FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3099651096> ?p ?o ?g. }

• W3099651096 endingPage "101783" @default.
• W3099651096 startingPage "101783" @default.
• W3099651096 abstract "•O2 pathways in the PTLs of a water electrolyzer were observed with operando X-ray CT•Periodicity of the O2 transport pathway observed here with the period of 400 μm•O2 is taking preferential pathways through PTLs at various water flow rates and currents•Merging of oxygen front in the middle of PTL is observed due to good in-plane transport Understanding the relationships between porous transport layer (PTL) morphology and oxygen removal is essential to improve the polymer electrolyte water electrolyzer (PEWE) performance. Operando X-ray computed tomography and machine learning were performed on a model electrolyzer at different water flow rates and current densities to determine how these operating conditions alter oxygen transport in the PTLs. We report a direct observation of oxygen taking preferential pathways through the PTL, regardless of the water flow rate or current density (1-4 A/cm2). Oxygen distribution in the PTL had a periodic behavior with period of 400 μm. A computational fluid dynamics model was used to predict oxygen distribution in the PTL showing periodic oxygen front. Observed oxygen distribution is due to low in-plane PTL tortuosity and high porosity enabling merging of oxygen bubbles in the middle of the PTL and also due to aerophobicity of the layer. Understanding the relationships between porous transport layer (PTL) morphology and oxygen removal is essential to improve the polymer electrolyte water electrolyzer (PEWE) performance. Operando X-ray computed tomography and machine learning were performed on a model electrolyzer at different water flow rates and current densities to determine how these operating conditions alter oxygen transport in the PTLs. We report a direct observation of oxygen taking preferential pathways through the PTL, regardless of the water flow rate or current density (1-4 A/cm2). Oxygen distribution in the PTL had a periodic behavior with period of 400 μm. A computational fluid dynamics model was used to predict oxygen distribution in the PTL showing periodic oxygen front. Observed oxygen distribution is due to low in-plane PTL tortuosity and high porosity enabling merging of oxygen bubbles in the middle of the PTL and also due to aerophobicity of the layer. The polymer electrolyte water electrolyzer (PEWE) is an essential part of a renewable energy economy. It uses renewable electricity to convert water into oxygen and hydrogen (Carmo et al., 2013Carmo M. Fritz D.L. Mergel J. Stolten D. A comprehensive review on PEM water electrolysis.Int. J. Hydrogen Energy. 2013; 38: 4901-4934Crossref Scopus (2460) Google Scholar). The product of interest from water electrolysis is hydrogen, a fuel that can be easily stored as compressed gas, transported via natural gas pipelines, and then combusted as a fuel or converted back to electricity in polymer electrolyte membrane (PEM) fuel cells. The produced hydrogen gas has a specific energy density of 120 MJ/kg, whereas gasoline has a specific energy density of 44 MJ/kg; this difference in specific energy density values is what makes hydrogen an attractive alternative fuel, as hydrogen has approximately 2.7 times the amount of energy per kilogram than liquid gasoline (Mazloomi and Gomes, 2012Mazloomi K. Gomes C. Hydrogen as an energy carrier: prospects and challenges.Renew. Sustain. Energy Rev. 2012; 16: 3024-3033Crossref Scopus (667) Google Scholar). Although water electrolysis is a well-established method for producing hydrogen, there remain several knowledge gaps to maximize efficiency in converting water to hydrogen and oxygen while minimizing production cost (Babic et al., 2017Babic U. Suermann M. Büchi F.N. Gubler L. Schmidt T.J. Critical review—identifying critical gaps for polymer electrolyte water electrolysis development.J. Electrochem. Soc. 2017; 164: F387-F399Crossref Scopus (228) Google Scholar). The overall electrochemical water splitting is divided into two half-cell reactions. When a potential is applied across a PEWE, an oxygen evolution reaction (OER) occurs on the anode and a hydrogen evolution reaction (HER) occurs on the cathode of the electrolyzer (Wang et al., 2019Wang J. Ji L. Teng X. Liu Y. Guo L. Chen Z. Decoupling half-reactions of electrolytic water splitting by integrating a polyaniline electrode.J. Mater. Chem. A. 2019; 7: 13149-13153Crossref Google Scholar). For the OER, water flows in via the anode and transports through the porous transport layer (PTL) to reach the catalyst layer (CL). Once in contact with the electrocatalyst, two moles of water are split into four protons, four electrons, and one mole of oxygen gas. Byproduct oxygen exists the PEWE either as dissolved oxygen or in a gas phase. Residual oxygen bubbles can prove to be problematic during the operation of an electrolyzer, as it can block water from reacting with the electrocatalyst, thus decreasing the overall electrolyzer efficiency. Figure 1 shows a schematic of transport processes within a PEWE. Particularly, at high current densities, the mass transport losses are mainly caused by inefficient removal of oxygen from the electrolyzer (Suermann et al., 2015Suermann M. Schmidt T.J. Buchi F.N. Investigation of mass transport losses in polymer electrolyte electrolysis cells.ECS Trans. 2015; 69: 1141-1148Crossref Scopus (36) Google Scholar). In order to enhance the mass transport, the mechanisms of oxygen transport in the PTL need to fully be understood. As an effort to better understand the two-phase transport behavior in the PTL, studies (Abdin et al., 2015Abdin Z. Webb C.J. Gray E.M. Modelling and simulation of a proton exchange membrane (PEM) electrolyser cell.Int. J. Hydrogen Energy. 2015; 40: 13243-13257Crossref Scopus (121) Google Scholar; Dedigama et al., 2014Dedigama I. Angeli P. Ayers K. Robinson J.B. Shearing P.R. Tsaoulidis D. ans Brett D.J.L. In situ diagnostic techniques for characterisation of polymer electrolyte membrane water electrolysers - flow visualisation and electrochemical impedance spectroscopy.Int. J. Hydrogen Energy. 2014; 39: 4468-4482Crossref Scopus (91) Google Scholar; García-Valverde et al., 2012García-Valverde R. Espinosa N. Urbina A. Simple PEM water electrolyser model and experimental validation.Int. J. Hydrogen Energy. 2012; 37: 1927-1938Crossref Scopus (132) Google Scholar; Han et al., 2016Han B. Mo J. Kang Z. Zhang F.Y. Effects of membrane electrode assembly properties on two-phase transport and performance in proton exchange membrane electrolyzer cells.Electrochim. Acta. 2016; 188: 317-326Crossref Scopus (61) Google Scholar; Kadyk et al., 2016Kadyk T. Bruce D. Eikerling M. How to enhance gas removal from porous electrodes?.Sci. Rep. 2016; 6: 1-14Crossref PubMed Scopus (50) Google Scholar; Kang et al., 2018Kang Z. Yang G. Mo J. Yu S. Cullen D.A. Retterer S.T. Toops T.J. Brady M.P. Bender G. Pivovar B.S. et al.Developing titanium micro/nano porous layers on planar thin/tunable LGDLs for high-efficiency hydrogen production.Int. J. Hydrogen Energy. 2018; 43: 14618-14628Crossref Scopus (29) Google Scholar; Kang et al., 2017Kang Z. Mo J. Yang G. Li Y. Talley D.A. Han B. Zhang F.Y. Performance modeling and current mapping of proton exchange membrane electrolyzer cells with novel thin/tunable liquid/gas diffusion layers.Electrochim. Acta. 2017; 255: 405-416Crossref Scopus (41) Google Scholar; Kim et al., 2020Kim P.J. Lee C. Lee J.K. Fahy K.F. Bazylak A. In-plane transport in water electrolyzer porous transport layers with through pores.J. Electrochem. Soc. 2020; 167: 124522Crossref Scopus (5) Google Scholar; Lee et al., 2020aLee C. Lee J.K. Zhao B. Fahy K.F. Bazylak A. Transient gas distribution in porous transport layers of polymer electrolyte membrane electrolyzers.J. Electrochem. Soc. 2020; 167: 024508Crossref Scopus (16) Google Scholar; Lee et al., 2017Lee C.H. Hinebaugh J. Banerjee R. Chevalier S. Abouatallah R. Wang R. Bazylak A. Influence of limiting throat and flow regime on oxygen bubble saturation of polymer electrolyte membrane electrolyzer porous transport layers.Int. J. Hydrogen Energy. 2017; 42: 2724-2735Crossref Scopus (51) Google Scholar; Leonard et al., 2020Leonard E. Shum A.D. Danilovic N. Capuano C. Ayers K.E. Pant L.M. Weber A.Z. Xiao X. Parkinson D.Y. Zenyuk I.V. Interfacial analysis of a PEM electrolyzer using X-ray computed tomography.Sustain. Energy Fuels. 2020; 4: 921-931Crossref Google Scholar; Leonard et al., 2018Leonard E. Shum A.D. Normile S. Sabarirajan D.C. Yared D.G. Xiao X. Zenyuk I.V. Operando X-ray tomography and sub-second radiography for characterizing transport in polymer electrolyte membrane electrolyzer.Electrochim. Acta. 2018; 276: 424-433Crossref Scopus (41) Google Scholar; Lopata et al., 2020Lopata J. Kang Z. Young J. Bender G. Weidner J.W. Shimpalee S. Effects of the transport/catalyst layer interface and catalyst loading on mass and charge transport phenomena in polymer electrolyte membrane water electrolysis devices.J. Electrochem. Soc. 2020; 167: 064507Crossref Scopus (39) Google Scholar; Schuler et al., 2019Schuler T. Schmidt T.J. Büchi F.N. Polymer electrolyte water electrolysis: correlating performance and porous transport layer structure: Part II. Electrochemical performance analysis.J. Electrochem. Soc. 2019; 166: F555-F565Crossref Scopus (58) Google Scholar; Seweryn et al., 2016Seweryn J. Biesdorf J. Schmidt T.J. Boillat P. Communication—neutron radiography of the water/gas distribution in the porous layers of an operating electrolyser.J. Electrochem. Soc. 2016; 163: F3009-F3011Crossref Scopus (49) Google Scholar; Suermann et al., 2017Suermann M. Takanohashi K. Lamibrac A. Schmidt T.J. Büchi F.N. Influence of operating conditions and material properties on the mass transport losses of polymer electrolyte water electrolysis.J. Electrochem. Soc. 2017; 164: F973-F980Crossref Scopus (54) Google Scholar; Zlobinski et al., 2020Zlobinski M. Schuler T. Büchi F.N. Schmidt T.J. Boillat P. Transient and steady state two-phase flow in anodic porous transport layer of proton exchange membrane water electrolyzer.J. Electrochem. Soc. 2020; 167: 084509Crossref Scopus (22) Google Scholar) have focused on using imaging techniques to investigate the evolution and transport of oxygen in electrolyzers, including optical, neutron, or X-ray imaging, as well as computational studies. Optical microscopy was utilized by Dedigama et al. (Dedigama et al., 2014Dedigama I. Angeli P. Ayers K. Robinson J.B. Shearing P.R. Tsaoulidis D. ans Brett D.J.L. In situ diagnostic techniques for characterisation of polymer electrolyte membrane water electrolysers - flow visualisation and electrochemical impedance spectroscopy.Int. J. Hydrogen Energy. 2014; 39: 4468-4482Crossref Scopus (91) Google Scholar) in order to study two-phase flow in an operating electrolyzer. With a 7,000 fps camera, as well as a transparent sheet and specific backlighting, they were able to capture anodic two-phase behavior with high temporal accuracy. Following this study, Lee et al. (Lee et al., 2017Lee C.H. Hinebaugh J. Banerjee R. Chevalier S. Abouatallah R. Wang R. Bazylak A. Influence of limiting throat and flow regime on oxygen bubble saturation of polymer electrolyte membrane electrolyzer porous transport layers.Int. J. Hydrogen Energy. 2017; 42: 2724-2735Crossref Scopus (51) Google Scholar) employed a microfluidic platform, termed PTL-on-Chip, to study the effect of microstructure on the growth of oxygen bubbles, using an optical microscope. It was concluded that the morphology of the PTL has a significant impact on the governing force of the oxygen gas cluster growth and dictates the flow regime during PEWE operation. Neutron radiography has been utilized by Seweryn et al. (Seweryn et al., 2016Seweryn J. Biesdorf J. Schmidt T.J. Boillat P. Communication—neutron radiography of the water/gas distribution in the porous layers of an operating electrolyser.J. Electrochem. Soc. 2016; 163: F3009-F3011Crossref Scopus (49) Google Scholar) to visualize steady-state oxygen distribution in the PTLs. They were able to observe oxygen residence time in a PTL, specifically porous titanium (Ti), and show an equilibrium in the two-phase flow in the PTL. Their observations surprising at a time, suggested that the PTL would always be saturated with water, regardless of the current density, and that oxygen saturation did not change with current density in the regime of 0.1–2.5 A/cm2. Lee et al. (Lee et al., 2020aLee C. Lee J.K. Zhao B. Fahy K.F. Bazylak A. Transient gas distribution in porous transport layers of polymer electrolyte membrane electrolyzers.J. Electrochem. Soc. 2020; 167: 024508Crossref Scopus (16) Google Scholar) have investigated the dynamic gas transport behavior in the anode PTL by using operando synchrotron X-ray imaging. When they applied the current with a steep ramp-up and a shallow ramp-down, they concluded that the oxygen responds more rapidly, which means that the gas saturation in the PTL reached a steady state quickly. Zlobinski et al. (Zlobinski et al., 2020Zlobinski M. Schuler T. Büchi F.N. Schmidt T.J. Boillat P. Transient and steady state two-phase flow in anodic porous transport layer of proton exchange membrane water electrolyzer.J. Electrochem. Soc. 2020; 167: 084509Crossref Scopus (22) Google Scholar) have studied the effect of the two-phase flow behavior within the PTLs under steady state and dynamic load of the electrolyzer by using a neutron imaging with high spatial resolution (6 μm) and relatively high temporal resolution (1 s exposure time). They concluded that the two-phase flow in the PTLs is purely capillary driven for a wide range of operating conditions and that viscous forces are negligible. They further confirmed the findings of Seweryn et al. that water and gas distribution is not affected by current density (from 10 to 2000 mA/cm2). Leonard et al. (Leonard et al., 2018Leonard E. Shum A.D. Normile S. Sabarirajan D.C. Yared D.G. Xiao X. Zenyuk I.V. Operando X-ray tomography and sub-second radiography for characterizing transport in polymer electrolyte membrane electrolyzer.Electrochim. Acta. 2018; 276: 424-433Crossref Scopus (41) Google Scholar) were able to observe oxygen bubble formation and transport with X-ray computed tomography (CT) and radiography using operando hardware. This work demonstrated that as the current density increased, the residence time of oxygen bubbles in the channel decreased; this was expected, as higher current density would result in more oxygen being formed, enabling faster oxygen bubbles detachment. With Ti PTLs currently, it is experimentally not possible to differentiate between water and oxygen in the PTL, as Ti is highly X-ray attenuating material. Therefore, limited information is known on microscale distribution of oxygen in the PTL and how PTL morphology can be tailored to remove oxygen more effectively, as the in-plane vs through-plane transport properties can be rationally designed. The aim of this work is to investigate oxygen content in the PTLs and its relation to both PEWE current density and water flow rate, as well as understand whether oxygen takes preferential pathways when transporting through the PTL. So, a PEWE was set up with model carbon fiber PTLs (with similar morphological properties to Ti PTL) in order to directly observe and quantify both the oxygen content and preferential pathways oxygen could take as it exits PTL using operando X-ray CT. Using carbon fiber PTLs and short experimental imaging time enables direct observation of the steady-state pathways for oxygen transport within the carbon PTL. This is unprecedented, as no previous study has seen the pore-scale observation of oxygen content within the PTL with operando techniques. In order to distinguish oxygen within the PTL, machine learning was utilized to analyze the amount of oxygen present in the PTL of the electrolyzers at varying flow rates and current densities. The direct three-dimensional model-based lattice Boltzmann method (LBM) was used to better understand the physics behind the two-phase transport in the PTL under different cell operating conditions. X-ray CT was conducted at the Advanced Light Source, on beamline 8.3.2, at the Lawrence Berkeley National Laboratory. The optics used were as follows: 50 μm LuAg:Ce scintillator, 5x lenses, a sCMOS PCO Edge camera, and a double multilayer monochromator. The X-ray energy selected was 26 keV. The resulting images had a voxel resolution of 1.3 μm and a horizontal field of view of 3.3 mm. The X-ray CT images required sample rotation (180 degrees), and therefore, special care had to be taken so that the water inlet and gas outlets, as well as the thermocouples, did not become tangled. These parameters resulted in a scan time of ∼6 min. The catalyst-coated membranes (CCMs) used in this experiment were provided by NEL Hydrogen, Wallingford, CT, and consisted of CCMs with catalyst loadings of 3 mg/cm2 of platinum (Pt) on the cathode and 3 mg/cm2 of iridium oxide (IrOx) on the anode. Nafion 117 was used as a PEM. The PTLs were treated and untreated Freudenberg carbon papers (Fuel Cell Store, College Station, TX) without a microporous layer (MPL), for anode and cathode, respectively. Treatment was done in a piranha solution for five hours to make the PTL surface hydrophilic. The piranha solution treatment resulted in irreversible wettability modification, where the contact angle of the PTL was reduced from 133 ± 4° to about 0°. The PTLs completely took in water, indicating the completely hydrophilic surfaces. Furthermore, the PTL hydrophilicity was also confirmed with X-ray imaging as will be discussed later. Two Freudenberg gas diffusion layers (GDLs) with 50 % compression on the anode side were used to simulate morphology of the Ti PTLs. The compressed GDLs should have pore sizes of less than 10 μm in diameter, which is what the typical pore sizes for sintered Ti PTLs. In this study, treated Freudenberg at this compression has pore sizes comparable to the conventional PTL, which has the average pore size about 10 μm in diameter, as reported in Supplementary Information (SI), Figures S1 and S2. The main morphological difference between the carbon fiber PTL used in this study and Ti-based PTL is the different through-plane vs. in-plane tortuosities. The PTL studied here has an in-plane tortuosity of 1.3 (Figure S1D), whereas Ti PTL has in-plane tortuosity of 3.3–3.9 (Leonard et al., 2020Leonard E. Shum A.D. Danilovic N. Capuano C. Ayers K.E. Pant L.M. Weber A.Z. Xiao X. Parkinson D.Y. Zenyuk I.V. Interfacial analysis of a PEM electrolyzer using X-ray computed tomography.Sustain. Energy Fuels. 2020; 4: 921-931Crossref Google Scholar). The cathode had untreated Freudenberg GDL. The anode and cathode bipolar plates (BPPs) were made of graphite, as operating time of electrolyzers was only several hours and therefore low corrosion currents were observed from using carbon PTL and carbon BPPs. Our earlier study proved that it is possible to operate a PEWE with carbon PTLs for short duration without significant carbon corrosion (Leonard et al., 2018Leonard E. Shum A.D. Normile S. Sabarirajan D.C. Yared D.G. Xiao X. Zenyuk I.V. Operando X-ray tomography and sub-second radiography for characterizing transport in polymer electrolyte membrane electrolyzer.Electrochim. Acta. 2018; 276: 424-433Crossref Scopus (41) Google Scholar). Furthermore, recent benchmarking study across leading PEWE laboratories in the world also used carbon layer instead of Ti for PTL to conduct the benchmarking round robin study (Bender et al., 2019Bender G. Carmo M. Smolinka T. Gago A. Danilovic N. Mueller M. Ganci F. Fallisch A. Lettenmeier P. Friedrich K.A. et al.Initial approaches in benchmarking and round robin testing for proton exchange membrane water electrolyzers.Int. J. Hydrogen Energy. 2019; 44: 9174-9187Crossref Scopus (46) Google Scholar). Figure 1 shows the 3D volume rendering and the cross-sectional tomographs for the PEWE configuration. The 3D volume rendering displays the composition and morphology of the PEWE, as shown in Figure 1A. At the anode side, water was transported from the channel to the catalyst layer, and the oxygen product was removed from the catalyst layer. It is difficult to quantify oxygen content in the PTL by using only 3D visualization. The cross-sectional tomographs were created for ease of visualization and quantification. The relevant planes examined in this study are labeled and highlighted in Figure 1. The y-z plane is used to understand how the oxygen content changed within the length of the PTL, as shown in Figure 1B. By area-averaging information in the y-z plane, oxygen content under land versus that under channel was found. It is used mostly to quantify land-channel effects on oxygen distribution. The x-y plane is the through-plane view where one can observe both channels, the land, the cathode GDL, the membrane, the catalyst layer, and the anode PTL in the same slice, as shown in Figure 1C. A tomograph “slice” is essentially a single reconstructed image that is used to build up the 3D data. This front-facing slice was used to determine the oxygen content along the length of the anode side PTL. The brightest spots near both sides of the PEM are the catalyst layers. This is due to the high X-ray attenuation of the catalyst materials (Ir and Pt) making them very distinguishable. At the x-z plane, oxygen appears as dark, as it has the lowest X-ray attenuation on the anode side, as shown in Figure 1D. Water and carbon have comparable X-ray attenuation, so it is difficult to render them separately; therefore, a single phase will be used to identify both in this study. The x-z plane, which is the in-plane or top-down view, was used to train the Weka machine learning algorithm in order to interpret the oxygen content throughout the PTL. In addition to using the x-z plane for interpreting the oxygen content in the PTL, it was also used with the z-project function in ImageJ to build an average of the amount of oxygen and the catalyst distribution. The X-ray CT scan of an electrolyzer under open-circuit voltage (OCV) condition was performed to observe whether there is any residual oxygen trapped within the electrolyzer. Figure S5 compares a cross-sectional image of an identical location of an electrolyzer at the OCV and that at 1 A/cm2, where all the PTL is filled with water. This image proves that piranha solution treatment of PTLs was successful and the PTL is mostly hydrophilic. From this figure, one can observe that the PTL under the land is more compressed than under the channel. This is to be expected, as the PTL was compressed by 50 % to achieve pore sizes comparable to those of Ti PTLs. The cathode PTL shows a lower degree of compression than cathode as expected given the target of 20 % compression on the cathode. Observing oxygen transport pathways is an important step toward understanding the transport of evolved oxygen and supplied water through the PTL. Figure 2 shows the x-y plane of the PEWE during cell operation at 1 A/cm2 and 4 A/cm2 with varied flow rates of water. The electrochemical data showing stable potentials at these current densities are shown in Figure S6 and overall polarization curves by Figure S7 at three water flow rates. The tomography cross sections were selected as representative of the whole domain and they reveal that the oxygen pathways are visible within the PTL. Figures 2A–2C present the x-y plane of a cell that operates at 1 A/cm2 with the flow rates of 1 mlpm to 3 mlpm, respectively. False coloring on these cross sections provided an example of the oxygen transport pathways in the PTL. The result shows that oxygen content in the PTL did not change as water flow rate increased from 1 to 3 mlpm. There are several locations where oxygen goes all the way through from the catalyst layer to the channel or land. However, there are some locations showing that the oxygen is trapped in the PTL under the land area. Similar to the study by Leonard et al. (Leonard et al., 2020Leonard E. Shum A.D. Danilovic N. Capuano C. Ayers K.E. Pant L.M. Weber A.Z. Xiao X. Parkinson D.Y. Zenyuk I.V. Interfacial analysis of a PEM electrolyzer using X-ray computed tomography.Sustain. Energy Fuels. 2020; 4: 921-931Crossref Google Scholar), there is more gas observed under the land areas. Comparing Figures 2A–2C at 1 A/cm2 to Figure S5 at OCV, there is a significant amount of oxygen present in the PTL after the current density is applied compared to that at OCV, which will be quantified in the later sections. Figures 2D–2F show the x-y cross-sectional tomographs of the PEWE that operates at 4 A/cm2 with the flow rates of 1 mlpm to 3 mlpm, respectively. Comparing these operating conditions to 1 A/cm2 at all flow rates, these figures appear virtually identical, although current densities are four times higher for Figures 2D-2F compared to Figures 2A–2C. Oxygen seems to follow the same pathways at 4 A/cm2 as at 1 A/cm2. No new pathways emerged for oxygen removal. The oxygen content for the PEWE operating at 2 A/cm2 and 3 A/cm2 is shown in Figure S8. Although these images are only 2D cross sections, the oxygen transport seems to follow the same pathways through the PTL, regardless of current density or water flow rate for the range studied. Oxygen bubbles may accumulate and become trapped in the pore space of the PTL, resulting in mass transport losses. Figure 3 shows averaged oxygen content within the PTL as a function of the PTL thickness. The PTL/catalyst layer interface is located at x = 0, whereas the flow field is located at x = 320 μm. The oxygen content in the channel is uncertain due to difficulty to quantify oxygen content within the interface of PTL and channel/land, showing no obvious trend among flow rates, yet this does not significantly impact the computed oxygen content in the PTL, especially near the catalyst layer. Figure 3A shows the comparison of oxygen content for 1 A/cm2 at different flow rates. The oxygen content near the catalyst layer is between 30 and 40 %. Then, in the middle of the PTL, it decreases to about 15 %, and near the channel, it varies with flow rate. For the cell measurements at 1 A/cm2, the cell was not fully conditioned, so more oxygen near the catalyst layer was observed, compared to operation at higher current densities. At 4 A/cm2, oxygen content near the catalyst layer was 20–30 % and it decreased to 5 % near the flow field, after which it increased again when entering the channel, as shown in Figure 3B. The results for the other cell operating conditions were reported in Figure S9. From the overall measurement of oxygen content as a function of PTL thickness, the results show that there is a high oxygen content at the catalyst layer interface with the PTL. Gas transport pathways merge approaching the middle of the PTL. As mentioned above, some of the oxygen is trapped in the PTL near the land area but due to low in-plane tortuosity (shown by SI, Figure S1); for this type of PTL, the majority of oxygen is laterally removed from under the land into the channel. Figure 3C shows the oxygen content comparison with varied currents at the same flow rate of 2 mlpm. Again, there seems to be no correlation between current density and oxygen content within the PTL. Zlobinski et al. (Zlobinski et al., 2020Zlobinski M. Schuler T. Büchi F.N. Schmidt T.J. Boillat P. Transient and steady state two-phase flow in anodic porous transport layer of proton exchange membrane water electrolyzer.J. Electrochem. Soc. 2020; 167: 084509Crossref Scopus (22) Google Scholar) and Seweryn et al. (Seweryn et al., 2016Seweryn J. Biesdorf J. Schmidt T.J. Boillat P. Communication—neutron radiography of the water/gas distribution in the porous layers of an operating electrolyser.J. Electrochem. Soc. 2016; 163: F3009-F3011Crossref Scopus (49) Google Scholar) were able to quantify water saturation in the PTL via neutron radiography. Zlobinski et al. confirmed the earlier Seweryn et al. study that oxygen content in PTLs did not change with current density from 0.1 to 2 A/cm2. They showed that near the catalyst layer, the water and oxygen saturations were 0.5. In this study, oxygen content which is equivalent to oxygen saturation is approximately 0.2–0.4 near the catalyst layer. Oxygen content near the flow field is close to 0, in agreement with Zlobinski et al. Note that their PTL thickness is 1 mm, about 3 times higher than that used in our study, which increases the transport path length and leads to more oxygen accumulation near the catalyst layer. In the study by Lee et al., a 250-μm PTL was used in neutron radiography study (Lee et al., 2020bLee C.H. Lee J.K. Zhao B. Fahy K.F. LaManna J.M. Baltic E. Hussey D.S. Jacobson D.L. Schulz V.P. Bazylak A. Temperature-dependent gas accumulation in polymer electrolyte membrane electrolyzer porous transport layers.J. Power Sourc. 2020; 446: 227312Crossref Scopus (29) Google Scholar), and they observed ∼0.3 oxygen saturation near the catalyst layer, which is close to the value obtained in this study. To better understand oxygen transport in the PTL, oxygen content within the three through-thickness portions of the PTL was investigated. Figure 4 shows the comparison of oxygen content distribution within the PTL at 1 A/cm2 with the flow rate of 2 mlpm. The PTL was separated into three portions: near the catalyst layer interface (CL/PTL), middle of the PTL (mid PTL), and near the land/channel interface (PTL/channel), as shown at the top of Figure 4A. The through-plane view of oxygen content generated with the z-project (averaged from 3D volume and projected onto a single slice) is depicted in Figure 4A, where the volumetric information of oxygen content in the PTL was combined into a single image. Figures 4C–4E show the in-plane view of oxygen content in the portions of catalyst layer/PTL, mid PTL, and PTL/channel, respectively. The location near the catalyst laye" @default.
• W3099651096 created "2020-11-23" @default.
• W3099651096 creator A5013719269 @default.
• W3099651096 creator A5015137668 @default.
• W3099651096 creator A5028857805 @default.
• W3099651096 creator A5031596874 @default.
• W3099651096 creator A5038846590 @default.
• W3099651096 creator A5042306641 @default.
• W3099651096 creator A5053385848 @default.
• W3099651096 creator A5059345639 @default.
• W3099651096 creator A5072214326 @default.
• W3099651096 date "2020-12-01" @default.
• W3099651096 modified "2023-10-12" @default.
• W3099651096 title "Observation of Preferential Pathways for Oxygen Removal through Porous Transport Layers of Polymer Electrolyte Water Electrolyzers" @default.
• W3099651096 cites W1218133119 @default.
• W3099651096 cites W2001063720 @default.
• W3099651096 cites W2020007931 @default.
• W3099651096 cites W2036699254 @default.
• W3099651096 cites W2049484041 @default.
• W3099651096 cites W2182351320 @default.
• W3099651096 cites W2206671913 @default.
• W3099651096 cites W2338084981 @default.
• W3099651096 cites W2535125805 @default.
• W3099651096 cites W2567575592 @default.
• W3099651096 cites W2591485117 @default.
• W3099651096 cites W2736335655 @default.
• W3099651096 cites W2737033326 @default.
• W3099651096 cites W2762833272 @default.
• W3099651096 cites W2802303295 @default.
• W3099651096 cites W2808795202 @default.
• W3099651096 cites W2921701333 @default.
• W3099651096 cites W2943613824 @default.
• W3099651096 cites W2946098965 @default.
• W3099651096 cites W2989864659 @default.
• W3099651096 cites W2993399759 @default.
• W3099651096 cites W3002423420 @default.
• W3099651096 cites W3008781013 @default.
• W3099651096 cites W3013344484 @default.
• W3099651096 cites W3017590398 @default.
• W3099651096 cites W3034886663 @default.
• W3099651096 cites W3081068307 @default.
• W3099651096 doi "https://doi.org/10.1016/j.isci.2020.101783" @default.
• W3099651096 hasPubMedCentralId "https://www.ncbi.nlm.nih.gov/pmc/articles/7691172" @default.
• W3099651096 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/33294791" @default.
• W3099651096 hasPublicationYear "2020" @default.
• W3099651096 type Work @default.
• W3099651096 sameAs 3099651096 @default.
• W3099651096 citedByCount "29" @default.
• W3099651096 countsByYear W30996510962021 @default.
• W3099651096 countsByYear W30996510962022 @default.
• W3099651096 countsByYear W30996510962023 @default.
• W3099651096 crossrefType "journal-article" @default.
• W3099651096 hasAuthorship W3099651096A5013719269 @default.
• W3099651096 hasAuthorship W3099651096A5015137668 @default.
• W3099651096 hasAuthorship W3099651096A5028857805 @default.
• W3099651096 hasAuthorship W3099651096A5031596874 @default.
• W3099651096 hasAuthorship W3099651096A5038846590 @default.
• W3099651096 hasAuthorship W3099651096A5042306641 @default.
• W3099651096 hasAuthorship W3099651096A5053385848 @default.
• W3099651096 hasAuthorship W3099651096A5059345639 @default.
• W3099651096 hasAuthorship W3099651096A5072214326 @default.
• W3099651096 hasBestOaLocation W30996510961 @default.
• W3099651096 hasConcept C105569014 @default.
• W3099651096 hasConcept C127413603 @default.
• W3099651096 hasConcept C138679309 @default.
• W3099651096 hasConcept C147789679 @default.
• W3099651096 hasConcept C171250308 @default.
• W3099651096 hasConcept C17525397 @default.
• W3099651096 hasConcept C178790620 @default.
• W3099651096 hasConcept C185592680 @default.
• W3099651096 hasConcept C192562407 @default.
• W3099651096 hasConcept C2986395496 @default.
• W3099651096 hasConcept C42360764 @default.
• W3099651096 hasConcept C521977710 @default.
• W3099651096 hasConcept C540031477 @default.
• W3099651096 hasConcept C6648577 @default.
• W3099651096 hasConcept C68801617 @default.
• W3099651096 hasConceptScore W3099651096C105569014 @default.
• W3099651096 hasConceptScore W3099651096C127413603 @default.
• W3099651096 hasConceptScore W3099651096C138679309 @default.
• W3099651096 hasConceptScore W3099651096C147789679 @default.
• W3099651096 hasConceptScore W3099651096C171250308 @default.
• W3099651096 hasConceptScore W3099651096C17525397 @default.
• W3099651096 hasConceptScore W3099651096C178790620 @default.
• W3099651096 hasConceptScore W3099651096C185592680 @default.
• W3099651096 hasConceptScore W3099651096C192562407 @default.
• W3099651096 hasConceptScore W3099651096C2986395496 @default.
• W3099651096 hasConceptScore W3099651096C42360764 @default.
• W3099651096 hasConceptScore W3099651096C521977710 @default.
• W3099651096 hasConceptScore W3099651096C540031477 @default.
• W3099651096 hasConceptScore W3099651096C6648577 @default.
• W3099651096 hasConceptScore W3099651096C68801617 @default.
• W3099651096 hasFunder F4320306084 @default.
• W3099651096 hasIssue "12" @default.
• W3099651096 hasLocation W30996510961 @default.
• W3099651096 hasLocation W30996510962 @default.
• W3099651096 hasLocation W30996510963 @default.
• W3099651096 hasLocation W30996510964 @default.

• next