Gas Diffusion Layer Materials Research Articles

Investigation of PEMFC Performance and Property of the Gas Diffusion Layers Utilizing the Numerical Model

The overpotential at the cathode is significantly large in the proton exchange membrane fuel cells; therefore, the oxygen concentration in the oxygen reduction reaction (ORR) area is vital. Condensed water in the flow field and the gas diffusion layer (GDL) reduces oxygen transport to the ORR area. The enhancement of the mass transport of liquid water and oxygen is critical to improve fuel cell performance particularly at the high current density operations. It is required to reduce the total active area in the fuel cell stack and also it is leading to the significant cost reduction. Experimental investigation of oxygen transport has been limited by an inability to resolve water saturation-dependent properties. The two-phase thermos-fluid dynamics model of GDL was developed with Lattice Boltzmann Method (LBM) and it was integrated with computational fluid dynamics (CFD) based proton exchange membrane fuel cell (PEMFC) model to simulate the fuel cell performance with given design parameters and operational conditions [1, 2]. This work, utilizing the developed model, investigates the correlations between fuel cell performance and GDL structures and material properties. Some empirical studies were pursued and structure and property of the micro porous layer (MPL) in the GDL is correlated to the fuel cell performance particularly high current densities [3-5]. The model simulates the fuel cell performance with various porosity and permeability of MPL in the cathode GDL to investigate those correlations. Also the two-phase model can show the presence of residual liquid water in the cathode GDL which effects on the fuel cell performance. It can distinguish the residual liquid water in the GLD underneath of the flow field channel ribs and under the channels. Figure 1 gives an concept of integrated CFD based PEMFC model with micro-scale GDL model with LBM. Figure 2 shows residual liquid water in the cathode GDL with various operating conditions. In this fuel cell design, residual liquid water is correlated with cathode conditions, relative humidity, rather than the current density. It shows the potential capability of a model-based investigation of mass-transport to find solution of design and operational conditions in the PEMFC.

On the Limitations of Volume-Averaged Descriptions of Gas Diffusion Layers in the Modeling of Polymer Electrolyte Fuel Cells

As a complement to experimentation, numerical modeling is a key tool to optimize the design of Polymer Electrolyte Fuel Cells (PEFCs). Moreover, it plays an important role for the study of multiphysics, multiphase and multiscale phenomena, which are difficult to explore experimentally due to the small characteristic dimensions of the Membrane Electrode Assembly (MEA). In particular, thorough understanding of the coupled mass, charge and heat transport processes that occur in thin, heterogeneous, anisotropic, mix-wettability Gas Diffusion Layers (GDLs) is essential to develop new MEA designs with improved performance and enlarged durability [1,2]. In the PEFC literature, the most extended approach to model GDLs is volume averaging, where effective properties are used to describe the multiple transport processes that take place in this component. However, the applicability of volume-averaged formulations when modeling thin GDL materials has been questioned many times, and the error introduced in the numerical predictions is unclear. In this work, we explore the limitations introduced by volume-averaged formulations of GDLs under single-phase conditions. To this end, Lattice Boltzmann (LB) simulations were performed on X-ray computed tomography images of various commercial GDLs to assess the existence of a Representative Elementary Volume (REV). The tomographic images of the dry GDLs were acquired at Lawrence Berkeley National Lab’s Advanced Light Source (ALS) synchrotron (Beamline 8.3.2) with a resolution of 1.3 µm/voxel, and then segmented into void and solid phases using the algorithms described in [1,3]; the model was implemented in the open-source LB solver Palabos. Relevant effective transport properties were determined in both the in- and through-plane directions by direct numerical simulation, including effective diffusivity, permeability and effective electrical/thermal conductivity. Figure 1(a) shows the results computed for the through-plane effective diffusivity and thermal conductivity on an untreated Toray® TGP-H-120 sample using boxes of material with different size in the x, y, z directions (L x, L y, L z); the center of the boxes was always taken in the middle position of the original GDL sample. The in-plane size to obtain representative and consistent effective properties was found to be around 1 mm, in agreement with the microstructural analysis reported by Zenyuk et al. [3]. This length scale is comparable to the rib/channel width used in PEFCs, which is typically around 1 mm. Regarding the through-plane direction, there is not a representative size much smaller than the thickness of the material. Numerical results are only similar to previous experimental data when the full thickness (or almost the full thickness) of the material is considered. Therefore, it can be concluded that a REV cannot be rigorously defined in a GDL. In other words, there is not a piece of material with representative properties, whose dimensions are much smaller than the characteristic size of the cell in any direction. Given the impossibility to define a REV in GDLs, the error introduced by volume-averaged formulations was then evaluated by using a non-isothermal, single-phase, steady-state model of a 2x1 mm2 differential PEFC implemented in ANSYS® Fluent. In particular, we compared the predictions of a fully volume-averaged model with those of a pore-scale model accounting for the GDL microstructure. Only the core region of the GDL was included, and a macroscopic description was considered for the remaining components of the MEA (catalyst layers, microporous layers and membrane). The operating conditions were established at 80 ○C, 50% relative humidity, 1.5 bar and 0.5 V. As shown in Figure 1(b), the average results predicted by both models are similar, even though the spatial distributions differ significantly. The impact of the GDL microstructure on the current density distribution arises from the lack of separation of scales discussed above. The preliminary results show that GDL volume-averaged models are a good approximation to determine overall cell performance, as it has been traditionally assumed in the literature. This holds, at least, when the effect of interfacial regions is not considered. However, the local distributions predicted by volume-averaged models can be quite different from the actual scenario at the pore-scale level, in which heterogeneities can lead, e.g., to the formation of hotspots that may affect MEA durability. Further implications of the use of GDL volume-averaged models under multiphase conditions should also be investigated in future work.

Evaporation of Water from Gas Diffusion Layers

Evaporation of water from gas diffusion layers (GDL) is an important process for water management in polymer electrolyte fuel cells. At temperatures around 80 °C and locally under-saturated conditions, evaporation can have a significant impact on the GDL saturation. However, little is known about the evaporation rates of water in gas diffusion layers. First ex-situ results at room temperature and for low saturations have been reported in [1]. But for a better understanding of the evaporation rates under real operating temperatures, generic data is required at different temperatures and saturation levels. Using a special setup, which is applicable also for X-ray imaging of water saturation, water surface and GDL structure, evaporation rates are determined for well-controlled gas speeds (over the surface of the GDL),, temperatures of up to 80 °C and relevant liquid saturations. Figure 1 shows data (including reproducibility) for Toray TGPH 060 material as function of temperature up to 80 °C, at a capillary pressure of 10 mbar and gas velocity of 6 m/s. [1] I. V. Zenyuk, A. Lamibrac, J. Eller, D. Y. Parkinson, F. Marone, F. N. Büchi, and A. Z. Weber, J. Phys. Chem. C, 2016, 120 (50), pp 28701–28711 Figure 1: Evaporation rates from Toray TGPH 060 gas diffusion layer material a) as function of temperature at a gas velocity of 6.0 m/s; b) as function of gas velocity on the GDL surface at 60 °C; both for a dry gas stream and capillary pressure of 10 mbar; data shown based on preliminary set-up. Figure 1

Superhydrophobic PAN nanofibers for gas diffusion layers of proton exchange membrane fuel cells for cathodic water management

Proton exchange membrane (PEM) fuel cells are considered to be promising alternatives to natural resources for generating electricity and various other powers. Optimal water management in the gas diffusion layer (GDL) is critical to the high performance of fuel cells. The basic function of the GDL includes transporting the reactant gas from flow channels to the catalyst effectively, draining liquid water from the catalyst layer to the flow channels, and conducting electrons with low humidity. In this study, poly-acrylonitrile (PAN) was dissolved in a solvent and electrospun at various conditions to produce PAN nanofibers prior to their stabilization at atmospheric pressure at 280 °C for 1 h and carbonization at 850 °C for one more hour. The surface hydrophobicity of the carbonized PAN nanofibers were adjusted using superhydrophobic and hydrophilic agents. The thermal, mechanical, and electrical properties of the new GDLs showed better results than the conventional ones. Water condensation tests (superhydrophobic and hydrophilic) on the surfaces of the GDLs showed a crucial step towards improved water management in fuel cells. This study may open up new possibilities for developing high-performing GDL materials for future PEM fuel cell applications.

Microstructure-property relationships in a gas diffusion layer (GDL) for Polymer Electrolyte Fuel Cells, Part II: pressure-induced water injection and liquid permeability

The performance of polymer electrolyte fuel cells (PEFC) strongly depends on a controlled water management within the porous layers. For this purpose we investigate liquid water transport in a commercial gas diffusion layer (SGL 25BA) on the pore scale. X-ray tomography experiments combined with pressure-induced water injection provide 3D images of the liquid water distribution inside the GDL at incremental pressure steps between 0 and 100mbar.The breakthrough behavior of the liquid phase is highly anisotropic. In through-plane (tp) direction first bubble points appear at the outlet plane already at 5mbar and the ‘breakthrough' then evolves continuously over an extended pressure range up to >30mbar. For in-plane (ip) direction the breakthrough is discontinuous and takes place at 27mbar. Simulations of the intrusion process reveal that the different breakthrough behaviors are mainly triggered by different ip- and tp-transport distances. Short tp-transport distances through the thin gas diffusion layer (ca. 100μm) are responsible for the characteristic continuous tp-breakthrough behavior, which is thus attributed to a so-called short-range effect.Dedicated methods for 3D-image analysis adapted to fibrous GDL microstructures were presented in part I. With these methods we quantify all microstructure characteristics that are relevant for liquid permeability. These characteristics of pore and liquid phases include size distributions of bulges and bottlenecks, connectivity, effective volume fractions, geodesic tortuosity, constrictivity and hydraulic radius. Quantitative relationships are established between these microstructure characteristics and the liquid permeability, which provide a better understanding of the underlying microstructure limitations for injection and liquid transport.For the in-plane direction the liquid permeability is limited to roughly a similar extent by tortuosity, constrictivity and effective volume fraction. In contrast, for through-plane direction relatively low volume fractions of the liquid phase put stronger limitations to the liquid permeability than tortuosity, constrictivity and hydraulic radius.The curves for relative permeability vs. saturation (and vs. capillary pressure, respectively) achieved from 3D-analysis reveal complex but characteristic (reproducible) shapes with concave, linear and convex segments. The shape of these segments can be attributed to distinct microstructure effects. In contrast, the conventional macroscopic descriptions from literature cannot capture these complex shapes and the underlying microstructure effects. Future investigations with different GDL materials are necessary in order to understand whether these complex shapes for the relative permeability represent a general feature of gas diffusion layers or if they are specific to the investigated SGL material.

A method for measuring relative in-plane diffusivity of thin and partially saturated porous media: An application to fuel cell gas diffusion layers

A new experimental technique, extended from similar work on dry materials, is presented for measuring the in-plane components of the relative diffusivity tensor for partially saturated porous media. The method utilizes a custom-built holder and measures the transient response to oxygen concentration changes at the boundaries of a porous sample placed between two plates surrounded by a cooling block. The apparatus is kept close to the freezing temperature of water to ensure stable saturation throughout the experiment. Fick's second law is used to fit the transient change in concentration to a numerical solution to obtain the diffusion coefficient for samples of differing saturation. As expected the effective gas diffusivity is found to decrease with increasing water saturation of the media as the porosity is reduced and the tortuosity of the diffusion pathways increased. After extensive validation, this new technique is used to determine the relative in-plane diffusivity of some common fuel cell gas diffusion layer materials. The results are found to follow a power-law function dependent on the saturation consistent with previous modelling work. Samples without hydrophobic treatment are found to have lower relative gas diffusivity, compared with treated samples for the same average saturation.

Pore network modeling to explore the effects of compression on multiphase transport in polymer electrolyte membrane fuel cell gas diffusion layers

Abstract Understanding how compression affects the distribution of liquid water and gaseous oxygen in the polymer electrolyte membrane fuel cell gas diffusion layer (GDL) is vital for informing the design of improved porous materials for effective water management strategies. Pore networks extracted from synchrotron-based micro-computed tomography images of compressed GDLs were employed to simulate liquid water transport in GDL materials over a range of compression pressures. The oxygen transport resistance was predicted for each sample under dry and partially saturated conditions. A favorable GDL compression value for a preferred liquid water distribution and oxygen diffusion was found for Toray TGP-H-090 (10%), yet an optimum compression value was not recognized for SGL Sigracet 25BC. SGL Sigracet 25BC exhibited lower transport resistance values compared to Toray TGP-H-090, and this is attributed to the additional diffusion pathways provided by the microporous layer (MPL), an effect that is particularly significant under partially saturated conditions.

Incorporating Embedded Microporous Layers into Topologically Equivalent Pore Network Models for Oxygen Diffusivity Calculations in Polymer Electrolyte Membrane Fuel Cell Gas Diffusion Layers

In this work, a voxel-based methodology is introduced for the hybridization of a pore network with interspersed nano-porous material elements allowing pore network based oxygen diffusivity calculations in a 3D image of a polymer electrolyte membrane (PEM) fuel cell gas diffusion layer (GDL) with an embedded microporous layer (MPL). The composite GDL is modeled by combining a hybrid network of block MPL elements with prescribed bulk material properties and a topologically equivalent network of larger discrete pores and throats that are directly derived from the 3D image of the GDL substrate. This hybrid network was incorporated into a pore network model, and effective diffusivity predictions of GDL materials with MPL coatings were obtained. Stochastically generated numerical models of carbon paper substrates with and without MPLs were used, and the pore space was directly extracted from this realistic geometry as the input for the pore network model. The effective diffusion coefficient of MPL-coated GDL materials was predicted from 3D images in a pore network modeling environment without resolving the nano-scale structure of the MPL. This method is particularly useful due to the disparate length scales that are involved when attempting to capture pore-scale transport in the GDL. Validation was performed by comparing our predicted diffusivity values to analytical predictions, and excellent agreement was observed. Upon conducting a mesh sensitivity study, it was determined that an MPL element size of 7μm provided sufficiently high resolution for accurately describing the MPL nano-structure.

Wettability of Gas Diffusion Layers for Proton Exchange Membrane (PEM) Fuel Cells

Water transport within the gas diffusion layer (GDL) depends on the material’s wettability. In a PEM fuel cell, hydrophobic substrates facilitate water transport at the cathodes under high current density operation. We report a new approach to quantify the relative, ex situ GDL wettability by measuring the solid- liquid (S-L) interfacial area. We applied electrochemical techniques to measure the double layer capacitance, and exploited the proportionality between this quantity and the S-L interfacial area. We combined this electrochemical approach with traditional techniques (e.g., contact angle, and capillary pressure) to characterize commercial GDL materials. The measurements demonstrate that the fibre structure and morphology affect the wetting behaviour of the GDL materials. We describe the experimental approach and the underlying physical mechanisms, as well as the method’s applicability to characterize water intrusion processes for GDL materials.

Experimental Study of Thermal Conductivity and Compression Measurements of the GDL-MPL Interfacial Composite Region

Through-plane thermal conductivity and thickness variation under different compaction pressures were measured for a composite region of commercial gas diffusion layer (GDL), namely Freudenberg H1410 GDL, as received as well as saturated with a custom-made carbon ink/microporous layer (MPL). 3D X-Ray pictures were taken of the materials. Thermal conductivity for the composite region is higher than for the GDL alone. The compressibility of the composite region is in the same order of magnitude as pure GDL material. The found results motivate further investigation of this composite region that has gotten very little attention in the literature. A PEMFC consists of several components, i.e. the membrane electrolyte assembly (MEA) sandwiched between a thin MPL and a somewhat thicker GDL on each side. The MEA consists of a membrane (PEM) coated with catalyst layers (CL) on each side. MPL and GDL are often treated as separate layers in the literature. However, there exists a considerable interfacial region where the two different materials are intertwined. The fine material of the MPL can intrude considerably into the fiber structure of the GDL material. It is this region we are interested in. The temperature difference across the PEMFC can reach several °C despite being less than a millimeter thick between the gas flow plates. Temperature differences arise mainly across the GDL. Several research efforts have led to a good understanding of the thermal conductivity of the GDL and how it changes with compression, temperature, PTFE content, different fabrics, and water content. This work presents an effort to determine the thermal conductivity as well as the compressibility of the aforementioned composite region. [2] At a compaction pressure of 9.2 bar the thermal conductivity of untreated Freudenberg H1410 (114 μm), was found to be 0.111±0.009 W K-1 m-1 and for the custom-MPL-drenched Freudenberg H1410 material it was 0.124±0.005 W K-1 m-1. The untreated Freudenberg H1410 material was compacted to 87% of its original thickness at 9.2 bar compaction pressure. The MPL-treated H1410 material was compacted to 77% of its original thickness at 9.2 bar compaction pressure. The MPL-treated Freudenberg H1410 has a larger compression from atmospheric pressure to first compaction pressure, but a less steep compression gradient with rising compaction pressure than the original H1410 material, see figure. The thermal conductivity at 9.2 bar compaction pressure of another untreated Freudenberg GDL, namely H2315 (182 μm), was reported earlier to be 0.15±0.02 W K-1 m-1, roughly 50% higher than for the H1410 we measured. This complies with a trend seen in Toray paper, where the thermal conductivity at 9.3 bar compaction increases from 0.53±0.03 W K-1 m-1 for Toray TGP-H-060 (165 μm) to 0.65±0.02 W K-1 m-1 for Toray TGP-H-090 (265 μm) and to 0.81±0.03 W K-1 m-1for Toray TGP-H-120 (333 μm). [1] SIGRACET GDL 10 AA, also an untreated GDL, was measured to have a thermal conductivity of 0.38±0.03 W K-1 m-1 at 9.2 bar compaction pressure, also much higher than the Freudenberg GDL. A SolviCore PTL had a very similar thermal conductivity of 0.36±0.08 W K-1 m-1 at 9.2 bar compaction pressure. [1] The way the GDL is produced, the Teflon content, and the binder content all have influence on the final value of the thermal conductivity for a GDL material. Toray paper has a higher thermal conductivity when the thickness increases. An increase in PTFE content will lower the thermal conductivity. In general, Freudenberg GDLs have a lower thermal conductivity than other commercially available GDLs. By depositing MPL material into the GDL, thermal contact between GDL fibers is enhanced, hence the overall thermal conductivity of the material increases. This could be a means to improve thermal conductivity of GDL materials with low thermal conductivity to achieve better temperature management in the fuel cell. Acknowledgment The Advanced Light Source is supported by the Director, Office of Science, Office of Basic Energy Sciences, of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231. The authors would like to acknowledge assistance of Dula Parkinson with the experimental set-up at the beamline.

(Invited) Imaging Fuel Cell Components: From Flow Field Channels to Catalyst Layers

Commercially viable polymer electrolyte membrane fuel cell (PEMFC) systems for transportation and stationary power applications require optimal combination of performance and durability at a competitive cost. Optimized water management across a fuel cell is key to improving performance and maintaining high conductivity of the ionomer in PEM and catalyst layers (CLs), while preventing excessive flooding in CLs, gas diffusion layers (GDLs), channels, and manifolds. Durability is also greatly affected by water transport as certain degradation mechanisms are accelerated depending on the humidity levels. As well, catalyst layers that suffered carbon corrosion due to long operation and/or high number of starts/stops are more susceptible to flooding. Imaging in situ and ex situ enables crucial insight into the microstructure of the fuel cell components and materials, associated water transport, and their effect on cell performance, dynamic response, and degradation mechanisms. Several case studies are presented to highlight the use of optical imaging, neutron imaging, and micro X-ray computed tomography (microXCT) to investigate fuel cell components across a range of length scales. Laboratory-scale microXCT instruments are used ex situ to resolve the three-dimensional structure of the GDL and electrodes, before and after various accelerated stress tests. Optical imaging in operating fuel cells offers a cost-effective combination of high temporal resolution (tens of frames per second) and high spatial resolution (several µm). The imaging requires special cell design with visual access to the imaged area (GDL surface and channels [1] or catalyst layer surface [2]). We employed optical visualization in a variety of studies, such as to visualize liquid water dynamics in the cathode and anode flow fields, or at the interface between the CL and the GDL. Besides water transport studies, optical imaging can quantify in-plane gas distribution in the flow fields during startups and shutdowns [3]. Neutron imaging is a powerful tool to visualize and quantify water transport in operating fuel cells. It has a high sensitivity to small amounts of water inside a cell, while having high transmission through the common fuel cell hardware. At a lower spatial resolution of 250 µm, neutron imaging is used to measure in-plane water distribution with high temporal resolution and large field of view of up to 20 cm by 20 cm. We studied the performance with novel NSTF (nano structured thin film) electrodes and GDL materials [4], and also visualized water and ice formation when cell was subjected to repeated starts at sub-freezing temperatures [5]. Since neutron imaging provides water content integrated along the neutron beam, we combined neutron imaging with the optical visualization to study water transport in different flow field geometries [6] as well as in an operating electrolyzer [7]. Such approach with simultaneous imaging provides additional information about the water location, and in certain situations allows distinguishing between channel and GDL water, or between anode and cathode channels and manifolds. High-resolution neutron imaging is able to measure water distribution across the cell thickness at spatial resolution as high as 10 µm. Image processing procedure is developed to measure the water uptake and observe Schroeder’s paradox in situ in PEMs [8,9]. Further, properties of the microporous layer (MPL) of the GDL were manipulated to prevent excessive flooding in cathode catalyst layers. Anode flooding is evidenced by optical visualization, simultaneous imaging, and high-resolution neutron imaging. Increased water transport across the membrane, from cathode to anode, can be detrimental as liquid water on the anode side may cause localized fuel starvation and cause irrecoverable degradation, similar to accelerated carbon corrosion during unassisted startups and shutdowns. However, for certain novel cathodes, which are prone to flooding, e.g. nano-structured thin film (NSTF) electrodes and non-precious group metal (non-PGM) catalysts, water removal through the anode has proven to be a viable option to improve the performance by reducing cathode water content. The authors acknowledge support from US Department of Energy EERE FCTO, Los Alamos National Laboratory LDRD (Laboratory Directed Research and Development) program, Federal Transit Administration, US Department of Commerce, and NIST Center for Neutron Research.

A Study on the Effect of Morphology and Wettability on Water Saturation Parameters of Gas Diffusion Layers of PEMFCs

Water transport within a Polymer Electrolyte Membrane Fuel Cell (PEMFC) is important for performance and lifetime optimization. Specifically within the porous structure of the Gas Diffusion Layer (GDL), liquid water agglomerations can impede gas transport to the catalyst layer and thus affect the electrochemical reactions, finally resulting in reduced cell performance. Consequently, it is important to understand the formation of water agglomerations within the pores both from an experimental point of view and for modelling purposes. Concerning the latter, correct predictions on cell behaviour and performance require a reliable description of the processes relevant for water and gas transport. In this study, we tackle this issue both with experimental and modelling techniques. The results of the different approaches clearly indicate the need of an overall description comprising all relevant processes. Employing a dedicated, custom-built experimental setup, the GDL specific water saturation parameters under varying hydrostatic pressure can be measured using an X-ray tomography apparatus. The results obtained can be used to parametrize a water saturation model for Computational Fluid Dynamics (CFD) simulations and thereby lead to more precise results of such simulations, since a realistic CFD simulation of water transport in the PEMFC requires the information of water saturation of the GDL depending on the generated water pressure. In a CFD simulation software such as ANSYS Fluent®, the water transport within the GDL is often described by a model based on the capillary pressure – water saturation relationship. This model was firstly developed by Leverett1 for oil reservoirs in sandstone and was later complemented by Udell2. However, the standard parameters of the Leverett-function suggested by Udell are material-dependent constants and not only might differ from those of recent state-of-the-art GDL materials in general, but also should to be determined individually for each GDL type. In this study, the dependence of the water saturation level of GDLs on the applied water pressure is investigated employing X-ray tomography. This method also allows for a locally resolved analysis of the liquid water distribution within the GDL pores. For the detection of water droplets with a resolution of down to 2.5 µm, a new apparatus has been developed which permits applying water pressure up to 500 mbar on the surface of the porous material while acquiring the tomography data. In Figure 1, such a tomogram of a GDL (through plane and in plane cross section) recorded under 100 mbar water pressure is shown. Following another path, cell- or single-channel level CFD calculations employing such integral models for the GDL can yield the local temperature and relative humidity distributions within the porous material, which can then be used as inputs for Monte Carlo (MC) simulations which work on locally resolved structural data of the µm-scale pores. This technique is used to study the dependency of pore saturation of the GDL on structural and operating parameters. In this work, MC simulations were employed to determine the liquid water distribution in the GDL pores under realistic cell operating conditions. This voxel-based MC model (Figure 2) takes into account not only the movements of droplets subject to the surface properties of the solid and liquid phases, but also their evaporation and condensation. These simulations are based on real GDL structures including the different material compressions under rib and channel, surface characteristics and operating conditions. Previous results obtained with the same technique have shown good reliability when compared to in-situ experiments3. The present results indicate that to obtain a realistic description of the water distribution in a GDL under real operating conditions, both pressure-driven liquid water transport and condensation effects should be reflected together with the structure and wettability of the porous material. This study was supported by the German Federal Government within the project Optigaa II (03ET6015F). The aim of this project is to enhance the performance of PEMFCs by systematic GDL optimization References M. C. Leverett, Trans. AIME, 142, 152–168 (1941).K. S. Udell, Int. J. Heat Mass Transfer, 28, 485–495 (1985).K. Seidenberger et al., J. Power Sources, 239, 628–641 (2013). Figure 1

Impact of Liquid Water on Oxygen Transport Resistance in a PEMFC Cathode Via Limiting Current Methods and Concurrent Synchrotron X-Ray Radiography

The accumulation of liquid water formed by the electrochemical reaction of hydrogen and oxygen in a polymer electrolyte membrane (PEM) fuel cell causes a gas diffusion layer (GDL) to become effectively less porous and more tortuous. To achieve designs directed for water and gas transport in PEMFC diffusion media, a nuanced understanding of the nature and extent of GDL water as an impedance to oxygen flux is required. In this work, a limiting current approach was used to measure the mass transport resistance in a custom PEM fuel cell. Using a small active area at high stoichiometric ratios, it was assumed that the concentration of gases was constant along the length of the reactant flow channels, creating a uniform concentration gradient across the entire diffusion medium. If the fuel cells are then operated at the limiting current density, where cell potential approaches zero, it can also be assumed that all the reactants are being consumed and the concentration of reactants at the reaction site is effectively zero. Consequently, oxygen transport resistance in these conditions can be inferred directly from the limiting current density using the following equation, where RT is the transport resistance in s/cm, c0 is the oxygen concentration at the reactant channel and iL is the limiting current: RT = 4Fc0/iL Previously, through a thorough set of limiting current experiments at a range of operating conditions, Baker et al.1 showed that the carbon-fiber diffusion medium is the dominant contributor to oxygen transport resistance. Owejan et al.2 combined limiting current methods with in situ neutron radiography to demonstrate a strong qualitative correlation between GDL saturation and oxygen transport resistance. Synchrotron radiography has been established as a powerful tool for visualizing water distribution in PEMFC diffusion media3. In this work, synchrotron X-ray radiographs were obtained with a pixel size of 6.5 µm with a spatial resolution of 10 μm. GDL materials were imaged in operando at limiting current over a range of relative humidities (set in the inlet cathode gas stream). Figure 1 (a) is an example radiograph of a cell operating at limiting current density processed according to the Beer-Lambert law5,6. A distinct in-plane variation in liquid water distributions is visible between regions over ribs and regions over channels (Figure 1 (b)). This work exploits the high spatial resolution for capturing liquid water accumulation trends, information which is essential for building towards a direct, quantitative empirical relation between oxygen transport resistance and GDL saturation.

Influence of Binder Porosity on Gdl Gas Phase Transport

In polymer electrolyte fuels (PEFC) carbon fiber based gas diffusion layers (GDL) are used for fine distribution of reactant gases from the flow field channels to the catalyst layer (CL). It is a key component for PEFC as it serves also for water removal from the CL as well as heat and electricity transport between CL and the flow field plates. GDL materials can be classified as woven, felts, papers, where the later type requires a binder to mechanically stabilize the fiber arrangement. While most GDL binders are solid, the binder of SGL GDLs appears porous and it’s influence on gas phase transport is unclear [1]. In SEM surface images of SGL, GDL binder domains of appear as flake like fine-structure with particles of about 100 to 300 nm (see Fig. 1a). Here, we apply multi-dimensional X-ray imaging techniques and multi-scale numerical simulations to clarify possible contributions of the binder porosity to overall GDL gas transport. Absorption contrast X-ray tomographic microscopy measurements at the TOMCAT beamline of the Swiss Light Source (SLS) with voxel sizes of minimum 0.16 µm revealed a high porosity (59 %) of SGL 24 BA binder domains of (Fig. 1b). Computational transport simulations on the segmented binder structures showed, that the binder domains provide isotropic effective diffusivity of 0.30. In order to quantify the contribution of the porous binder to the overall fluid transport in the GDL, the obtained binder diffusivity values were used in multi-scale transport simulations of ternary segmented XTM data of representative GDL samples with 2.2 µm voxel size. If the binder was treated as solid, through-plane effective diffusivity of the GDL of 0.1 was obtained. If we applied a multi-scale simulation approach, that treats the binder domain as porous and uses the transport parameters of the binder domain as input values for the coarse scale simulations of the whole GDL domain, the through-plane effective diffusivity increased to 0.4 (+300 %). These numbers will be verified by X-ray ptychograptic computed tomography measurements of the binder fine structure at the cSAXS beamline of the SLS that proved resolutions of up to 16 nm [2].

Accelerated Degradation of Polymer Electrolyte Membrane Fuel Cell Gas Diffusion Layers: Performance Degradation and Steady State Liquid Water Distributions with in Operando Synchrotron X-ray Radiography

The polymer electrolyte membrane (PEM) fuel cell is a promising alternative power source for electric vehicles, forklifts, and stationary backup power due to its high power density, low operating temperature, and clean emissions. The gas diffusion layer (GDL) facilitates the transport of heat, reactant gases, water vapor, and liquid water in the PEM fuel cell. The GDL is a highly porous material composed of dense arrays of carbon fibers, resins, and polytetrafluoroethylene (PTFE). Additionally, it is typically coated with a micro porous layer (MPL), which results in the reduction of contact resistance at the catalyst layer interface and the overall improvement in water management. Effective water management must be achieved within the GDL over the entire operating lifetime in order to improve the commercial success of PEM fuel cells. Investigations into the effects of ageing on water management within the GDLs can be useful tools to guide this success. Water is required to maintain polymer electrolyte membrane hydration and ionic conductivity; however, excess water tends to accumulate in the catalyst layer, GDL, and flow channels. This accumulation limits mass transport and leads to performance losses. Therefore, effective water management in the GDLs is essential for improving cell performance, durability, and stability. Most GDL research has been focused on the impact of GDL materials and microstructure design on overall performance, rather than durability. Decreased GDL hydrophobicity has been found after only a few hundred hours of operation (1, 2). Therefore, a clear understanding of the influence of GDL degradation on the liquid water behavior is essential. As a tool to visualize this behaviour, synchrotron X-ray radiography is a powerful technique for visualizing liquid water transport behavior within an operating PEM fuel cell. High intensity X-rays generated at the BMIT-BM beamline of the Canadian Light Source were used for identifying liquid water inside the GDLs and capturing the multiphase flow behavior in an operating PEM fuel cell. Figure 1 is an example synchrotron X-ray radiograph of the fuel cell in operando, illustrating the spatial distribution of liquid water. The observed pixel brightness was correlated to the presence of liquid water, and the saturation profile across the GDL thickness was obtained. In this study, as-received fresh GDLs were immersed in a 35wt % solution of H2O2 at 90o C for 12 hours in order to perform an accelerated artificial ageing process. As-received fresh and artificially aged GDLs were then used to assemble specialized fuel cells for synchrotron imaging. Both the electrochemical performance and liquid water transport behavior were observed for a range of current densities. Additionally, the effect of changes in relative humidity on performance was compared to investigate the effect of artificial ageing. Notable differences in cell performance and liquid water transport behavior were observed at steady state operating conditions. The cell with aged GDLs showed lower cell voltage and higher quantities of liquid water than those of as-received fresh GDLs. The impacts of aging on the liquid water distributions within the GDL and on cell performance will be presented in this work. The objective of current work is to highlight the potential degrading effect on liquid water management in long term fuel cell operation.

Self-Standing Carbonaceous Sheets Composed of Micro and Nanometer Fibers Derived Bamboo and Application to PEFC and Edlc

Bamboo is a sustainable natural resource and the forests are considered to act as large carbon sinks, giving bamboo an advantage over other sustainable bioresources. However, it concerned that the rapid growth of bamboo may lead to biodiversity loss. This could be considered a serious disadvantage of bamboo in terms of its environmental impact. Bamboo has a fibrous microstructure and is composed mainly of cellulose, lignin, and hemicellulose, giving the isolation of cellulose fiber with relative ease. Owing to the unique microstructure of bamboo, we attempted to prepare self-standing carbonaceous fiber sheet by using bamboo as a starting material and have been investigated its application to polymer electrolyte fuel cells (PEFC) as the gas diffusion layer (GDL) [1].The Bamboo (Phyllostachys edulis, moso bamboo) cut in Japan was employed in this study. The basic preparation procedure of the self-standing carbonaceous bamboo fiber sheet (CBFS) was reported in [1]. In brief, CBFS was prepared viadelignification, defibration, molding, and carbonization. The bamboo fiber was obtained after the defibration process and then it was subjected to molding by using a plastic mold. After molding, the resultant fiber sheet was immersed in polyvinyl alcohol (PVA) aqueous solution for addition of PVA into the sheet. We also prepared CBFS without PVA. Finally, the sheet was subjected to carbonization under argon atmosphere. Here, the CBFS added PVA is denoted as CBFS w/ PVA. The microstructure of the CBFS was observed by field emission scanning electron microscope (FE-SEM) and was analyzed by X-ray diffraction (XRD) and Raman spectroscopy to elucidate the carbonization degree. The both of the in-plane and the through-plane of electrical conductivity were measured and the contact angle for a water droplet was measured to evaluate hydrophobicity of CBFS.A gas diffusion electrode was prepared by using CBFS, CBFS w/ PVA, the commercially available Pt-loaded carbon catalyst (Pt/C, TEC10E50E, Tanaka Kikinzoku Kogyo K. K., Japan) and Nafion ionomer solution by general manner. A single PEFC operation test was carried out by using Nafion117 membrane as the electrolyte and JARI’s standard PEFC single cell (the geometric electrode surface area was restricted to 1 cm × 1 cm).FE-SEM observation image shows CBFS is composed by aggregation of the carbonaceous fibers. The average diameter of fiber in the CBFS is ca. 6 μm irrespective to PVA addition, which is very close to that in commercially available carbon paper-based GDL as TGP-H-120 (Toray). In the case of CBFS w/ PVA, thin films derived from PVA were observed between each fibers and the surface of the fiber was seemed to be smooth, suggesting that PVA was partially covered with the bamboo fiber. The electrical conductivity was surely increased by addition of PVA.Figure 1 shows PEFC operation test results by using TGP-H-120, CBFS, and CBFS w/ PVA as GDL. The maximum power of CBFS was 135 mW cm-2, which is lower than that of TGP-H-120. On the other hand, the value was increased by addition of PVA and reached at 155 mW cm-2. Thus, the bamboo is expected as a potential GDL material. We also prepared cellulose nanofibrils (cellulose nanofibers; named as NFC or CFS) from the bamboo through chemical and physical treatments. Then, self-standing carbonaceous nanofiber sheet have been already obtained. The preparation procedure and its application to electric double layer capacitor (EDLC) will be also presented.Reference[1] T. Kinumoto et al., ACS Sustainable Chemistry & Engineering, Vol. 3(7), 1374−1380 (2015). AcknowledgementThis study was supported by The Environment Research and Technology Development Fund (ERTDF) from The Ministry of Environment, JAPAN, and the Venture Business Support Program from Oita University. The authors express gratitude to Mr. Yasuji Imahori Mr. Hideki Nakamizo, Mrs. Mayumi Nakamizo, and Mr. Yasuki Imahori of Silverloy Co. Ltd for supplying the bamboo strips. Figure 1

Heterogeneous porosity distributions of polymer electrolyte membrane fuel cell gas diffusion layer materials with rib-channel compression

In this study, X-ray Computed Tomography (X-CT) is used to study the structure of the bi-layered gas diffusion layer (GDL) of polymer electrolyte membrane (PEM) fuel cells. This work presents a unique, calibrated segmentation procedure developed in-house to identify the distinct components of the bi-layer GDL, isolating the carbon fibre, the microporous layer and the void regions as individual phases. The novel use of the areal mass and areal volume of the GDL facilitates calibrated, reliable and repeatable multi-component segmentation and removes the high degree of subjectivity that might otherwise be encountered in typical thresholding procedures. Samples with and without MPLs are studied for the impact of rib – channel compression on porosity profiles. Under the channel, the porosity profile is nearly identical to the uncompressed profile. Under the land region, there is a significant decrease in the GDL thickness. It is also observed that the majority of the compression is exhibited by the substrate region while the MPL undergoes significantly reduced compression.

Gas diffusion layer properties on the performance of proton exchange membrane fuel cell: pc-s relationship with K-function

Accurate capillary pressure-liquid water saturation relationship (pc-s) plays a key role in predicting the performance of proton exchange membrane (PEM) fuel cells and the two-phase flow characteristics in gas diffusion layer (GDL). In this work, a validated Leverett function (K-function) based on the experimental pc-s data for SGL24 and SGL10 series carbon paper GDL materials is incorporated into a three-dimensional and two-phase PEM fuel cell model to investigate the effects of gas diffusion layer properties on the cell performance. Special emphasis is put on Polytetrafluoroethylene (PTFE) loading (5–20 wt%) in the GDL, compression pressure (1–4 MPa) and micro-porous layer (MPL). The results show that the cell exhibits better performance when a higher PTFE loading of 20% and a lower compression pressure of 1 MPa are adopted. However, liquid water is hard to drain from the GDL with the 5 wt% PTFE loading due to the low hydrophobicity, especially under the ribs; for the high compression pressure of 4 MPa, small effective porosity caused by the pressure becomes the major resistance of water drainage. The existence of MPL is also very helpful to reduce the water saturation in the GDL.

Microscale characterisation of stochastically reconstructed carbon fiber-based Gas Diffusion Layers; effects of anisotropy and resin content

A novel process-based methodology is proposed for the stochastic reconstruction and accurate characterisation of Carbon fiber-based matrices, which are commonly used as Gas Diffusion Layers in Proton Exchange Membrane Fuel Cells. The modeling approach is efficiently complementing standard methods used for the description of the anisotropic deposition of carbon fibers, with a rigorous model simulating the spatial distribution of the graphitized resin that is typically used to enhance the structural properties and thermal/electrical conductivities of the composite Gas Diffusion Layer materials. The model uses as input typical pore and continuum scale properties (average porosity, fiber diameter, resin content and anisotropy) of such composites, which are obtained from X-ray computed microtomography measurements on commercially available carbon papers. This information is then used for the digital reconstruction of realistic composite fibrous matrices. By solving the corresponding conservation equations at the microscale in the obtained digital domains, their effective transport properties, such as Darcy permeabilities, effective diffusivities, thermal/electrical conductivities and void tortuosity, are determined focusing primarily on the effects of medium anisotropy and resin content. The calculated properties are matching very well with those of Toray carbon papers for reasonable values of the model parameters that control the anisotropy of the fibrous skeleton and the materials resin content.

Understanding Water Transport in Polymer Electrolyte Fuel Cells Using Coupled Continuum and Pore‐Network Models

AbstractWater management remains a critical issue for polymer electrolyte fuel cell performance and durability, especially at lower temperatures and with ultrathin electrodes. To understand and explain experimental observations better, water transport in gas diffusion layers (GDLs) with macroscopically heterogeneous morphologies was simulated using a novel coupling of continuum and pore‐network models. X‐ray computed tomography was used to extract GDL material parameters for use in the pore‐network model. The simulations were conducted to explain experimental observations associated with stacking of anode GDLs, where stacking of the anode GDLs increased the limiting current density. Through imaging, it is shown that the stacked anode GDL exhibited an interfacial region of high porosity. The coupled model shows that this morphology allowed more efficient water movement through the anode and higher temperatures at the cathode compared to the single GDL case. As a result, the cathode exhibited less flooding and hence better low temperature performance with the stacked anode GDL.