3D Co-Ni-C Network from Milk as Competitive Bifunctional Catalysts for Methanol and Urea Electrochemical Oxidation

: Methanol oxidation (MOR) and urea oxidation (UOR) have been considered for new types of fuel cells, but the lack of highly active nonnoble metal catalysts restricts such cells. A NiCo-modiﬁed biomass carbon (milk as the carbon source)-based catalyst with a 3D structure is synthesized by using salt templates. The results show that 3D-C-NiCo (1:1) exhibits excellent MOR and UOR properties with a potential of 1.33 V vs. RHE and 1.35 V vs. RHE at 10 mA cm − 2 , respectively. MOR and UOR reactions not only can replace the oxygen evolution reaction (OER) in consumption of electrolytic water but also can effectively degrade wastewater pollution rich in methanol and urea.


Introduction
The search for new green energy is an effective way to address the increasingly serious energy crisis and environmental issues [1,2]. Novel fuel cells, such as methanol fuel cells (DMFCs) and urea fuel cells (DUFCs), have attracted extensive interest from scientists due to their high energy density and official sources [3][4][5]. Methanol oxidation (MOR) and urea oxidation (UOR) are the key reactions of these fuel cells [5,6]. In the preparation process of hydrogen energy, the reaction kinetics of hydrogen evolution from electrolytic water are slow; the overpotential required for the reaction is large, and the total energy consumption of the reaction is high [7]. Therefore, the reaction rate of hydrogen evolution from the cathode is limited to a certain extent. The MOR and UOR replacing the oxygen evolution reaction (OER) could significantly decrease the total energy consumption of the water electrolysis reaction and accelerate the hydrogen evolution reaction [8,9]. MOR and UOR can also alleviate the pollution of methanol wastewater produced in the organic process [10,11].
Pt-based materials have good properties in MORs and UORs, but they are difficult to apply on a large scale because Pt-based materials are precious metals [4,12]. Therefore, nonnoble metals have been used to replace noble metal catalysts for MORs and UORs. In the last several years, nickel-based catalysts have been widely available for oxygen evolution reactions (OERs), hydrogen evolution reactions (HERs), UORs, and MORs [13][14][15][16][17][18][19]. Many experiments have shown that the performance of MORs or UORs could be improved by the combination of nickel-based catalysts with cobalt or other elements, such as 3D NiCo 2 O 4 architectures [20], nickel phosphates [21], and Mn 2+ -doped Ni 3 S 2 [22]. There are some Ni and Co-based materials that exhibit excellent catalytic activity for MORs and UORs, which are shown in the Tables S1 and S2.
Because of the excellent oxidation properties of Ni, Co-based and alloy materials, we designed the use of salt templates and Ni and Co precursors after carbonization to prepare 3D structured NiCo alloy-modified carbon-based materials. At the same time, different proportions of catalysts were prepared by calcination at the same temperature. The catalyst with Ni:Co (1:1) has the best UOR and MOR performance.

Results and Discussion
All catalysts were synthesized by calcining salt and milk, as shown in Figure 1a. It shows the SEM of the samples, indicating a 3D porous structure in Figure 1b. In Figure 1b and Figure S1, the element distribution diagram of the sample contains Co, Ni, P, C, and N. 3D-C-CoNi (1:1) has a layered structure, showing a relatively thin state (Figure 1c). In transmission electron microscopy (TEM), the local structure information of the sample was observed, and the surface was porous (Figure 1d). The lattice structure of the sample can be clearly seen by high resolution transmission electron microscope (HR-TEM) characterization in Figure 1e. The NiCo is uniform distribution with the 3D carbon, and a clear structure is shown in the TEM (Figure 1c-e). The carbon layer is tightly combined with the metal to ensure good conductivity in 3D-C-NiCo (1:1) catalysts. The structural stripe is 0.211 nm, corresponding to CoNi (111), which is the same as that reported in the literature [23,24]. Further characterization shows that the sample has a clear diffraction ring in Figure 1f, indicating that the CoNi alloy in the sample has a good crystalline state.
Catalysts 2021, 11, x FOR PEER REVIEW 2 of 3D structured NiCo alloy-modified carbon-based materials. At the same time, differen proportions of catalysts were prepared by calcination at the same temperature. The cata lyst with Ni:Co (1:1) has the best UOR and MOR performance.

Results and Discussion
All catalysts were synthesized by calcining salt and milk, as shown in Figure 1a. I shows the SEM of the samples, indicating a 3D porous structure in Figure 1b. In Figure 1 and Figure S1, the element distribution diagram of the sample contains Co, Ni, P, C, and N. 3D-C-CoNi (1:1) has a layered structure, showing a relatively thin state (Figure 1c). In transmission electron microscopy (TEM), the local structure information of the sampl was observed, and the surface was porous (Figure 1d). The lattice structure of the sampl can be clearly seen by high resolution transmission electron microscope (HR-TEM) char acterization in Figure 1e. The NiCo is uniform distribution with the 3D carbon, and a clea structure is shown in the TEM (Figure 1c-e). The carbon layer is tightly combined with the metal to ensure good conductivity in 3D-C-NiCo (1:1) catalysts. The structural strip is 0.211 nm, corresponding to CoNi (111), which is the same as that reported in the litera ture [23,24]. Further characterization shows that the sample has a clear diffraction ring in Figure 1f, indicating that the CoNi alloy in the sample has a good crystalline state.  The X-ray powder diffraction (XRD) pattern of the catalyst (Figure 2a and Figure S2) shows diffraction peaks of the CoNi alloy at 44.5 • (111), 51.5 • (200) and 76.0 • (220) [23]. It shows that 3D-C-CoNi (1:1) metal exists in the form of an alloy. The electronic structures of the elements in 3D-C-CoNi (1:1) were analyzed by X-ray photoelectron spectroscopy (XPS). In the full spectrum ( Figure 2b and Table S3), there were distinct C, N, P, Ni, and Co elements. The 284.80 and 286.30 eV are graphene-like C and C in C-O, respectively ( Figure S3). The phosphorus is rich in phospholipids of Milk, which could be doped in the carbonization process. In the high-resolution graph, the binding energy corresponds to zero valence Ni at 855.64 (Ni 2p 3/2 ) and 872.99 eV (Ni 2p 1/2 ), while the binding energy is fitted for Ni 2p 3/2 (861.95 eV) and Ni 2p 1/2 (879.70 eV) in Ni 2+ (Figure 2c) [24]. The binding energy at 780.93 eV is the zero valence state of Co (Figure 2d), while the 786.49 and 796.74 eV are fitting peaks, indicating Co 2p 3/2 and 2p 1/2 for C O 2+ , respectively [24]. In Figure 2e, it reveals the presence of four N species in 3D-C-CoNi (1:1), namely, pyridine nitrogen (398.42 eV), metal-coordinated N (399.27 eV), pyrrole nitrogen (400.82 eV), and oxidized nitrogen (403.21 eV); moreover, P-C (133.27 eV) and P-O (134.42 eV) were confirmed to be present in the P2p spectrum [25,26]. Pyridine-N, metal-N, and graphite-N of 3D-C-CoNi (1:1) are the active centers, which potentially improve its electrocatalytic activity.
that 3D-C-CoNi (1:1) metal exists in the form of an alloy. The electronic structures of the elements in 3D-C-CoNi (1:1) were analyzed by X-ray photoelectron spectroscopy (XPS). In the full spectrum ( Figure 2b and Table S3), there were distinct C, N, P, Ni, and Co elements. The 284.80 and 286.30 eV are graphene-like C and C in C-O, respectively ( Figure  S3). The phosphorus is rich in phospholipids of Milk, which could be doped in the carbonization process. In the high-resolution graph, the binding energy corresponds to zero valence Ni at 855.64 (Ni 2p3/2) and 872.99 eV (Ni 2p1/2), while the binding energy is fitted for Ni 2p3/2 (861.95 eV) and Ni 2p1/2 (879.70 eV) in Ni 2+ (Figure 2c) [24]. The binding energy at 780.93 eV is the zero valence state of Co (Figure 2d), while the 786.49 and 796.74 eV are fitting peaks, indicating Co 2p3/2 and 2p1/2 for CO 2+ , respectively [24]. In Figure 2e, it reveals the presence of four N species in 3D-C-CoNi (1:1), namely, pyridine nitrogen (398.42 eV), metal-coordinated N (399.27 eV), pyrrole nitrogen (400.82 eV), and oxidized nitrogen (403.21 eV); moreover, P-C (133.27 eV) and P-O (134.42 eV) were confirmed to be present in the P2p spectrum [25,26]. Pyridine-N, metal-N, and graphite-N of 3D-C-CoNi (1:1) are the active centers, which potentially improve its electrocatalytic activity.  The cyclic voltammetry (CV) and linear sweep voltammetry (LSV) curves measured in different electrolytes are exhibited in Figure 3a,b. The potential of the OER for 3D-C-CoNi (1:1) in 1 M KOH electrolyte is 1.55 V vs. RHE (@10 mA cm −2 ). The 3D-C-CoNi (1:1) has a potential of 1.35 V vs. RHE (@10 mA cm −2 ), which indicates that UOR could proceed smoothly. The comparison of LSV curves for urea oxidation performance of all catalysts calcined at different proportions (Co, Ni) is shown in Figure 3c. The results show that the overpotential of 3D-C-CoNi (1:1) is the lowest at the same current density, indicating that 3D-C-CoNi (1:1) has the best urea oxidation performance in Figure 3c and Figure S4a. The 3D-C-CoNi (1:1) electrocatalyst shows fast UOR reaction kinetics because of the smaller Tafel slopes (128 mV dec −1 ) compared to other samples (Figure 3d).
To study the ECSA, we also measured the C dI of 3D-C-CoNi (1:1) in Figures S5 and S6, which is slightly higher than that of other catalysts, indicating that the active site exposure was caused by metal-coordinated N and nanoalloyed cobalt-nickel. Electrochemical impedance spectroscopy (EIS) was tested (@ 1.4 V) in Figure 3e. The results show that 3D-C-CoNi (1:1) has the fastest catalytic kinetic process because of the smallest semicircle radius (Figure 3e). The increase in surface area and the 3D structure of 3D-C-CoNi (1:1) are conducive to the function of active sites, and the structure also accelerates the charge transfer in UOR performance. A stability test of urea oxidation performance is shown in Figure 3f. Significantly, 3D-C-CoNi (1:1) showed highly enhanced catalytic activity for UOR, but 3D-C-CoNi (1:2) could stably catalyze urea oxidation for 4000 s, which means that the increase in nickel content could enhance the catalytic performance. To study the ECSA, we also measured the CdI of 3D-C-CoNi (1:1) in Figures S5 and  S6, which is slightly higher than that of other catalysts, indicating that the active site exposure was caused by metal-coordinated N and nanoalloyed cobalt-nickel. Electrochemical impedance spectroscopy (EIS) was tested (@ 1.4 V) in Figure 3e. The results show that 3D-C-CoNi (1:1) has the fastest catalytic kinetic process because of the smallest semicircle radius (Figure 3e). The increase in surface area and the 3D structure of 3D-C-CoNi (1:1) are conducive to the function of active sites, and the structure also accelerates the charge transfer in UOR performance. A stability test of urea oxidation performance is shown in Figure 3f. Significantly, 3D-C-CoNi (1:1) showed highly enhanced catalytic activity for UOR, but 3D-C-CoNi (1:2) could stably catalyze urea oxidation for 4000 s, which means that the increase in nickel content could enhance the catalytic performance.   The EIS of MOR (at 1.4 V vs. RHE) was measured in Figure 4e. All semicircles of the samples have the smallest radius, and the semicircles have the largest radius. A small semicircle radius implies that the charge transfer impedance for methanol oxidation is  The EIS of MOR (at 1.4 V vs. RHE) was measured in Figure 4e. All semicircles of the samples have the smallest radius, and the semicircles have the largest radius. A small semicircle radius implies that the charge transfer impedance for methanol oxidation is small, which may be related to the synergism between 3D carbon and CoNi to accelerate charge transfer [27]. The electrochemical stability of 3D-C-CoNi (1:1) is also a very significant factor to be evaluated by chronoamperometry, as shown in Figure 4f. The results show that 3D-C-CoNi (1:1) has the highest oxidation performance. The current density at 4000 s was 20.2 mA g −1 , which was 1.9-and 2.1-fold greater than that observed for 3D-C-CoNi (1:2) and 3D-C-CoNi (2:1), respectively.

Experimental Materials and Preparation of the Catalysts
Milk (from cow) was produced at Inner Mongolia Shenniu Emulsion Development Co., Ltd. (Hohhot, China). Deionized water was homemade in the laboratory. All experimental reagents were used directly without further treatment, and detailed information is provided in the Supporting Information.
We weighed 10 g of NaCl, 11 g of milk and 0.4441 g of CoCl 2 ·6H 2 O with an analytical balance. The above materials were mixed evenly and dissolved with deionized (40 mL) water. The mixture was lyophilized in liquid nitrogen and dry a freeze-drying machine for 24 h with an internal temperature of −50 • C and a vacuum of 19 Pa. The above samples were heated at 5 • C/min and kept 800 • C for 2 h, and the above operation was carried out in an Ar atmosphere (40 cm 3 /min) in a tubular furnace. Finally, the NaCl was washed off by centrifugation with deionized water, and the samples were dried for 12 h. Samples were collected and named 3D-C-Co. Samples with different mass ratios of Ni and Co (denoted as 3D-C-Co:Ni (1:2), 3D-C-Co:Ni (2:1), and 3D-C-Co:Ni (1:1) and 3D-C-Ni were prepared in the same way as 3D-C-Co. Figure 1a shows the synthetic process of the material. More detailed information is shown in the Supporting Information.

Characterizations of the Catalysts
XRD was performed using a Rigaku UItima IV (Tokyo, Japan) with Cu Kα radiation (λ = 1.5406 Å) at 40 kV and 40 mA. Field emission transmission electron microscopy (FE-TEM) was performed using a Tecnai G2 F20 (Hillsboro, OR, USA) with a voltage of 200 kV, and Cu grids served as the substrate. XPS was performed using an ESCALAB 250Xi (Waltham, MA, USA), and Al was used as the anode XR50 source (hν = 1486.6 eV) at 150 W.

Electrochemical Test
The three-electrode system was using a CHI 760E (Shanghai, China) electrochemical workstation with Pt mesh, Hg/HgO and carbon paper (CP). The catalyst slurry was a mixture of water, ethanol and Nafion prepared by ultrasonication for 30 min at a volume ratio of 1:9:10. The catalyst and slurry were prepared at a ratio of 5 mg per 1 mL and treated with ultrasonication for 30 min. Twenty-five milliliters of the catalyst slurry was evenly coated on carbon paper (0.5 cm × 0.5 cm), and the catalyst load was 0.5 mg cm −2 .

Conclusions
The results show that 3D-C-NiCo (1:1) is an inexpensive, efficient, and persistent catalyst for MOR and UOR with a potential of 1.33 and 1.35 V vs. RHE at 10 mA cm −2 . The high performance of 3D-C-NiCo (1:1) is due to the three-dimensional and nanocrystalline composite structure, thus providing more active sites for the reaction. These kinds of nonnoble metal catalysts provide a new method for MOR and UOR in fuel cells, as well as the degradation of organic pollutants and the reduction of potential water decomposition.