1500hr耐久性测试增湿器串漏造成燃料电池电解质膜失效，探测针孔的泄露法、红外热成像和电压弛豫法对比和局限性
Understanding Membrane Failure in PEMFC:Comparison of Diagnostic Tools at Different Observation Scales
G. De Moor
C. Bas
N. Charvin
E. Moukheiber
F. Niepceron
N. Breilly
J. André
E. Rossinot
E. Claude
N. D. Albérola
L. Flandin
Abstract
Polyelectrolyte membrane used in protonexchange membrane fuel cell (PEMFC) should act as a barrier to hydrogen andelectrons. In the present work, fast and robust specific tools were developedboth to characterize these properties and localize defects, such as pinholes orshort-circuits, within the membrane electrode assembly (MEA). Theidentification of defective cells without disassembling the Axane's stack wasperformed in a single step through a passive electrical technique so-called “relaxometry”and by internal gas leak measurements. The flaws were further trackedinside MEA using Infrared Thermography in order to obtain their preciselocation. The latter, performed after stack disassembly, permitted to revealthe defects responsible for the Fuel Cell shutdown. A third step dealt with SEMexaminations of the flaws in order to identify the origin and nature of themembrane failure.
Fig. 1 Current profile of the stack duringthe 1,500 h of service.
Fig. 2 Schematic representation of therelaxometry setup for a synchronous measurement of the entire stack.
Table 1 Flaw size classification forthermal camera measurement.
Fig. 3 Overview of the cells identified asdefective by internal leak test within the stack at Fuel Cell end of life.
The stack design is built with external gascollector individually removable. A portable device developed by Axane isplugged to the external collector of each individual cell and 250 mbar isapplied at anode or cathode side. The pressure drop is then measured over time(30 s).
这个似乎是Axane特有的，在他们公司网站上我没有看到了理解这个过程是如何操作的，如何在不拆解电堆能够对每一节单独供气进行串漏测量。对某一节拆解下来测漏也是比较繁琐的一件事。
1Good, the cell does not show any pressuredrop after 30 s;
2Intermediate, the cell reveals a slightpressure drop not critical for the application (few mbar); 3Defective, thepressure drop is complete over a short time which would mean a high gascrossover through membrane.
Fig. 4 Phase 2 voltage discharge curve as afunction of time: definition of 3 possible states (Good, Intermediate, andDefective).
Fig. 5 Summary of the defective MEA asrevealed by simultaneous relaxometry on entire stack at Fuel Cell end of life(a: Stack 1; b: Stack 2).
Fig. 6 Identification of the minimumpinhole size detectable by relaxometry.
这件事情做得很严谨，通过对膜打孔，判断电压弛豫法探测针孔的检测限，200微米。
Fig. 7 Distribution of pinhole sizes asdetected by Infrared Thermography at Fuel Cell end of life (a: Stack 1; b:Stack 2).
Fig. 8. Superposition of the picturesrepresenting the flaws detected within MEA in a given stack at Fuel Cell end oflife (top and bottom represent Stack 1 and 2, respectively).
Measurements on fresh MEA (not shown)revealed that flaws of size 1 and 2 may be originally present beforeassembling. It was therefore not possible to conclude on the amount of smallflaws that appeared during ageing.
这里应该显示一下新鲜电极的测量结果，装配前就有针孔的尺寸、数量直方图统计、位置和耐久性之后的对比，1500hr后针孔没有统计就失去了跟踪的意义。判断多大的针孔是在装配初期是允许的。
the flaws with sizes 4 and 5, responsiblefor the shutdown, were all located in the air outlet region of the cell.
电堆1失效的原因：增湿器失效欠空气，Thesystem was shut down for safety reason when auxiliary humidifier failed onStack 1 and a gas leakage was detected.
意思是系统零部件失效过程造成了4、5级的针孔出现，可能由于氢空压差过大。
Fig. 9 Calibration of IR thermographytechnique by measure of pinhole with various sizes down to 75 um.
红外热成像的检测下限75微米。
When applied to real system, this techniquemay not be suitable to identify the possible formation of pinhole or cracks closeto one another, mainly due to the good thermal conductivity of the GDL.
Fig. 10 Localization of the defective cellswithin the stack as revealed by three macroscopic techniques leak test,relaxometry, and infrared thermal camera.
这个图非常有趣。
For H2 leak test, flaw size 1, 2, and 3were considered as intermediate cells.
A main part of the stack was degraded,typically at the bottom side of the stack.，也就是Polar+
就像核酸检测一样，电压弛豫法会出现假阳性和假阴性，会影响这种方法的应用。
这三种方法的结果也不是一一对应的。泄露法检测出来的失效节在红外热成像中分级可能会不同，因为泄露法是一个总包性的结果，可能多点均匀泄露。
Fig. 11 Comparison of the sensitivity of IRThermography versus relaxometry for the two (55 cells) stacks studied(Configuration 1: defects in air outlet region solely; Configuration 2: defectsin the whole MEA except in air outlet; Configuration 3: defects both in airoutlet and in the rest of the MEA).
Fig. 12 Basic reason for stack failure:cracks in membrane identified with the help of the three techniques
Fig. 13 Influence of external parameters onfalse negative obtained with relaxometry: measured voltage after 500 s in phase2 in various operating conditions.
Fig. 14 Schematic representation of falsenegative samples in the relaxometry measurements.
Table 2 Comparison between the techniquesto characterize flaws in the MEA.
