I samband med den globala uppvärmningen ökar efterfrågan för rena och förnybara bränslen alltmer i dagens samhälle. Eftersom flygindustrin idag är ansvarig för samma mängd växthusgaser som all motortrafik i Sverige, skulle ett byte till en avgasfri energikälla för flygfarkoster vara ett stort framsteg. Därför har projektet genom modellering framtagit ett hybridsystem av ett batteri och en bränslecell och undersökt hur kombinationen av olika storlekar på dem presterar i en driftcykel. Då batterier har hög specifik effekt men är tunga, kompletteras de med fördel av bränsleceller, som är lättviktiga och bidrar med uthållig strömförsörjning. På så sätt blir hybriden optimal för flygfarkoster. Kandidatarbetet är en del av projektet Green Raven, ett tvärvetenskapligt samarbete mellan instutitionerna Tillämpad Elektrokemi, Mekatronik och Teknisk Mekanik på Kungliga Tekniska Högskolan. Driftcykelmodelleringen gjordes i Simulink, och flera antaganden gjordes beträffande effektprofilen, samt bränslecellens mätvärden och effekt. Tre olika energihushållningsscheman skapades, vilka bestämde bränslecellseffekten beroende på vätgasnivån och batteriets laddningstillstånd. Skillnaden på systemen var vilka intervall av laddningstillstånd hos batteriet som genererade olika effekt hos bränslecellen.  Det bästa alternativet visade sig vara 0/100-systemet, eftersom det var det enda som inte orsakede någon degradering av bränslecellens kapacitet. / In today’s society, with several environmental challenges such as global warming, the demand for cleanand renewable fuels is ever increasing. Since the aviation industry in Sweden is responsible for the sameamount of greenhouse gas emissions as the motor traffic, a change to a non-polluting energy source forflying vehicles would be considerable progress. Therefore, this project has designed a hybrid system of abattery and a fuel cell and investigated how different combinations of battery and fuel cell sizes perform ina drive cycle, through computer modelling. As batteries possess a high specific power but are heavy, thefuel cells with high specific energy complement them with a sustained and lightweight power supply,which makes the hybrid perfect for aviation. The bachelor thesis is a part of Project Green Raven, aninterdisciplinary collaboration with the institutions of Applied Electrochemistry, Mechatronics andEngineering Mechanics at KTH Royal Institute of Techology. The drive cycle simulations were done inSimulink, and several assumptions regarding the power profile, fuel cell measurements and power weremade. Three different energy management strategies were set up, determining the fuel cell powerdepending on hydrogen availability and state of charge of the battery. The strategies were called 35/65,20/80 and 0/100, and the difference between them was at which state of charge intervals the fuel cellchanged its power output. The best strategy proved to be 0/100, since it was the only option which causedno degradation of the fuel cell whatsoever.

A novel method has been implemented for measuring the concentration of gas species, water vapor and carbon dioxide, within a narrow channel flow field non-invasively using tunable diode laser absorption spectroscopy (TDLAS) in conjunction with a laser modulated at a high frequency [Wavelength Modulation Spectroscopy (WMS)] tuned to the ro-vibrational transition of the species. This technique measures the absorption profile which is a strong function of the species concentration across short path lengths and small time spans, as in PEM fuel cells during high load cycles. This method has been verified in a transparent circular flow 12 cm path length and a 12 mm rectangular flow channel. Distinct absorption peaks for water vapor and carbon dioxide have been identified, and concentrations of water vapor and carbon dioxide within the test cells have been measured in situ with high temporal resolutions. A comparison of the full width at half maximum (FWHM) of the absorption lineshapes to the partial pressure of water vapor and carbon dioxide showed a predominantly linear relationship, except in the lower partial pressure regions. Test section temperature was observed to have very minimal impact on these curves at low partial pressure values. A porous media like a membrane electrode assembly (MEA) similar to those used in PEM fuel cells sandwiched between two rectangular flow channels was also tested. Some of the scattered radiation off the MEA was observed using a photodiode at high gain, allowing for more localized species detection. The technique was used to monitor the humidity on either side of the MEA during both temperature controlled and super-saturated conditions. The measurements were observed to be repeatable to within 10 %.

One of the important factors determining the lifetime of polymer electrolyte membrane fuel cells (PEMFCs) is membrane degradation and failure. The lack of effective mitigation methods is largely due to the currently very limited understanding of the underlying mechanisms for mechanical and chemical degradations of fuel cell membranes. In order to understand degradation of membranes in fuel cells, two different experimental approaches were developed; one is fuel cell testing under open circuit voltage (OCV) with bi-layer configuration of the membrane electrode assemblies (MEAs) and the other is a modified gas phase Fenton's test. Accelerated degradation tests for polymer electrolyte membrane (PEM) fuel cells are frequently conducted under open circuit voltage (OCV) conditions at low relative humidity (RH) and high temperature. With the bi-layer MEA technique, it was found that membrane degradation is highly localized across thickness direction of the membrane and qualitatively correlated with location of platinum (Pt) band through mechanical testing, Infrared (IR) spectroscopy, fluoride emission, scanning electron microscopy (SEM), transmission electron microscopy (TEM), and energy dispersive spectroscopy (EDS) measurement. One of the critical experimental observations is that mechanical behavior of membranes subjected to degradation via Fenton's reaction exhibit completely different behavior with that of membranes from the OCV testing. This result led us to believe that other critical factors such as mechanical stress may affect on membrane degradation and therefore, a modified gas phase Fenton's test setup was developed to test the hypothesis. Interestingly, the results showed that mechanical stress directly accelerates the degradation rate of ionomer membranes, implying that the rate constant for the degradation reaction is a function of mechanical stress in addition to commonly known factors such as temperature and humidity. Membrane degradation induced by mechanical stress necessitates the prediction of the stress distribution in the membrane under various conditions. One of research focuses was on the developing micromechanism-inspired continuum model for ionomer membranes. The model is the basis for stress analysis, and is based on a hyperelastic model with reptation-inspired viscous flow rule and multiplicative decomposition of viscoelastic and plastic deformation gradient. Finally, evaluation of the membrane degradation requires a fuel cell model since the degradation occurs under fuel cell operating conditions. The fuel cell model included structural mechanics models and multiphysics models which represents other phenomena such as gas and water transport, charge conservation, electrochemical reactions, and energy conservation. The combined model was developed to investigate the compression effect on fuel cell performance and membrane stress distribution.

Anhydrous proton conducting polymers are highly sought after for applications in high temperature polymer electrolyte membrane fuel cells (PEMFCs). N-heterocycles (eg. imidazole, triazole, and benzimidazole), owing to their amphoteric nature, have been widely studied to develop efficient anhydrous proton transporting polymers. The proton conductivity of N-heterocyclic polymers is influenced by several factors and the design and development of polymers with a delicate balance among various synergistic and competing factors to provide appreciable proton conductivities has been a challenging task. In this thesis, the proton transport (PT) characteristics of polymers functionalized with two diverse classes of functional groups - N-heterocycles and phenols have been investigated and efforts have been made to develop the molecular design criteria for the design and development of efficient proton transporting functional groups and polymers. The proton conduction pathway in 1H-1,2,3-triazole polymers is probed by employing structurally analogous N-heterocyclic (triazole, imidazole, and pyrazole) and benz-N-heterocyclic (benzotriazole, benzimidazole, and benzopyrazole) polymers. Imidazole-like pathway was found to dominate the proton conductivity of triazole and pyrazole-like pathway makes only a negligible contribution, if any. Polymers containing benz-N-heterocycles exhibited higher proton conductivity than those with the corresponding N-heterocycles. Pyrazole-like functional groups, i.e. the molecules with two nitrogen atoms adjacent to each other, were found not to be good candidates for PT applications. A new class of proton transporting functional groups, phenols, has been introduced for anhydrous PT. One of the highlighting features of phenols over N-heterocycles is that the hydrogen bond donor/acceptor reorientation can happen on a single -OH site, allowing for facile reorientational dynamics in Grotthuss PT and enhanced proton conductivities in phenolic polymers. Unlike the case of N-heterocycles, comparable conductivities were achieved between poly (3,4,5-trihydroxy) styrene and the corresponding small molecule, pyrogallol. This observation suggests that reorientation should be considered as a crucial design parameter for PT functional groups. The PT characteristics of phenol-based biaryl polymers are studied and compared with the analogous phenol-based linear styrenic polymers. The two-dimensional disposition of -OH moieties in biaryl polymers, although resulted in lower apparent activation energies (Ea), did not improve the net proton conductivity due to the accompanying increase in glass transition temperature (Tg). Thus, the ease of synthesis and lower Tg values of phenol-based styrene polymers make the styrenic polymer architecture preferable over the biaryl architecture. Finally, the synthesis of a series of poly(3,4-dihydroxy styrene)-b-polystyrene block copolymers has been demonstrated via anionic polymerization. These block copolymers will provide an opportunity to systematically investigate the effect of nanoscale morphology on proton transport.

No description available.

No description available.

The typical architecture of the proton exchange membrane fuel cell (PEMFC) contains a layer called the microporous sublayer (MSL). The MSL is a mixture of carbon black and polytetrafluoroethylene (PTFE), which is typically applied to the gas diffusion layer (GDL). The composition (wt.% PTFE) and loading (mg/cm2) can be varied to optimize the electrochemical performance of the PEMFC and the overall adhesion of the layers within the PEMFC. This research establishes correlations that characterize the performance and adhesion of the layers within the PEMFC based on composition, loading, fabrication pressure, and fabrication time. MSL loading was varied from 1.5-4 mg/cm2, composition was varied from 10-50 wt.% PTFE, fabrication pressure was varied from 3.45-10.34 MPa, and fabrication time was varied from 2-8 minutes. Using these four factors, correlations were created, and optimal solutions for each response were identified. The adhesion correlation identifies a low MSL loading, mid-range MSL composition, high fabrication pressure, and high fabrication time as desirable factors. The performance correlation suggests that the PEMFC performance is enhanced with low MSL loadings, low MSL PTFE content, and a low fabrication pressure and does not find fabrication time to be a significant factor in the correlation. / Master of Science

Bipolar plate material and fabrication costs make up a significant fraction of the total cost in a polymer electrolyte membrane fuel cell stack. In an attempt to reduce these costs, a novel manufacturing method was developed for use with composite materials. Conductive fillers were mixed with a polypropylene binder and molded into single cell monopolar plates. A fuel cell test stand, capable of testing six cells simultaneously, was used for long-term corrosion testing. In-situ tests took place in 5 cm2 active area fuel cells with cathode humidification. Using data from test cells containing graphite monopolar plates as a baseline, two composite formulations, were able to produce power at 66-79% of the baseline power. Power output from one cell remained in this range for over 200 hours, while the other sample experienced surface oxidation and eventually failed. With improvements in part conductivity coming from conductive polymers, this manufacturing technique holds the promise of producing monopolar and bipolar plates that could eventually be scaled up for use in fuel cell stacks. / Master of Science

Afin de pouvoir utiliser les piles à combustible du type PEMFC dans une application automobile, leur coût doit être diminué et leur durée de vie doit être augmentée. De nombreux résultats montrent que la gestion de l'eau dans les piles PEMFC est essentielle sur ces aspects et qu’une meilleure maitrise contribuera a développer des piles plus performantes. La couche active cathodique (CCL, Cathode Catalyst Layer) est le lieu de production de l'eau ce qui en rend l'optimisation importante pour assurer une bonne gestion de l'eau. Dans ce travail, la méthode réseau de pores a été adaptée pour modéliser le transport diphasique dans la structure poreuse de la CCL. Dans l'état de l'art actuel, le modèle développé est le seul permettant d’analyser l'effet des proprietes locales de la CCL (structure, mouillabilite…) sur les mecanismes de transport diphasique. Cet outil de compréhension constitue également une base pour proposer des améliorations de la CCL afin d'améliorer les performances des piles. Les algorithmes d'invasion développés ont été analysés d'une façon détaillée. Le transport fluidique (gaz et liquide) est couplé avec le transport des charges (électrons et protons) par un modèle de réaction électrochimique. Les mécanismes de capillarité, de diffusion gazeuse et d’evaporation sont integres au modele afin d'avoir une représentation la plus complète possible du fonctionnement de la CCL. La description de la structure poreuse par un réseau de pore régulier, l'algorithme d'invasion de l'eau liquide et le modèle de la diffusion des gaz ont été validés par des comparaisons avec des résultats expérimentaux de la littérature ou spécifiques de ce travail. Le modèle est ensuite exploité pour analyser l'effet des paramètres de la CCL tels que la mouillabilité et la taille des pores sur les performances de la couche active. Les résultats permettent d’analyser de premieres idees de modifications de la CCL pour ameliorer la gestion de l’eau et les performances des PEMFC. / In order to use PEM fuel cells in an automotive application, their cost must be reduced and their lifetime must be increased. Many results show that water management is a critical issue in PEMFC optimization. The water is produced in the cathode active layer (CCL) which makes the optimization of this component very important to ensure a better water management in the PEMFC. In this work, the pore network method has been adapted to model the two-phase transport in the porous structure of the CCL. Considering the state of the art, this is the only model developed to analyze the effect of local properties of the CCL (structure, wetting ...) on the two-phase transport mechanisms. This model is proposed as a scientific tool to help understanding the fundamentals behind the transport phenomena inside the CCL and also to help in the conception of the future CCL. The liquid invasion algorithms developed in this work were analyzed in details. The fluids transport (gas and liquid) is coupled with the charges transport (electrons and protons) using an electrochemical reaction model. The capillary driven liquid transport, the gas phase diffusion and the evaporation process are all integrated into the model in order to have the most possible complete description of the CCL. The description of the porous structure by a regular network, the liquid invasion algorithm and the gas diffusion model all have been validated by comparisons with experimental results from literature or specific work . The model is then exploited to analyze the effect of parameters such as the CCL wettability and pore size distribution on the performance. The results allow analysis of initial ideas that can help in the conception of the CCL in order to improve the water management and the performances of the PEMFC.

Nous avons étudié les mécanismes de dégradation de catalyseurs Pt3Co/C en conditions réelles (stacks 16 cellules, hydrogène/air, stationnaire et intermittent, t > 1000 heures). Des modifications de la structure atomique, de la morphologie et de la composition chimique des catalyseurs ont été mises en évidence grâce à des techniques à résolution atomique, tels que la microscopie HAADF ou encore la spectroscopie d'absorption de rayons X. En plus d'être sujets à la maturation d'Ostwald 3D, ces catalyseurs perdent continuellement et irréversiblement les atomes de cobalt contenus dans le matériau « natif », ce qui conduit à la formation de nanoparticules « creuses » de Pt. Nous avons montré l'effet d'une contamination de l'électrode par des cations métalliques (Co2+). Des hétérogénéités de vieillissement de ces électrodes, à la fois « dans le plan » et « à travers le plan », ont été mises en évidence, en utilisant des marqueurs structuraux caractéristiques des électrodes. Des différences locales des cinétiques et des mécanismes de dégradation ont été confirmées grâce à des tests en monocellule PEMFC à cathode segmentée. / The durability of Pt3Co/C PEMFC cathode catalysts is investigated under real operating conditions (16-cell short stacks, hydrogen/air, constant current or start/stop, ageing time > 1000 hours). Using atomically resolved physical techniques such as HRSTEM-HAADF, and XAS, a detailed picture of how atomic structure, chemical composition and morphology of these cathode catalysts are changing over time has been drawn. In addition to 3D Ostwald ripening, these Pt-alloy catalysts undergo irreversible decrease of their cobalt content upon aging, yielding formation of “hollow” Pt/C nanoparticles. In the meantime, a great amount of Co2+ species is released within the MEA, which influences the catalyst surface reactivity and its ORR activity. Finally, structural markers of the degradation of the cathode catalyst have been used to unveil aging heterogeneities within the MEA: “through-the-plane” heterogeneities of aging (i.e. from the PEM/cathode interface to the cathode/GDL interface), and “in-the-plane” heterogeneities of ageing (i.e. from the gas inlet to the gas outlet) have been evidenced. The latter was confirmed using a cathode catalytic layer segmented in 20 segments along the gas flow channel.