Identification of critical parameters for PEMFC stack performance characterization and control strategies for reliable and comparable stack benchmarking

https://doi.org/10.1016/j.ijhydene.2016.08.065Get rights and content

Highlights

  • Identification of critical parameters for PEMFC stack benchmarking.

  • Definition of control sensor positions for critical parameters.

  • Parameter variation strategies for sensitivity tests.

  • Summary of test operating parameters for all PEMFC applications.

  • Concept of “steady-state” polarization curve.

Abstract

This paper is focused on the identification of critical parameters and on the development of reliable methodologies to achieve comparable benchmark results. Possibilities for control sensor positioning and for parameter variation in sensitivity tests are discussed and recommended options for the control strategy are summarized. This ensures result comparability as well as stable test conditions. E.g., the stack temperature fluctuation is minimized to about 1 °C. The experiments demonstrate that reactants pressures differ up to 12 kPa if pressure control positions are varied, resulting in an average cell voltage deviation of 21 mV.

Test parameters simulating different stack applications are summarized. The stack demonstrated comparable average cell voltage of 0.63 V for stationary and portable conditions. For automotive conditions, the voltage increased to 0.69 V, mainly caused by higher reactants pressures.

A benchmarking concept is introduced using “steady-state” polarization curves. The occurring 20 mV hysteresis effect between the ascending and descending polarization curve can be corrected calculating the mean value of both voltages. This minimizes the influence of preceding load levels, current set points, and dwell times.

Introduction

Polymer electrolyte membrane fuel cell (PEMFC) stacks are the main components in fuel cell systems and convert the chemical energy from the fuel to electrical and thermal energy usable for different applications. Compared with other energy converters, PEMFC systems are considered as one of the most effective and promising power systems due to high fuel-to-electricity conversion efficiency and low to zero emission [1]. Nevertheless, several challenges need to be overcome for the commercialization of this technology. At the moment, the required investments for PEMFC stacks and systems are still too high due to the costs of the used components, the manufacturing costs and the complexity of the system [2], [3]. Apart from the costs, the main challenges for competitive fuel cell systems are stack performance and durability [4], [5]. All these aspects are examined in many laboratories and improvements are ongoing by component [6], [7] and stack design modifications [8], [9], [10]. The reliability of the stack characterization and of the benchmark results is thereby crucial because many parameters have significant impact on the results and may vary using different test hardware and test procedures. Therefore, the use of standardized procedures and clear definitions of all influential parameter sensors are of high importance to assure reliable and comparable test results for different test objects as well as for different test facilities. Consequently, the harmonization of PEMFC tests at the stack level is important in order to accelerate stack development and to benchmark stacks. The resulting procedures are of high interest for stack manufacturers, system integrators and for academia.

A search for already existing test procedures for PEMFC results in many procedures for the component and single cell level, but only a few procedures are dedicated to the stack level. The available stack procedures are typically limited to a certain application and, consequently, focus on testing specific goals of this application. The methodology of these tests is mainly similar to single cell characterization. However, the procedures have to be defined more precisely to cover the inhomogeneous distribution of parameters within a stack.

An international standard exists for the single cell level, IEC/TS 62282-7-1 [11], including definitions for nomenclature, test hardware, cell assembling, test procedures, and test reporting. The test procedures cover different aspects: (i) constant current and voltage tests, (ii) high frequency resistance (HFR) and mass transport limited current, (iii) current–voltage characteristics using different reactant concentrations and compositions including impurities, (iv) sensitivity tests to reactants stoichiometric ratios, pressures and humidity as well as cell temperature, and (v) durability tests using constant load, load cycling and start-stop cycling. A respective standard for the PEMFC stack level has not been defined so far.

The U.S. Drive Fuel Cell Tech Team [12] and the U.S. Department of Energy (DoE) [13] have published quite similar test protocols mainly focused on component characterization for automotive application, but procedures for the measurement of polarization curves and for durability tests using a dynamic load cycle are also given. Polarization and durability tests address the single cell and the stack levels, and the information given in the document is limited to the test operating conditions and the test sequence. A definition of the parameter control strategy is not included in the document and, consequently, reliable stack characterization seems to be challenging using these procedures.

The EU-funded projects FCTESTNET [14] and FCTESQA [15] have developed test procedures for single cells, stacks and systems for different applications and different types of fuel cells. These procedures cover the field of PEMFC, SOFC and MCFC. The procedures defined for the PEMFC stack level deal with sensitivity tests for reactant pressure and stoichiometry, polarization curve measurements, and durability test using constant load and dynamic load cycling. Positions of the control sensors and control strategies are partially defined in the available documents and are included in the discussion of these aspects in the present paper.

Besides the mentioned procedure definitions, two industrial standards are available for PEMFC stack performance characterization, a Japanese Industrial Standard (JIS) and a standard by the Society of Automotive Engineers (SAE). The JIS C 8832 standard [16] describes performance test procedures for the characterization of stationary PEMFC stacks. Procedures for current–voltage curves and tests of sensitivity regarding stack temperature, reactant utilization, and reactant humidity are given. Due to the focus on stationary applications including the use of reformate fuel, the impact of impurities is one of the main objectives of this standard. Thereby, the effects of carbon monoxide and ammonia concentration in the fuel, as well as the effect of nitrogen dioxide, sulfur dioxide, ammonia, and toluene concentration in the oxidant are covered.

The SAE J2617 standard summarizes test procedures for performance tests of PEMFC stacks for automotive application [17]. These procedures cover: (i) open circuit voltage (OCV) tests, (ii) polarization curves using different reactant concentrations, (iii) sensitivity sweeps for reactants stoichiometric ratios, humidity, and pressures as well as coolant temperature and flow rate, and (iv) voltage response tests in constant load and dynamic load cycling.

The characterization of PEMFC stacks has to consider existing parameter gradients across the stack, like stack temperature and reactant pressure. This requires identification of critical parameters with high impact on the test results as well as clear definitions of the positions of all parameter control sensors and of the required parameter control strategies. These definitions are not consistent or not included in the above listed procedures and standards. The EU-funded project Stack-Test was initiated by the European fuel cell OEMs to create universal standards for operating conditions and test reference procedures to assure reliable benchmarking of different stacks. The benchmark results can be used to identify the best stack design for a certain application. This paper summarizes the achievements of the project and points out the requirements for reliable and comparable stack performance characterization and benchmarking. The prepared test procedures, covering not only performance, but also durability and safety aspects, can be obtained at the project website [18]. Within the presented work, the possibilities for the positioning of the parameter control sensors and for the variation of parameters in sensitivity tests are discussed in detail and recommendations for the resulting control strategy are given based on these discussions. Test operating conditions (TOCs) for benchmark tests for different applications are also summarized to enable the comparison of research achievements, even if these conditions do not totally match the conditions in the target system. Furthermore, a concept to obtain a “steady-state” polarization curve without actual steady-state measurements is introduced. The achievements presented in this paper thereby remove an existing lack of definitions needed for reliable and comparable PEMFC stack performance characterization. Consequently, the developed benchmark methodology can be a powerful tool for system integrators to identify the best stack for a certain application, for stack manufacturers for quality assurance and stack development, and for academia to assure reliable comparison of research results.

Section snippets

Experimental

A liquid-cooled PEMFC stack manufactured by ZSW (Zentrum für Sonnenenergie- und Wasserstoff-Forschung Baden-Württemberg) was used for the tests presented in this work. This stack contained graphitic composite bipolar plates with 10 cells and an active electrode area of 96 cm2. It represents a typical stationary stack with nominal electrical power output of 480 W.

All tests were carried out using an in-house manufactured 1 kW test station using hydrogen (grade 5.0 by Linde AG) and compressed

Sensor definition for parameter control

The stack is frequently considered as a black box with inlet and outlet ports for benchmark tests. Consequently, the aim of these tests is not to internally diagnose what is happening inside the stack, but to reliably measure its performance under certain prefixed conditions. While the available test procedures mainly define the methodology and the parameter set points for this purpose, the positions of the used sensors for controlling these parameters are usually not defined. Certainly, the

Control strategies for sensitivity tests using variable parameter set points

Not only the sensor position for parameter control has a high impact on benchmark results, also the direction of parameter changes can influence the test results. This influence is obvious for the most common tool for PEMFC stack characterization, the measurement of the stack performance as a function of the stack current or current density in the form of a two-way polarization curve. It is well known that the results from polarization curves determined by current increase (ascending part) and

Set of test operating conditions for different applications

Even if the TOCs for specific tests by system integrators depend on the target system, harmonized TOCs are important to identify the capability of a stack design (including flow field design and components) for a certain application. Additionally, the performance test results for PEMFC stacks presented in scientific literature cannot be compared because the tests were realized using different conditions. Harmonized TOCs can allow the reliable result comparison from different facilities using

Reliable stack characterization by polarization curve

The most common tool for PEMFC characterization regarding the impact of load levels on the stack performance is the polarization curve, but also the comparison of different polarization curves can be challenging. The results can significantly differ, even if the TOCs are the same. This can be caused by the choice of the test points and the variation of their dwell time. Furthermore, it is not always clearly indicated if the polarization curve was measured using an ascending or descending

Conclusion

This work is focused on the identification of critical parameters for PEMFC stack performance characterization and on the development of reliable benchmark methodologies to achieve comparable test results using different test equipment. Due to different possibilities for the positioning of the control sensors, their positions were clearly defined for all test operating parameters considering different aspects of stack safety, parameter stability, test duration, and comparability to an actual

Acknowledgements

The research leading to the presented results has received funding from the European Union Seventh Framework Programme (FP7/2007–2013) for the Fuel Cells and Hydrogen Joint Technology Initiative under grant n° 303445 (Stack-Test: Development of PEM fuel cell stack reference test procedures for industry). The Stack-Test project consortium is gratefully acknowledged for all discussions and testing activities to create test methodologies and test procedures for reliable and comparable PEMFC stack

References (29)

Cited by (18)

  • Influence of the dwell time in the polarization hysteresis of polymer electrolyte membrane fuel cells

    2022, Electrochimica Acta
    Citation Excerpt :

    It also generates difficulties when it comes to comparing polarization curves, as hysteresis will appear regardless of the dwell times used in the experiment, set-points defined for the current, or previous load levels (ascending or descending currents). Indeed, Mitzel et al. [5] introduced the concept of “steady-state” polarization curves for a more suitable benchmarking among different cells or stacks, consisting on calculating the average cell voltage (for each current) between the ascending and descending polarization curves in order to obtain a fair polarization curve and benchmarking with a minimum influence of hysteresis. This was also included in harmonized test protocols for fuel cells such as the one by the Joint Research Centre of the European Commission [6,7].

  • Review on mechanisms and recovery procedures for reversible performance losses in polymer electrolyte membrane fuel cells

    2021, Journal of Power Sources
    Citation Excerpt :

    Hence, detailed understanding of reversible degradation effects and procedures for recovery of related performance losses are of particular interest and are still under discussion in the literature. The development of reliable and efficient recovery procedures can thereby enable: (i) high system efficiency if recovery phases can be included in the operation strategy [17], (ii) reliable performance evaluation for component benchmarking and differentiation between reversible and irreversible degradation rates [18], and (iii) understanding of reversible degradation effects which supports the development of new materials [19,20]. When designing recovery procedures not only for the laboratory but also for specific system operations, it is important to consider the constraints of a PEMFC system [21].

View all citing articles on Scopus
View full text