English

• 0.   Introduction

  Nowadays, considerable attention to seeking rene-wable resources has been paid to reduce or even stop the rapid consumption of fossil fuels and the incr-easing emissions of automobile exhaust^[1-3]. As environ-mental friendly electrochemical energy devices, fuel cells have received extensive attention in view of their outstanding performances. Among them, direct formic acid fuel cell (DFAFC) possesses such significant advantages as lower operating temperatures, higher theoretical open circuit voltage and energy density, safer storage and transportation and less crossover rate of formic acid through Nafion membrane compared with direct methanol fuel cell (DMFC)^[4-11]. There are many factors affecting the performance of DFAFC, one of which is the electroactivity of the anode catalyst. A great number of reports have demonstrated that Pt and Pt-based catalysts exhibit excellent catalytic activity for formic acid oxidation in alkaline and acidic circumstances, including Pt/C^[12] and Pt-M (M=Au, Li, Sb, Ru, Ir, Bi)^[13-20] and so on. Nevertheless, extremely low natural reserve and the exorbitant cost of Pt severely limit the large-scale commercial applications of Pt or Pt-based catalysts in DFAFC^[21-22]. A large numbers of studies have shown that pure Pd or Pd-based catalysts are found to be more promising alternative anode electrocatalysts for DFAFC owing to lower price and greater abundance of the metal Pd on the earth′s crust compared with metal Pt^[23-27]. Li et al.^[28] synthesized phosphorus doped carbon supported Pd catalyst (Pd/P-C) by liquid reduction method, which exhibits the better electrocatalyst activity (0.8 A·mg^-1) and stability for formic acid oxidation in acid medium. It is generally accepted that formic acid oxidation on Pd catalyst takes place primarily through a direct pathway, leading to the direct formation of CO₂. However, there is still a small amount of formic acid being oxidized by an indirect pathway, resulting in the gradual accumulation of intermediates on the surface of the catalyst during the process of formic acid oxidation. These intermediates will make the catalyst got poisoning and lead to the decline in catalyst activity and long-term stability^[29-31]. Inspired by this, an effective strategy has been implemented by alloying of other transition metals with Pd to enhance both the electrocatalytic activity and long-term stability of the Pd catalyst such as bimetallic Pd-Pt^[32], Pd-Ag^[33], Pd-Sn^[34-35], Pd-Ni^[36], Pd-Au^[37], trimetallic Pd-Ni-Cu^[29], Pd-Ni-Ag^[38] and Pd-Pt-Ni^[39]. Typically, Lu et al.^[40] have successfully synthesized nanoneedle-covered Pd-Ag nanotubes through a galvanic displacement reaction with Ag nanorods at 100 ℃ (PdAg-100) and room temperature (PdAg-25) and obtained higher catalytic activity and stability than bulk Pd. Liu and co-workers^[41] have reported Pd-Sn nanoparticles supported on Vulcan XC-72 carbon by a microwave-assisted polyol process, and results indicate that Pd₂Sn₁/C and Pd₁Sn₁/C catalysts exhibit higher current density for formic acid oxidation compared with the prepared Pd/C catalyst. Carbon supported ternary PdNiCu catalyst was prepared by Hu et al.^[29] and exhibit an increased electroactivity for formic acid oxidation compared to that of binary Pd-Ni and Pd-Cu catalysts. Multi-walled carbon nanotube (CNT) supported Pd₁Cu₁Sn₁ ternary-metallic nanocatalyst was also studied by Zhu et al.^[42] through chemical reduction with NaBH₄ as a reducing agent and it reveals a higher mass activity of 534.83 mA·mg_Pd^-1 towards formic acid oxidation compared with bimetallic PdCu/CNTs and PdSn/CNTs. These studies have revealed that Pd-based bimetallic and ternary-metallic catalysts show a superior electro-chemical activity and stability for formic acid oxidation compared with pure Pd catalyst in virtue of synergistic effect between metals, electronic or surface effects^{[22, 43]}. However, ternary-metallic catalysts are more effective than the corresponding bimetallic catalysts in tuning the electronic properties and composition of catalytic surfaces. Furthermore, ternary Pd-based cata-lysts are able to further improve Pd usage efficiency and enhance their electrocatalytic performances. Consequently, it is necessary to develop novel ternary Pd-based catalysts with less cost and higher performance for formic acid oxidation. In addition, the choice of suitable support for Pd-based catalyst is also significant to reduce the Pd loading and improve the dispersion of catalyst nanoparticles, such as carbon black, graphene, carbon nanotubes, conductive poly-mers and so on. Recently, carbon nanotubes (CNTs) as a potential carbon carrier have been reported by many researchers for Pd-based catalysts^{[42, 44-46]}. As a support, CNTs possess unique structure and properties like high specific surface area, outstanding electronic conductivity and high chemical stability, which would be conducive to the dispersion and stability of Pd-based catalyst particles and further enhance their electroactivity^{[44, 47]}. Therefore, carbon nanotubes are a prominent support in the development of electro-catalysts.

  In this study, carbon nanotubes supported Pd and Pd-based binary/ternary catalysts (Pd/CNT, PdAg/CNT, PdSn/CNT, PdAgSn/CNT) were successfully synthesized by the NaBH₄ reduction method. The electrochemical activities of the prepared catalysts towards formic acid oxidation in both acidic and alkaline media were evaluated by cyclic voltammetry (CV) and chronoamperometry (CA) techniques. The results demonstrate that ternary Pd-Ag-Sn catalysts exhibit much higher electrochemical activity and stability towards formic acid oxidation in both acid and alkaline media.

  1.   Experimental conditions

  1.1   Chemicals

  Palladium chloride, stannous chloride, silver nitrate, sodium borohydride, ethylene-glycol, formic acid, sodium hydroxide and sulfuric acid were analy-tical purity grade and used as received without further purification. Water was deionized water subjected to the double distillation. Before used, multi-walled carbon nanotubes (CNTs, >90% (w/w), outside diameter: 10~20 nm, length: 5~20 μm) were added to a mixture of concentrated H₂SO₄ and concentrated HNO₃ (the volume ratio was 3:1), and heated at 60 ℃ under stirring for 8 h to obtain the acidified CNTs.

  1.2   Catalyst synthesis

  Catalysts were synthesized according to our recent report^[48]. Typically, the ternary Pd₇Ag₁Sn₂/CNT catalyst was prepared via the following steps: A metal precursor composed of 8.9 mg PdCl₂, 1.2 mg AgNO₃ and 3.2 mg SnCl₂ was added to the mixing solvent of 12 mL ethylene glycol and 4 mL water. Then the solid salts were fully dispersed for 30 min with ultra-sonication to make them be completely dissolved. Then, 30 mg of the acidified carbon nanotubes was added to the resulting solution and the mixture was further treated with ultra-sonication to obtain a uniform black ink. 3 mL of 50 g·L^-1 NaBH₄ dissolved in ethylene glycol was added dropwise to it under stirring to reduce the metal ions, and the mixture was stirred for 5 h. Finally, the resulting suspension was filtered, washed with water and dried at 40 ℃ in vacuum for 10 h to obtain the Pd₇Ag₁Sn₂/CNT catalyst. Other cata-lysts (Pd/CNT, Pd₇Ag₃/CNT, Pd₇Sn₂/CNT, Pd₇Ag₂Sn₂/CNT and Pd₇Ag₃Sn₂/CNT) were prepared according to this procedure by adjusting the metal molar ratio in the metal precursors. For synthesis of Pd/CNT catalyst, the precursor was composed of 8.9 mg PdCl₂. For Pd₇Ag₃/CNT catalyst, the precursor was composed of 8.9 mg PdCl₂ and 3.6 mg AgNO₃. For Pd₇Sn₂/CNT catalyst, the precursor was composed of 8.9 mg PdCl₂ and 3.2 mg SnCl₂. For Pd₇Ag₂Sn₂/CNT catalyst, the precursor was composed of 8.9 mg PdCl₂, 2.4 mg AgNO₃ and 3.2 mg SnCl₂. For Pd₇Ag₃Sn₂/CNT catalyst, the precursor was composed of 8.9 mg PdCl₂, 3.6 mg AgNO₃ and 3.2 mg SnCl₂.

  1.3   Characterization

  In order to further explore the microstructure and particle size distribution of the prepared catalysts, transmission election microscopic (TEM) images were recorded with a JEM-2100F. The X-ray diffraction (XRD) profiles of the prepared catalysts were collected to analyze the compositions of the samples in a D/MAX2500X diffractometer (Japan) operating with Cu Kα radiation generated at 40 kV and 250 mA (λ=0.154 18 nm) and 2θ=20°~90°. The elemental composi-tions and valence states of the samples were investig-ated by X-ray photoelectron spectroscopy (XPS) operated with an ESCALAB 250Xi spectrometer (VG Scientific Ltd., England). Inductively coupled plasma (ICP-AES-7510, Shimadzu) data of the nanoparticles were acquired to determine the Pd loading relative to the total mass of the catalyst. XRD profiles, XPS and ICP of the prepared catalysts were also investigated in our recent work^[48].

  1.4   Electrochemical measurements

  All electrochemical measurements of the prepared catalysts for formic acid oxidation in both acid and alkaline media were conducted in a conventional three-electrode system using an AutoLab PGSTAT30/FRA electrochemical workstation (Eco Chimie, The Netherlands). The counter electrode was a Pt sheet. A Ag/AgCl in saturated KCl solution was used as the reference electrode, and all potentials reported in this work were quoted versus the Ag/AgCl reference. The working electrode was a glassy carbon (GC) coated with a film of catalyst, which was fabricated as follows: the glassy carbon (GC, 3 mm diameter, from LanLiKe, TianJing, China) was firstly polished with a 0.3 μm alumina suspension to give a mirror surface. Then, 5 mg of the as-synthesized catalyst was dispersed ultrasonically in the mixed solution containing 0.94 mL of ethanol and 60 μL of 5% (w/w) Nafion solution to obtain a homogeneous ink. Finally, 15 μL of this ink was dropped onto the top surface of the polished GC disc by a micropipette and dried at room tempera-ture to get the working electrode. The blank CVs of the electrocatalysts were recorded in both 0.5 mol·L^-1 H₂SO₄ solution and 1.0 mol·L^-1 NaOH solution, and the corresponding electrocatalytic activities towards formic acid oxidation were investigated both in the solution of 0.5 mol·L^-1 H₂SO₄ in the presence of HCOOH and in 1.0 mol·L^-1 NaOH containing HCOOH. For the sake of comparison, electroactivity of the com-mercial Pd/C for formic acid oxidation was also examined under the same conditions. All measure-ments were performed at room temperature ((22±2) ℃).

  2.   Results and discussion

  2.1   Physical characterization

  Fig. 1(a~d) show the TEM images of the prepared Pd/CNT, Pd₇Ag₃/CNT, Pd₇Sn₂/CNT and Pd₇Ag₂Sn₂/CNT catalysts as the typical samples. The corresponding inset is the particle size distribution histogram of the catalyst sample. It is evident from the images that the metallic nanoparticles have been successfully decorated on the surface of multi-walled CNTs for all prepared catalysts. In addition, Pd/CNT, Pd₇Ag₃/CNT and Pd₇Sn₂/CNT catalysts exhibit obvious agglomera-tion between the nanoparticles and some particles are even stacked together to form clumps as shown in Fig. 1(a~c), and their average particle sizes (D_average) are 3.6, 4.7 and 3.7 nm, respectively. For the ternary Pd₇Ag₂Sn₂/CNT catalyst, however, most of the nano-particles are well uniformly dispersed on the surface of CNTs except for a small amount of agglomeration as indicated in Fig. 1d. Furthermore, the ternary Pd-Ag-Sn catalyst exhibits a smaller average particle size of 2.3 nm compared to Pd/CNT and binary Pd-Ag (or Pd-Sn) catalysts. Results indicate that an appropriate amount of Ag and Sn additives can effectively improve the dispersion of the Pd nanoparticles in the ternary Pd-Ag-Sn catalysts.

  图 1

  

  图 1  TEM images and the corresponding size distributions of the Pd/CNT (a), Pd₇Ag₃/CNT (b), Pd₇Sn₂/CNT (c) and Pd₇Ag₂Sn₂/CNT (d) samples

  Figure 1.  TEM images and the corresponding size distributions of the Pd/CNT (a), Pd₇Ag₃/CNT (b), Pd₇Sn₂/CNT (c) and Pd₇Ag₂Sn₂/CNT (d) samples

  下载: 全尺寸图片 幻灯片

  XRD patterns and XPS data of the prepared catalysts were recorded as indicated in Fig. 2. Fig. 2a shows that the peaks at 40.1°, 46.6°, 68.1° and 82.1° are attributed to characteristic diffraction peaks of face-centered cubic (fcc) crystalline Pd for Pd/CNT catalyst. However, a slight negative shift is observed with regard to the angle position of the Pd diffraction peaks on the Pd₇Ag₃/CNT and Pd₇Ag₂Sn₂/CNT catalysts compared to the Pd/CNT catalyst, while the Pd₇Sn₂/CNT catalyst does not show such a shift. This reveals that the alloy formation between Pd and Ag arises in the binary Pd₇Ag₃/CNT and ternary Pd-Ag-Sn catalysts. As is shown in Fig. 2b, the binding energies at 335.7 and 340.4 eV are ascribed to Pd3d_3/2 and Pd3d_5/2 spin orbit states of zero-valent Pd^[49]. But the other two distinct peaks located at 337.3 and 342.7 eV are related to Pd3d_3/2 and Pd3d_5/2 peaks of Pd(Ⅱ), which is indexed to the Pd oxide. These results indicate that the prepared Pd-based catalysts contain the metal Pd and Pd oxide. Fig. 3b shows the XPS spectra of 3d for Pd₇Ag₃/CNT and Pd₇Ag₂Sn₂/CNT catalysts, and the two obvious peaks centered at 367.9 and 373.9 eV are related to Ag3d_5/2 and Ag3d_3/2 respectively^[50], revealing that Ag ions are reduced completely during the preparation of the catalysts. Similarly, as indicated in Fig. 3c, Sn3d XPS spectra are divided into two peaks located at 486.8 and 487.4 eV, which are associated with Sn and SnO₂^[51], confir-ming that the metal Sn in the Pd₇Ag₂Sn₂/CNT and Pd₇Sn₂/CNT catalysts exists in the form of Sn and SnO₂.

  图 2

  

  图 2  XRD (a) and XPS (b~d) spectra of the Pd/CNT, Pd₇Ag₃/CNT, Pd₇Sn₂/CNT and Pd₇Ag₂Sn₂/CNT samples: (b) Pd3d, (c) Ag3d, and (d) Sn3d

  Figure 2.  XRD (a) and XPS (b~d) spectra of the Pd/CNT, Pd₇Ag₃/CNT, Pd₇Sn₂/CNT and Pd₇Ag₂Sn₂/CNT samples: (b) Pd3d, (c) Ag3d, and (d) Sn3d

  下载: 全尺寸图片 幻灯片

  图 3

  

  图 3  CV curves of the samples in 0.5 mol·L^-1 H₂SO₄ (a) and in 1.0 mol·L^-1 NaOH (b) at a scan rate of 50 mV·s^-1

  Figure 3.  CV curves of the samples in 0.5 mol·L^-1 H₂SO₄ (a) and in 1.0 mol·L^-1 NaOH (b) at a scan rate of 50 mV·s^-1

  下载: 全尺寸图片 幻灯片

  2.2   Electrochemical performance analysis

  Fig. 3a shows CV curves of the prepared catalysts and Pd/C in 0.5 mol·L^-1 H₂SO₄ solution. All catalysts reveal a similar CV curve to Pd/C in acidic solution. A well-defined hydrogen adsorption/desorption peaks around 0 V arises on all samples, and the cathode characteristic reduction peak (r_p) of the Pd oxides produced during the forward potential scan is vividly observed at ca. 0.48 V for all the catalysts. Also, the r_p peak current density on the Pd/C, Pd/CNT, Pd₇Ag₃/CNT, Pd₇Sn₂/CNT, Pd₇Ag₁Sn₂/CNT, Pd₇Ag₂Sn₂/CNT, Pd₇Ag₃Sn₂/CNT catalysts is 6.6, 6.9, 10.4, 8.6, 11.5, 15.2 and 12.2 mA·cm^-2, respectively. Fig. 3b shows cyclic CV curves of the prepared catalysts and Pd/C in 1.0 mol·L^-1 NaOH solution. Similarly, the cathode reduction peak (r_n) at ca. -0.41 V is attributed to the formation of Pd oxides during the forward-going, and the r_n peak current density is 14.3, 17.0, 20.5, 24.6, 20.2, 31.4 and 24.0 mA·cm^-2 for Pd/C, Pd/CNT, Pd₇Ag₃/CNT, Pd₇Sn₂/CNT, Pd₇Ag₁Sn₂/CNT, Pd₇Ag₂Sn₂/CNT Pd₇Ag₃Sn₂/CNT catalysts, respectively. Based on the charge of PdO reduction peak in each CV, the electrochemical active surface area (ECSA) of Pd for the samples can be calculated by using the methods reported in the literature and corresponding results are listed in Table 1^{[42, 52-53]}. Results reveal that Pd₇Ag₂Sn₂ /CNT catalyst possesses the largest ECSA value of 9.56 m²·g^-1 in H₂SO₄ solution and 15.34 m²·g^-1 in NaOH solution among the prepared catalysts and Pd/C, which is consistent with the results observed from TEM images.

  表 1

  

  表 1  ECSA values of Pd/C and the prepared samples in both 1 mol·L^-1 NaOH and 0.5 mol·L^-1 H₂SO₄ solution

  Table 1.  ECSA values of Pd/C and the prepared samples in both 1 mol·L^-1 NaOH and 0.5 mol·L^-1 H₂SO₄ solution

  下载: 导出CSV

  	Solution	Catalyst
  		Pd/C	Pd/CNT	Pd7Ag3/CNT	Pd7Sn2/CNT	Pd7Ag1Sn2/CNT	Pd7Ag2Sn2/CNT	Pd7Ag3Sn2/CNT
  ECSA/	H2SO4	5.22	4.60	6.39	5.99	7.28	9.56	7.24
  (m2·g-1)	NaOH	11.13	9.73	8.75	11.83	9.96	15.34	11.08

  Electrocatalytic activity of the prepared catalysts for formic acid oxidation was measured in 0.5 mol·L^-1 H₂SO₄ solution containing 0.5 mol·L^-1 formic acid by CV as indicated in Fig. 4a. A characteristic anodic peaks j_f1 caused by formic acid oxidation is observed for all catalysts. In general, all the as-synthesized Pd-based catalysts exhibit better electrocatalytic activity for formic acid oxidation than Pd/C. Further, the ternary Pd₇Ag₂Sn₂/CNT catalyst shows the largest j_f1 peak current density of 108.8 mA·cm^-2, which is 6.7 times higher than the Pd/C catalyst. Also, the j_f1 peak current density on the binary Pd₇Ag₃/CNT and Pd₇Sn₂/CNT catalysts is 2.7 and 2.3 times larger than that of Pd/C catalyst respectively. This may be contributed to the synergistic effect between Pd and Ag/Sn^{[2, 26]}. Furth-ermore, ternary Pd₇Ag₁Sn₂/CNT and Pd₇Ag₂Sn₂/CNT catalysts display an onset potential (OP) of ca. -0.06 V for formic acid oxidation in acidic media, which presents a negative shift compared to that of ca. -0.045 V on the other catalysts. Fig. 4b shows the CV curves of all samples in 1.0 mol·L^-1 NaOH solution containing 0.5 mol·L^-1 formic acid. Compared to formic acid oxidation in acidic solution (Fig. 4a), the formic acid oxidation in alkaline solution (Fig. 4b) presents a much negative onset potential of ca. -0.82 V. Fig. 4b also shows that during the forward-going scan, formic acid oxidation current density displays an almost linear increment with the positive shift of the anodic potential until an anodic peak j_f2 at ca. 0.2 V arises. The anodic current density for formic acid oxida-tion in alkaline medium follows the order: Pd₇Ag₂Sn₂/CNT>Pd₇Ag₁Sn₂/CNT>Pd₇Ag₃/CNT>Pd₇Ag₃Sn₂/CNT>Pd/CNT>Pd₇Sn₂/CNT>Pd/C. Obviously, the ternary Pd₇Ag₂Sn₂/CNT catalyst presents the best electroca-talytic activity for formic acid oxidation in alkaline medium among the prepared catalysts.

  图 4

  

  图 4  CV curves of the samples in 0.5 mol·L^-1 H₂SO₄ containing 0.5 mol·L^-1 HCOOH (a) and in 1.0 mol·L^-1 NaOH containing 0.5 mol·L^-1 HCOOH (b) at a scan rate of 50 mV·s^-1

  Figure 4.  CV curves of the samples in 0.5 mol·L^-1 H₂SO₄ containing 0.5 mol·L^-1 HCOOH (a) and in 1.0 mol·L^-1 NaOH containing 0.5 mol·L^-1 HCOOH (b) at a scan rate of 50 mV·s^-1

  下载: 全尺寸图片 幻灯片

  It is generally considered that there are two possible parallel pathways for the oxidation of formic acid^{[42, 44]}: (ⅰ) a "direct pathway" in which formic acid is directly oxidized to CO₂ without production of any intermediate and (ⅱ) an "indirect pathway" which involves two steps including the dehydrogenation of formic acid and the adsorption of CO intermediate on the Pd catalyst surface. The choice of the pathway greatly depends on the properties of the catalyst used. Normally, the direct pathway prevails for formic acid oxidation on Pd-containing catalysts^[54-55]. The processes of formic acid oxidation in acidic solution are based on the following equations (1)~(3)^[44]:

  $ {\rm{HCOOH + Pd}} \to {\rm{HCOO-Pd + }}{{\rm{H}}^{\rm{ + }}} + {{\rm{e}}^{\rm{-}}} $

  $ {\rm{HCOO-Pd}} \to {\rm{Pd-H + C}}{{\rm{O}}_{\rm{2}}} $

  ${\rm{Pd-H}} \to {\rm{Pd + }}{{\rm{H}}^{\rm{ + }}} + {{\rm{e}}^{\rm{-}}}\left( {{\rm{dehydrogenation}}} \right) $

  During formic acid oxidation, the adsorbed HCOO_ad (HCOO-Pd) species are firstly formed via the adsorption of formic acid molecules on the surface of Pd-based catalysts and subsequent break of O-H bond in the adsorbed HCOOH (HCOOH_ad) (Equation (1)). Then, decomposition of the HCOO_ad species produces CO₂ by breaking C-H bond (equation (2)). In alkaline media, electro-oxidation of formic acid on Pd-based catalysts follows a similar mechanism to that in acidic media except that the adsorbed HCOO_ad species on Pd can be formed easier because of the neutralization reaction between HCOOH and NaOH, leading to the much negative onset potential of formic acid oxidation. In general, the addition of others metal or metal oxide to Pd catalyst can significantly improve its electroactivity due to the synergistic effect of different metals. It is noticed from Fig. 4 that the Pd₇Ag₂Sn₂/CNT catalyst displays the highest current density of formic acid oxidation in both acidic and alkaline media among the prepared catalysts, reflecting that the addition of proper amount of Ag or Sn is conducive to enhance the electrochemical activity. It is known from the XRD data of the prepared catalysts that the alloying between Pd and Ag arises. Based on the so-called bifunctional mecha-nism therefore, the adsorption bond of intermediates like absorbed CO (CO_ad) and COOH (COOH_ad) at the surface of catalysts, produced by formic oxidation during the forward scan, can be efficaciously weakened by the Pd-Ag bimetallic alloy. This makes the decomposition of formic acid to CO₂ go into easier. Furthermore, the presence of Pd-Ag alloy can also prominently reduce the accumulation of poisoning-intermediates on the catalyst surface and release more Pd active sites. The presence of SnO₂ observed from the XPS data may also contribute to the removal of toxic intermediates to accelerate the adsorption and desorption of formic acid on the catalyst surface.

  Effect of formic acid concentration on the kinetic characterization of formic acid oxidation was further investigated. Fig. 5a shows the CV curves of ternary Pd₇Ag₂Sn₂/CNT catalyst in 0.5 mol·L^-1 H₂SO₄ solution with different formic acid concentrations at 50 mV·s^-1, and Fig. 5b depicts the relationship between the anodic peak current density and HCOOH concentra-tion. As can be seen from Fig. 5b, the j_p1 peak current density exhibits a rapid rise with the formic acid concentration in the range of 0.5 to 1.8 mol·L^-1, while it displays a decrease from 1.8 to 2.5 mol·L^-1. In addition, the j_p1 peak potential shifts to more positive direction at the higher concentration of formic acid. In 1 mol·L^-1 NaOH solution, dependence of the j_p2 peak current density upon HCOOH concentration is also studied as indicated in Fig. 6(a, b). A similar changing trend of the j_p1 peak current density vs HCOOH concentration to Fig. 5b is observed, revealing that the HCOOH concentration has the same effect on electroactivity of the ternary Pd₇Ag₂Sn₂/CNT catalyst in both acidic and alkaline media. At high concen-trations of HCOOH, the j_p1 peak current density for the oxidation of formic acid on the Pd₇Ag₂Sn₂/CNT catalyst decreases. This may be related to the saturated adsorption of HCOOH on Pd active sites at high concentrations of HCOOH. On the other hand, high concentrations of HCCOH may result in partial decomposition of HCOOH to produce CO (equation (4)), which is absorbed on the surface of the catalyst and reduce the electroactivity of the catalyst.

  $ {\rm{HCOOH = }}{{\rm{H}}_{\rm{2}}}{\rm{O + CO}} $

  图 5

  

  图 5  (a) CV curves of ternary Pd₇Ag₂Sn₂/CNT catalyst in 0.5 mol·L^-1 H₂SO₄ solution with different formic acid concentrations at 50 mV·s^-1; (b) Relationship between the anodic peak current density and HCOOH concentration (C_HCOOH) for ternary Pd₇Ag₂Sn₂/CNT catalyst

  Figure 5.  (a) CV curves of ternary Pd₇Ag₂Sn₂/CNT catalyst in 0.5 mol·L^-1 H₂SO₄ solution with different formic acid concentrations at 50 mV·s^-1; (b) Relationship between the anodic peak current density and HCOOH concentration (C_HCOOH) for ternary Pd₇Ag₂Sn₂/CNT catalyst

  下载: 全尺寸图片 幻灯片

  图 6

  

  图 6  (a) CV curves of ternary Pd₇Ag₂Sn₂/CNT catalyst in 1.0 mol·L^-1 NaOH solution with different formic acid concentrations at 50 mV·s^-1; (b) Relationship between the anodic peak current density and HCOOH concentration for ternary Pd₇Ag₂Sn₂/CNT catalyst

  Figure 6.  (a) CV curves of ternary Pd₇Ag₂Sn₂/CNT catalyst in 1.0 mol·L^-1 NaOH solution with different formic acid concentrations at 50 mV·s^-1; (b) Relationship between the anodic peak current density and HCOOH concentration for ternary Pd₇Ag₂Sn₂/CNT catalyst

  下载: 全尺寸图片 幻灯片

  Fig. 7(a, b) displays the CV curves for the oxidation of pre-adsorbed carbon monoxide (CO) on Pd/C, Pd/CNT and Pd₇Ag₂Sn₂/CNT catalysts in order to inves-tigate the anti-poisoning intermediates ability of the catalysts. It is shown from Fig. 7a recorded in 0.5 mol·L^-1 H₂SO₄ solution that an intense stripping peak of CO_ad is observed on the catalysts. The peak potential of CO stripping on Pd/C, Pd/CNT and Pd₇Ag₂Sn₂/CNT catalysts is 0.792, 0.726 and 0.722 V, respectively, reflecting that the prepared catalysts in this work have more negative CO stripping peak potential values than that of the Pd/C catalyst. The lower potential displays the weaker binding energy between Pd and CO_ad on Pd/CNT and Pd₇Ag₂Sn₂/CNT catalysts. Notably, the onset potential of CO oxidation for Pd₇Ag₂Sn₂/CNT catalyst is measured at 0.67 V, showing a negative shift compared to that for Pd/C (0.72 V) and Pd/CNT (0.70 V). Results indicate that the CO_ad on the surface of the Pd₇Ag₂Sn₂/CNT catalyst can be more easily removed. Furthermore, the CO stripping is also tested for Pd/C, Pd/CNT and Pd₇Ag₂Sn₂/CNT catalysts in mol·L^-1 NaOH solution as indicated in Fig. 7b. It is worth noting that the Pd/CNT and Pd₇Ag₂Sn₂/CNT catalysts also exhibit a more negative CO stripping peak potential at ca. -0.19 V compared to Pd/C catalyst (ca. -0.139 V). These results show that the ternary Pd₇Ag₂Sn₂/CNT catalyst possesses much better resistance to CO_ad poisoning than the Pd/C and Pd/CNT catalysts.

  图 7

  

  图 7  CO stripping curves of the Pd/C, Pd/CNT and Pd₇Ag₂Sn₂/CNT catalysts in 0.5 mol·L^-1 H₂SO₄ (a) and 1.0 mol·L^-1 NaOH (b) at a scan rate of 50 mV·s^-1

  Figure 7.  CO stripping curves of the Pd/C, Pd/CNT and Pd₇Ag₂Sn₂/CNT catalysts in 0.5 mol·L^-1 H₂SO₄ (a) and 1.0 mol·L^-1 NaOH (b) at a scan rate of 50 mV·s^-1

  下载: 全尺寸图片 幻灯片

  The long-term electrocatalytic activity of the Pd/C, Pd/CNT and Pd₇Ag₂Sn₂/CNT catalysts is evaluated by CA measurement in 0.5 mol·L^-1 H₂SO₄ solution containing 0.5 mol·L^-1 HCOOH at different potentials as depicted in Fig. 8. The current density of the studied catalysts exhibits a continuous decay in the initial stage at both 0.05 (Fig. 8a) and 0.1 V (Fig. 8b). This may be attributed to the adsorption of CO-like intermediates on the surface of the catalysts, leading to the decline on the number of the active sites^{[22, 56]}. However, the Pd₇Ag₂Sn₂/CNT catalyst exhibits a significantly slower decay rate of current density than the Pd/C and Pd/CNT catalysts. Additionally, at the end of electrolysis (at 3 600 s), the current density on the ternary Pd₇Ag₂Sn₂/CNT catalyst is 5.8 mA·cm^-2 at 0.05 V or 14.2 mA·cm^-2 at 0.1 V, which is still the highest among the studied catalysts. Fig. 9 also shows CA curves of the Pd/C, Pd/CNT and Pd₇Ag₂Sn₂/CNT catalysts in 1 mol·L^-1 NaOH solution containing 0.5 mol·L^-1 HCOOH at the potentials of -0.75 and -0.45 V. Apparently, the current density of ternary Pd₇Ag₂Sn₂ /CNT catalyst is 13.7 mA·cm^-2 at -0.75 V after 3 600 s as shown in Fig. 9a, which is almost 3.3 and 7.2 times larger than that of the Pd/C (4.2 mA·cm^-2) and Pd/CNT (1.9 mA·cm^-2) catalysts, respectively. In addition, it can be also observed from Fig. 9b that ternary Pd₇Ag₂Sn₂/CNT catalyst has the highest current density among the studied catalysts at the potential of -0.45 V after 3 600 s. The above results demonstrate that the as-synthesized ternary Pd₇Ag₂Sn₂/CNT catalyst displays excellent electrocatalytic activity and more outstanding durability towards formic acid oxidation in both acidic and alkaline media, which is consistent with the results derived from CV analyses.

  图 8

  

  图 8  Chronoamperometric responses of the Pd₇Ag₂Sn₂/CNT catalyst in 0.5 mol·L^-1 H₂SO₄ comtaining 0.5 mol·L^-1 HCOOH at 0.05 V (a) and 0.1 V (b)

  Figure 8.  Chronoamperometric responses of the Pd₇Ag₂Sn₂/CNT catalyst in 0.5 mol·L^-1 H₂SO₄ comtaining 0.5 mol·L^-1 HCOOH at 0.05 V (a) and 0.1 V (b)

  下载: 全尺寸图片 幻灯片

  图 9

  

  图 9  Chronoamperometric responses of the Pd₇Ag₂Sn₂/CNT catalyst in 1.0 mol·L^-1 NaOH comtaining 0.5 mol·L^-1 HCOOH at -0.75 V (a) and -0.45 V (b)

  Figure 9.  Chronoamperometric responses of the Pd₇Ag₂Sn₂/CNT catalyst in 1.0 mol·L^-1 NaOH comtaining 0.5 mol·L^-1 HCOOH at -0.75 V (a) and -0.45 V (b)

  下载: 全尺寸图片 幻灯片

  Fig. 10 shows the CV profiles of the prepared catalysts where the current density is based on the mass of Pd to show the Pd usage efficiency for formic acid oxidation. It can be found from Fig. 10a that during the forward-going scan the anodic peak mass current density of the Pd/C, Pd/CNT, Pd₇Ag₃/CNT, Pd₇Sn₂/CNT, Pd₇Ag₁Sn₂/CNT, Pd₇Ag₂Sn₂/CNT and Pd₇Ag₃Sn₂/CNT catalysts towards formic acid oxidation in 0.5 mol·L^-1 H₂SO₄ solution containing 0.5 mol·L^-1 HCOOH is 153, 274, 474, 431, 1 030, 1 364 and 767 mA·mg_Pd^-1, respectively, which indicates that the ternary Pd₇Ag₂Sn₂/CNT catalyst has the highest Pd mass current density among the prepared catalysts. Zhu et al.^[42] prepared the ternary PdCuSn/CNTs catalyst with the mass current density of 534.8 mA·mg_Pd^-1. Binary PdCo/CFC catalyst was also synthesized with the mass current density of 1 220 mA·mg_Pd^-1 by Vafaei and co-workers^[57]. Compared with the reported cataly-sts, the ternary Pd₇Ag₂Sn₂/CNT catalyst prepared in this work exhibits excellent electrocatalytic activity and higher Pd usage efficiency for formic acid oxidation. Additionally, the ternary Pd₇Ag₂Sn₂/CNT catalyst also displays the highest Pd mass current density in 1 mol·L^-1 NaOH solution containing 0.5 mol·L^-1 HCOOH as indicated in Fig. 10b, which is as high as 2 640 mA·mg_Pd^-1. These results reveal that the ternary Pd₇Ag₂Sn₂/CNT catalyst can be applied to DFAFCs as a promising anodic catalyst for formic acid oxidation in both acidic and alkaline media due to the synergetic effect between Pd and Ag/Sn.

  图 10

  

  图 10  CV curves based on the Pd mass current density in 0.5 mol·L^-1 H₂SO₄ containing 0.5 mol·L^-1 HCOOH (a) and in 1.0 mol·L^-1 NaOH containing 0.5 mol·L^-1 HCOOH (b) based on Fig. 4

  Figure 10.  CV curves based on the Pd mass current density in 0.5 mol·L^-1 H₂SO₄ containing 0.5 mol·L^-1 HCOOH (a) and in 1.0 mol·L^-1 NaOH containing 0.5 mol·L^-1 HCOOH (b) based on Fig. 4

  下载: 全尺寸图片 幻灯片

  3.   Conclusions

  In summary, carbon nanotube-supported Pd-based catalysts including Pd/CNT, binary Pd-Ag/CNT and ternary Pd-Ag-Sn/CNT were synthesized by the conventional NaBH₄ reduction method. The metal nanoparticles of ternary Pd₇Ag₂Sn₂/CNT catalyst are uniformly dispersed on the surface of the carbon nanotubes with an average size of about 2.3 nm. Among the catalysts investigated, the ternary Pd₇Ag₂Sn₂ /CNT catalyst has the best catalytic performance and stability towards formic acid oxidation in both acidic and alkaline media. These outstanding features may be ascribed to the formation of Pd-Ag alloy and the presence of SnO₂, which are conducive to the reducing of the accumulation of poisoning-intermediates and the releasing of the active sites of Pd during the formic acid electro-oxidation. Meanwhile, the ternary Pd₇Ag₂Sn₂/CNT catalyst exhibits the highest Pd mass current density of 1 364 mA·mg^-1 in H₂SO₄ solution or 2 640 mA·mg^-1 in NaOH solution, showing the ultra high usage efficiency of Pd in the prepared Pd-based catalysts towards formic acid oxidation. Results imply that the ternary Pd-Ag-Sn catalyst may be a very promising anodic electrocatalyst for direct formic acid fuel cells.

• 1. [1]

     Lu X Y, Zheng L, Zhang M S, et al. Electrochim. Acta, 2017, 238:194-201 doi: 10.1016/j.electacta.2017.03.115

  2. [2]

     Xu H, Yan B, Zhang K, et al. Appl. Surf. Sci., 2017, 416:191-199 doi: 10.1016/j.apsusc.2017.04.160

  3. [3]

     Budischak C, Sewell D, Thomson H, et al. J. Power Sources, 2013, 225(3):60-74 http://linkinghub.elsevier.com/retrieve/pii/S0378775312014759

  4. [4]

     Miao K H, LuoY, Zou J S, et al. Electrochim. Acta, 2017, 251:588-594 doi: 10.1016/j.electacta.2017.08.167

  5. [5]

     El-Nagar G A, Dawood K M, El-Deab M S, et al. Appl. Catal., B, 2017, 213:118-126 doi: 10.1016/j.apcatb.2017.05.006

  6. [6]

     Yi Q F, Zou T, Zhang Y Y, et al. J. Power Sources, 2016, 321:219-225 doi: 10.1016/j.jpowsour.2016.04.134

  7. [7]

     Yu X W, Pickup P G. J. Power Sources, 2008, 182(1):124-132 doi: 10.1016/j.jpowsour.2008.03.075

  8. [8]

     Rhee Y W, Ha S Y, Masel R I. J. Power Sources, 2003, 117(1/2):35-38 https://www.cheric.org/research/tech/periodicals/view.php?seq=458327

  9. [9]

     邹涛, 易清风, 张媛媛, 等.高等学校化学学报, 2017, 38(1):101-107 doi: 10.7503/cjcu20160457ZOU Tao, YI Qing-Feng, ZHANG Yuan-Yuan, et al. Chem. J. Chinese Universities, 2017, 38(1):101-107 doi: 10.7503/cjcu20160457

  10. [10]

      Winjobi O, Zhang Z Y, Liang C H, et al. Electrochim. Acta, 2010, 55(13):4217-4221 doi: 10.1016/j.electacta.2010.02.062

  11. [11]

      Yi Q F, Zhang J J, Chen A C, et al. J. Appl. Electrochem., 2008, 38(5):695-701 doi: 10.1007/s10800-008-9490-x

  12. [12]

      Zhu C X, Liu D, Chen Z, et al. J. Colloid Interface Sci., 2018, 511:77-83 doi: 10.1016/j.jcis.2017.09.109

  13. [13]

      Cabello G, Davoglio R A, Hartl F W, et al. Electrochim. Acta, 2017, 224:56-63 doi: 10.1016/j.electacta.2016.12.022

  14. [14]

      Han Y, Ouyang Y J, Xie Z H, et al. J. Mater. Sci. Technol., 2016, 32(7):639-645 doi: 10.1016/j.jmst.2016.04.014

  15. [15]

      Awaludin Z, Okajima T, Ohsaka T. Electrochem. Commun., 2013, 31:100-103 doi: 10.1016/j.elecom.2013.03.020

  16. [16]

      Yang H Z, Dai L, Xu D, et al. Electrochim. Acta, 2010, 55(27):8000-8004 doi: 10.1016/j.electacta.2010.03.026

  17. [17]

      Yi Q F, Chen A C, Huang W, et al. Electrochem. Commun., 2007, 9(7):1513-1518 doi: 10.1016/j.elecom.2007.02.014

  18. [18]

      Choi J H, Jeong K J, Dong Y J, et al. J. Power Sources, 2006, 163(1):71-75 doi: 10.1016/j.jpowsour.2006.02.072

  19. [19]

      Yi Q F, Li L, Yu W Q, et al. J. Alloys Compd., 2008, 466(1/2):52-58 doi: 10.1007/s11467-013-0324-x

  20. [20]

      张丽娟, 夏定国.无机化学学报, 2006, 22(6):1085-1089 http://www.wjhxxb.cn/wjhxxbcn/ch/reader/view_abstract.aspx?file_no=20060621ZHANG Li-Juan, XIA Ding-Guo. Chinese J. Inorg. Chem., 2006, 22(6):1085-1089 http://www.wjhxxb.cn/wjhxxbcn/ch/reader/view_abstract.aspx?file_no=20060621

  21. [21]

      Li R X, Ma Z Z, Zhang F, et al. Electrochim. Acta, 2016, 220:193-204 doi: 10.1016/j.electacta.2016.10.105

  22. [22]

      Xu H, Zhang K, Yan B, et al. J. Power Sources, 2017, 356:27-35 doi: 10.1016/j.jpowsour.2017.04.070

  23. [23]

      Xin Z L, Wang S H, Wang J, et al. Electrochem. Commun., 2016, 67:26-30 doi: 10.1016/j.elecom.2016.03.008

  24. [24]

      Wang Y R, He Q L, Wei H G, et al. Electrochim. Acta, 2015, 184:452-465 doi: 10.1016/j.electacta.2015.10.046

  25. [25]

      Krishna R, Fernandes D M, Marinoiu A, et al. Int. J. Hydrogen Energy, 2017, 42(37):23639-23646 doi: 10.1016/j.ijhydene.2017.04.263

  26. [26]

      陈滢, 唐亚文, 高颖, 等.无机化学学报, 2008, 24(4):560-564 http://www.wjhxxb.cn/wjhxxbcn/ch/reader/view_abstract.aspx?file_no=20080411CHEN Ying, TANG Ya-Wen, GAO Ying, et al. Chinese J. Inorg. Chem., 2008, 24(4):560-564 http://www.wjhxxb.cn/wjhxxbcn/ch/reader/view_abstract.aspx?file_no=20080411

  27. [27]

      沈娟章, 唐亚文, 陆天虹.无机化学学报, 2012, 28(2):326-330 http://www.wjhxxb.cn/wjhxxbcn/ch/reader/view_abstract.aspx?file_no=20120220&flag=1SHEN Juan-Zhang, TANG Ya-Wen, LU Tian-Hong. Chinese J. Inorg. Chem., 2012, 28(2):326-330 http://www.wjhxxb.cn/wjhxxbcn/ch/reader/view_abstract.aspx?file_no=20120220&flag=1

  28. [28]

      Li J D, Tian Q F, Jiang S Y, et al. Electrochim. Acta, 2016, 213:21-30 doi: 10.1016/j.electacta.2016.06.041

  29. [29]

      Hu S Z, Munoz F, Noborikawa J, et al. Appl. Catal., B, 2016, 180:758-765 doi: 10.1016/j.apcatb.2015.07.023

  30. [30]

      Jiang K, Zhang H X, Zou S Z, et al. Phys. Chem. Chem. Phys., 2014, 16(38):20360-20376 doi: 10.1039/C4CP03151B

  31. [31]

      Haan J L, Stafford K M, Masel R I. J. Phys. Chem. C, 2010, 114(26):11665-11672 doi: 10.1021/jp102990t

  32. [32]

      Yi Q F, Huang W, Liu X P, et al. J. Electroanal. Chem., 2008, 619(1):197-205 http://www.sciencedirect.com/science/article/pii/S1003632613625820

  33. [33]

      Liu D, Xie M L, Wang C M, et al. Nano Res., 2016, 9(6):1590-1599 doi: 10.1007/s12274-016-1053-6

  34. [34]

      Tu D D, Wu B, Wang B X, et al. Appl. Catal., B, 2011, 103(1/2):163-168 https://www.journals.elsevier.com/applied-catalysis-b-environmental

  35. [35]

      Yi Q F, Chen Q H, Yang Z. J. Power Sources, 2015, 298:171-176 doi: 10.1016/j.jpowsour.2015.08.050

  36. [36]

      Zhang Y Y, Yi Q F, Zou T, et al. Ionics, 2017, 23:3169-3176 doi: 10.1007/s11581-017-2113-y

  37. [37]

      Suo Y G, Hsing I M. Electrochim. Acta, 2011, 56:2174-2183 doi: 10.1016/j.electacta.2010.12.037

  38. [38]

      Yurderi M, Bulut A, Zahmakirana M, et al. Appl. Catal., B, 2014, 160(7):514-524 doi: 10.1016/j.apcatb.2014.06.009

  39. [39]

      Zhang J M, Wang R X, Nong R J, et al. Int. J. Hydrogen Energy, 2017, 42(10):7226-7234 doi: 10.1016/j.ijhydene.2016.05.198

  40. [40]

      Lu Y Z, Chen W. J. Phys. Chem. C, 2010, 114(49):21190-21200 doi: 10.1021/jp107768n

  41. [41]

      Liu Z L, Zhang X H. Electrochem. Commun., 2009, 11(8):1667-1670 doi: 10.1016/j.elecom.2009.06.023

  42. [42]

      Zhu F C, Ma G S, Bai Z C, et al. J. Power Sources, 2013, 242(22):610-620 doi: 10.1007/s12274-016-1076-z

  43. [43]

      Shen Y Y, Sun Y, Zhou L N, et al. J. Mater. Chem. A, 2014, 2(9):2977-2984 doi: 10.1039/c3ta14502f

  44. [44]

      Marinšek M, Šala M, Jančar B. J. Power Sources, 2013, 235(8):111-116 http://www.sciencedirect.com/science/article/pii/S0360319915007910

  45. [45]

      Yi Q F, Chu H, Tang M X, et al. J. Electroanal. Chem., 2015, 739:178-186 doi: 10.1016/j.jelechem.2014.12.029

  46. [46]

      Cazares-Ávila E, Ruiz-Ruiz E J, Hernández-Ramírez A, et al. Int. J. Hydrogen Energy, 2017, 42(51):30349-30358 doi: 10.1016/j.ijhydene.2017.08.156

  47. [47]

      Carmo M, Paganin V A, Rosolen J M, et al. J. Power Sources, 2005, 142(1):169-176 doi: 10.1134/S1023193510080045

  48. [48]

      Zhang Y Y, Yi Q F, Deng Z L, et al. Catal. Lett., 2018, 148:1190-1201 doi: 10.1007/s10562-018-2335-2

  49. [49]

      Xiong B, Zhou Y K, Zhao Y Y, et al. Carbon, 2013, 52(2):181-192 doi: 10.1007%2Fs12274-014-0644-3

  50. [50]

      Yang Z Z, Wang X L, Kang X, et al. Electrochim. Acta, 2017, 236:72-81 doi: 10.1016/j.electacta.2017.03.165

  51. [51]

      Lewera A, Barczuk P J, Skorupska K, et al. J. Electroanal. Chem., 2011, 662(1):93-99 doi: 10.1016/j.jelechem.2011.03.038

  52. [52]

      Kakaei K, Dorraji M. Electrochim. Acta, 2014, 143:207-215 doi: 10.1016/j.electacta.2014.07.134

  53. [53]

      Li S S, Wang A J, Hu Y Y, et al. J. Mater. Chem. A, 2014, 2(43):18177-18183 doi: 10.1039/C4TA04164J

  54. [54]

      Ha S, Larsen R, Masel R I. J. Power Sources, 2005, 144(1):28-34 doi: 10.1016/j.jpowsour.2004.12.031

  55. [55]

      Wang Y Y, Qi Y Y, Zhang D J. Comput. Theor. Chem., 2014, 1049:51-54 doi: 10.1016/j.comptc.2014.09.020

  56. [56]

      Zhang S X, Qing M, Zhang H, et al. Electrochem. Commun., 2009, 11(11):2249-2252 doi: 10.1016/j.elecom.2009.10.001

  57. [57]

      Vafaei M, Rezaei M, Tabaian S H, et al. J. Solid State Electrochem., 2015, 19(1):289-298 doi: 10.1007/s10008-014-2561-5

• 

  Figure 1  TEM images and the corresponding size distributions of the Pd/CNT (a), Pd₇Ag₃/CNT (b), Pd₇Sn₂/CNT (c) and Pd₇Ag₂Sn₂/CNT (d) samples

  下载: 全尺寸图片 幻灯片
  

  Figure 2  XRD (a) and XPS (b~d) spectra of the Pd/CNT, Pd₇Ag₃/CNT, Pd₇Sn₂/CNT and Pd₇Ag₂Sn₂/CNT samples: (b) Pd3d, (c) Ag3d, and (d) Sn3d

  下载: 全尺寸图片 幻灯片
  

  Figure 3  CV curves of the samples in 0.5 mol·L^-1 H₂SO₄ (a) and in 1.0 mol·L^-1 NaOH (b) at a scan rate of 50 mV·s^-1

  下载: 全尺寸图片 幻灯片
  

  Figure 4  CV curves of the samples in 0.5 mol·L^-1 H₂SO₄ containing 0.5 mol·L^-1 HCOOH (a) and in 1.0 mol·L^-1 NaOH containing 0.5 mol·L^-1 HCOOH (b) at a scan rate of 50 mV·s^-1

  下载: 全尺寸图片 幻灯片
  

  Figure 5  (a) CV curves of ternary Pd₇Ag₂Sn₂/CNT catalyst in 0.5 mol·L^-1 H₂SO₄ solution with different formic acid concentrations at 50 mV·s^-1; (b) Relationship between the anodic peak current density and HCOOH concentration (C_HCOOH) for ternary Pd₇Ag₂Sn₂/CNT catalyst

  下载: 全尺寸图片 幻灯片
  

  Figure 6  (a) CV curves of ternary Pd₇Ag₂Sn₂/CNT catalyst in 1.0 mol·L^-1 NaOH solution with different formic acid concentrations at 50 mV·s^-1; (b) Relationship between the anodic peak current density and HCOOH concentration for ternary Pd₇Ag₂Sn₂/CNT catalyst

  下载: 全尺寸图片 幻灯片
  

  Figure 7  CO stripping curves of the Pd/C, Pd/CNT and Pd₇Ag₂Sn₂/CNT catalysts in 0.5 mol·L^-1 H₂SO₄ (a) and 1.0 mol·L^-1 NaOH (b) at a scan rate of 50 mV·s^-1

  下载: 全尺寸图片 幻灯片
  

  Figure 8  Chronoamperometric responses of the Pd₇Ag₂Sn₂/CNT catalyst in 0.5 mol·L^-1 H₂SO₄ comtaining 0.5 mol·L^-1 HCOOH at 0.05 V (a) and 0.1 V (b)

  下载: 全尺寸图片 幻灯片
  

  Figure 9  Chronoamperometric responses of the Pd₇Ag₂Sn₂/CNT catalyst in 1.0 mol·L^-1 NaOH comtaining 0.5 mol·L^-1 HCOOH at -0.75 V (a) and -0.45 V (b)

  下载: 全尺寸图片 幻灯片
  

  Figure 10  CV curves based on the Pd mass current density in 0.5 mol·L^-1 H₂SO₄ containing 0.5 mol·L^-1 HCOOH (a) and in 1.0 mol·L^-1 NaOH containing 0.5 mol·L^-1 HCOOH (b) based on Fig. 4

  下载: 全尺寸图片 幻灯片

  Table 1.  ECSA values of Pd/C and the prepared samples in both 1 mol·L^-1 NaOH and 0.5 mol·L^-1 H₂SO₄ solution

  	Solution	Catalyst
  		Pd/C	Pd/CNT	Pd7Ag3/CNT	Pd7Sn2/CNT	Pd7Ag1Sn2/CNT	Pd7Ag2Sn2/CNT	Pd7Ag3Sn2/CNT
  ECSA/	H2SO4	5.22	4.60	6.39	5.99	7.28	9.56	7.24
  (m2·g-1)	NaOH	11.13	9.73	8.75	11.83	9.96	15.34	11.08

  下载: 导出CSV
•