时间:2024-09-03
Gu Xianrui; Wu Yuchao; Wang Houpeng; Rong Junfeng
(SINOPEC Research Institute of Petroleum Processing, Beijing 100083)
Abstract: Fuel cells are considered to be one of the ideal alternatives to traditional fossil energy conversion devices.Membrane electrodes are the core components in the hydrogen fuel cells. Our work reported the synthesis of the Pt/C catalysts with different Pt loading, and by changing the Nafion content, hot pressing temperature and hot pressing pressure,the catalyst coated membrane (CCM) spraying process was optimized. Moreover, the three-dimensional structure model of the single battery membrane electrode was studied quantitatively, and the porous membrane electrode with gradient distribution was fabricated under optimized processing conditions, with excellent electrical performance.
Key words: hydrogen fuel cell; membrane electrode; Pt/C catalyst; polarization curve; power density; single cell test
Hydrogen fuel cell is an electrochemical device that converts chemical energy stored in hydrogen (H2) into electricity and heat[1-2]. There are different kinds of fuel cells, and the most common and practical one is the proton exchange membrane fuel cell (PEMFC)[3-4].PEMFC is applied in the field of automobiles on account of its high efficiency[5], low working temperature and zero emission[6]. The membrane electrode is the core component of the hydrogen fuel cell[7], where the electrochemical reaction occurs, and the electrons and protons are formed[8]. This process involves the mass transfer, heat transfer, and momentum transfer of each component, and thus the membrane electrode is the“reactor” for chemical reactions[9]. Therefore, the design and optimization of membrane electrodes not only are the key topic in scientific area, but also can provide an important technical support for stack industrialization[10].The working mechanism of the Pt/C catalyst in the proton membrane is complicated, with the process technology involving mass transfer, adsorption, electrochemical reaction, and desorption. Gas adsorption, electron reception, proton conduction and other three-phase processes occur on the anode side of catalyst surface[11].The cathode side is more complicated, including gas adsorption, electron acceptance, proton acceptance,water generation, and other four-phase processes. As a consequence, the process for fabrication of membrane electrodes is very critical. For the same batch of catalysts,different fabrication processes, such as spraying, scraping or transfer printing could lead to a large discrepancy in the performance of membrane electrodes. The common process of fabricating membrane electrode is the catalyst coated membrane (CCM) method[12], which sprays the Pt/C catalyst slurry evenly on the surface of proton membrane or the carbon paper[9]. Therefore, the membrane electrode may suffer from the non-uniform oxygen distribution due to limited oxygen diffusion, which would affect the Pt mass activity of the membrane electrode[13].
This work focuses on the optimization of the CCM spraying process. For that purpose, the Pt/C catalysts were prepared with different Pt loading (20%, 40%, and 60%),and the samples were used for the optimization of process parameters and spraying of gradient film electrodes for the preparation of single battery.
Vulcan XC72 carbon (600 mg) was dispersed in deionized water and was then placed in a 500 mL round-bottomed flask. The H2PtCl6solution with a concentration of 2.05×10-3mol/L was added into the round-bottomed flask and was ultrasonicated for 20 min. The pH value of the above solution was adjusted to 8—14. The temperature of the solution was raised to 60—80 ℃ under stirring at a speed of 300 rad/min. A certain amount of the reducing agent and the stabilizer was dispersed in 5 mL of deionized water, and then was slowly added into the dispersion. The reaction time was four hours. The above dispersion was filtered to obtain the filter cake. The above filter cake was washed with deionized water for 3 times with ultrasonication. The filter cake was then dried (at 100 ℃ for 8 h) to obtain a powdered solid catalyst.
8 mg of Pt/C catalyst were placed in a glass bottle, with addition of water and isopropanol. Then a certain amount of Nafion solution (5%) was added into the above galss bottle. Finally, isopropanol was added with ultrasonication for 30 min.
The spray gun was cleaned with anhydrous ethanol,with the sprayed quantity being adjusted. The catalyst mixture was sealed and wrapped in the spray gun solution cylinder. The catalyst solution was sprayed onto the proton membrane slowly and uniformly at a constant injection speed. The injection time was about 0.5 hours,and then the anode was weighed. For the cathode, the same procedures were repeated.
In this work, the membrane electrode was fabricated by the hot pressing process. The specific process parameters covered: a hot pressing pressure of 5 kg/cm2, a temperature of 120 ℃, and a hot pressing time of 2 min.
The Pt mass on the anode of the fuel cell is 0.1 mg/cm2,and the Pt mass of the cathode is 0.2 mg/cm2. The type of the diffusion layer used in the cell is SGL 28BC, equipped with a #211 proton exchange membrane from the DuPont Company. The size of a single cell membrane electrode is 1 cm×1 cm. When the single cell is tested, the inlet gas of anode is hydrogen (99.99% pure), while the inlet gas of the cathode is oxygen (99.99% pure). For this system, the back pressure is 0.2 MPa, and the gas flow rates for both the anode and the cathode are 50 mL/min.
The Pt/C catalyst is the widely used catalyst in hydrogen fuel cells, which can influence the performance of membrane electrodes directly. We prepared three types of Pt/C catalysts with different loadings (20%, 40%, and 60% of Pt loading) by using the liquid-phase reduction method. The size distribution of Pt particles in Pt/C samples with different Pt loadings may vary greatly,leading to a significant change in the mass activity of Pt,which can further influence the efficiency of the gradient film electrodes. Judging from the transmission electron microscope (TEM) images, it was found that the average particle sizes of the three types of catalysts (with 20%,40%, and 60% of Pt loading) were 4.6 nm, 5.1 nm, and 6.0 nm, respectively (Figure 1a-c), with an uniform particle size distribution and no obvious agglomeration.We have also measured the X-ray diffraction (XRD)patterns of the three Pt/C samples (Figure 1d), and all of them showed single elemental Pt XRD diffraction signals[14-16]. The X-ray photoelectron spectroscopy (XPS)characterization results (Figure 1e) are consistent with the XRD analysis, where the Pt binding energy of all three catalysts was 71.8 eV, which is consistent with the results of metallic Pt mentioned in the literature[17-20]. The results of redox reaction (ORR) performance tests showed that the half-slope potential of the catalysts with a Pt loading of 20%, 40%, and 60% was close to 0.90V, 0.90V, and 0.89V, respectively. In comparison, we had also tested a commercial catalyst (JM’s commercial catalyst with a Pt loading of 20%) and its half-slope potential was 0.88V(Table 1). The mass specific activity of the catalyst with a Pt loading of 20%, 40%, and 60% is 0.18 m2/g-Pt,0.16 m2/g-Pt, and 0.11 m2/g-Pt, respectively. According to the TEM characterization and the ECSA results, when the Pt loading amount increases, the average particle size of the Pt NPs in the catalyst inevitably becomes larger, which can also leads to a decrease in mass specific activity. Notably, the decrease of the specific activity is not severe, notwithstanding the Pt loading of samples varies by three times between 20% and 60%. Therefore,we consider that these samples can be used for spraying the gradient film electrode.
The performance of the membrane electrode could be influenced by several factors, such as the amount of Nafion in syrup materials, and temperature and pressure of heat pressing during the fabrication of membrane electrodes. These optimized variables from the Pt/C catalyst (with a Pt loading of 40%) could be applied to the technology of gradient membrane electrode fabrication. Firstly, the percentage of Nafion in the syrup materials was studied in the range of 20% to 50% under the standard membrane electrode fabrication process. As shown in Figure 2a and 2b, the power density and output voltage of single battery reaches a highest performance when Nafion content is 38%. In terms of the heat pressing temperature, Figure 2c and 2d reveal that the output performance (power density and output voltage) of single battery reaches the optimal condition when the heat pressing temperature is 120 °C. Further study was carried out where the single battery was prepared from a different heat pressing pressure of 4 kg/cm2, 5 kg/cm2, and 6 kg/cm2, respectively. The output performance (power density and output voltage) of a single battery reached its optimal condition when the heat pressing temperature is 5 kg/cm2(Figure 2e and 2f). We will also combine thepolarization curve of the single battery when reviewing the membrane electrode performance, calculating the current consumption, exchanging the current density,limiting the current density and interior resistance used on the model as well as the linear regression equation,which are important parameters in membrane electrode modulation.
Table 1 ORR performance tests of JM commercial Pt/C, and catalyst samples with 20% — 60% of Pt loadings
Figure 1 Particle size distribution (a,b,c), XPS characterization (d) of Pt 4f, XRD characterization (e), and LSV test resultsof catalysts with a Pt loading between 20%—60% (f) under acidic conditions
Figure 2 Power density curves (a, b and c) and polarization curves (b, d and f) of single battery with varied Nafion content,heat pressing temperature and pressure(a and b) ■—20 wt.% Nafion; ●—26 wt.% Nafion; ▲—30 wt.% Nafion; ▼—38 wt.% Nafion; ◄—50 wt.% Nafion
There are three gas components on the membrane electrode in a working single battery, which are nitrogen, oxygen and water[21]. Among these factors, the concentration of oxygen would directly affect the catalytic efficiency at cathode, thereby determining the power of single battery[22]. When the gas concentration increases,the Fick diffusion theory could no longer be applied in the mass transfer of three components, while the Maxwell-Stefan equation can apply under the circumstance that the oxygen concentration is not well distributed on the surface of membrane electrode.Figure 3 shows a 3-dimensional (3D) image of the fuel cell cathode, which is built for the study of mass transfer, current distribution, and reaction distribution within the membrane electrode[23]. The circle in Figure 3a represents the electrode as it is porous. The upper layer is cathode catalyst layer, and the bottom layer is the free electrolyte proton film. The air and water enter the cathode and trigger the electrochemical reaction through that circle. The marked area from the cathode surface in Figure 3b is the entrance for gas inlet, the rest area is parallel to the collection fluid body. This single battery (Figure 3c) composed of a porous cathode (top layer) and free electrolytes (bottom layer) is divided by tetrahedron grid with a total of 5498 deltas and 18941 tetrahedrons[24]. Figure 3d shows the oxygen distribution in the catalyst layer at 0.6V over potential (standard load in single battery test). Notably, there is a huge difference in oxygen concentration at the xy plane, and the oxygen concentration is reduced to 0 in very small distance on the xy plane.
Furthermore, both Figure 3e and 3f can verify this phenomenon. The partial over potential is evenly distributed while the oxygen concentration is not,implying an uneven reaction speed in the reaction layer (Figure 3e). In Figure 3f, there is a significant difference of ion flow density distribution at the bottom boundary of electrolyte. Therefore, the oxygen concentration is the highest one surrounding the proton film holes, which gives a maximum reaction speed, while the oxygen concentration decreases rapidly far apart from the hole area, giving the lowest reaction speed. Meanwhile, the oxygen concentration also displays a gradient changes at the Z-axis direction of the film. To solve this problem, we not only applied the concentration gradient coating in the catalyst preparation but also introduced porous structures to further distribute the oxygen, thereby reducing the potency difference polarization.
Based on result of the calculation, we introduced volatile additives into the syrup materials during the fabrication of membrane electrode, which could be decomposed into gas at high temperature, leaving small holes in the catalyst layer during heating.
Figure 3 Structure of (a) porous membrane electrode; (b) membrane electrode; (c) divided grids of membrane electrode; (d)the mass distribution of oxygen on the porous electrode; (e) partial over potential; (f) current density of electrolyte
The test result of the prepared membrane electrode is shown in Figure 4a. The power density of the membrane electrode increases under the same current as compared to reference sample after pores creating, and so does the output voltage. Upon further comparing with the reference sample (Figure 4b and Table 2), the porous membrane electrode shows a greater circulating current density, implying an increased number of active sites in Pt/C catalyst owing to the pore creating. In addition, the oxygen is more well-distributed, reducing the barrier of oxygen transfer, which can cause the augmentation of diffusive electric density of the working membrane electrode. In the mean time, based on the calculation, the electric potential, current density and oxygen from the diffuse layer to the proton film is not well-distributed,and displaysing a gradient change during the fuel battery working process. To be more specific, the closer to the proton film, the higher the over potential and current density can be achieved with fewer oxygen. Therefore,the gradient membrane electrode is designed with a higher mass ratio of catalysts coated near the proton film for generating more Pt active sites, thereby compensating the reduction of dynamic velocity due to the reduced oxygen supply. And at the far end of the proton film, less catalyst is coated with less Pt active sites, which would reduce the dynamic velocity due to the excess oxygen. The result of designed single battery is shown in Figure 4a, the output power of gradient membrane electrode (0.1 mg of Pt for anode + 0.2 mg of Pt for cathode) is higher than the reference membrane electrode, and exchange current density is also higher than the reference membrane electrode (Figure 4b and Table 2).
Figure 4 The density curve (a) and polarization curve (b) of the single battery with high porosity and gradient coating; (c)Scheme of optimized membrane electrode; (d) Density curve and polarization curve of optimized membrane electrode(a and b) ■—gradient; ●—porous; ▲—normal
Table 2 Parameters of normal, porous, and gradient single battery polarization curve equation
After the optimization of the above-mentioned technology parameters to the pre-synthesized 20%, 40% and 60%loading of Pt/C catalyst, the membrane electrode was constructed with (Figure 4c) high porosity and gradient coating. The maximum power peak value is about 1.41W,with a total loading of 0.25 mg of Pt and the power peak value can reach 5.64 kW/g of Pt. In addition, the current of the single battery can reach 2.15 A under a working voltage of 0.60 V, the catalyst out-performances the industry standard and can be further applied in electric pile assembling.
We reported the influence of CCM technology on the single battery performance of hydrogen fuel cell based on the pre-synthesized Pt/C catalysts with different Pt loading.
(1) The technology parameters of membrane electrodes including the percentage of Nafion, heat pressing temperature and pressure have been optimized to reach a power density and output voltage.
(2) The 3D structural model of single battery membrane electrode is built. Furthermore, the mass distribution,electric charge distribution, and momentum distribution of oxygen on the membrane electrode are established for better understanding of the interior working status of battery.
(3) The porous membrane electrode is fabricated using gradient coating, and the assembled single battery showed better performance than the reference battery. The results shown in this work provides a guidance on batch production of membrane electrodes in order to assemble higher performance electric pile.
Acknowledgement: This work was financially supported by China Petrochemical Corporation (ST 20006-1, ST 20006-2).
我们致力于保护作者版权,注重分享,被刊用文章因无法核实真实出处,未能及时与作者取得联系,或有版权异议的,请联系管理员,我们会立即处理! 部分文章是来自各大过期杂志,内容仅供学习参考,不准确地方联系删除处理!