1. Introduction

Direct methanol fuel cells (DMFC) have been widely studied as a promising power source for portable electronic devices due to the ease of carry and high specific energy density ^[1-5]. The noble metal Pt is still the most efficient anode catalyst because of its high catalytic activity, but the high cost and low tolerance to poisoning limit its use in fuel cells. So the improvement of Pt-based catalyst has great practical significance ^[6].

Alloying a non-noble metal with pure platinum has become a popular method to increase the electrochemical activity and corrosion resistance ^[7-10], wherein tin as a non-noble alloyed metal component received extensive attention ^{[11, 12]}. Studies suggested that Sn, as an alloy component is beneficial for the decomposition of H₂O on the surface of tin, thus, enhanced CO oxidation into CO₂. However, its bonding with Pt weakened the adsorption of oxide species on Pt by altering electronic properties of Pt ^[13-15]. Current methods of synthesis of Pt-Sn alloy mostly are complex ^[16-18], for example, high temperature organic phase co-reduction method ^[19], in which precursor reduced at 180-280 ℃ in 1-octadecene with oleylamine as reducing agent and oleylamine/oleic acid as surfactants. The Pt/oxide-C materials show increasing electrochemical activity, for instance, presence of Sn-oxide on conventional Pt/C exhibits remarkable promotional effects by reducing CO poisoning, where CO oxidation on Pt/SnO₂- C samples take place at lower electrode potentials compared to Pt/ C due to changes in their electronic properties of Pt caused by the interaction of Pt and the oxide ^{[20, 21]}.

In this paper, we successfully prepared PtSn₂-SnO₂/C nanocatalyst by co-reduction of Pt and Sn precursor using NaBH₄ as the reducing agent at 15 8C, integrating advantages of alloy and oxide. Electrochemical tests proved its enhanced resistance to CO poisoning and the methanol oxidation activity was up to 3 times than that of commercial Pt/C.

2. Experimental 2.1. Chemicals

Analytical grade hexachloroplatinic (Ⅳ) acid hexahydrate (H₂PtCl₆·6H₂O, 99.9%, 0.0193 mol L^-1 aqueous solution) and anhydrous ethanol were obtained from Tianjin Chemical Reagents, China. Stannous chloride (SnCl2·2H₂O), sodium borohydride (NaBH₄) and the Pt/C (20 wt%. Pt nanoparticles on Vulcan XC-72 carbon support) catalyst was purchased from Alfa Aesar (Ward Hill, MA, USA). Nafion (5% in a mixture of lower aliphatic alcohols and water) was purchased from Sigma Aldrich. All the chemicals were used as received without further purification.

2.2. Preparation of catalyst

Carbon black (0.05 g), stannous chloride (0.0344 g), chloroplatinic acid (1.32 mL), together with ethanol (5 mL), were stirred at 15 ℃ for 2 h, followed by addition of 0.04 g sodium borohydride aqueous solution and stirred for another 2 h. Then centrifuged, dried at 80 ℃ to obtain PtSn₂-SnO₂/C catalyst. Reduce the amount of stannous chloride, and cut off the air, we can obtain PtSn₂/C. As a comparison, samples prepared without Sn or Pt precursors were denoted as Pt/C and Sn/C, respectively, with all the other parameters the same as above.

2.3. Physical characterization

XRD was performed using a Brucker D8 Focus diffractometer with Cu Ka radiation (λ = 0.154 nm, 40 kV, 40 mA). The transmission electron microscopy (TEM) investigations were performed using a 200 kV TEM Philips Tecnai G2F20 instrument. X-ray dispersion spectroscopy (EDX) can be used to determine the composition of the sample. The elemental composition of the surfaces was measured with XPS (ESCALAB 250Xi spectrometer, XR6 X-ray source).

2.4. Electrochemical measurement

The test was performed at r.t. on the CHI 660D electrochemical work station measurement. The electrodes were glassy carbon electrodes (support for the working electrodes with a diameter of 3 mm) with Ag/AgCl as the reference electrode and Pt wire as the counter electrode. All potentials used in tests are versus Ag/AgCl. Five microliter catalyst in water suspension (2 mg of catalyst with 1 mL water) was deposited on a glassy carbon electrode (3 mm in diameter, mirror polished) and dried at room temperature. Five microliter Nafion was deposited over the catalyst and dried at r.t. Then, these three electrodes are placed in the well-prepared electrolytic cell mixture (2 mol L^-1 methanol 3 mL, 0.1 mol L^-1 HClO₄ 3 mL). The Electrochemical Impedance Spectra (EIS) was also carried on the CHI660D electrochemical workstation measurement at open circuit potential with a 5 mV sine perturbation. The measuring frequency range was 10⁵-10^-2 Hz.

3. Results and discussion 3.1. Physical characterization

Fig. 1 shows XRD patterns of PtSn₂-SnO₂/C and Pt/C in comparison with reference data of PtSn₂ and SnO₂. In the case of PtSn₂-SnO₂/C, the characteristic diffraction peaks at 2θ = 24.05°, 39.65°, 46.875° and 57.31°, which correspond to the (1 1 1), (2 2 0), (3 1 1) and (4 0 0) planes of PtSn₂ structure (JCPDS PDF #65-2991), and the diffraction peaks at 2θ = 33.89°, 51.78° correspond to (1 0 1) and (2 1 1) planes of SnO₂ standard (JCPDS PDF #71-652). These results indicate the formation of PtSn₂-SnO₂/C, being consistent with the results of TEM test which will be showed in detail in the following part.

Structure analysis of the samples prepared above is examined by TEM measurements as shown Fig. 2. According to the statistical survey results, the average diameter of particles is about 2-6 nm (Fig. 2c). As confirmed by the EDS in Fig. 2d, the constitution of PtSn₂-SnO₂/C nanoparticles is measured to be 16.34 at% Pt and 83.66 at% Sn. The composition is also supported by EDS elemental mapping of Pt, Sn and O (Fig. 2e and f). As can be seen, nanoparticles are composed by Pt, Sn and O, meanwhile, some SnO₂ homogeneously distribute at the surface of carbon black. The XPS was employed to further analyze the compositions and atom binding states of PtSn₂-SnO₂/C, it revealed that Pt and partial Sn are in zero valence state while the other part of Sn is tetravalent (Fig. 2h and i).

3.2. Electrochemical measurement

Fig. 3a shows the comparison of CV curves for PtSn₂-SnO₂/C, PtSn₂/C, Pt/C Sn/C and commercial Pt/C catalysts in the 0.1 mol L^-1 HClO₄ and 1.0 mol L^-1 CH₃OH at a scan rate of 20 mV s-1. Here, the electro-catalytic current density is normalized in references to the loading amount of Pt. The oxidation of methanol molecules on the electrode surface occurs in the first anodic peak, which is usually used to evaluate the catalytic performance of the catalysts. During the forward scan, some methanol is incompletely oxidized resulting in an increase in the catalyst surface coverage by adsorbed CO intermediates, which can be oxidized when the kickback continues, forming another oxidation peak. Therefore, the current intensity ratio of these two oxidation peaks often act as an indicator to estimate the poison tolerance of catalysts. The forward peak potential (E_f), forward scan current intensity (J_f), reverse scan peak potential (E_b), reverse scan peak current intensity (J_b) and J_f/J_b of catalysts are analyzed and their values are shown in Table 1. It is obvious that the current strength of PtSn₂-SnO₂/C (533.7) is 3.02 times high than that of commercial Pt/C (176.9). When Sn and Pt are alloyed, maybe Sn promotes H₂O decomposition to enhance CO oxidation into CO₂, which would increase methanol oxidation effectively ^[19]. The Jf/Jb of PtSn₂-SnO₂/C (1.12) is high than that of PtSn₂/C (1.10), and commercial Pt/C (0.99), which proves the enhanced ability of resistance to oxygen species poison.

The CO stripping experiments further prove the catalyst activity (Fig. 3b). The electrochemical cell (0.1 mol L^-1 HClO₄) is purged for 30 min with N₂ before experiment to avoid the impact of oxygen, and then introduces into CO (10% CO, 90% He) for 15 min. As can be seen from the figure, commercial Pt/C catalyst has a sharp stripping CO oxidation peak at 0.72 V. While the peak position of PtSn₂/C and PtSn₂-SnO₂/C catalyst has a negative shift to 0.551 V and 0.534 V, respectively. The negative direction indicates that the catalyst is capable of oxidizing CO at a lower voltage, implying SnO₂ has a positive effect on the stripping removal of CO absorbed on the Pt surface, which will lower CO poisoning effectively. ^[22]

To further investigate the stability of the catalyst, chronoamperometry (CA) experiments were performed (Fig. 4a). As can be seen, the current density PtSn₂-SnO₂/C catalyst declined more slowly than that of commercial Pt/C, after 5 h it still remains a large current density.

Fig. 4b displays the hydrogen adsorption and desorption test of catalysts in 0.1 mol L^-1 HClO4. The electrochemical cell is purged for 30 min with N₂ before experiment. The cyclic voltammetry curves have four distinct potential zones. -0.2-0.1 V, 0.1-0.3 V, 0.3-0.6 V and 0.6-1.2 V are respect for hydrogen adsorption dissociation interval, oxygen adsorption and oxide reduction zone, double plateau region and oxygen precipitation range, respectively. The process of catalyst’s adsorption of H+ in the acid solution shows hydrogen adsorption and desorption of PtSn₂-SnO₂/C catalyst are weakened. It can also prove that when Sn is added, the adsorption of the oxygen-containing species such as CO was weakened.

The electrochemical impedance spectra (EIS) is shown in Fig. 4c. As can be seen, the depressed semicircle in each curve is related to the double-layer capacitance between the interface of the electrode and the electrolyte ^[23]. The smaller arc diameter of PtSn₂-SnO₂/C indicates the better ability of charge transfer and ion diffusion rate than that of commercial Pt/C.

4. Conclusion

PtSn₂-SnO₂ nanocatalyst is prepared by a relatively simple method. XRD and TEM of the nanocatalyst show the formation of metal alloy. The alloying of PtSn₂ and existence of SnO₂ improves the electrical properties of methanol oxidation and enhanced the resistance ability to CO oxidation, which are proved by stripping voltammetry curves from electrochemical measurements. This synthetic method is convenient, strong, stable and environment friendly.

Acknowledgment

This work was supported by NSFC (Nos. 21373116 and 21421001), the Tianjin Natural Science Research Fund (No. 13JCYBJC₁₈₃₀₀), RFDP (No. 20120031110005) and the MOE Innovation Team (No. IRT13022) of China.