Test methods for the quality assurance during the production of PEM fuel cells

Contents

Zusammenfassung

Abstract

Contents

Illustration directory

Table directory

1. Fundamentals of the fuel cell technology
1.1 History of fuel cells
1.2 Applications
1.3 Types
1.4 Principals
1.5 Structure and components of a PEMFC
1.5.1 Proton exchange membrane
1.5.2 Cathode/Anode catalyst layer
1.5.3 Gas diffusion layer
1.5.4 Bipolar plates
1.5.5 Front/End plates

2. Control Plan
2.1 Methodology
2.2 FMEA
2.3 Control parameters
2.3.1 Thickness variation of the MEA
2.3.2 Electrolyte cluster
2.3.3 Delamination
2.3.4 Catalyst cluster
2.3.5 Humidification
2.3.6 Cracking
2.3.7 Flow field structure in the bipolar plates
2.3.8 Tightness of the fuel cell
2.3.9 Temperature
2.3.10 Impedance

3. Control methods and plan
3.1 Cluster and cracks detection in the catalyst layer
3.1.1 IR thermography
3.1.2 Cluster detection with scanning electron microscope (SEM)
3.2 Flow field structure control of the bipolar plates
3.2.1 Image processing
3.3 Tightness control
3.4 Delamination control
3.5 Impedance control
3.5.1 High frequency resistance method
3.5.2 Current interrupt method
3.6 Membrane thickness control

4. Measurement uncertainties
4.1 Thickness measurement uncertainties
4.2 Impedance measurement uncertainties
4.3 Temperature measurement uncertainties

5. Conclusion and outlook
5.1 Conclusion
5.2 Outlook

Bibliography

Zusammenfassung

Mit den steigenden Energiekosten, dem zunehmenden Umweltbewusstsein und der Klimaerwärmung gewinnen die alternativen Antriebe immer mehr an Bedeutung. Dabei kann die Brennstoffzellentechnik als saubere und zuverlässige Methode der Energieerzeugung einen großen Beitrag leisten.

Zahlreiche Automobilkonzerne wie Hyundai, Daimler und Toyota arbeiten intensiv an serientaugliche Fahrzeuge mit Brennstoffzellenantrieb. Toyota hat es im Dezember 2014 gelungen, das erste serienmäßige Auto mit Wasserstoffantrieb, das Toyota Mirai, auf den Markt zu bringen und sie rechnet in den 2020er Jahren bereits mit zehntausenden Fahrzeuge jährlich[1]. Daher werden steigende Absatzzahlen erwartet, wodurch der Bedarf, einhergehend mit einer Massenproduktion von Brennstoffzellen, ansteigen wird.

Die Brennstoffzellentechnik basiert sich auf dem Prinzip der Elektrolyse. Um Strom zu erzeugen werden Wasserstoff und Sauerstoff der Brennstoffzelle zugeführt, wo sie zu Strom, Wasser und Wärme reagieren. Daher haben Autos mit Brennstoffzellenantrieb viele Vorteile gegenüber Autos mit herkömmlichen Energiequellen, da sie umweltfreundlich sind und die langen Ladezeiten wie bei Elektroautos ersparen.

Die heutige Produktionsart erfüllt jedoch noch nicht die Anforderungen einer Massenproduktion. Bis heute werden die Komponenten der Brennstoffzelle oft ohne ausreichende Prüfung zu einem Stack zusammengebaut, was das Fehlerpotenzial deutlich steigert. Nach dem Zusammenbau wird ein End-of- Line Test durchgeführt, bei dem die Brennstoffzelle und ihre Funktionen bis zu 24 Stunden geprüft werden. Um höhere Produktionsmenge zu erreichen, muss die Dauer des End-of-line Tests auf wenigen Minuten reduziert werden. Dies erfordert ein vollumfängliches Qualitätsmanagement und eine fertigungsbegleitende Qualitätssicherung, um eine schnelle und fehlerfreie Produktion zu erreichen. Das Ziel dieser Arbeit ist die Verkürzung der Dauer des End-of-Line Tests durch die Erarbeitung eines montagebegleitenden Prüfplans.

Der erste Teil beschäftigt sich mit den Grundlagen der Brennstoffzellentechnik. Dabei werden die wichtigsten historischen Ereignisse der Entwicklung der Brennstoffzelle erwähnt. Um ihre Funktionsweise zu verstehen, werden die chemischen und physikalischen Grundlagen dargelegt und die Funktionen der einzelnen Komponenten beschrieben.

Im zweiten Teil wird eine Fehlermöglichkeits- und -einflussanalyse (FMEA) basierend auf einer Literaturrecherche durchgeführt, um die wichtigsten Fehler zu bestimmen. Daraus leiten sich die Prüfmerkmale ab, worauf bei der Fertigung der Brennstoffzelle geprüft werden muss.

Im dritten Teil der Arbeit werden geeignete Prüfmethoden zu den ausgewählten Prüfmerkmalen dargestellt. Bei der Auswahl der Kontrollmethoden werden vor allem die Dauer der Prüfung und die Eignung zur Massenfertigung beachtet. Dabei werden verschiedene Technologien zum Stand der Technik angewendet.

Zum Schluss werden die wichtigsten Messunsicherheiten dargestellt. Bei den Prüfmethoden, wo die Messunsicherheiten eine Wirkung auf die Fertigung haben, wird der Zusammenhang mit den Messtoleranzen dargestellt. Dies ermöglicht die Auswahl und die Bestimmung der Auflösung der verwendeten Messgeräte.

Abstract

With the rising cost of energy, the increasing environmental awareness and the global warming, alternative drive systems are becoming increasingly important. Here, the fuel cell technology makes as clean and reliable method of energy production a great contribution.

Many automotive companies like Hyundai, Daimler and Toyota are working intensively on fuel cell vehicles. Toyota has succeeded in December 2014 to bring the first car powered by hydrogen, the Toyota Mirai, on the market and it expects in the 2020s already with tens of thousands of vehicles annually[1]. Therefore, the sales numbers will be increasing, which will require a mass production of fuel cells.

Fuel cell technology is based on the principle of electrolysis. In order to generate electricity, hydrogen and oxygen are supplied to the fuel cell, where they react to electricity, water and heat. Thus cars powered by fuel cells have many advantages over cars with conventional sources, because they are environmentally friendly and save the long load time characterizing electric cars.

However, the current manufacturing way does not meet the requirements of mass production. To date, the components of the fuel cell are often assembled to a stack without sufficient testing, which significantly increases the potential for errors. After assembly, an end-of-line test is performed, wherein the fuel cell and its functions are checked up for up to 24 hours. To achieve higher production volume, the duration of the end-of-line tests must be reduced to a few minutes. This requires a total quality management and quality assurance during production in order to achieve a quick and error-free one. The aim of this work is to shorten the duration of the end-of-line test by the development of a montage accompanying control plan.

The first part deals with the basics of the fuel cell technology. The most important historical events of the development of fuel cells are mentioned. In order to understand their functioning, the chemical and physical principles are explained and the functions of the individual components are described.

In the second part, a Failure mode and effect analysis (FMEA) is performed. Based on a literature review, this analysis aims at determining the most important errors in the components of the fuel cell. Its results deduce the most necessary test characteristics to be checked during the manufacturing of the fuel cell.

In the third part of the thesis, suitable test methods for the selected inspection characteristics are presented. In the selection of control methods, especially the duration of the test and the suitability for a mass production are observed. In the control plan, various technologies to the state of the art are employed.

Finally, the most relevant measurement uncertainties are explained. In the test methods, where the measurement uncertainties have an effect on the production, its relationship to the measurement tolerances is represented. This allows the selection and determination of the resolution of the employed measuring instruments.

Illustration directory

Figure 1: Diagram of a PEM fuel cell[7]

Figure 2: Fuel cell V-I curve[8]

Figure 3: Schematic of a MEA[2]

Figure 4: Fuel cell stack[2]

Figure 5: Chemical structures of Nafion[12]

Figure 6: Schematic diagram of a fuel cell electrode[18]

Figure 7: Micrographs (scanning electron microscope (SEM)) of three different GDL supports (a: cloth, b: felt, c: paper)

Figure 8: examples of bipolar plates with identical flow field (a: non-layered bipolar plate; b: gold-coated bipolar plate; c: milled graphite-composite bipolar plate[14]

Figure 9: Fuel cell stack schematic[22]

Figure 10: Rule of 10[24]

Figure 11: Thickness variations in the CCM[27]

Figure 12: Scanning electron micrograph of Nafion cluster (highlighted circle)[27]

Figure 13: Delamination between electrolyte and catalyst layer in the MEA[27]

Figure 14: Catalyst cluster in a CCM: (a) × 20 000; (b) × 100

Figure 15: Scanning electron micrograph of cracking in catalyst layer of Ion Power 522B[27]

Figure 16: Flow field structure in the bipolar plates[30]

Figure 17: Experimental configuration for line flow experiment using a slotted gasket

Figure 18: (a) Schematic of the RFT (reactive flow-through) experimental setup with computational domain marked by dashed line. (b) Full computational domain[31]

Figure 19: Steady state heat signatures at flowrates of (a) 500 sccm, (b) 1000 sccm, and (c) 1500 sccm

Figure 25: Principal of the dark field microscopy[33]

Figure 26: a capture of the flow field structure in the bipolar plates using the dark field microscopy[34]

Figure 27: Laser profilometer[35]

Figure 28: Laser triangulation of the flow field structure in the bipolar plates[34]

Figure 29: Flow field structure with defects

Figure 30: measurement lines[34]

Figure 31: Integration of the dark field microscopy in the control station (1)[34]

Figure 32: Integration of the dark field microscopy in the control station (2)[34]

Figure 33: Tightness control station

Figure 34: Principal of the EMAT[36]

Figure 35: signal variation after the detections of defects[36]

Figure 36: Amplitude of through transmission guided wave signal when the sensors are scanned through the sample with tapered defect[37]

Figure 37: An in-line inspection system for composite strip inspection using guided waves: (a) inspection system and (b) EMAT sensor assembly[37]

Figure 38: Simplified, idealized equivalent circuit of a PEMFC[38]

Figure 39: Equipment set-up for HFR measurement techniques[38]

Figure 40: Idealized voltage waveform during current interrupt event[38]

Figure 41: Diagram of transmittance of light through a cuvette[39]

Figure 42: a schematic diagram of a system for measuring thickness using transmission[40]

Figure 43: a diagram of the specular transmission spectra of two Nafion® membranes of the following thicknesses: (a) 1 mils, (b) 2 mils and (c) 4 mils[40]

Table directory

Table 1: Currently developed types of fuel cells, their characteristics and applications[2]

Table 2: Rating scale for risk assessment[25]

Table 3: FMEA of a fuel cell[26][27]

Table 4: Conductivity at different relative humidities [%] for E- and N-form Nafion 117 membranes (E- form: no heat-treatment, N-form: heat treatment at 85°C and 105°C)[28]

1. Fundamentals of the fuel cell technology

A fuel cell is a power source that converts chemical energy into electrical energy as a result of chemical reaction between fuel and oxygen. It is mainly composed of two electrodes around a layer of material called the electrolytes.

Fuel cells have similarities with batteries sharing the electrochemical nature of power generation and with engines continuously consuming fuel while operating. The difference to both batteries and engines is that the fuel cell does not need to be recharged, works quietly and efficiently and generates only power and water[2].

1.1 History of fuel cells

The basic principle of the fuel cell was discovered by Christian Friedrich Schönbein (1799-1868), a German scientist, in 1840 and then the first experiment was realized by the British scientist Sir William R. Grove (1811-1896). Unfortunately the fuel cell was forgotten for many years until the late 1950s, when the National Aeronautics and Space Administration (NASA) developed it for manned space missions. One result of this development was the proton exchange membrane fuel cell (PEMFC) that was used later in the mid-1960s in the Gemini space program. The International Fuel Cells (IFC, later UTC Power) developed a 1.5 kW alkaline fuel cell (AFC) to be used in the Apollo space missions. It provided electrical power and drinking water for the astronauts.

Increasing environmental awareness, concerns over air pollution and OPEC oil embargoes, led governments, businesses and individuals in the 1970s to rethink their energy politic and to search for alternative energy sources. Then several German, Japanese and US vehicle manufacturers began experimenting with fuel cell electric vehicle and developing the PEMFC, in order to increase its power density and to develop hydrogen fuel storage systems. In the 1990s, carmakers such as DaimlerChrysler, General Motors, and Toyota continued investing in alternative powertrains especially in PEMFC. Companies other than carmakers, such as Ballard, one of the important suppliers of PEMFC units, conducted research on PEMFC for automotive and stationary clean power.

In the last decade, there has been a renewed focus on fundamental research to reduce the production costs of fuel cells and to improve their performance to make them competitive with conventional technology. The European Union, Canada, Japan, South Korea and the United States have funded pilot projects, primarily of stationary and transport fuel cells and their associated fuelling infrastructure. For example, tens of fuel cell buses were deployed in the mid-2000s as part of the HyFleet/CUTE project in Europe, China and Australia.

In the last five years, the supply chain has been steadily growing and the number of fuel cell system manufacturers has been increasing[3].

In the last years, Mercedes-Benz has been testing its F-Cell fuel cell vehicle and plans, with other carmakers such as Ford and Nissan, the introduction of fuel cell cars in 2017[4].

1.2 Applications

The fuel cells are used in both stationary and mobile field. In the stationary field we can differentiate between the large stationary, such as hospitals, wastewater treatment plants, government buildings, and jails, and the small stationary, such as telecommunications, residential, and small commercial buildings.In the mobile field, the fuel cells are used in small power suppliers, such as in laptops, mobile phones, and battery chargers, and in the transportation sector, such as passenger vehicles, buses, and campers[5].

1.3 Types

illustration not visible in this excerpt

Table 1: Currently developed types of fuel cells, their characteristics and applications[2]

1.4 Principals

The PEMFC is the most used fuel cell in cars and it has the more promising technology, because of its simplicity, high power density and long lifetime. Its name is derived from the polymer membrane, the proton exchange membrane[2][6].

The PEMFC consists mainly of 3 active components: a fuel electrode (anode), an oxidant electrode (cathode), and an electrolyte. In the next section, the different components are deeply analyzed.

At first, the hydrogen ଶ is oxidized on the fuel electrode into reaction:

illustration not visible in this excerpt

The hydrogen ions migrate through the electrolyte membrane, while the electrons are transferred into the external circuit through an electrical receptor to the cathode.

At the cathode, the hydrogen ions and the electrons react with the oxygen, which is conducted from the surrounding air into the fuel cell, to water following the next reaction:

illustration not visible in this excerpt

The two reactions can be combined to the following reaction:

illustration not visible in this excerpt

The chemical energy created due to the electrochemical reaction changes into water, electricity and heat[7].

illustration not visible in this excerpt

illustration not visible in this excerpt

Figure 1: Diagram of a PEM fuel cell⁷

1.5 Structure and components of a PEMFC

The main part of the PEMFC is the MEA (Membrane Electrode Assembly). It delivers, when supplied with fuel and air, a cell voltage of around 0.7 V and has a power density of up to 1W…².

The following figure shows the performance curve of a PEM fuel cell.

illustration not visible in this excerpt

Figure 2: Fuel cell V-I curve[8]

The performance of the PEMFC depends on the performance of the MEA and the costs of the MEA represent 75% of the total costs of the PEMFC. The development of the PEMFC is always closely related to the development of the membrane electrode assembly[6].

The components of the MEA are:

The proton exchange membrane (PEM) The gas diffusion layer (GDL)

The catalyst layer

Figure 3: Schematic of a MEA[2]

The other components that are important for the operation of a fuel cell are:

The bipolar plates

The End/Front plates

In order to obtain higher voltages, many cells (MEA & Bipolar plates) can be connected in series. The product of the serial connection of many cells is called stack. It consists of several cells, front/end plates and tie bars as shown in the next figure.

illustration not visible in this excerpt

Figure 4: Fuel cell stack[2]

In the next chapter, each of the components is explained and analyzed. In addition to that, their functions and requirements are described to determine their most important characteristics.

The operating temperature of a PEMFC varies between 60°C and 120°C (typically 80°C) and its electrical efficiency is about 50% to 68 %[9]. The water evaporates at 120°C and its evaporation should not be faster than its production in the fuel cell in order to maintain the membrane hydrated.

The advantages of the PEMFC are[9][10]:

Fast start up and shut down:

A PEMFC reaches quickly the operating temperature und provides instantly power and heat. It can likewise quickly be shutdown.

High power density (1 W/…cm³ ):

The power provided by the fuel cell is high compared to its volume. High power weight (0.25W/g):

Long life time

A fuel cell has a long life time. Mercedes-Benz has reached with a fuel cell car a record of 300000 Km[11].

The disadvantages are[9]:

CO sensitivity

The electrodes of the fuel cell are sensitive to CO. The carbon monoxide contaminates the platinum in the fuel cell electrodes[9].

Drying and freezing of the membrane:

The membrane has to be always hydrated to assure the optimal operation of the fuel cell. The water in the membrane can freeze or evaporate at very low or very high temperature.

1.5.1 Proton exchange membrane

The proton exchange membrane (PEM) is the most important component in a PEMFC. The most used material in the PEM production is Nafion, which was developed by DuPont in the 1970s. The Nafion membrane is a chemically stable proton-exchange membrane based on polytetrafluoro-ethylene.

illustration not visible in this excerpt

Figure 5: Chemical structures of Nafion[12]

The PEM has three main roles:

- charge carrier for protons
- separate the reactant gases
- electronic and physical insulator between the two compartments[13]

The PEM has to separate hydrogen and oxygen in the two compartments of the fuel cell. If the gases come in contact, it can produce an explosion.

The membrane has also to be chemically resistant to hydrogen, oxygen, and water, which is a product of the reaction[14].

The performance of the PEM depends on the level of hydration and thickness of the membrane. The level of hydration or humidity of the membrane increases the conductivity of the PEM and then its performance. By reduced thickness, it is easier to hydrate the membrane and its material costs are lower. However, the reduction of the thickness has limits because of difficulties with durability and fuel by-pass. The thickness of the PEM expands by up to 20% with the hydration[13].

The working temperature is also an important issue for the proton exchange membrane. Its conductivity is 10 times lower at 80°C than at 60°C [Abbildung in dieser Leseprobe nicht enthalten][6][15].

1.5.2 Cathode/Anode catalyst layer

illustration not visible in this excerpt

The electrocatalyst accelerates both oxidation and reduction. It increases the reaction rate without being consumed during the transformation and enables favorable reaction pathways with low activation barriers. It also serves to transport the electrons from the catalyst surface out of or in the PEMFC.

The most used electrocatalyst material in PEMFC is platinum. In order to increase the reaction rate, more platinum can be used. In fact, higher platinum loading results in voltage gain. However, Pt is rare and expensive. The estimated resources of platinum are 40 000t worldwide[16]. The power density of Pt in a PEMFC is about 0.85-1.1 g/kW, it means nearly 72-94 g of platinum in a ca. 85 kW fuel cell stack[17].

illustration not visible in this excerpt

Figure 6: Schematic diagram of a fuel cell electrode[18]

1.5.3 Gas diffusion layer

The gas diffusion layer is a porous structure that consists of a Micro Porous Layer MPL and a hydrophobic treated component (generally polytetrafluoroethylene PTFE, called Teflon). It is constructed of a sheet of electrically conductive macro-porous substrate, such as a non-woven carbon paper, a woven carbon cloth or a carbon felt. The GDL is positioned between the catalyst layer and the bipolar plate (flow-field)[19][20][14].

The main functions of the GDL are:

Uniform distribution of the reaction gases

The reaction gases have to be uniformly distributed in the fuel cell to use the whole catalyst surface and assure a uniform heat development.

Evacuation of the reaction products

Reaction products, heat and water, has to be evacuated out of the fuel cell. An excess in water or heat can damage the fuel cell components.

Ensuring the electrical contact between the catalyst layer and the bipolar plates[21]:

The electrical courant produced in the fuel cell has to be transported to the bipolar plates and then to end plates.

Ensuring the heat transfer from the catalyst layer to the bipolar plates[14]:

The heat produced on the membrane and the catalyst layer has to be eliminated. The GDL assures the heat transport from the catalyst layer to the bipolar plates.

Guarantee that the PEM is constantly humidified with water in order to increase its conductivity

illustration not visible in this excerpt

Figure 7: Micrographs (scanning electron microscope (SEM)) of three different GDL supports (a: cloth, b: felt, c: paper)

1.5.4 Bipolar plates

The bipolar plates are a multi-functional component in a PEMFC that is responsible for the water, fuel and heat management. It consists of diverse flow fields for fuel, in PEMFC hydrogen, oxygen, also air is used, and cooling medium, often water based coolants are used[14]. Generally, metallic and graphite composite materials are used in the bipolar plates of PEMFCs, because they are resistant to corrosion. If typical metals are used such as aluminum, steel, titanium or nickel, they would corrode in the fuel cell and the dissolved metal ions would enter into the proton exchange membrane and affect negatively its proton conductivity. Because a corrosion layer would increase significantly the electrical resistance of the bipolar plates, they are coated with non-corrosive but electrically conductive layer, like graphite, conductive polymer, etc.[22].

illustration not visible in this excerpt

Figure 8: examples of bipolar plates with identical flow field (a: non-layered bipolar plate; b: gold-coated bipolar plate; c: milled graphite-composite bipolar plate[14]

The bipolar plates have to fulfill the following functions:

Connecting the cells electrically to form a stack

Separating the gases in the adjacent cells

Conducting the heat out of the active cells

Providing a structural support for the stack

As a result of those functions, the bipolar plates have to meet the following requirements:

Thermal conduction

Electrical conduction Gas impermeability

Strong and light structure[22]

1.5.5 Front/End plates

The front/end plates have an important role in a fuel cell stack. They provide uniform distribution of the pressure on the cells of the stack and guarantee that all components of the fuel cell are in contact in order to decrease the ohmic resistance as low as possible, which increases the efficiency of the fuel cell stack. They also unitize the different fuel cell parts to be a stack, provide passages for reactant gases and cooling medium and ensure good sealing at various interfaces.

There are two categories of end plates materials:

- Non-metallic material: engineering plastic, polysulfons, etc.
- Metallic material: steel, aluminum, titanium, etc.

The end plates have to fulfill the following requirements:

Strength
Rigidity
High power-weight ratio Resistance to corrosion
Electrical insulation[23]

illustration not visible in this excerpt

Figure 9: Fuel cell stack schematic[22]

2 Control Plan

2.1 Methodology

In order to determine the control parameters and to build a control plan, the failures that occur during the manufacturing of the fuel cell or the defects in the different components have to be identified. Therefore, it is important to carry out an FMEA (Failure Mode and Effects Analysis) to evaluate and compare the failures and their severity. Finally, the most important proprieties and characteristics that influence the performance of the PEMFC are deduced. This proprieties and characteristics help deducing the control parameters.

2.2 FMEA

FMEA is a quality engineering tool that helps to detect possible failures or errors and their effects during the conception and development phase of a product. Because of the increasing requirements and laws, such as the product liability law, the FMEA gain in importance and is frequently used in the product development.

The basic idea of the FMEA is that, if an error does not occur, it has not to be repaired or controlled. The FMEA helps to detect failures in the early development phases and to avoid their occurrence. So it is possible to improve the fulfillment of the product requirement and to reduce the failure repairing costs.

The figure 10 shows the increase of the failure repairing cost in the different phases of the product lifecycle.

illustration not visible in this excerpt

Figure 10: Rule of 10[24]

The advantages of an FMEA are[24]:

- Identification of weak points of the product
- Early identification and elimination of possible errors
- Estimation and quantification of risk errors
- Reduction in development costs and time
- Improved reliability

In this thesis, the FMEA will be used to determine the possible errors and failures that occur in the PEMFC and to distinguish the most critical parameter that affect the performance of the PEMFC.

To create an FMEA it is important to go through the following steps:

1. Structure analysis
2. Functional analysis
3. Error analysis
4. Risk assessment
5. Optimization

The risk estimation is carried using three criteria:

- Severity
- Probability
- Detection

These three criteria are multiplied by each other and result in a risk priority number (RPN). The rating for each risk criteria must be defined at the beginning of the FMEA[25].

illustration not visible in this excerpt

Table 2: Rating scale for risk assessment[25]

The rating scale from 1 to 5 is used in this thesis instead of the known scale from 1 to 10, because the evaluation and the comparison of failures and their effects are easier. The lack of information, especially about the probability of the failures, makes the scale 1-5 more suitable for this FMEA. In addition to that, this scale was mentioned and recommended in the literature (see[25]).

Structure analysis

illustration not visible in this excerpt

Functional analysis

illustration not visible in this excerpt

[...]