Abstract
The successful commercialization of the direct methanol fuel cell (DMFC) is inseparable from the development of advanced Pt-based anode catalysts with high electrocatalytic activity and acceptable manufacturing cost. Here, we present a robust bottom-up strategy to anchor ultrasmall Pt nanocrystals with an average diameter of only 2.3 nm onto carbon nanohorn-decorated three-dimensional (3D) graphene networks (Pt/CNH-G) through a controllable self-assembly process. The as-derived 3D Pt/CNH-G catalysts manifest a series of distinctive architectural advantages, such as interconnected porous frameworks, large accessible surface areas, plentiful active cones, highly dispersed Pt nanoparticles, and good electron conductivity. Consequently, the optimized Pt/CNH-G catalyst is endowed with exceptional methanol oxidation properties with a large electrochemical active surface area of 128.6 m2 g-1, a high mass activity of 1626.0 mA mg-1, and excellent long-term stability, which are significantly superior to those of conventional Pt catalysts supported by carbon black, carbon nanotube, carbon nanohorn, and graphene matrices.
1. Introduction
In the past few decades, the rapid development of human society is accompanied by the surge of energy consumption and the intensification of environmental pollution, which highlight the importance of various advanced clean energy technologies [1–3]. In this context, direct methanol fuel cell (DMFC), as an emerging energy-conversion system, has shown great application potentials in the fields of electric vehicles, mobile electronic equipment, and aerospace due to its high energy-conversion efficiency, low working environment temperature, miniature size, and low pollution emission [4, 5]. However, as the key semireaction of DMFC, the methanol oxidation reaction (MOR) has intrinsically slow electrochemical kinetics that has seriously hindered the commercial use of DMFC devices [6, 7]. In recent years, enormous efforts have been devoted to the design and fabrication of efficient MOR catalysts, which are expected to accelerate the reaction efficiency and thereby boost the overall output power of DMFC [8, 9].
It is widely known that platinum- (Pt-) based materials are considered to be the most outstanding MOR catalysts [10, 11]. Unfortunately, the exorbitant price of metallic Pt as well as its poor antitoxic capacity towards CO have posed great challenges to the large-scale practical application [12–14]. Aiming at these problems, various types of carbon materials, including carbon black, porous carbon, carbon nanotubes, and graphene, have been used as supports for dispersing Pt nanoparticles, which are able to improve the Pt utilization efficiency and meanwhile alleviate the toxic effects of CO [15, 16]. Among them, graphene has been regarded as a high-quality matrix for the loading of Pt nanoparticles due to its large surface area, high chemical stability, and good electron conductivity [17, 18]. However, the presence of van der Waals force usually induces the longitudinal restacking between the neighboring graphene nanosheets, rendering insufficient exposure of the catalytically active sites [19, 20]. To overcome this barrier, our recent studies have demonstrated that the spatial construction of three-dimensional (3D) graphene-based networks is very beneficial to circumvent the restacking of graphene nanosheets and effectively promote the transport of reactants during the catalytic process [21, 22]. Therefore, the combination of noble metal nanocrystals with 3D graphene support opens up a new avenue for the development of high-efficiency MOR electrocatalysts.
Despite these improvements, it should be noted that the inherent chemical inertness of single graphene material makes it lack of effective metal growth sites, which largely limits the synergistic coupling effect between noble metal and carrier [23, 24]. In this aspect, one sensible solution is to introduce another highly active carbon material into the graphene system [25, 26]. As a new type of carbon nanomaterial, single-walled carbon nanohorn (CNH) has recently attracted extensive research interests in the fields of energy and environment [27, 28]. Specifically, CNH is a unique horn-shaped carbon material with abundant open holes formed by the expansion of hexagonal graphite structure with a pentagon ring limiting the cone apex [29, 30]. This intriguing geometric configuration endows CNHs with rich channels and extra chemical activity, which can largely facilitate the uniform loading of noble metal nanoparticles [31, 32]. Considering the strong π-π supramolecular interaction between different carbon materials, it is great significance to construct a 3D multiple-carbon nanoarchitecture consisting of CNH-decorated graphene nanosheets and investigate its application as a supporting material for Pt nanocrystals. As far as we know, neither preparation nor performance of this attractive nanostructure has been reported up to now.
In this study, we put forward a simple and scalable bottom-up strategy to anchor ultrasmall Pt nanocrystals onto CNH-decorated 3D graphene networks (Pt/CNH-G) through a controllable stereo-assembly process. As illustrated in Figures 1 and S1, pristine CNHs with high surface free energy tend to form a dahlia-like aggregation structure, which were first dispersed in graphene oxide (GO) solution with the help of ultrasonic treatment. Notably, GO could serve as an amphiphilic surfactant due to its hydrophobic region (unoxidized benzene rings) and hydrophilic region (diverse oxygenated groups) [33], leading to the well dispersion of separated CNHs on GO surface. Subsequently, the above CNH-GO mixture was moved to an autoclave and solvothermally heated at 180°C for 24 h, during which a CNH-decorated 3D porous graphene (CNH-G) hydrogel was formed. After a freeze-drying treatment, the as-generated CNH-G aerogel was further exposed in an ethylene glycol solution containing potassium tetrachloroplatinate (K2PtCl4) at 120°C for a period of time. Because of the rich nucleation and growth sites on the surfaces of CNHs and the porous structure of graphene networks, the Pt salt ions will be reduced by ethylene glycol and then uniformly anchored onto the 3D CNH-G frameworks, thus obtaining the desired 3D Pt/CNH-G hybrid. It is worth noting that the compositions of the Pt/CNH-G catalysts can be easily controlled by changing the mass ratios of CNH and GO in the above synthesis process, and here, we adopt four different CNH/G ratios including 1 : 9, 3 : 7, 5 : 5, and 7 : 3, which are named as Pt/(CNH)1-G9, Pt/(CNH)3-G7, Pt/(CNH)5-G5, and Pt/(CNH)7-G3, respectively. By virtue of the remarkable structural characteristics such as 3D cross-linked porous nature, large specific surface area, optimized electronic structure, and uniform Pt dispersibility, the resulting Pt/CNH-G catalyst possesses a significantly enhanced MOR performance compared with that of traditional Pt/carbon black (Pt/C), Pt/carbon nanotube (Pt/CNT), Pt/graphene (Pt/G), and Pt/carbon nanohorn (Pt/CNH) catalysts.

2. Materials and Methods
2.1. Synthesis of 3D Pt/CNH-G Catalysts
Commercial graphite powder was first converted into graphite oxide by an improved Hummer’s method [34]. Typically, the 3D Pt/CNH-G catalysts with different CNH/G mass ratios were prepared by a bottom-up strategy. Taking the CNH/G ratio of 5 : 5 as an example, the specific synthesis process is as follows: 10 mg of CNH powder (XFNANO Materials Tech Co., Ltd) was dispersed into 5 mL of GO solution (2 mg mL-1) by sonication for 1 h to achieve a uniform black mixture. Afterwards, the above mixture was sealed into a stainless-steel autoclave and reacted at 180°C for 24 h to generate a 3D CNH/G hydrogel. Subsequently, the obtained hydrogel was dialyzed and washed for 3-5 days and then transformed into an aerogel structure through a freeze-drying treatment. Finally, 10 mg of the as-prepared CNH-G aerogel was exposed to an ethylene glycol solution containing 5.3 mg of K2PtCl4 at 120°C for 12 h, thus giving birth to the Pt/(CNH)5-G5 catalyst. In addition, conventional Pt/C, Pt/CNT, Pt/G, and Pt/CNH catalysts were also prepared with the use of carbon black, CNT, GO, and CNH as supports through the similar synthesis steps, respectively. For the fairness of comparison, the feeding ratio of the carbon support to metal Pt was kept at 4 : 1; thus, the overall Pt content in all prepared catalysts was controlled at 20.0 wt%.
2.2. Characterizations
The 3D porous nanostructure and morphology of the Pt/CNH-G hybrids were systematically observed by field emission scanning electron microscopy (FE-SEM, JEOL 6500F), transmission electron microscopy (TEM, JEOL 2100F), and high-angle annular dark field scanning TEM (HAADF-STEM, JEOL 2100F). Powder X-ray diffraction (XRD, Rigaku Ultima IV) and Raman microscopy (Horiba LabRAM HR Evolution) were used to study the phase compositions and crystal structure of the Pt/CNH-G hybrids. The element valences and bonding relationships of the Pt/CNH-G hybrids were explored by X-ray photoelectron spectroscopy (XPS, Thermo Scientific K-Alpha). The specific surface area and pore characteristics of the Pt/CNH-G hybrid were examined through a Micromeritics ASAP 2020 Plus system.
2.3. Electrochemical Measurements
The electrocatalytic MOR performances of the Pt/CNH-G and other control catalysts were tested on the CHI 760E electrochemical workstation. The as-employed three-electrode system consisted of a catalyst-coated glass carbon disk (3 mm) as the working electrode, a Pt wire as the counter electrode, and a saturated calomel electrode (SCE) as the reference electrode. The specific preparation steps for the working electrode can be referred to our previous work [9]. The overall Pt loading density on the working electrode surface was kept at 0.028 mg cm-2. The electrochemical tests were carried out in a N2-purged acidic electrolyte containing 0.5 M H2SO4 and 1 M methanol solution.
3. Results and Discussion
The 3D porous feature and micromorphology of the newly developed Pt/CNH-G hybrid were carefully investigated by FE-SEM and TEM. It can be seen from Figure 2(a) that the Pt/CNH-G hybrid has a well-defined 3D interconnected network, accompanied by abundant continuous macropores with the sizes between hundreds of nanometers and several microns, which are able to separate the adjacent graphene nanolayers and provide a large number of crossed channels to accelerate the transport of electrolyte during the MOR. At the meantime, a number of small-sized CNHs with sizes ranging from 50 to 100 nm are found to be randomly distributed on the graphene planes due to the strong π-π supramolecular interaction, which can offer extra anchoring sites for the immobilization of Pt nanocrystals (Figure 2(b)). Under close inspection (Figures 2(c)–2(e)), numerous ultrasmall Pt nanoparticles with an average diameter of around 2.3 nm are uniformly dispersed on the surface of the CNH carrier, while only a small number of Pt nanoparticles are anchored on graphene nanosheets. This implies that there are more plentiful nucleation and growth sites on CNHs because of their higher chemical activity arising from the unique conical structure. In contrast, the Pt particles supported by traditional carbon black, CNT, graphene, and CNH materials are prone to form large aggregates (Figure S2), mainly due to less porous nature or insufficient growing sites for these matrixes. In addition, typical lattice stripes of Pt nanoparticles and carbon matrix are clearly observed in high-resolution TEM (HR-TEM) images (Figures 2(f) and 2(g)), indicating that the crystal structures of these components can be well maintained during the synthesis process. It is noteworthy that the crystal plane spacings of Pt nanoparticles are 0.221 and 0.194 nm, which correspond to the (111) and (200) planes of face-centered cubic (fcc) Pt crystals, respectively. Besides, the lattice fringe with an interplanar spacing of 0.340 nm corresponds to the carbon (002) planes of CNHs. Furthermore, the HAADF-STEM image and the corresponding element mapping results show that the Pt/CNH-G hybrid is mainly composed of C, O, and Pt components, and these three elements are evenly distributed on the whole 3D framework (Figures 2(h)–2(k)).

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(i)

(j)

(k)
The interior phase composition and defect density of the 3D Pt/CNH-G hybrid were further explored by powder XRD and Raman analysis. As shown in Figures 3(a) and S3, the diffraction peak at 2θ = ~25.0° in the XRD pattern of Pt/CNH-G originates from the carbon (002) plane, while the sharp characteristic peak of GO at 2θ = ~10.0° is totally disappeared, unraveling that GO could be successfully reduced to graphene after the solvothermal reaction with the assistance of ethylene glycol [35]. Noteworthily, when compared with the Pt/CNH pattern, the carbon (002) peak intensity in the case of Pt/CNH-G is much weaker, which further verifies that the 3D porous graphene networks can effectively avoid the restacking of CNHs. At the same time, the other four diffraction peaks of metallic Pt are situated at 2θ =39.2°, 46.2°, 67.5°, and 82.4°, corresponding to the (111), (200), (220), and (311) crystal faces of fcc Pt structure, respectively, evidencing the successful anchoring of Pt nanoparticles on the 3D CNH/G frameworks. Figure 3(b) shows the typical Raman spectra of the Pt/CNH-G, Pt/CNH, and Pt/G samples. Obviously, two prominent Raman scattering peaks are located at about 1325 and 1585 cm-1, corresponding to the D and G bands of graphitic carbon materials, respectively. It is well known that the D band arises from the disordered vibration of carbon atoms, while the G band belongs to the in-plane stretching vibration of sp2-hybridized carbon atoms [36]. As calculated, the / ratio of the Pt/CNH-G hybrid is up to 1.62, much higher than that of Pt/CNH (0.97) and Pt/G (1.03), suggesting that the generation of 3D CNH-G networks could induce the formation of plentiful topological defects in the carbon skeletons, which may provide more catalytic active sites for the MOR. Moreover, as shown in Figures 3(c) and 3(d), N2 adsorption and desorption isotherms of the 3D Pt/CNH-G hybrid have a typical IV-type hysteresis loop, and the corresponding pore sizes are mainly ranging from 2 to 50 nm, indicating its well-defined mesoporous structure with a specific surface area of 113.2 m2 g-1. This value is much larger than that of GO (11.2 m2 g-1) and similar to that of reported 3D graphene aerogel-based nanomaterials [37, 38].

(a)

(b)

(c)

(d)
The chemical valence and bonding relationship information of the Pt/CNH-G hybrid was then systematically studied by XPS measurements. As displayed in Figure 4(a), the XPS survey spectrum of Pt/CNH-G confirms that the hybrid is mainly composed of C (89.3 at%), O (9.9 at%), and Pt (0.8 at%) components without any impurity, which is consistent with the EDX test result (Figure S4). Moreover, in the complex C 1s spectrum (Figure 4(b)), four different C signals appear at 284.6, 285.4, 287.4, and 289.7 eV, corresponding to the sp2 C-C, C-OH, C-O-C, and HO-C=O configurations, respectively. The proportions of oxygen-containing groups for Pt/CNH-G are much lower than those for the pristine GO (Figure S5), proving the efficient conversion from GO to graphene after the solvothermal assembly reaction. Meanwhile, a series of characteristic O signals including HO-C=O (531.7 eV), C=O (532.7 eV), and C-OH (533.6 eV) can be observed by fitting the high-resolution O 1s spectrum (Figure 4(c)). In addition, as shown in Figure 4(d), there are two chemical states of Pt in the Pt/CNH-G hybrid. The two characteristic peaks with lower binding energies of 71.3 and 74.6 eV are associated with metal Pt, while the other two peaks with higher binding energies of 71.9 and 75.8 eV are caused by a small amount of Pt oxide in the sample, attesting that the Pt component mainly exists in the form of zero-valent metal in the Pt/CNH-G hybrid.

(a)

(b)

(c)

(d)
In order to explore the practical application of the 3D Pt/CNH-G hybrids in DMFC devices, we carried out various electrochemical tests to evaluate their MOR properties in the acidic media. Figure 5(a) shows the typical cyclic voltammetry (CV) curves of the Pt/CNH-G catalysts with diverse CNH/G ratios in a N2-purged 0.5 M H2SO4 solution. Clearly, characteristic current peaks for hydrogen adsorption and desorption are detected in these CV curves within the potential region between -0.2 and 0.1 V. Generally, the electrochemical active surface areas (ECSAs) of Pt-based catalysts are commonly estimated by calculating the hydrogen adsorption peak area during the negative potential scanning [39]. According to the calculating data, the Pt/(CNH)5-G5 catalyst is found to have the largest ECSA value of up to 128.6 m2 g-1, followed by the Pt/(CNH)3-G7 (106.1 m2 g-1), Pt/(CNH)7-G3 (97.8 m2 g-1), and Pt/(CNH)1-G9 (82.9 m2 g-1). In addition, we also compared the electrocatalytic activity of Pt/(CNH)5-G5 with that of traditional Pt/CNH, Pt/G, Pt/CNT, and Pt/C catalysts under the same test condition. As presented in Figures 5(b) and 5(c) and Table S1, the ECSA value of the Pt/(CNH)5-G5 catalyst is 2.6, 3.3, 4.2, and 4.4 times that of Pt/CNH (49.0 m2 g-1), Pt/G (39.2 m2 g-1), Pt/CNT (30.7 m2 g-1), and Pt/C (29.3 m2 g-1) catalysts, respectively. Furthermore, our Pt/(CNH)5-G5 catalyst also has a larger ECSA value than that of recent state-of-the-art Pt-based nanostructures (Table S2), which testifies that the utilization of 3D CNH-G support is beneficial to improve the Pt use efficiency.

(a)

(b)

(c)

(d)

(e)

(f)
The mass and specific activities of different Pt/CNH-G as well as reference catalysts were next investigated by CV in a mixed solution containing 0.5 M H2SO4 and 1 M CH3OH. As depicted in Figures 5(d)–5(f), all recorded CV curves show a significant positive current peak near 0.7 V, corresponding to the oxidation of methanol molecules on the electrode surface, while the reverse current peak near 0.5 V is attributed to the oxidation of CO-like by-products [40, 41]. Consistent with the aforementioned ECSA results, the Pt/(CNH)5-G5 catalyst exhibits a high mass activity of 1626.0 mA mg-1, which is significantly better than that of Pt/(CNH)3-G7 (1126.5 mA mg-1), Pt/(CNH)7-G3 (1028.3 mA mg-1), and Pt/(CNH)1-G9 (829.0 mA mg-1), verifying that a proper CNH/G ratio can give full play to the electrocatalytic functions of Pt. Meanwhile, the mass activity of the Pt/(CNH)5-G5 catalyst increases by 252~564% compared with that of Pt/CNH (462.3 mA mg-1), Pt/G (398.2 mA mg-1), Pt/CNT (306.1 mA mg-1), and Pt/C (244.7 mA mg-1) catalysts. Moreover, the ECSA-normalized specific activity of the Pt/(CNH)5-G5 catalyst is determined to be 1.26 mA cm-2 (Figure S6), more competitive than other Pt/CNH-G and reference catalysts (0.83~1.06 mA cm-2), indicative of its highest intrinsic catalytic ability. Besides, linear sweep voltammetry (LSV) measurements further disclose that the Pt/(CNH)5-G5 electrode needs relatively lower electrode potential to obtain the same oxidation current compared with other control electrodes (Figure S7), and simultaneously, its Tafel slope (210 mV dec-1) is also the smallest among all investigated electrodes (Figure S8), demonstrating that the fastest MOR kinetics can be achieved on the Pt/(CNH)5-G5 electrode. The dramatically enhanced electrocatalytic capacity of the newly designed Pt/(CNH)5-G5 catalyst should be attributed to the following two aspects: on the one hand, the 3D porous cross-linked networks of the Pt/(CNH)5-G5 catalyst provide abundant channels for ion transport, leading to sufficient catalytic reaction interfaces in the hybrid system; on the other hand, the well-dispersive Pt nanoparticles as well as numerous conical tips of separated CNHs can serve as efficient catalytic centers to boost the dynamic processes of the MOR.
The long-term durability of an anode catalyst for DMFC is another critical issue in its large-scale commercial application. In this aspect, the electrocatalytic stability of the selected Pt/(CNH)5-G5 catalyst was studied by the chronoamperometric measurements, together with the Pt/CNH, Pt/G, Pt/CNT, and Pt/C catalysts for comparison. With a fixed electrode potential of 0.5 V, the MOR current on each electrode shows a downward trend as time goes on (Figure 6(a)), mainly due to the accumulation of CO-like by-products and the inevitable formation of Pt oxide on the electrode surface [42]. Remarkably enough, the Pt/(CNH)5-G5 electrode has a high initial oxidation current and maintains a slow decay rate of only 28.9% after 3000 s (Figure 6(b)), which is much better than that of the reference Pt/CNH, Pt/G, Pt/CNT, and Pt/C as well as previously reported Pt-based catalysts [43, 44], thus affording an exceptional long-term MOR stability. This significant improvement is not only linked to the 3D interconnected frameworks of Pt/(CNH)5-G5 for the rapid separation of CO species but also originated from the strong interfacial interaction between Pt and CNH-G matrix that could prevent the Pt nanoparticles from aggregation and dissolution during the catalytic process.

(a)

(b)

(c)

(d)

(e)

(f)
We further carried out the chronopotentiometric tests to compare the antipoisoning ability of the Pt/(CNH)5-G5 catalyst with that of the reference catalysts. With a constant anodic current, the CO-poisoning species would adsorb on the catalyst surface and decrease the activity [45]. To satisfy the current, a higher electrode potential was required to facilitate the water decomposition process. It is clearly seen from Figure 6(c) that the Pt/(CNH)5-G5 electrode can maintain a relatively low potential level for a long time of ~1220 s, which is significantly superior to the Pt/CNH (~450 s), Pt/G (~180 s), Pt/CNT (~110 s), and Pt/C (~75 s) electrodes. The test results strongly prove that the rich porous channels and optimized electronic structure make the Pt/(CNH)5-G5 catalyst have a much better poison tolerance towards CO intermediates during the MOR process. Besides, the semiconductor character of nanosized Pt particles usually results in unsatisfactory electron conductivities, while the introduction of carbon matrixes can lower the interfacial electrode resistances of the Pt-based catalysts. In this study, the electron conductivities of diverse Pt-based catalysts were estimated by the alternating-current (AC) impedance method. As depicted in Figures 6(d) and 6(e), it is found that the Pt/(CNH)5-G5 electrode shows the smallest semicircle diameter at the high-frequency region, suggesting the lowest charge-transfer resistance among these electrodes. The equivalent circuit used to fit the impedance spectra is shown in Figure 6(f). In this circuit, and correspond the resistances of electrolyte and electrocatalyst, respectively, is a constant phase element, is related to the semi-infinite diffusion at the interfaces between electrolyte and electrocatalyst, and and represent the resistance and capacitance of the Nafion-carbon films, respectively. The fitting results confirm that the charge-transfer resistance of the Pt/(CNH)5-G5 electrode is about 4.6 Ω, which is significantly lower than that of Pt/CNH (7.6 Ω), Pt/G (10.3 Ω), Pt/CNT (14.0 Ω), and Pt/C (7100.4 Ω) electrodes. Therefore, the existence of the CNH-G carrier can effectively promote the electron transport rate in the catalytic system, thus offering more three-phase reaction boundaries during the MOR.
4. Conclusions
In summary, a convenient and scalable bottom-up method has been proposed to anchor ultrasmall Pt nanoparticles onto the carbon nanohorn-decorated 3D graphene frameworks via a facile stereo-assembly process. Strikingly, the 3D cross-linked CNH-G networks are able to create abundant pore channels and growth sites for the immobilization of Pt nanocrystals, while the incorporation of CNHs enhances the electronic coupling with Pt atoms and thereby improves the synergistic effects. As a consequence, the optimized Pt/CNH-G catalyst expresses excellent MOR performance with a large ECSA value of 128.6 m2 g-1, a high mass activity of 1626.0 mA mg-1, and outstanding long-term stability, far surpassing the traditional Pt/CNH, Pt/G, Pt/CNT, and Pt/C catalysts. The design concept presented in this work is anticipated to expand the variety of 3D nanocarbon-supported noble metal catalysts and promote their practical application in the energy-storage and -conversion fields.
Data Availability
Data is available on request.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Authors’ Contributions
B. S. and H. Y. contributed equally to this work.
Acknowledgments
This work was supported by the National Natural Science Foundation of China (22209037), the Fundamental Research Funds for the Central Universities (B220202042), the Postgraduate Research and Practice Innovation Program of Jiangsu Province (KYCX23_0674), and the Project on Excellent Post-graduate Dissertation of Hohai University (422003483).
Supplementary Materials
Figure S1: the synthetic process for the 3D CNH-G hydrogel. Figure S2: representative FE-SEM of (a) Pt/C, (b) Pt/CNT, (c) Pt/G, and (d) Pt/CNH, respectively. Figure S3: typical XRD patterns of Pt/CNH-G catalyst and GO nanosheets. Figure S4: EDX spectrum of the Pt/CNH-G catalyst. Figure S5: high-resolution C 1s spectrum of the pristine GO nanosheets. Figure S6: the ECSA-normalized CV curves of different Pt/CNH-G electrodes and Pt/CNH, Pt/G, Pt/CNT, and Pt/C electrodes in 0.5 M H2SO4 and 1 M CH3OH mixture at 50 mV s-1. Figure S7: linear sweep voltammetry curves of different Pt/CNH-G electrodes and Pt/CNH, Pt/G, Pt/CNT, and Pt/C electrodes in 0.5 M H2SO4 and 1 M CH3OH mixture at 50 mV s-1. Figure S8: the Tafel plots and histograms of the Pt/(CNH)5-G5, Pt/CNH, Pt/G, Pt/CNT, and Pt/C electrodes. Table S1: methanol oxidation behaviors on different catalysts. Table S2: comparison of methanol oxidation behavior of the Pt/(CNH)5-G5 catalyst with various state-of-the-art Pt-based electrocatalysts. Table S3: the charge-transfer resistance (Rct) of different catalysts. (Supplementary Materials)