Skip to main content

MembrAnEs for STationary application with RObust mechanical properties

Final Report Summary - MAESTRO (MembrAnEs for STationary application with RObust mechanical properties)

Executive Summary:
The primary objective of MAESTRO was to improve the mechanical properties of low equivalent weight (EW) state of the art perfluorosulfonic acid (PFSA) membranes using chemical and thermal ionomer processing and fibre network and filler reinforcement methodologies.
Benchmark MEAs prepared using benchmark state-of-the-art short-side-chain Aquivion ionomer of EW 790 g/mol decayed significantly in performance as the temperature was increased to 100 and 110 °C. Further, accelerated durability testing via an open circuit voltage (OCV) hold test showed that the benchmark membrane failed after only 100-200 hours and highlighted the need for significant improvements to the robustness of the low EW membranes.
A range of different approaches was used to prepare membranes having robust mechanical properties using an ionomer of EW lower than that of the benchmark, 700 g/mol. These approaches included ionomer cross-linking during emulsion polymerisation, which leads to non-linear ionomer molecules with high molecular weight, and cross-linking during the process of membrane casting by use of a reagent to cross-link through a small fraction of the sulfonic acid groups. Attempts were also made to associate these two approaches (during polymerisation/casting). Electrospinning was used to fabricate organic and inorganic fibres to be employed as mechanical reinforcements of low EW Aquivion. In all cases the nanofibre reinforcement led to a significant improvement of mechanical properties of the final composite membranes, and the conductivity was higher than that of the benchmark membrane. The possibility to mechanically reinforce Aquivion by ionic crosslinking induced by incorporation of inorganic nanoparticles was also investigated, and composite membranes prepared using filler types of different hydrophobicity. All the membranes showed strongly enhanced elastic modulus and yield stress in comparison with the benchmark Aquivion membrane.
First durability testing at the 18 month stage of the project exhibited significant enhancements in lifetime from the new membrane approaches.
In the second phase of the project, three configurations of the most prospective individual membrane stabilisation routes were associated, cross-linked EW 700 ionomer being used in all new membrane developments, and membrane formulation and thickness were optimised to develop improved robustness membranes.
The month 26 metric to have improved tensile properties by 50% was exceeded with several of the novel membrane materials developed in MAESTRO. Further, the approaches developed to increase mechanical stabilisation did not compromise conductivity and, in some cases, improved conductivity compared with the project start point was observed. Critical assessment of the results of in situ and ex situ characterisation led to final down-selection of three most promising candidates. Although the original project plan had been to select one membrane only for a long-term accelerated durability test, the promising results allowed the partnership to extend the ambition of the project to stack testing, using a stack comprising membrane electrode assemblies (MEAs) incorporating different membrane types. The stack testing protocol developed to simulate conditions realistic for micro-CHP type application included continuous operation and load cycling as well as aggressive accelerated ageing conditions of stop/start cycling. Under these conditions, cross-linked Aquivion membranes showed greater durability than benchmark membranes, while cross-linked Aquivion with an electrospun nanofibre reinforcement showed the most significant improvement in durability; these MEAs displaying less than 3% voltage loss after 2,000 hours of operation, including 500 hours of cycling between 12 hours operation/12 hours stop, and a further 250 hours of cycling between 4 hours operation/4 hours stop, thus comfortably achieving the final project objective.
Non-confidential results have been communicated at the most important fuel cells/electrochemistry conferences in Europe and overseas. Four papers have been published to date. The website www.maestro-fuelcells.eu lists dissemination activities carried out for the project, and the password-protected partners-only area collects all project documentation, presentations and reports.

Project Context and Objectives:
Project Context
The European Strategic Energy Technology (SET) Plan has identified fuel cells and hydrogen among the technologies needed for Europe to achieve the targets for 2020 - 20% reduction in greenhouse gas emissions, 20% share of renewable energy sources in the energy mix, and 20% reduction in primary energy use – as well as to achieve the long-term vision for 2050 towards decarbonisation. The challenge facing fuel cells and hydrogen technologies is of great complexity, and their contribution to Community policies, in particular energy, environment, transport and industrial competitiveness is very important. FCH-JU1 objectives for stationary power generation and combined heat and power (CHP) were ambitious both in terms of volume (100 MW installed electric capacity) and cost (€ 4,000 – 5,000 kW for micro-CHP, €1,500 – 2,500/kW for industrial units) targets for 2015.
Uninterruptible Power Supply (UPS) and power supply to residential buildings are important fuel cell markets requiring units of ca. 5 kW and 1-2 kW respectively. Generation of electricity and heat at the site of demand can save a significant amount of primary energy compared to the central generation of electricity and the generation of heat on site: electricity is lost during transmission and distribution , being >6% in Europe and North America and >10% in developing countries, and the heat generated at central production is lost or wasted. A survey makes known that in 2008, deliveries of fuel cells for small stationary applications rose by 79% to reach 4,000 units, the majority of the systems being sold for UPS applications but there were advances too in the use of small units for µ-CHP applications. While both solid oxide fuel cells (SOFC) and PEMFC are in development for small stationary applications, over 95 per cent of the units delivered in 2008 were based on proton exchange membrane (PEM) electrolytes. The same survey reports that during 2008 North America (USA and Canada) produced nearly two thirds of all units and Asia produced a further 25%. Europe produced the rest, which leaves enormous room for improvement in the worldwide small stationary market for European industry, which is of strategic importance given the forecast made for a significant increase in units shipped to over six million units, with a fairly even split between UPS and µ-CHP, by 2019.
An issue clearly highlighted in the FCH-JU1 Multi-Annual Implementation Plan in the stationary application area was the need to address lifetime requirements of 40,000 hours for cell and stack, and the call for new or improved materials leading to step change improvements over existing technology in terms of performance, endurance, robustness and cost. In general, failure mechanisms of PEMFC membranes are of two main types: chemical, of which the best understood at present arises from attack by peroxide radicals on susceptible polymer end groups and side chains, and mechanical, which originates from weak intermolecular interactions between polymer chains. While methods of chemically stabilising the polymer end groups have been developed, and other work has significantly advanced on incorporation of radical scavengers in the fuel cell membrane electrode assembly to avoid polymer chain scission events, including of the side-chain functional groups, failure due to inadequate membrane mechanical properties limits cell and stack lifetime. The problem is exacerbated by the trend in use of membranes of much reduced thickness (<30 µm, compared with the use of membranes of ca. 175 µm some 10 years ago) which negatively impacts membrane strength, thin membranes being required for their low area specific resistance and for their enhanced water back diffusion properties. In particular in conditions of use of stop/start (as in summer season use in residential application for example) or load cycling (as in spring/autumn seasons), the variation in relative humidity causes membrane swelling and contraction that ultimately leads to membrane failure, in particular in areas of the fuel cell where the membrane is subject to greatest compression.
Furthermore, higher temperatures of operation are required to increase the overall efficiency and approach the FCH-JU1 targets of electrical efficiency of >80% for CHP units. Low equivalent weight ionomers are required to reach the membrane conductivity at higher temperature and lower humidity, and the MEA performance targets for stationary operation, to enable stationary PEMFC systems to achieve superior overall system yield to competitive technologies. In stationary applications, where the situation of deep MEA dehydration and frequent open circuit voltage events can be reasonably avoided, the most relevant failure mode in extended life time operation is associated with membrane mechanical failure. Such high ion exchange capacity (low equivalent weight) polymers show increased tendency to dimensional variation under wet/dry cycling and increased mechanical instability.

Project Objectives
The MAESTRO project aimed to establish methods to increase the mechanical stability of state-of-the-art short-side-chain perfluorosulfonic acid (PFSA) membranes for stationary application of proton exchange membrane fuel cells (PEMFC) to increase their durability and cell lifetime. Such membranes were operated at temperatures up to 110-120 °C in stacks developed in the FP6 Autobrane "automotive fuel cell membranes" project, which involved all of the technical partners of MAESTRO. The membranes were therefore already known to represent a viable option for operation at higher temperature/lower relative humidity for PEMFC small stationary applications, whereas their long-term durability required improvement.
MAESTRO proposed to develop solutions to the above bottlenecks by developing and screening a range of approaches to improve the mechanical stability of short-side-chain PFSA type PEM fuel cell membranes. To achieve these objectives, the project partners identified five routes to develop robust membranes (i) by increasing polymer molecular weight to increase inter-chain entanglement; (ii) by increasing molecular weight and associating original polymer cross-linking approaches, carried out during polymerisation and/or during membrane casting, to avoid membrane dimensional change; (iii) by tailoring membrane thermal annealing at identified temperature/relative humidity couples; (iv) by embedding electrospun nanofibre mats providing mechanical stability into a PFSA matrix, and by associating electrospun inorganic oxide fibres in cast PFSA; (v) by ionic interaction between PFSA and a dispersed inorganic phase. The final project target for the membrane was to have increased the tensile strength (compared with the benchmark material at the project beginning) by 50%, with a milestone at the mid-term stage of improvement by 20-25% - without detriment to the membrane conductivity. MAESTRO further intended to characterise stabilised membranes for their ex situ properties and to integrate selected candidate membrane materials into MEAs and validate them by evaluating single cell performance and durability under accelerated stress testing conditions designed to enhance chemical and/or mechanical degradation. The objective of the second phase of the project was to associate the most prospective individual approaches, and then down-select most promising candidate membranes on the basis of the membrane proton conductivity and tensile properties, and MEA fuel cell performance and durability on OCV hold and in wet-dry cycling. The final aim of the project was to submit these candidate membranes, after MEA preparation, to accelerated durability testing over 1000 hour periods comprising repeated stop-start events and voltage cycling, in conditions simulating those encountered in a micro-CHP application, with a target durability indicator of achieving voltage loss below 10 percent of that at beginning of life.

Project Results:
See attached file

Potential Impact:
See attached file
List of Websites:
www.maestro-fuelcells.eu