voltage of 0.55 V. This indicates that the prepared AuCeO 2 /C catalyst demonstrated significantly high stability during the ORR process in the alkaline medium compared to commercial Pt/C as well as to Au/C and CeO 2 /C. pared with our earlier reported AuCeO2/C (0.80 V) [30] and AuCo3O4/C (0.88 V) [29] catalysts. Figure 8b reveals the comparison of mass and specific activities for AuCeO2/C and Au/C calculated from LSV at a potential value of 0.8 V at a rotation speed of 1600 rpm. As seen from the data obtained, the mass and specific activity are approximately 3.5 and 12 times higher, respectively, at the AuCeO2/C than that obtained at Au/C. From the data obtained can be assumed that when Au and CeO2 are together in the composition the electrocatalytic activity of the AuCeO2/C catalyst towards the oxygen reduction reaction could be significantly enhanced. This enhancement could be attributed to the synergistic effect of present metal and metal oxide in the catalyst composition. The durability of the prepared CeO2/C, Au/C, and AuCeO2/C catalysts was evaluated by CA measurements at a potential value of 0.55 V for 1900 s. (Figure 8c). Initially, in the first 100-200 s, a decrease in current density in the ORR was observed for all investigated catalysts. However, after approximately 200 s, the current density settled down and remained constant throughout the experiment (Figure 8c). Figure 8d presents the normalized ORR current densities (percentage of current density retained vs. operation time) for each catalyst after 200 s. It was calculated that investigated catalysts can maintain 94.0, 92.0, 87.0, and 80% in the order Pt/C, AuCeO2/C, Au/C, and CeO2/C, respectively, of the initial current density under continuous operation for 1900 s under a constant voltage of 0.55 V. This indicates that the prepared AuCeO2/C catalyst demonstrated significantly high stability during the ORR process in the alkaline medium compared to commercial Pt/C as well as to Au/C and CeO2/C.  Finally, the performance of the NaBH 4 -H 2 O 2 fuel cell was evaluated using the AuCeO 2 /C and Au/C catalysts as the anode material. The fuel cell polarization curves and the corresponding power densities against the current density at temperatures in a range of 25-55 • C are given in Figure 9. The fuel cell maintained an open circuit voltage approximately at 1.55 V for the Au/C catalyst (Figure 9a) and 1.7 V for the AuCeO 2 /C catalyst (Figure 9b). It can be seen that NaBH 4 -H 2 O 2 fuel cell power density values increased exponentially with increasing temperature. The achieved peak power density of 69.8 mW cm −2 at 25 • C for AuCeO 2 /C was two times greater as compared with that (31.6 mW cm −2 ) obtained for Au/C. Increasing the temperature from 25 to 55 • C, the power density increases approximately 1.5 and 2 times for the AuCeO 2 /C and Au/C catalysts, respectively. More detailed electrochemical parameters of the NaBH 4 -H 2 O 2 fuel cell using Au/C and AuCeO 2 /C catalysts as an anode are listed in Table 1. In general, power density peak values obtained in this study were higher compared with those for fuel cells described in the literature [45][46][47]. Finally, the performance of the NaBH4-H2O2 fuel cell was evaluated using the AuCeO2/C and Au/C catalysts as the anode material. The fuel cell polarization curves and the corresponding power densities against the current density at temperatures in a range of 25-55 °C are given in Figure 9. The fuel cell maintained an open circuit voltage approximately at 1.55 V for the Au/C catalyst (Figure 9a) and 1.7 V for the AuCeO2/C catalyst (Figure 9b). It can be seen that NaBH4-H2O2 fuel cell power density values increased exponentially with increasing temperature. The achieved peak power density of 69.8 mW cm −2 at 25 °C for AuCeO2/C was two times greater as compared with that (31.6 mW cm −2 ) obtained for Au/C. Increasing the temperature from 25 to 55 °C, the power density increases approximately 1.5 and 2 times for the AuCeO2/C and Au/C catalysts, respectively. More detailed electrochemical parameters of the NaBH4-H2O2 fuel cell using Au/C and AuCeO2/C catalysts as an anode are listed in Table 1. In general, power density peak values obtained in this study were higher compared with those for fuel cells described in the literature [45][46][47].

Preparation of Catalysts
The AuCeO 2 /C catalyst was prepared using the microwave irradiation method as described in [31]. Briefly, the CeO 2 /C substrate was obtained by mixing dry powder of CeO 2 with carbon powder (the mass ratio being 1:1) in a 2-propanol solution by ultrasonication for 30 min with further desiccation of the mixture. Then, 100 mg of CeO 2 /C substrate was used for the deposition of AuNPs. The CeO 2 /C was added to the 6 mL solution containing 1.9 mM of HAuCl 4 , 1 M of glycerol, and 14 mM NaOH and sonicated for 1 h. The prepared mixture was heated at a temperature of 170 • C for 30 s in the microwave reactor Monowave 300 (Anton Paar). The synthesized AuCeO 2 /C catalyst powder was filtered and dried in a vacuum oven at 80 • C for 2 h. The Au/C catalyst was prepared in the same way.

Characterization of Catalysts
The crystal structure and lattice parameters of the synthesized catalysts were characterized using powder X-ray diffraction. XRD patterns of the prepared catalysts were recorded using a D8 diffractometer (Bruker AXS) equipped with a Cu Kα radiation using a Ni/graphite monochromator. A step-scan mode was used in the 2-theta range from 20 • to 90 • with a step length of 0.02 • and a counting time of 5 s per step.
The shape, size, and dispersion of Au nanoparticles on the substrate were examined by TEM analysis using a transmission electron microscope Tecnai G2 F20 X-TWIN equipped with an EDAX spectrometer with an r-TEM detector. For microscopic examinations, 10 mg of sample was first sonicated in 1 mL of ethanol for 1 h, and then the obtained mixture was deposited on the Cu grid covered with a continuous carbon film.
The elemental composition and state of the prepared catalysts were explored by X-ray photoelectron spectroscopy. The XPS analysis was carried out using an "ESCALABMKII" spectrometer (VG Scientific, UK) equipped with an Al Kα X-ray radiation source (1486.6 eV) operated at a fixed pass energy of 20 eV.
The surface morphology of the synthesized catalysts and the distribution of elements in the catalyst was analyzed by SEM using a scanning electron microscope (TM4000Plus, Hitachi). The measurements were made using double-sided adhesive tape (12 mm OD, Pelco Tabs, Ted Pella), which was glued to a pallet for measurements. The surface of the adhesive tape was coated with the catalyst powder.
The Au loading in the catalysts was estimated employing Inductively Coupled Plasma Optical Emission Spectroscopy (ICP-OES). The ICP optical emission spectra were recorded using an ICP optical emission spectrometer Optima 7000 DV (Perkin Elmer).

Electrochemical Measurements
Electrochemical measurements were undertaken on the electrochemical workstation AUTOLAB and rotating disc electrode (RDE) in a conventional three-electrode cell. A Pt sheet was used as a counter electrode and an Ag/AgCl/KCl electrode was used as a reference. The glassy carbon RDE with a geometric area of 0.07 cm 2 coated by catalyst ink was used as the working electrode. The catalyst ink was obtained according to the following steps: at first, 10 mg of the investigated catalysts were dispersed ultrasonically for 1 h in a 100 µL 2% polyvinylidenefluoride in N-methyl-2-pyrrolydone solution. With the aim to keep a similar amount of Au during the measurement, 71 and 78 µg cm −2 for AuCeO 2 /C and Au/C, respectively, the required amount of the prepared suspension mixture was pipetted onto the polished surface of a glassy carbon electrode and dried in an oven at a temperature of 80 • C for 4 h.
The cyclic voltammetry (CV) measurements were performed in a 0.5 M H 2 SO 4 solution for the determination of electrochemically active surface areas of the prepared catalysts. The electrode potential was cycled in a range of 0-1.5 V with a scan rate of 50 mV s −1 . The electro-oxidation of sodium borohydride (BH 4 − ) in an alkaline medium was investigated by CV in a 0.05 M BH 4 − + 1 M NaOH solution in a range of −0.9-0.6 V at a sweep rate of 50 mV s −1 . Chronoamperometric measurements (CA) were carried out in the same solution at a constant potential value of −0.2 V for 1900 s. The electrode potential values are quoted vs. the silver/silver chloride electrode (Ag/AgCl/KCl sat ).
The CV measurements of ORR were recorded in an Ar or O 2 saturated 0.1 M NaOH solutions in the electrode potential range from 0.1 to 1.2 V at a scan rate of 5 mV s −1 . A 0.1 M NaOH solution was saturated with Ar or O 2 for 10 min before each CV test of ORR. An RDE linear sweep voltammetry curves were recorded in an O 2 -saturated 0.1 M NaOH solution (saturated with O 2 for 10 min before each ORR test) in the electrode potential range from 1 to 0.2 V in the cathodic direction at a scan rate of 5 mV s −1 , varying rotation speed from 0 to 2000 rpm. CA for ORR was recorded at a potential value of 0.55 V in an O 2 -saturated 0.1 M NaOH solution for 1900 s. The electrode potential values for all ORR measurements are quoted vs. the reversible hydrogen electrode (RHE).
Koutecky-Levich (K-L) Equations (2)-(4) were used for calculation of the electron transfer number per oxygen molecule (n) in ORR: j d = 0.62nFD 2/3 Co 2 V −1/6 ω 1/2 (3) where, j, j k, and j d are the measured current density, kinetic current density, and the diffusion limiting current density (mA cm −2 ), respectively; n is the number of electrons transferred in the reaction; F is the Faraday constant (F = 96,485 C mol −1 ); D is the diffusion coefficient of the reactant (1.9 × 10 −5 cm 2 s −1 ); Co 2 is the concentration of the reactant in the bulk electrolyte (1.2 × 10 −3 mol L −1 ); V is the kinetic viscosity of the electrolyte (1.13 × 10 −2 cm 2 s −1 ); w is the rotation rate; A is the slope of the linear plot of j −1 vs. w −1/2 (K-L plot) [48].
All measurements were carried out at a temperature of 25 • C. All solutions were deaerated with Ar before each measurement except after O 2 saturation for ORR measurements. To facilitate comparison of the obtained data with those described in the literature, the electrode potential for the BH 4 − oxidation measurements is quoted vs. Ag/AgCl. Electrode potential for ORR measurements is quoted vs. RHE. The presented current densities are normalized concerning the geometric area (0.07 cm 2 ) of the working electrode.

Fuel Cell Test Measurements
The performance of direct borohydride-hydrogen peroxide fuel cell (NaBH 4 /H 2 O 2 ) was evaluated using an electrochemical fuel cell testing system at a temperature in a range of 25-55 • C. A 0.7 cm 2 geometric area glassy carbon electrode coated with the AuCeO 2 /C catalyst was employed as the anode. Catalyst loading was calculated to be the same as for CV measurements. A Pt sheet was used as the cathode. The anode and cathode chambers were separated using a Nafion N117 membrane and filled with 100 mL of the corresponding aqueous electrolyte. The anolyte was composed of an alkaline mixture of 0.05 M BH 4 − in 4 M NaOH. The catholyte was composed of 5 M H 2 O 2 in 1.5 M HCl. The anolyte and catholyte solutions were prepared immediately before the measurements. The presented current densities are normalized to the geometric area of the anode electrode. The cell testing was performed using a Zennium electrochemical workstation (ZAHNER-Elektrik GmbH & Co. KG).

Conclusions
In summary, we deposited AuNPs on the CeO 2 /C and C substrates preparing the AuCeO 2 /C and Au/C catalysts. Because of the synergistic effect between CeO 2 and Au in the AuCeO 2 /C catalyst, this catalyst exhibited an enhanced electrocatalytic activity toward the BOR and ORR compared with that of the bare Au/C and Pt/C catalysts. It has been found that the AuCeO 2 /C catalyst showed approximately 4.5 times higher electrocatalytic activity toward the oxidation of BH 4 − . Ca. 1.5 times greater power density values in the direct NaBH 4 -H 2 O 2 fuel cell were obtained by employing AuCeO 2 /C as the anode compared with those for Au/C. Moreover, the synthesized AuCeO 2 /C catalyst demonstrated similar onset potential (0.96 V vs. 0.98 V) of ORR and maintenance of initial ORR current density (92% vs. 94%) compared to commercial Pt/C.