Below is the text version for the "Fuel Cell Durability and Performance through FC-PAD, a Multilab Collaboration" webinar, held on March 28, 2019.
Eric Parker, Fuel Cell Technologies Office
Good day, everyone, and welcome to the US Department of Energy's Fuel Cell Technology's Office webinar series for March. Today, we've got a great presentation from Los Alamos National Laboratory on FC-PAD, or Fuel Cell Performance and Durability Consortium.
My name is Eric Parker. I provide program support within the Fuel Cell Technologies Office, and I'm the organizer for today's meeting. We'll begin in just a second, but first, I have a few housekeeping items to tell you about.
Today's webinar, as you heard, is being recorded, and the recording, along with the full slide deck, will be posted online in the coming weeks. All attendees will be on mute throughout the webinar, so please submit your questions via the chat box that you should see in your WebEx panel on slide two. We will cover those questions during the Q&A at the end of the presentation.
With that, I would like to introduce today's DOE webinar host, Gregory Kleen, who is joining us here at Golden Field Office. Welcome, Greg.
Gregory Kleen, Fuel Cell Technologies Office
Thank you, Eric. So I guess first, I just want to remind everyone that the Fuel Cell Technology Office has two funding opportunity announcements open. The first one, which includes a topic on fuel cells for medium duty and heavy duty trucks, requires that applicants work with FC-PAD. That concept paper is due tomorrow, March 29th. And the concept papers for the H2@SCALE funding opportunity announcement are due April 8th. As a reminder, you must submit a concept paper in order to be allowed to submit a full application.
And with that, I'd like to introduce today's speaker. So Dr. Rod Borup has been a scientist at Los Alamos National Laboratory since 1999, starting as a postdoctoral researcher in 1994. Rod is the Los Alamos program manager for the fuel cell and vehicle technologies programs. He has worked on fuel cells for transportation at both Los Alamos and General Motors. He has 13 US patents, authored over 120 papers related to fuel cell technology, with over 8,500 citations and an H factor of 35. He has received numerous awards, including the 2016 DOE Fuel Cell Technology Office Annual Merit Review Award for fuel cells.
Rod is the director of FD-PAD, which includes five national laboratories: Los Alamos, Argonne, Lawrence Berkeley, Oak Ridge, and the National Renewable Energy Laboratory. So with that, I'd like to turn it over to Rod, and thank you for being here.
Rod Borup, Los Alamos National Laboratory
All right. Thank you, Greg and Eric, for the introduction. As Greg mentioned, I'm the director for FC-PAD, and part of this presentation will be given by Adam Weber from Berkeley National Lab, who's the deputy director. I want to make it clear, this is just not Berkeley and Los Alamos information. FC-PAD is a joint consortium between Los Alamos, Argonne, Berkeley, National Renewable Energy Lab, and Oak Ridge, and the capabilities and data in this presentation reflects all of FC-PAD. And so you'll see work from all the different members.
So in terms of the FC-PAD consortium, it's named Fuel Cell Performance and Durability, so we're not hiding what we're doing. Obviously, our objective is to improve performance and durability. That's why it's named that way. And primarily, we're looking at polymer electrolyte membrane fuel cells. Our work is predominantly at a pre-competitive level, so the things we do, we will most likely present and publish as we go along.
This consortium was formed in 2016, and a lot of the goal of the FC-PAD is to make this consortium interact with the different labs, different developers, component suppliers, to try to harmonize activities related to performance and durability for the Fuel Cell Technologies Office, and if I didn't say it, obviously, our funding comes from DOE EERE, the Fuel Cell Technologies Office.
So in terms of how we're structured in FC-PAD, I already mentioned the core consortium team. We are also working with four industrial projects that DOE awarded out of FOA 1412, and those are led by 3M Company, General Motors, United Technologies Research Center, and Vanderbilt. We work primarily on the MEA components, so when we've broken our work structure down, we have three components, thrusts, which include catalysts and supports, electrode layer, membranes, ionomer, GDLs, those types of different components.
And then we have cross-cutting thrusts, which includes modeling and validation of the different components, testing, performance, durability performance, and we do substantial amounts of component characterization in that. And I want to make sure people understand, these are not standalone thrusts. Each thrust is integral to the other parts of thrusts. So as an example, if we're taking catalysts and we do testing on them, we also do modeling on those components, and we also do characterization on those components, so no thrust is a standalone thrust. It's integrated in terms of both labs and components and cost cutting thrusts, in terms of how we work together.
In terms of some of the. people, I mentioned the fact that we have six thrusts. We do have management from DOE, which is led by Greg Kleen and Demetrius Papageorgopoulos. We take input from the U.S Drive Fuel Cell Tech Team, and we also have an associate steering committee. The steering committee members are made up of those four FOA projects, so sitting on our steering committee at the moment is Peter Pintauro from Vanderbilt, Mike Perry from UTRC, Andrew Haug from 3M, and Swami from General Motors.
In terms of our component leads, we – in terms of the thrust leads, we have thrust coordinators from all five of the different labs, including Deborah Myers from Argonne National Labs, K. C. Neyerlin from NREL, Ahmet Kusoglu from Berkeley National Lab, Karren More from Oak Ridge, Mukun from Los Alamos, and Rajesh Ahluwalla from Argonne National Lab.
So in terms of how we do our work structure, all of these people participate in terms of setting our priorities, and we do take the Fuel Cell Tech Team and the associate steering committee input on what it is we're working on.
So between these five labs, we have a lot of different capabilities, and all five of these labs are big scientific labs, so somewhere in these laboratories, we tend to have most all of the different capabilities that are out there. But really, what's important to FC-PAD are the people that are working in it, and I don't have time to go through all the different names, but I wanted to make sure we acknowledge all the people that work in FC-PAD. There's quite a number of staff scientists from each of the five different labs. We also have some research technologists. And then we use a lot of students and postdocs, so a lot of our staff members only participate part time in this, but that way, we can bring a lot of skills and background to cover all of the different components and characterization techniques.
So where does FC-PAD sit in terms of the landscape? We view our goal as understanding how the different components interact with each other, and how they do in terms of their own performance and durability. We work on a regular basis with OEMs and with component manufacturers. A lot of the goal there is to get the current state of the art materials. We want to hopefully be able to analyze these materials and understand exactly how they work and how they degrade, so that we can improve performance as things go.
We also work a lot with the academic world. In terms of the more fundamental research, we use a number of Office of Science user facilities, so that we can leverage the EERE capabilities with what already exists at the different national labs, and the whole goal of this is to eventually come out with understanding of these multi-component assemblies, how they interact between each other, and come out with design rules that then developers can use to either make new, better performing components, or more durable components, or OEMs can use so that they can make more commercializable systems. And I didn't say it, but generally, we work on the technology readiness levels of about two to four, and then once it's gone beyond four, that typically goes more to the industrial developers.
So in terms of our objectives, again, it's in our name. We work on performance and durability, recognizing that cost reduction is a very big part of the FCTO programs, so we're trying to improve performance, improve durability, and then simultaneously reduce cost. And most of our work is really working on the MEA components at a pre-competitive level. So our information, as we can we will publish and present work that we do for the FOA projects. That's up to them. But it can be done on a proprietary basis, and we do have agreements with those projects in terms of what we can say about the different components and our work for them.
So in terms of our knowledge base, PEM fuel cell materials and components, a lot of it, we're trying to understand the science of how you integrate these different components. Examples include things like how does the ionomer and the catalyst layer interact with the catalyst, and that includes the platinum and the carbon support for platinum. And then what goes on between the interfaces, between like the electrodes and the GDL, and electrodes and the membrane?
And what should come out of this is improved electrode structures where we have higher performance, reduced mass transport losses, and components that have improved durability. For example, things like enhanced membrane stabilization and more durable catalysts that have less degradation as a function of operating time.
So I mentioned our steering committee earlier. We did go through a process this last year where the fuel cell tech team, DOE, our associate steering committee members, all came up with suggestions, and then as a group we rated them all. And I'm happy to say that there was really very good agreement between the tech team, the FOA projects, and the national labs, on what should be our highest priorities. I'm not listing all of them, but this gives you an example of where we see the emphasis for FC-PAD should be going in the future, and that includes things like the catalyst layer structure, correlating the catalyst layer microstructure to the performance, using our characterization results, and then modeling to understand where all the different components in the catalyst layer are, and then evaluate the electrode transport properties.
We want to be able to look at the cathode catalyst layer and the various parameters using multiple methodologies, and to actually get these methodologies to agree, because we see a lot of characterization results that don't necessarily correlate with each other, so we want to understand what are the right characterization techniques, and what are the most consistent results we can get out. Therefore, we can make the best models to predict performance going forward.
We do a lot of understanding and characterization of the catalyst layer, such as FTEM mapping of ionomer and the catalyst layer and where it is for different systems, including different catalysts, alloy catalysts, different carbons. And then looking at starting with an ink and how the ink composition can change what processing and fabrication methods can do, and how these things lead into what eventually is the electrode microstructure and your PEM fuel cell.
In terms of performance and durability, we're looking at primarily alloy catalysts now, and trying to understand some of the things in terms of both degradation of the alloy and what some of those transition metals leaching out do to durability of the electrode catalyst, and then things like understanding resistances for oxygen and hydrogen diffusion, doing a lot of oxygen rate limiting measurements and hydrogen rate limiting measurements, and again, doing those for those different systems, so that we can come up with the information to design better electrode structures.
We always get a lot of interest at the national labs in new capabilities and new models, so there's emphasis on coming up with new both in situ characterization techniques and ex situ characterizations techniques, and different methods in situ testing-wise that can lead to more information that lead us to better performing, lower cost, more durable, high performing fuel cells. Those include things like higher resolution ionomer imaging, different spectroscopy methods to understand the catalyst layer better, better integrated predictive models that can be used to help people understand what it is they need to do to make better performing MEAs.
So I just wanted to touch briefly on FOA 2044. Greg mentioned this on his opening slides. This is out, and I guess concept papers are due tomorrow. And in terms of FC-PAD, similar to FOA 1412, we expect these capabilities in FC-PAD to be used to help those projects. However, at the bottom of the slide, I have to say that we are excluded to participate in these proposals and proposal development, so we're not supposed to discuss the open FOA. So we're trying to get information out to people so that they understand what capabilities exist in FC-PAD, so I'm suggesting three different things. You can go to our website. We've been trying to update that in terms of some of our capabilities. Obviously, we're doing this webinar today, and hopefully, these slides will be made available in a week or so to cover a lot of the different capabilities.
We did do a webinar in 2016. That presentation is out there on the web. And all of our presentations in terms of the Fuel Cell Technologies Office AMR, are published out there on the web under the FCTO website, so you can go take a look at those.
I'll also say that we have a very good publication record in FC-PAD. During 2018, I believe our count was something like 35 publications, so you can look at the primary staff members in FC-PAD and go see a lot of what we've published out there. So hopefully, there's a lot of information out there in terms of what our capabilities are.
This is a reasonably old slide. We did update it a little bit just recently, just to give you an idea of some of our capabilities that we've gotten a lot of interest in. And again, it varies from the catalyst support, electrodes and MEAs, the membrane and ionomer, and overall cell modeling and evaluation, and trying to look at it in terms of modeling and theory versus characterization versus in situ testing.
We do do a lot of TEM at Oak Ridge National Laboratory. They do a whole lot of microscopy, ionomer mapping, and that's used for most of the projects that we support. We have access to a couple of different beam lines, including the APS and ALS. And this is talking about some of the SAXS measurements that are done at APS, at Argonne National Lab, looking at particle size and carbon agglomerates, and they also do a lot of online analysis, testing of catalyst degradation at NREL. They do a lot of MEA fabrication, a lot of advanced MEA diagnostics, including a lot of surface area analysis and understanding effects of ionomer coverage and oxygen limiting and hydrogen limiting current measurements at Los Alamos.
We again do characterization of electrode structures. This is some ASM work. We also do some MEA fabrication, component diagnostics, and do a lot of performance and durability testing.
And then the modeling primarily exists at Berkeley National Lab and Argonne National Lab, and you shouldn't take this list of pictures as all the capabilities at the labs. It's just meant to give you an overview. And if I go back to the 2016 webinar, this was a great big long list of capabilities. As I go through this list, I'm not going to read it all, it should give people an idea of some of the capabilities. A lot of these we have been using in FC-PAD, but there are some that I've looked at that really haven't been used recently, so there's a lot more capabilities out there that we actually have in these five national labs that we don't necessarily use.
This was a good case study for this discussion. So in the last year to year and a half, we've had a collaboration with USCAR. So this is sort of the home of the Fuel Cell Tech Team. And this was some stuff they asked us to do with DOE's permission. They gave us some relatively new cells and aged cells out of commercially bought Toyota Mirais, and they did their own testing, and then they gave us cells so that we could understand what these state of the art materials look like.
And this is to help DOE benchmark what these materials are out there, and I'll have a one slide brief summary of it. But this is to help developers understand this is where we already are, so if you want to make a better performing catalyst, it really needs to be benchmarked across – against some of these commercial state of the art materials that are already for sale out there.
So in our work with USCAR, these are actually the list of techniques that we used just for this little collaboration. So it gives you an idea of the various techniques, such as SEM and EDAX mapping of components and TEM and TEM particle-size distributions of the MEAs, so I've got the different components listed here so you can see what we're addressing how, and then what type of information we're trying to get out of it. So this, again, is not necessarily a complete list, but these were all the techniques that were just used in that one collaboration. So hopefully, it gives you an idea of the list of capabilities that FC-PAD has in these five laboratories.
So just some results from this collaboration with USCAR, looking at the commercial Toyota Mirai materials, they are using a platinum cobalt alloy catalyst at a loading of approximately 0.32, and we're listing what the cathode catalyst layer was. We did look at cells that have been tested for about 300 hours versus about 3,000 hours, and Toyota has done a very good job of operating their vehicle systems so that the durability of the catalysts and things like the membranes are very good, so they've got a lot of system mitigation strategies out there. We also did some testing with some of the Fuel Cell Tech Team ASTs, and we presented some of these results in last year's Annual Merit Review, also at the ECS in a couple of our presentations. In terms of things like the anode, it looks like a reasonably standard anode platinum catalyst. In terms of GDLs, they're doing some things slightly different. Primarily, they're putting things like seria in the micro-porous layers to most likely get enhanced membrane durability.
In terms of the membrane, it appears to be a nafion type side chain on a (inaudible) acid with an expanded PTFE, or what looks like expanded PTFE, mechanical reinforcement. And it's a 10 micron thick membrane, so that's something developers should be aware of. It is using very, very thick membranes is probably not very competitive with some of the commercial materials that are out there.
And then of course Toyota has published this themselves in terms of what their flow fields look like, and that they're using titanium foil and titanium mesh for the flow field distributions. And we did do some imaging with the – with NIST, National Institute of Standards and Technologies, which is Department of Commerce lab, just looking at where water is sitting in terms of the flow field. And I don't want to go into too much detail about that, but it gives us an idea of where liquid water is sitting, so we can take a look at those different components in terms of degradation.
So I'll get into more of FC-PAD's core research. In this slide, we've sort of tried to divided it up between characterization and diagnostics and modeling, and where we're trying to look at the fundamental critical phenomena that we feel is controlling performance in the electrode layers, and that includes doing a lot of characterization of the catalyst by things like TEM and by XAPS, in this case, understanding the platinum-platinum and platinum-cobalt bonding, and how much of the surface is oxidized by the beam line at APS, looking at thin-film interactions of ionomer between ionomer on platinum and ionomer on carbon, and this is some grazing incident scattering at ALS. Ionomer mapping is something that is very critical to understanding how to get the best performance in terms of catalyst layers. Understanding things like cation migration in terms of cations leaching out of catalysts, and cations that we use for radical scavenging to extend membrane lifetimes, doing both ex situ and in situ monitoring of cation migration.
We do do a lot of testing, including both performance and long term ASTs and drive cycle testing. And then this data, again, is modeled from XCT analysis, understanding the carbon agglomerates, what the nanoscale looks like, and then using those structures to understand what the component transport level is, all the way up to the cell level, and then understanding durability, including things like interacting durability mechanisms, such as membranes swelling and chemical degradation in membranes.
So work that we've done this last year is related to conditioning and recovery, and so this is some of those fundamental results. And looking at a conditioning protocol, we'll see that the performance didn't increase a whole lot, but after we did voltage recovery protocols, you can see that we got much better performance out of this, and this was a platinum cobalt on high surface area carbon. The catalyst was from Umicore, in this case. And when we correlate that to some of the transport limiting mechanisms, this is the non-[inaudible] resistance. During the break-in procedure, we saw a big reduction in the transport limiting resistance, even though we didn't really see a big improvement in terms of the overall performance. But then once we did the voltage recovery steps, that's when the performance really improved in terms of performance.
So we like to do these types of tests and do characterization as we go along, and so this is some of the TEM that came out of Oak Ridge National Laboratory, and SAXS out of Argonne National Laboratory, looking at the particles during these types of things. And we do see that there's a decrease in the electrochemical surface area for both platinum on Vulcan and platinum on high surface area carbons. BOL is beginning of life, and EOT is end of test. So we did see ECSA loss, although we got a much higher performance after we did the conditioning and recovery protocols.
And that's backed up by the SAXS, and you can see the particle sizes increases in terms of the number of fraction and diameter for both the Vulcan support and the high surface area support. And looking at the data, including the microscopy, it appears that a lot of this is due to particle size coarsening through platinum dissolution reprecipitation and the Ostwald ripening and coarsening of the different particles.
So in terms of durability, there's a lot of information on this slide in terms of what we do, but we do do a lot of durability testing. That includes long term drive cycles, shutdown/startup simulations, and accelerated stress tests. Data from that does go to things like the modeling, where they do the voltage loss breakdown, model the degradation rate, and evaluate transport properties. We try to take the most up to date state of materials from the different component thrusts.
And these materials, both before the testing and after the testing, go through our suite of characterization techniques, so we can best understand what's going on with the materials. We're always looking at trying to define and develop new accelerated stress tests, so that they're life predictive. And this is probably some work scope that we will do related to medium and heavy duty vehicles going forward. We haven't refined AST as much in the last year, but we've done that previously.
And then, again, going back to people would like new operando evaluation capabilities, and so we've spent time doing that, and a couple of the different things that we've done this year that'll be talked about in our annual merit review this year, but not in this presentation, is things like confocal XRF, looking at cation migration, and surface coverage by CO displacement, looking at how much the ionomer covers the electrodes.
So just an example of some of the durability measurements, looking at platinum cobalt, in this case, from different loadings of .15 down to .1 down to .05. You can see there's much higher degradation at the very low loadings. You can see that mass activity and ECS decrease as a function of these AST cycles, and these are the square wave catalyst AST cycles that are recommended by the Fuel Cell Tech Team.
And then we take this data and model it, and this is a voltage loss breakdown that comes out of Argonne National Lab, where they've modeled the different loadings and tried to understand what's going on in terms of how much of it is kinetics and how much of it is total mass transport. And this is primarily representing the total mass transport, and so you can see that the biggest change really is at the very low loading and the ionomer film resistance, that transport resistance goes way up. And again, at the beginning of life, you have a bigger overall mass transport resistance for the lower loadings than the higher loadings.
So with that, I'm going to turn it off to Adam Weber for the next set of slides, and he can talk about the electrode layers and a lot of the modeling that's going on in FC-PAD. Adam?
Adam Weber, Lawrence Berkeley National Laboratory
Thanks, Rod. And so as you saw with the durability, we also take a very holistic and kind of systematic approach when we look at things like catalyst layers and catalyst layer formation. So that's everything from model studies of films and inks to advanced characterization diagnostics. And then we kind of enter into an iterative scheme with the multiscale modeling, the performance diagnostics, to really understand and optimize what the overall structure is looking like.
So if you go to the next slide, so I'll go through a couple of examples here, at least of kind of our approach. So one of them that we've taken recently is really to start to understand what aggregation and agglomerates are, and not only kind of in the catalyst layer, where we can visualize them to a certain extent, but also to overall how they form and where they start from.
So this is an example of just saying, you know, if we take a simple system, which at least in the fuel cells is already a three component system, then ionomer and ink, so on the previous slide, and so what you can see is we can see as we do the cryo-electron microscopy, as we change the solvent ratios, we're getting different aggregation of just the ionomer chains.
And then we've actually been able to do this in an operando way at the advanced light source to where we can actually look at the film formation properties, and so that's kind of where crystallinity and where is the onset of crystallinity, what happens as we go through solvent drying, gelation, and eventually annealing in the system. So we're getting kind of a better handle on formation.
And then, of course, we can go to more complicated systems, which is shown on this slide, of when we start looking at actual catalyst layers and actual inks, and so now we're at perhaps a five or six component type system. And so we have different ways that we've looked at that. We have the electron microscopy, to look at things like the ionomer and the platinum distribution, atomic force microscopy, as well to look at kind of the larger ionomer aggregates. And then x-ray computed tomography to look at kind of the overall structures and then reconstructing those to see what they might look like.
In the SAXS, we have a wide range of link scales that we actually probe, so everything from kind of carbon-carbon agglomerates or aggregates to agglomerates of those aggregates, to actually just the platinum particles as well, in kind of more the XRD type of range.
And so if we look at kind of things that we've done recently, and so I'll just provide a couple of quick case studies, one is looking at kind of the solvent ratio and looking at its effect on performance. And so when we saw with just the ionomer studies as well as with some of the ink studies, as we're going from an alcohol rich to a water rich, we're getting kind of different distributions of kind of the carbon-carbon coming together, where the ionomer might sit, such that if we're in kind of very alcohol-rich, we have these kind of – these structures that are kind of inverted micelles, and then as we start to add water, it starts to open up the side chains, and those allow more preferential interactions.
And so you see there in the data, with the CO displacement, we're getting an increase in kind of surface coverages and surface interactions as we go to more water-rich cases, and then if we actually just look at this as a performance at a given voltage, we can see that we're getting a peak performance in kind of a certain water ratio, and this is all with high surface-area carbon.
We can actually look at this in terms of the non-(inaudible) resistance as well, or kind of the very local resistance, and once again, we see that tradeoff to where the way these inks and the way the formation and the conditioning is causing changes to the aggregation, such that we need kind of more connected pathways. But of course, if we're getting too much interactions, we start to shut down the very local transport resistance as well.
And so this is just an example of kind of the way FC-PAD's approached kind of in a systematic way to look at things like ink solutions and solvents and ionomer/carbon ratios.
And so as mentioned, we can take a lot of the diagnostic information, and we can actually put that together to reconstruct at least virtually what the catalyst layer looks like. And so we have information of kind of what we actually put into the inks in terms of weight loadings. We have information in terms of the microscopy. We know the platinum distributions, the platinum size distributions. From XCT, we kind of have at least at the nanoscale some of the larger link scales, and then we can put that into a model type system to reconstruct what kind of the fine features might be that we can't currently image, but something that we're actually trying to go towards to image.
And so then what do we do with kind of these structures once we actually have them? And so one thing that we can do is microstructural modeling, and this is actually looking at transport through one of these agglomerates or multiple of these agglomerates at the kind of – that have been reconstructed, and at that kind of scale. And so what we can do is basically say at different operating conditions what the effectiveness factor is, so how much utilized is the actual agglomerate structure, whether that's in terms of the platinum activity, or whether that's also accounting for transport resistances in the film.
And so we can take that information, which is at the very local scale, and start to scale that up to be able to predict overall performance. And so we take that effectiveness factor information, we take that with the agglomerate size distributions that we can pull off from kind of those reconstructed catalyst layers, based on the experimental data, and then we have of course the distributions of porosity or ionomer films. And all of that can be upscaled into a more macroscopic model, where we're accounting for all the water transport, where we're accounting for the thermal management across multiple layers. And then we can start to predict performance.
And so what you see here is a nice way that we've taken kind of microstructural data that's very detailed and figured out ways to upscale that to macroscopic type observable, to really understand what is limiting in terms of performance and durability under different operating conditions.
Now of course, we can look at the inverse of that, too, and use the models to help guide what optimal structures might be within catalyst layers. And so as you see on the bottom here, it's very difficult, and as we saw, it's very difficult to really look at some of these distributions, and it's difficult to really determine what we actually want.
And so we've come up with ways to think about, well, what if we have engineered pathways for optimal water transport, for optimal gas transport, for optimal ion transport, and try and minimize things, like ionomer film resistance, or actually really any kind of local resistance that might be occurring.
And so what you see on the next slide is kind of different structures that we've come up with, so either an array electrode or a kind of nanowire-type electrodes. And so FC-PAD is currently working on kind of optimizing the synthesis of these, but you can see on the right that we can make these on kind of the micron link scales that we know we're going to need for the catalyst layers, and where we can actually start to then dial in the actual transport processes that we want, such that we can mitigate some of the limitations that we see that occur within the operating parameters that we're operating with.
And so now I'll pivot a little bit, and so as Rod kind of mentioned, we have been supporting four FOA projects led by 3M, GM, UTRC, and Vanderbilt. And so what we wanted to do at kind of the tail end of the presentation here is to give people a flavor for what we're actually doing very specific to those actual projects.
And so if we start on this slide of looking at what we're doing for the 3M project, it's a lot of component characterization as well as cell analysis. And so what you see in the table in the upper right, we're looking at processing conditions and the catalyst layer structure, when we go towards things like their dispersed NSTF, and so this isn't the traditional NSTF. This is kind of dispersing it within a thicker catalyst layer.
And then also using their advanced ionomers, so whether that's like a PFIA or kind of multiple (inaudible) side chain ionomers, what we see when we characterize these, we see that within the electrode and within the inks, we're getting much wider distributions of agglomeration and kind of larger agglomerates.
And so shown down on the bottom middle is we can actually show that as you're moving towards the dispersed NSTF and these PFIA ionomers, you're actually getting better or lower resistances at the very local level for transport. And so what we see then is from a diagnostic standpoint, what it looks like is these layers from the impedance analysis seem to be a little bit more proton limiting, rather than traditional oxygen-infusion limiting, so that's giving ideas for kind of ways to optimize these things.
And then as I mentioned, what we saw with the better transport resistance with the PFIAs, what we see specifically is when we characterize the catalyst layer resistance, we have kind of an interfacial component which is due to the way the ionomer is interacting with the platinum sites, as well as kind of the transport perhaps through an ionomer film. And so that's what's shown on the bottom left in terms of the hydrogen transport resistance.
And then we've actually been able to correlate that to things like how much it swells, so the fact that the PFIA has higher free volume seems to correlate very well with perhaps better oxygen transport, and the higher domains means it's interacting a little bit more with kind of the water-type domains within the catalyst layer, or within the ionomer thin films and the ionomer structures, rather than the ionomer side chains. And so we are seeing the benefits from the thin film studies to rationalize what we see from the overall performance studies, and this is providing 3M with critical information in terms of what to look for in terms of next generation ionomers, as well as what to optimize in terms of their catalyst layers.
So one of the other projects we're looking at is supporting Vanderbilt, and Vanderbilt's project is related to making novel membrane electrode assembles with ultra-low platinum loadings using nanofiber electrodes. So they're electrospinning nanofiber electrodes, so a lot of the different measurements that FC-PAD has been supporting Vanderbilt in is understanding what these electrodes look like, and then helping support them with some of our in situ diagnostics.
So what's here on the left is quantifying the kinetics and transport losses by impedance, looking at the beginning of life, and then conducting, again, various accelerated stress durability tests, and then looking at how those durability changes as a function of time. In their case, they're using pretty novel nanofiber electrodes, so there's been a lot of microscopy done on them, and understanding where the platinum is going, and the fiber is going. And again, understanding just what the nanofiber diameter is like at the beginning of life and at the end of life.
And then, again, we've supported some of their work by using the neutron imaging beam at National Institute of Standards and Technology, doing high resolution water visualization, where water is visualized in terms of more standard electrodes versus these nanofiber electrodes, so that we can understand better the water transport that's going on in these nanofiber electrodes, because that's really some of the advantage of them, is to get better water management and better durability of these nanofibers, which we have seen in terms of how this project has gone forward.
One of the other projects is United Technologies Research Center, and a lot of their project is looking at making high performance dispersed-type electrodes, but really taking a look at the current electrodes, really understanding the carbon agglomerates and where ionomer is sitting in them, and then developing at a pretty basic level different structures in terms of how the ionomer interacts with the carbon.
So a lot of the FC-PAD support to the United Technologies project has been things like taking a look at these electrode structures, and we've done things like basic electrodes, and then diluted electrode structures, and then digitizing the images so that we can understand what the carbon agglomerates and the aggregates look like in those structures, both from a delimited basis and with standard electrode structures. And then that data has been going back to UTRC's modeling efforts, where they're developing better models in terms of predicting what the structure should look like, and then developing those electrodes in their process.
And again, what's here on the right is looking at the different agglomerates, equivalent diameter and number of agglomerates that you get during – in some of these structures, and then what the pore volume and how that changes from different MEA structures to MEA structures. And again, looking and measuring what the catalyst layer porosity is, in this case, using AFM to try to understand what the ionomer thickness is on some of these carbon aggregates.
And then there's a modeling component that we are, as FC-PAD, also helping UTRC with their modeling, and some of that is modeling the porous carbons to try to optimize and get the most out of the catalyst that you're putting into your electrode structures.
So the last FOA project that FC-PAD is supporting right now is out of General Motors, and they're looking at more durable high-power membrane electrode assemblies, again, with low platinum loading. So some of the measurements that we have supported GM with include things like understanding catalyst durability through platinum in solution and half-cell measurements, where we're looking at platinum in solution as a function of potential.
Part of GM's project is just trying to understand drive-cycle effects and holding at different potentials, and how you want to mechanize your system. We don't get into that in FC-PAD. That's part of GM's project. But those are measurements that we have been supporting the GM project. And then we're doing things like rate-limiting measurements and supporting them with, again, modeling in terms of catalyst durability and surface area, beginning of test, and the catalyst AST, and then how different conditions affect that, and helping them model that.
And then one of the final things that GM has been interested in is impact of local shorting on membrane degradation, and that's things like fibers penetrating MEAs, and so you can see that we've done a lot of tomography in terms of trying to help them understand what happens if you get fiber penetrating part of your ionomer structure or the mechanical layer, and how that affects your overall MEA durability.
So with that, I've just got a couple of slides left. I wanted to touch, again, on the – where we're going in the future. We are looking at doing more work related to medium- and heavy-duty applications, and obviously, some of the differences there are going to much longer lifetimes. For example, a million miles and 25,000 to 30,000 hours of durability. We expect there to be significantly different drive cycles with these, including things like operating conditions and possibly higher temperatures, which could lead to better kinetics.
And yeah, they're probably going to have to do high current densities, but there's probably going to be more of an emphasis on improved efficiency, and more time at higher voltage. So that's some of the things that we're starting to look at this year, is what some of this effect is. There's probably more time at idle time for some of these vehicles. There's probably a lot fewer short stops, especially for the long-haul vehicles. Efficiency for these long-term, long-haul vehicles becomes much more important. And cost of ownership, you want to use much less fuel than light duty, where in light duty, initial cost is a bigger factor of it.
So these are all things that FC-PAD is starting to look at related to heavy-duty vehicles, and we'll be refining our models and characterization and diagnostic tools more to line up with this heavy-duty, and possibly, we'll be looking at refining ASTs to make them better life prediction for these heavy-duty vehicles.
So with that, I'll try to summarize. Hopefully, we've given a good idea what we're trying to do in FC-PAD in terms of the fundamental understanding of membrane electrode assemblies and the components therein, and what our approach is. It is heavy related to understanding the core phenomena of these. I really don't want to go much into the technical accomplishments. We've spend more time on capabilities, but we've got a lot of the technical accomplishments in our publications and our various AMR presentations. We've got more in this year's AMR presentation.
But this is some of the things that we've looked at, including platinum-cobalt loading effects, the conditioning and recovery protocols, looking from the catalyst ink all the way up to the MEA structure, and modeling of all these structures.
So future work, again, we'll probably be supporting new heavy-duty FOA projects from DOE, depending on how those come out. But ourselves and our core work, we'll probably do more of an emphasis on heavy-duty applications.
But yet a lot of the work we do in light duty is still applicable to the heavier-duty applications, so some of that core research will probably continue, just shifting a little bit in terms of some of the materials and operating conditions we look at.
So with that, I'd like to acknowledge DOE EERE Fuel Cell Technologies Office for the funding for FC-PAD, including Greg Kleen and Demetrios Papageorgopoulos, who's our program managers out of DOE. We have collaborated with a huge number of organizations, so I can't really take time to list them here. I did mention all the FOA projects, but we worked with a lot of other people, especially in terms of some of the component suppliers out there and some of the universities we've worked with. We have made good use of the different Office of Science user facilities out there at the various different labs, including Berkeley, Argonne, and Oak Ridge and Los Alamos, and we've also used the Department of Commerce Lab in terms of National Institute of Standards and Technology and their imaging facility there.
And now I'll hand it back to Greg, and he can close out the webinar.
Eric Parker
Okay. Thanks, Rod, Adam. This is Eric. I just want to remind everyone that the Annual Merit Review is coming up for the Fuel Cell Technologies Office, from April 29th to May 1st, in Crystal City, Virginia. Please check that out on our website and sign up for our newsletter. And if you could advance to the next slide, please.
As a reminder, everyone please submit your questions if you've just thought of them at the end of the presentation now, and we'll do our best to get to the in the next 10 or 15 or so minutes. And with that, I'll have Greg kick off the Q&A.
Gregory Kleen
Okay. Thank you. So we have gotten a few questions. Like Eric said, please feel free to submit additional questions here. So I guess the first question we got in, Rod, is is there a global alignment with EU and Asia-Pacific efforts? I guess could you talk about some of the work you've done there?
Rod Borup
So in terms of – we are always happy to collaborate with other people out there if it makes sense. We have had discussions working with what's called the IDFAST project that is funded by EU and has a number of organizations, including CEA in it. That project is related to AST development.
We've also had a lot of discussion about harmonizing new test hardware, and people have been moving towards more of a differential cell testing approach, and away from integral cell testing approach. And inside FC-PAD, we have been doing both integral cell and differential cell.
In terms of the harmonization testing hardware that is coming out of EU, there is a little bit of a different side effort that's sort of coming out of the Fuel Cell Tech Team. And so yes, we've been interacting with both of those organizations. We have had discussions with some component suppliers out of Asia. But I would say at the moment, the vast majority of our work is really North-American-related. As it makes sense, we will work as we can with similar projects out of EU. We've had more recent discussions with them than we have had some of the Pacific people. We've talked with more component suppliers out of Asia recently, but we are happy to hold those conversations. But we really need to go back then to the steering committee and DOE and make sure all of our interactions make sense and are value-added to the North American base and the DOE program in general. Hopefully, that tells you some of the things that we're looking at in terms of that question.
Gregory Kleen
Okay. All right. Thank you. So the next question is how do you make a fuel cell robust enough for a quick start in freezing conditions?
Rod Borup
That's I'm going to say sort of an interesting and pretty specific question. I don't know that I can give exact, specific details. We did do some cold temperature testing in FC-PAD two to three years ago, but the truth is, the steering committee sort of de-emphasized cold testing.
I think our feedback has been, from the Fuel Cell Tech Team and the OEMs, that they think they're at a point where they think the technology has progressed enough that it's really more – it's less pre-competitive work at this stage. So we have not done a lot of freeze testing in the last year or two, at least.
In terms of my comments on how you do it, obviously, you need to self-start and you need to balance heat generation inside the fuel cell with heating it up before you make too much ice and saturate your MEA component so that you can't get gases in. And so a lot of that is getting more into the system-level mitigation strategies, as opposed to the fundamental MEA portions, which is really why we've put less emphasis on it recently.
So we're happy to look at the more fundamental level measurements with freeze, but in terms of system levels, I think that probably doesn't belong to the national labs there. Does anybody else, in terms of the thrust coordinators, want to make a comment there?
Adam Weber
Yeah. At the component level, we've done a lot of work previously, and are still interested in understanding the water management, both from kind of a modeling as well as a catalyst layer design, and overall GDL design standpoint. But Rod's right. In terms of kind of system mitigation and other strategies, that's typically where we would hand off those kind of knowledge and design roles to somebody to actually design that system.
Gregory Kleen
Okay. Yeah. Thank you. I guess this is probably a quick answer here. Is there a publication concerning your work on the impact on local shorting on membrane degradation?
Rod Borup
Well, that work primarily belongs out of the GM project. I'm not familiar with GM actually publishing that. I will say that we did get permission from those four FOA projects, so that we could show that bit. But if there's a publication, that's probably going to be first authored by GM. That work was actually – in terms of the shorting, was actually done by Argonne, and it's probably Debbie that did it, and Debbie unfortunately is not on this call. But Rajesh, do you know if that has been published yet? Or maybe K. C.?
Adam Weber
I don't think it's been published, that specifically. There is some previous work in a book chapter by GM that covers some of kind of the soft shorts to hard short kind of analysis.
Gregory Kleen
Okay. All right. So the next question, when studying the durability of fuel cell catalysts, can you apply a voltage higher than 1.2 volts?
Rod Borup
So some of the original ASTs, if you go way back, did use a much higher voltage, and the Fuel Cell Tech Team recommended AST for evaluating carbon support, is actually a triangle wave that goes from 1 to 1.5. We do that in a hydrogen-nitrogen atmosphere. So I guess the short answer is yes, we do that all the time, in terms of the carbon – the platinum support AST, apply voltages much higher than that. But that was the recommended 1 to 1.5 volt AST, which we do on a regular basis.
Gregory Kleen
Okay. And I guess speaking of ASTs, here's another question asking if durability test data is available or shared, and along with that, what is your rationale in designing ASTs, and how do you compare AST results to driving cycles?
Rod Borup
Okay, so the last question I'll start with, in terms of how do we do ASTs. We really try to understand the fundamental degradations first. And actually, some of us in FC-PAD have been doing this for quite a while, over a decade. And when we looked at cycling between open circuit say down to .6 volts, and then the equilibrium concentration of platinum, which we showed one slide on, we do see that there is an equilibrium concentration of platinum that goes into solution.
That equilibrium platinum that goes into solution varies by almost two orders of magnitude from 0.6 volts to 0.95 volts. So when we looked at developing ASTs, we used that known mechanism and designed the AST around the mechanism. Membrane ASTs, again, we understand there's a very good synergy or should I say bad synergy between R8 swelling of membranes and chemical degradation of mechanisms, so the latest AST that has been recommended is a combined R8 cycling at open-circuit potential. And Berkeley has done modeling related to understanding what the impact of swelling, de-swelling, plus chemical degradation, because unfortunately, the two combined is much worse than the other ones.
So to try to summarize that, our methodology is really looking at the fundamental degradation mechanisms, and then designing an AST that attacks those degradation mechanisms, and generally, most of the time, the ASTs are single component ASTs. We're trying to look at one degradation mechanism at a time. There's a couple that look at more than that. We won't argue that the ASTs are perfect, but they are a good standardized set of protocols, so that people can compare components.
So then the other part of the question was how do we use those AST results to do life prediction. That's a really hard question, and we've tried to do that. And part of FC-PAD did have a project, and what we did is we looked at field data and materials that were in the field, and we conducted ASTs on those exact materials, and then we compared the field data, when they were out there actually in buses driving around, to what we were getting in terms of the ASTs, and then came up with an acceleration factor.
Again, it's not perfect, but it's what we could do in terms of getting some of that. In terms of what we're going to do for heavy duty, we're probably going to have to look at some of those different conditions, but that's what we're trying to do. And I actually did mention the EU funded IDFAST meeting. They, again, are trying to extend that work to get better life predictions.
So we do have some life-prediction ability. We do believe that it could be better. It's probably not perfect. But we also have to be careful in that different OEMs operate under different operating conditions, so we're trying to make the protocols generic, so that they work for everybody. I'm not sure if anybody else has a different answer to that. Mukun is not on the phone today, but we have the rest of our thrust coordinators on. And that partly was some of Mukun's work in the past.
Gregory Kleen
Okay. So we're nearing the end of the hour. I want to ask one more question here. So the question is, is there any coordination between FC-PAD and ElectroCat, and how? And I guess to start off the answer on that, for those that aren't aware, ElectroCat is the other fuel cell consortium, lab consortium that the FCTO office has. That one's focused on PGM-free catalysts, whereas FC-PAD is working on low-PGM catalysts. And there is some overlap between some of the members, so I think there's definitely coordination, because there's people that are involved in both FC-PAD and ElectroCat.
But if any of the thrust coordinators, or Rod, if you'd like to talk any more about this question, please feel free.
Rod Borup
I mean, my short answer is Debbie Myers is one of our thrust coordinators in FC-PAD, and she's one of the directors in ElectroCat, so there's very good communication between ElectroCat and FC-PAD, but at the same time, I will say FC-PAD concentrates on noble-metal MEAs, and ultra-low loaded MEAs, where ElectroCat concentrates on precious-group-metal-free electrodes. So there's good communication between them, but we have not been testing PGM-free catalysts as part of our work scope in FC-PAD, at least as of present. And I think that's probably the way FC-PAD will continue until we're told otherwise.
Adam Weber
And I would just further that ElectroCat as an EMN, an energy material network, is focused very much on kind of developing and utilizing new materials, whereas FC-PAD is focused more on kind of the science of integration and how we put kind of materials together.
Gregory Kleen
Okay. Thank you for everyone's questions and those answers there, and that's going to conclude the Q&A portion and our webinar for today. If we didn't get to your question, definitely feel free to email the presenter directly or the DOE webinar inbox, and we'll try to address them. And with that, I'd just like to thank everyone for joining and presenting today, and remind everyone again that this full recording and the webinar slides will be available online, and I encourage everyone to sign up for our monthly newsletter, so you get notifications for future webinars and events like these. And with that, have a great rest of your week, everyone, and goodbye.
[End of Audio]