Introduction

The direct methanol fuel cell (DMFC) [1], as a portable fuel cell, has received widespread attention [2], and the methanol oxidation reaction (MOR) is the most critical step to enhance the energy conversion efficiency [3, 4]. It is found that both precious metals and transition metals show highly catalytic performance for MOR [5, 6]. Among them, the precious metal platinum (Pt) has the optimal catalytic performance [7], but the high cost and low utilization of platinum precious metal restrict the development of platinum-based catalysts [8]. In addition, the slow electrooxidation kinetics and CO intoxication of Pt electrocatalysts during MOR usually result in a rapid loss of electrocatalytic performance [9,10,11].

Therefore, developing low-cost alternative catalysts with high activity, high stability, as well as high durability is important for DMFC [12, 13]. At present, the platinum content of commercial platinum–carbon catalysts used in fuel cells is more than 20 wt%, but its catalytic activity and utilization rate are not satisfactory. Numerous efforts have been devoted to overcome this major obstacle and accelerate the commercialization of DMFC [14,15,16,17,18]. On the one hand, platinum-based catalysts can be designed with a core–shell structure [19, 20] or a porous structure to decrease the platinum content by increasing the active sites [21]; on the other hand, according to the synergistic effect between multiple metals, platinum nanocrystals can be combined with transition metals to form multi-element alloy catalysts, which can improve the catalytic efficiency with reducing the platinum content [22,23,24,25]. In addition, non-precious metal catalysts can be explored to replace platinum-based catalysts, and according to the d-band center theory [26], for example, a series of transition metal oxides [27,28,29,30], sulfides [31, 32], and carbides [33] are used to explore non-noble metal catalysts in electrocatalysis applications [34]. The electronic structure of transition metal carbides is connected with the interaction between metal atoms and carbon [35,36,37,38,39,40,41]. Therefore, by inhibiting the agglomeration of the active phase during the carbonization process, more catalytic sites are exposed to enhance the catalytic efficiency of the carbide [42].

Due to immediate availability and better charge storage properties, numerous nanocomposites show important application prospects in the fields of photovoltaics, sensors, energy conversion, and energy storage [43, 44]. Chitosan with NaI was synthesized by Rahman et al. and applied as a solid polymer electrolyte for dye-sensitised solar cells to improve their ionic conductivity [45]. Composite gel polymer electrolyte (CGPE) was used by Amici et al. to enhance lithium ion conductivity in safer Li-O2 batteries [46]. Carbonized composite nanostructured silicon-sulfur composites were synthesized by Alidoost et al. and used to replace lithium metal anodes in sulfur batteries [47]. A laser-induced graphene (LIG) network captures and interconnects activated carbon (AC) particles and was used by Reina et al. to increase the electrical double-layer capacitance of LIG electrodes [48]. Polymanganese oxide was coated on the surface of porous carbon nanofibers by Abdah et al. to increase catalytic activity and buffer volume changes during cycling [49].

MOFs have a porous framework structure [50], which is a good functional material to prepare carbide composites [22]. MOFs were used as the key precursors [51] for low-cost and controlled synthesis of nano-structures [52]. Zeolitic imidazolate frameworks (ZIFs) [53] are considered to be one of the most suitable precursors [22, 54,55,56,57]. For example, ZIF-L [58] and ZIF-M-X [31] are metal–carbon hybrid precursors with large surface areas [54]. In addition, it is difficult for MOFs to be used directly as electrocatalysts because of its poor conductivity. In order to solve the disadvantage, several MOFs were designed to obtain high-performance electrocatalysts with good electrical conductivity and chemical stability through forming highly conductive carbon [59] and N-doped metal nanohybrids [60]. It is also possible to prepare carbon-supported metal oxides, phosphides, or metal nanoparticles through a special post-treatment process [59].

Here, through simple microwave-assisted reduction, a small amount of Pt nanoparticles was deposited on the surface of the Mo/Co@NC, and multi-dimensional Pt–Mo/Co@NC nanocomposites with low platinum content (6.5 wt%) are finally obtained to explore their application in methanol oxidation.

Experimental section

Reagents and materials

Cobaltous nitrate (Co(NO3)2·6H2O), 2-methylimidazole (C4H6N2), ammonium molybdate tetrahydrate ((NH4)6Mo7O24·4H2O), polyvinylpyrrolidone (PVP), ethanol (C2H5OH), potassium hydroxide (KOH), methanol (CH3OH), and ascorbic acid (AA) were purchased obtained Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Nafion (5wt%) and K2PtCl4 were purchased from Sigma-Aldrich.

Catalyst preparation

Co-ZIF NPs

The Co-ZIF NPs were prepared based on the literature. Firstly, 4 mmol Co(NO3)2·6H2O and 16 mmol PVP were evenly dispersed in 20 mL methanol to form solution I and sonication for 10 min. Then, 8 mmol of 2-methyl imidazole was uniformly dispersed in 10 mL of methanol to obtain solution II. Then, the solution II was added to the solution I, and, after evenly stirring, the reaction was hydrothermally reacted at 120 °C for 2 h. Finally, the obtained products were centrifuged, washed, and dried.

Mo/Co-ZIF NPs

Firstly, 10 mg (NH4)6Mo7O24·4H2O was uniformly dispersed in the aqueous solution to form solution A, and 100 mg of Co-ZIF was uniformly dispersed 20 ml of methanol to form solution B. Secondly, solution A was injected to B and evenly stirred for 2 h. Finally, Mo/Co-ZIF NPs were obtained after being dried at 60 °C for 12 h under vacuum.

Mo/Co@NC NPs

The as-prepared Mo/Co-ZIF precursor was pyrolyzed at 900 °C for 2 h in an N2 atmosphere, and then Mo/Co@NC nanocomposite could be obtained. In addition, Co@NC nanocomposite materials could be obtained by the same method.

Pt–Mo/Co@NC NPs

A total of 20 mg of Mo/Co@NC was uniformly dispersed in 20 mL of deionized water, and then 50 mg of AA was added. After ultrasonic stirring, 10 mg of K2PtCl4 evenly dispersed in 5 ml of deionized water was added to the mixed solution, which then was irradiated by microwave (400 W) at 85 °C for 10 min. After centrifugal washing, the resulting black substance was freeze-dried to obtain the Pt–Mo/Co@NC. For comparation, the Pt–Co@NC nanocomposite materials could be obtained by the same method.

Physicochemical characterization

The surface structure and micromorphology of the as-prepared catalysts were analyzed by transmission electron microscopy (TEM) and HRTEM (Tecnai Talos, Netherlands). X-ray diffraction (XRD) was implemented using a D8-Advance diffractometer (Germany) with a Cu Ka radiation source (β = 0.15418 nm). Element distribution was detected by energy-dispersive X-ray spectroscopy (EDX, JSM6510LV, Japan) and field emission scanning electron microscopy (FE-SEM, JSM7100F). The elemental composition of the as-prepared catalysts was analyzed through X-ray photoelectron spectroscopy (XPS; Thermo Fisher Scientific Escalab250Xi).

Electrochemical characterization

The CHI760E (Shanghai Chenhua, China) was used for electrochemical analysis at room temperature in a three-electrode configuration. A modified glass carbon electrode (GCE) (3 mm in diameter), a platinum electrode, and a saturated calomel electrode (SCE) were used as the work, counter, and reference electrodes, respectively. Each specimen was ultrasonically treated with 500 μL of ethanol to obtain a uniform catalyst ink of 2 mg mL−1. After ultrasonic cleaning for 60 min, a 2 μL droplet of the resulting mixture was pressed onto the polished GCE and dried at 25 ℃ for 1 h. After ultrasonic cleaning for 60 min, a 2 μL catalyst ink was pressed onto the cleaned GCE and dried at room temperature. N2 was blown into the electrolyte for 30 min to avoid the influence of air for the test results. The electrolyte solution for the methanol oxidation test consisted of 1.0 M KOH and 1.0 M CH3OH.

Results and discussion

The synthesis method of Pt–Mo/Co@NC

The preparation steps of Pt–Mo/Co@NC nanocomposites are shown in Scheme 1. Firstly, the as-prepared Co-ZIF was dispersed in a methanol solution. Secondly, a moderate ammonium molybdate solution was added and stirred for 1 ~ 2 h. During this process, the surface Co2+ ions of the Co-ZIF were released into the solution and the molybdate ions in the solution could form coordination bonds with the unetched Co2+ to obtain Mo/Co-ZIF composite materials [41]. Then, the Mo/Co-ZIF precursors were calcinated to obtain the Mo/Co@NC composite material. Finally, Pt–Mo/Co@NC nanocomposite material was obtained by a simple microwave-assisted reduction method.

Scheme 1
scheme 1

The schematic illustration of the synthesis process of Pt–Mo/Co@NC

Full size image

Surface morphology and chemical structure characterization

The XRD pattern of the Co-ZIF and Mo/Co-ZIF is analyzed in Fig. 1a. The XRD diffraction peaks of Co-ZIF polyhedrons suggest the successful preparation of Co-ZIF crystals [31]. Based on the XRD analysis, the peak positions of Mo/Co-ZIF and Co-ZIF were obviously overlapped, but the peak intensity of Mo/Co-ZIF reduced a lot, which shows that the crystal structures of Mo/Co-ZIF and Co-ZIF were basically the same and the decrease of the peak intensity might be due to the separation of Co2+ in Co-ZIF. In Fig. 1b, the peak positions of the Mo/Co@NC nanocomposites at 44.2°, 51.4°, and 75.8° corresponded to the (111), (200), and (220) planes of element Co (PDF#15–0806), respectively. Besides, there was a clear peak at around 25°, which corresponded to the (002) plane of derived carbon. Compared with the Co@NC derived from the pure Co-ZIF polyhedron, it could be found that the peak intensity of the Mo/Co@NC was slightly weakened, but the overall crystal structure was the same with Co@NC. Finally, after a small amount of Pt nanoparticles were deposited over Mo/Co@NC surface, XRD analysis of Pt–Mo/Co@NC showed that the 2θ diffraction peaks at 39.7°, 46.2°, and 67.4° can be matched to the (111), (200), and (220) planes of metallic Pt (PDF#04–0802), respectively. At the same time, the crystal structure of Mo/Co@NC did not change; the decrease of the peak intensity might be attributed to that part of the cobalt was etched away. For comparison, the XRD analysis on Pt–Co@NC was performed, whose crystal structure was consistent with Pt–Mo/Co@NC, which showed that Co@NC was still the foremost composition.

Fig. 1
figure 1

a XRD patterns of the Co-ZIF and Mo/Co-ZIF. b XRD patterns of Co@NC, Mo/Co@NC, Pt–Co@NC, and Pt–Mo/Co@NC

Full size image
Fig. 2
figure 2

a XPS survey spectrum of Pt–Mo/Co@NC and high-resolution XPS spectra of b Co 2p, c Mo 3d, d Pt 4f, e N 1 s, and f C 1 s regions

Full size image

The crystal structures and electronic state for Pt–Mo/Co@NC and Pt–Co@NC were characterized by XPS analysis (Figs. 2 and S1). As revealed in Fig. 2a, the survey spectrum showed Pt–Mo/Co@NC was composed of Mo, Co, N, C, O, and Pt elements. The high-resolution Co 2p spectrum at the binding energies of 778.2, 796.9.0/781.9, and 780.0 eV [38] could be regarded as Co0, Co2+, and Co3+ (Fig. 2b), and the presence of Co2+ and Co3+ might be due to the oxidized surface of the metal Co when the sample was exposed to the air. Besides, the peaks at 803.1 and 786.2 eV were the satellite peaks of the metal Co. The Mo 3d spectrum (Fig. 2c) displayed different peaks at 228.6, 229.4, 232.1, and 235.6 eV, manifesting that Mo existed in multiple states, which might be regarded as Mo2+, Mo3+, and Mo6+, respectively. The peaks at 235.6 eV could be regarded as Mo6+ from MoO3 [34]. The presence of MoO2 and MoO3 might be caused by partial oxidation on the surface of the molybdenum carbide due to contact with air [61]. As revealed in Fig. 2d, the peaks of the Pt 4f spectrum for Pt–Mo/Co@NC located at 71.44 and 74.81 eV. Compared with the earlier reports where the peaks of the commercialized Pt/C were located at 71.9 and 75.2 eV [62], the binding energies of our samples decreased, manifesting that Pt obtained many electrons from the Mo/Co@NC carriers, which might be attributed to the electronic effect [14]. On the basis of d-band theory, the electrons enriched on the surface of Pt will weaken the occupation of active sites by toxic substances such as CO, thus the catalytic activity of alcohols oxidation was significantly improved compared with the commercially available Pt/C catalyst [3]. In Fig. 2e, the N 1 s spectrum exhibited pyridinic-N (397.4 eV), Co-Nx (398.1 eV), pyrrolic-N (399.9 eV), and graphitic-N (401.2 eV). In Fig. 2f, the main peak of the C 1 s spectrum is at 284.4 eV as sp2 hybridized carbon; other peaks at 285.0, 286.0, and 288.0 eV can be assigned to sp3 hybridized carbon, indicating the existence of C = C, C–C, C–N, and C–O, separately.

Fig. 3
figure 3

a SEM images of the Co-ZIF, b SEM images of the Co@NC, c SEM images of the Pt–Co@NC, d SEM images of the Mo/Co-ZIF, e SEM images of the Mo/Co-NC, f SEM images of the Pt–Mo/Co@NC

Full size image

FE-SEM was used to analyze the surface morphology of a series of composite materials. In Fig. 3a, the surface of Co-ZIF particles was smooth and polyhedral. After molybdate ion treatment, the surface structure of Mo/Co-ZIF had subtle changes, forming some lamellar structures (Fig. 3d), which also verified that the change of the Mo/Co-ZIF surface structure will cause the decrease of the XRD peak value. As shown in Fig. 3e, after the Mo/Co-ZIF precursor was reduced at a high temperature, the surface of the Mo/Co@NC nanoparticles was coated with a small amount of molybdenum carbide nanoparticles, effectively protecting the inner cobalt nanoparticles from agglomeration and oxidation. As shown in Fig. 3b, there was the Co@NC obtained from the Co-ZIF precursor, which showed a very obvious difference that the Co nanoparticles derived from the Co-ZIF had a very obvious agglomeration without the protection of molybdenum particles. Finally, the morphology analysis of Pt–Mo/Co@NC showed that the surface of Mo/Co@NC was loaded with a small amount of Pt nanoparticles, which was consistent with our expected results (Fig. 3f). Similarly, the topography analysis of the Pt–Co@NC also showed a similar surface structure (Fig. 3c). N2 adsorption–desorption isotherms for the Mo/Co@NC and Co@NC are shown in Fig. S7. The specific surface areas of Mo/Co/@NC and Co@NC are 242.68 m2 g−1 and 107.99 m2 g−1, respectively, indicating that the doping of molybdenum has significantly increased the surface area.

Fig. 4
figure 4

a,b TEM images of the Pt–Mo/Co@NC, the inset of b is the particle size distribution histogram of Pt nanoparticles in the Pt–Mo/Co@NC. c–g HR-TEM images and SAED pattern of Pt–Mo/Co@NC

Full size image

The architecture and the distribution of the compound of the Pt–Co@NC and Pt–Mo/Co@NC composites were further evaluated by TEM, HRTEM, and EDX elemental mapping (Figs. S2 and 4). Firstly, the TEM tests of the Pt–Mo/Co@NC composite (Fig. 4a, b) revealed a clear three-dimensional frame structure. At the same time, after statistical analysis, Pt nanocrystals with a mean particle size of 2.0 ± 1 nm were shown uniformly distributed in Fig. 4b. Such a surprising result might be attributed to the interface effect of metal–carbon support [14]. In Fig. 4c–e, the HRTEM image showed three clear lattice fringes with a spacing distance of 0.23 nm, 0.2 nm, and 0.34 nm, corresponding to Pt (111) plane, Co (111) plane, and C (002) plane, separately, which was consistent with the result of XRD analysis. In addition, as shown in Fig. 4f, the interplanar crystal spacing of 0.24 nm reflects the (002) plane of Mo, which implied that there was molybdenum carbide in composite materials. The SAED pattern analysis (Fig. 4g) also showed results consistent with the HR-TEM.

Fig. 5
figure 5

a–f Elemental mapping images of Pt–Mo/Co@NC

Full size image

Elemental mapping images of Pt–Mo/Co@NC are shown in Fig. 5b–f. Pt, Co, Mo, N, and C manifested a uniform distribution, and it was consistent with the results of the TEM lattice pattern. In addition, the SEM EDX spectrum analysis of Pt–Mo/Co@NC and Pt–Co@NC showed that the quality ratio of Pt in the composite material was about 6.5 wt% (Fig. S3a, b), which was consistent with our expectation. According to the interface effect of the metal carrier, a small amount of Pt nanocrystals could be grown on the Mo/Co@NC surface by a simple microwave-assisted reduction. In Fig. S4, the elemental mapping images of the Pt–Co@NC are shown, which also exhibited a uniform distribution of Pt, Co, N, and C.

Fig. 6
figure 6

a,c Cyclic voltammograms curves for MOR catalyzed by Co-ZIF, Mo/Co-ZIF, Co@NC, Mo/Co@NC. b,d Cyclic voltammograms curves for MOR catalyzed by Pt–Co@NC and Pt–Mo/Co@NC. The CVs were obtained at a scan rate of 50 mV s−1 in an alkaline solution of 1.0 M KOH and 1.0 M CH3OH

Full size image

Electrochemical performance characterization

The MOR activities of different composite materials were evaluated by cyclic voltammetry (CV). The cyclic voltammograms of the Pt–Mo/Co@NC and Pt–Co@NC catalysts in 1 M KOH at a potential scan rate of are shown in Fig. S8. For the Pt–Mo/Co@NC catalyst, however, a noticeable increase in the double-layer capacitance is observed that may be attributed to the formation of oxygenated molybdenum species on the surface.

The CV analysis of Co-ZIF, Mo/Co-ZIF, Co@NC, and Mo/Co@NC was performed in a 1 M KOH + 1 M methanol solution with a potential range of − 0.5 ~ 0.5 V at a scan rate of 50 mV/s for 10 cycles. As could be distinctly seen in Fig. 6a, c, this series of non-noble metal catalysts exhibited a certain catalytic activity, and the highest specific activity and mass activity of the Mo/Co@NC had a peak current of 50 mA cm−2 and 883 mA mg−1, separately. The result corresponded to the previous series of characterizations, and the doping of molybdenum had greatly improved the catalytic activity of methanol. As indicated in Fig. 6b, d, Pt–Mo/Co@NC exhibited the highest mass activity and area activity, reaching 1727 mA mg−1 and 98 mA cm−2 in the potential range of − 0.1 ~ 0.5 V, respectively. It was around 1.9 and 2.2 times of Mo/Co@NC and Pt–Co@NC, respectively, compared with the peak efficiency of the as-prepared catalyst shown in Fig. 7a, b. In Fig. S6a, MOR curves of the Pt–Mo/Co@NC before and after 10,000 cycles show that the peak mass density drops from 1727 to 1167 mA mg−1, reducing of efficiency by 30%, which is normal for our as-prepared catalyst and proves that the Pt–Mo/Co@NC catalyst has preferable stability. Regarding the comparison of catalyst performance, in Table S1, the Pt–Mo/Co@NC was compared with the Pt/C catalyst and the other catalysts. The mass catalytic activity of Pt–Mo/Co@NC was about 4.3 times improved by the commercially available Pt/C catalyst, indicating the superior anti-CO poisoning effect of Pt–Mo/Co@NC. In addition, the active sites of the Pt–Mo/Co@NC catalyst are mainly Pt nanoparticles, Mo2C on the surface, and Co on the inner layer. The improvement of the as-prepared catalyst activity and stability was mainly because of the following factors: (a) the interface effect of the metal carrier made Pt nanoparticles gain more electrons, which reduced the CO poisoning effect, increasing the catalytic activity and stability; (b) the multi-metal synergistic effect of the metal carrier Mo/Co@NC with certain catalytic activity and Pt nanoparticles promoted the methanol oxidation reaction; (c) Mo/Co@NC with a three-dimensional structure provided sufficient attachment sites for Pt nanoparticles, making Pt nanoparticles have excellent dispersion and uniformity; and (d) the doping of high-valence Mo regulated the crystalline and electronic structures of the catalyst, which increased the catalytic sites and accelerated the electron transport, thus improving the electrocatalytic performance of MOR[63].

Fig. 7
figure 7

a–b Mass activity and specific activity of the as-prepared catalysts for methanol oxidation. c–d Chronoamperometric curve for MOR catalyzed by the Co-ZIF, Mo/Co-ZIF, Co@NC, Mo/Co@NC, Pt–Co@NC, and Pt–Mo/Co@NC at 0.3 V vs in 1.0 M KOH and 1.0 M CH3OH

Full size image

To further verify the CO tolerance of the catalyst, CO-stripping voltammetry was performed in a 1 M KOH solution (Fig. S9). Compared with Pt–Co@NC (− 0.296 V), the peak potential of Pt–Mo/Co@NC (− 0.4754 V) is more negative; that is, the CO oxidation potential is more negative, indicating that Pt–Mo /Co@NC is easier to oxidize CO at a lower potential. This is consistent with the previous results and further demonstrates that Pt–Mo/Co@NC has better CO tolerance.

The stability of our catalysts was measured by chronoamperometry (CA). The stability measurements of Co-ZIF, Mo/Co-ZIF, Co@NC, and Mo/Co@NC were carried out at 0.3 V vs. SCE for 3000 s; the Pt–Mo/Co@NC and Pt–Co@NC were carried out at − 0.1 V vs. SCE for 3000 s. As shown in Fig. 7c and d, for all catalysts, the curve began with an ultra-rapid descent and then gradually reached an approximate steady state. In Fig. S5b, the chronoamperometric curves of the Pt–Mo/Co@NC show that the attenuation of Pt–Mo/Co@NC was smaller than 20% Pt/C, indicating that, although the Pt content of as-prepared Pt–Mo/Co@NC was only 6.5%, its stability was much higher than 20% Pt/C. However, the decay rate of the Pt–Mo/Co@NC electrode was much slower than other catalysts, and according to the interface effect of the metal carrier [62, 64], there were abundant electrons around Pt nanoparticles when they deposited over the Mo/Co@NC surface, which could weaken the ability of adsorption of CO-poisoning species. Thus, the catalytic stability of the Pt–Mo/Co@NC was better than other as-prepared catalysts. The electron transfer mechanism of the Pt–Mo/Co@NC catalyst surface in alkaline media during MOR probably abides by the steps below [65, 66]:

$$\mathrm{Pt}-\mathrm{Mo}/\mathrm{Co}@\mathrm{NC}+{\mathrm{CH}}_3\mathrm{OH}\rightarrow\mathrm{Pt}-\mathrm{Mo}/\mathrm{Pt}-\mathrm{Mo}/\mathrm{Co}@\mathrm{NC}-({\mathrm{CH}}_3\mathrm{OH})_{\mathrm{ads}}$$
(1)
$$\mathrm{Pt}-\mathrm{Mo}/\mathrm{Co}@\mathrm{NC}+\mathrm{OH}^-\rightarrow\mathrm{Pt}-\mathrm{Mo}/\mathrm{Co}@\mathrm{NC}-{(\mathrm{OH})}_{\mathrm{ads}}+\mathrm e^-$$
(2)
$$\mathrm{Pt}-\mathrm{Mo}/\mathrm{Co}@\mathrm{NC}-({\mathrm{CH}}_3\mathrm O{\mathrm H)}_{\mathrm{ads}}+\mathrm{OH}^-\rightarrow\mathrm{Pt}-\mathrm{Mo}/\mathrm{Co}@\mathrm{NC}-({\mathrm{CH}}_3\mathrm O{\mathrm H)}_{\mathrm{ads}}+{\mathrm H}_2\mathrm O+\mathrm e^-$$
(3)
$$\mathrm{Pt}-\mathrm{Mo}/\mathrm{Co}@\mathrm{NC}-({\mathrm{CH}}_3\mathrm O{\mathrm H)}_{\mathrm{ads}}+\mathrm{OH}^-\rightarrow\mathrm{Pt}-\mathrm{Mo}/\mathrm{Co}@\mathrm{NC}-({\mathrm{CH}}_2\mathrm O{\mathrm H)}_{\mathrm{ads}}+{\mathrm H}_2\mathrm O+\mathrm e^-$$
(4)
$$\mathrm{Pt}-\mathrm{Mo}/\mathrm{Co}@\mathrm{NC}-({\mathrm{CH}}_2\mathrm O{\mathrm H)}_{\mathrm{ads}}+\mathrm{OH}^-\rightarrow\mathrm{Pt}-\mathrm{Mo}/\mathrm{Co}@\mathrm{NC}-{\left(\mathrm{CHO}\right)}_{\mathrm{ads}}+{\mathrm H}_2\mathrm O+\mathrm e^-$$
(5)
$$\mathrm{Pt}-\mathrm{Mo}/\mathrm{Co}@\mathrm{NC}-{\left(\mathrm{CHO}\right)}_{\mathrm{ads}}+\mathrm{OH}^-\rightarrow\mathrm{Pt}-\mathrm{Mo}/\mathrm{Co}@\mathrm{NC}-{\left(\mathrm{CO}\right)}_{\mathrm{ads}}+{\mathrm H}_2\mathrm O+\mathrm e^-$$
(6)
$$\mathrm{Pt}-\mathrm{Mo}/\mathrm{Co}@\mathrm{NC}-{\left(\mathrm{CO}\right)}_{\mathrm{ads}}+\mathrm{Pt}-\mathrm{Mo}/\mathrm{Co}@\mathrm{NC}-{(\mathrm{OH})}_{\mathrm{ads}}\rightarrow\mathrm{Pt}-\mathrm{Mo}/\mathrm{Co}@\mathrm{NC}-{(\mathrm{COOH})}_{\mathrm{ads}}+\mathrm{Pt}-\mathrm{Mo}/\mathrm{Co}@\mathrm{NC}$$
(7)
$$\mathrm{Pt}-\mathrm{Mo}/\mathrm{Co}@\mathrm{NC}-{(\mathrm{COOH})}_{\mathrm{ads}}+\mathrm{OH}^-\rightarrow\mathrm{Pt}-\mathrm{Mo}/\mathrm{Co}@\mathrm{NC}+{\mathrm{CO}}_2+{\mathrm H}_2\mathrm O+\mathrm e^-$$
(8)

These MOF-derived nanocomposites have great performances as valid electrocatalysts for DMFC. Firstly, a single metal Co-ZIF was prepared. Then, a molybdate precursor containing Mo metal was adsorbed on the surface and pores of Co-ZIF. After pyrolyzed at an appropriate temperature and atmosphere, the cobalt@nitrogen-doped carbon nanocomposites with a small amount of molybdenum carbide (marked as Mo/Co@NC) were prepared, and it was often referred to as an armor catalyst. The molybdenum carbide on the surface provided a protective coat of armor for the cobalt elements in the inner layer. The strong coupling effect between the metal and the carrier not only makes the molybdenum carbide have a certain methanol oxidation performance but also provides abundant electrons for the surface Pt nanoparticles.

Conclusions

A small amount of Pt nanoparticles (6.5%) were deposited over Mo/Co@NC surface by microwave-assisted method, then a small amount of Pt modified Mo/Co@NC composite material was obtained. A great deal of electrons would accumulate around the Pt nanoparticles on the surface of Mo/Co@NC via the strong interaction between the metal and the metallic carbon carrier, which greatly reduced the CO poisoning effect, improving catalytic activity and stability. In this method, non-noble metals and platinum nanocrystals were ingeniously combined, effectively solving the problems of low stability of non-noble metal catalysts and the high cost of platinum catalysts, which provide a good idea for the synthesis of low platinum content catalysts.