金属双极板燃料电池短堆2000hr耐久性测试和失效分析

The durability investigation of a 10－cell metal bipolar plate proton exchange membrane fuel cell stack

Kailin Fu

Tian Tian

Yanan Chen

Shang Li

Chao Cai

Yu Zhang

Wei Guo

Mu Pan

Summary

The metal bipolar plates （BPs） have replaced the graphite BPs in vehicle－used proton exchange membrane fuel cell （PEMFC） stack because of their high volume power density． To investigate the durability of metal BP stack， this paper performed a durability test of 2000 hours on a 10－cell metal BP fuel cell stack． The degradations of the average voltage and individual cell voltage in fuel cell stack were analyzed． To investigate the degradation mechanism， the stack was disassembled and the morphologies and compositions of no． 1， no． 5， and no． 10 cells after 2000 hours were characterized by SEM， TEM， and ASS（atomic absorption spectrometer）． The results indicated that at 800 mA／cm2， the voltage decay rate is 42．303 μV／hour and the voltage decay percentage of the stack is 14．34％ after 2000 hours according to the linearly fitting result． According to the US Department of Energy （DOE） definition of fuel cell stack life， only the voltage decay rate of OCV and the tenth cell is lower than the maximum voltage degradation rates of 10 000 hours． The decreases of homogeneity of stack were the main reason for its performance degradation． Especially for the tenth cell， its performance has almost no drop． The main failure reason of this metal BP stack is structural design rather than metal corrosion． The losses of Pt catalyst and C supporting are the main reason of performance degradation．

FIGURE 1 The test protocol of one dynamic driving cycle for the PEMFC stack （rated power （1 kW） was normalized to 1）

FIGURE 2 Polarization curves of fresh stack and that measured after 2000 hours， the test condition was the same as that of durability test listed in Table 1

FIGURE 3 Changes of the stack average voltage at different current density

TABLE 1 The durability test conditions of PEMFC stack

TABLE 2 The decay rates and decay percentages of voltage under
different current densities

FIGURE 4 Schematic diagram of the fuel cell stack

TABLE 3 The voltage decay rate of projected hours

FIGURE 5 The voltages of individual cell before and after 2000 hours， at 800 mA／cm2 current density

表现出进出气口更容易衰减。这个现象非常典型。

TABLE 4 The voltage decay rate （μV／hour） of individual cell after 2000‐hour testing

FIGURE 6 The performance decay of no． 1， no． 5， and no． 10 cell at different current densities

FIGURE 7 The surface of metal plate after 2000‐hour testing

FIGURE 8 SEM cross‐sectional micrograph of A， initial cell， B， no． 1 cell， and C， no． 5 cell， D， no． 10 cell after 2000‐hour testing

FIGURE 9 TEM image of A， initial CL， B， cathode and C， anode CL of no． 1 cell after 2000 hours， D， cathode and E， anode CL of no． 5 cell
after 2000 h， and F， cathode and G， anode CL of no． 10 cell after 2000 hours

TABLE 5 Parameter of the TEM and ASS of the MEA before and
after 2000‐hour test

In the cathode catalyst layer， much bigger pores and highly textured structure are formed in no． 1 and no． 5 cells than no． 10 cell． This is the main reason for the increase of ohmic resistance． All the above indicated that the no． 1 and no． 5 cells experienced more severe carbon corrosion than no． 10 cell．

这个工况怠速时长较长，但是电位并不高，因为并没有严格意义上的启停氢氧截面，为什么会出现碳腐蚀？

The proportion of idle， rated power， and high load power running time to total cycle time is 60％， 30％， and 10％， respectively

氢气侧的催化层厚度逐渐下降。

The anode catalyst layers of no． 1 and no． 5 cells are thinner and looser
than that of no． 10 cell．

文章中图9和表5不对应，图上看B是第1节的阴极，铂粒子出现长大。而表5第10节的阴极，铂粒子出现长大。文章对粒子大小只对阳极和阴极做了比较。我感觉是表5错了。

表5中第1节铂催化剂的总量流失最明显。

CONCLUSION

A 2000‐hour durability test was achieved on a 10‐cell metal plate stack under actual running conditions of fuel cell vehicles in urban areas． The Pt loading of whole MEA was 0．4 mg／cm2， and the area of each electrode was approximately 276 cm2． At 800 mA／cm2， the voltage decay rate is 42．303 μV／hour， and the voltage decay percentage of the stack is 14．34％ after 2000 hours according to the linearly fitting result． According to the DOE＇s definition of fuel cell stack life， only the voltage decay rate of OCV and the tenth cell is lower than the maximum voltage degradation rates of 10 000 hours． The voltage decay rate of no． 4， no． 6， no． 7， no． 8， and no． 9 cells is in the projected hour of 5000 hours at 150 mA／cm2． The decreases of homogeneity of stack were the main reason for its performance degradation． Especially for the tenth cell， its performance has almost no drop． The main failure reason of this metal bipolar plate stack is its structure design， which is the same as graphite bipolar plate stacks．
In addition， after 2000‐hour testing， there is no obvious corrosion on the surface of the metal plate． The characteristics of the MEA illuminated that the losses of Pt catalyst and C supporting are the main reason of performance degradation．

常规说的Pt催化剂的衰减是指ECSA的下降，出现这么严重的铂流失很难理解。是否观测到铂流失到什么位置？膜电极上重新分布还是流失到堆外？

第一节不同区域是否存在着性能不同和失效现象的不同？

第一节的问题是阳极的问题还是阴极的问题？因为氢气侧的催化层厚度逐渐下降。

Therefore， the main failure reason of this metal bipolar plate stack is its structure design， which is the same as graphite bipolar plate stacks．

具体是什么设计问题，这里都没有介绍。

The durability test of 2000 hours was performed for more than 8 months without trouble．

没有更详细的log，失效也可能和实际操作有关，这个耐久性拖得时间比较长。