Video Url

Below is the text version of the webinar titled "NREL's Fuel Cell Contaminant Database," originally presented on May 27, 2014. In addition to this text version of the audio, you can access the presentation slides.

Alli Aman:
I'm just going to go through a few housekeeping items before I turn it over to today's speaker. Today's webinar is being recorded. So a recording along with slides will be posted to our website in about 10 days. I will send an email out once those have posted to our website but I definitely encourage you to check back. Everyone is on mute but we encourage you to submit questions during the presentation today because we do have a Q&A session at the end of the presentation.

Please submit your questions via the question function on GoToMeeting and we will cover those the last 10–15 minutes of the presentation. I also encourage you to check back to our website for future webinars as we do host webinars monthly. So definitely check back to our website. I also encourage you to sign up for our monthly newsletter that we send out monthly, which will also keep you informed of monthly webinars and just other activities and projects going on in the Fuel Cell Technologies Office.

And on that I'm going to turn it over to Dave Peterson. He is the fuel cells technology manager for the Fuel Cell Technologies Office and he will be presenting today's speaker. Dave?

Dave Peterson:
Thanks Alli and welcome to all of our webinar attendees. Before we get to our speaker, also here is Chris Ainscough who's on detail from NREL to the Fuel Cell Technologies Office. He helped develop the interactive website tool that Huyen Dinh will be speaking about and will be available during the Q&A session to answer any questions.

Chris Ainscough:
Hello everyone.

Dave Peterson:
And our speaker today is Dr. Huyen Dinh. She is a senior scientist at the National Renewable Energy Laboratory and has over 15 years of experience in direct methanol, hydrogen polymer electrolyte membrane, and zinc/air fuel cells. She received her B.S. in applied chemistry and Ph.D. in electrochemistry from the University of Calgary, Canada. She has worked at three different fuel cell startup companies and is now the NREL lead for hydrogen production and delivery projects.

She is a co-editor and co-author of a book on "Photoelectrochemical Water Splitting: Standards, Experimental Methods, and Protocols." Her current research focuses on fuel cell catalysis and system contaminants R&D. And with that let me turn it over to Huyen.

Huyen Dinh:
Thank you Dave for the introduction. Good morning everyone. I hope that you all had a nice Memorial Day weekend and thank you for joining us today. I'd like to thank DOE EERE Fuel Cell Technologies Office for the opportunity to talk to you all about our work. This work stems from a DOE-funded project on system-derived contaminants that started in 2009. The team consists of NREL as the lead with GM, the University of South Carolina, Colorado School of Mines, University of Hawaii, and Los Alamos National Lab as project partners.

I'm very excited today to share this important work with you and making it readily available to the public. I don't think that you have—oh, there it is.

[Next slide]

I don't think you have seen a tool or database like this. I hope that you find it useful. This is the outline of my talk. I will start by giving a little motivation of why we do what we do in this project, talk about how we select the materials for our work, the methodology we use to screen all these balance of plant materials, and provide you with some screening result examples to give you some idea of how we interpret the data, and share some of the learnings that we had during the project. And then we'll go into a live demo of the website and the interactive materials data tool.

[Next slide]

So I want to start by prefacing this with the fact that I think most of you know that fuel cell electric vehicles are coming. As you can see from this slide they're coming as soon as 2015. So the need for low cost and durable and high performing fuel cells is greater than ever.

[Next slide]

To enable the commercialization of fuel cell vehicles our project objective is to increase the fuel cell performance and durability by limiting contamination-related losses and decrease overall fuel cell system costs.

As you can see on the right side up here the balance of plant material now has risen and is now greater than 50 percent of the fuel cell system cost and that the overall—the need to lower the balance of plant cost has risen in importance with decreasing stack costs. The overall cost of the balance material needs to come down in order to reduce overall cost of the system. I want to mention that this particular plot will be updated. There's an updated version of this and we'll put it in before we publish this presentation online.

To date we have a few hundred FCEVs on the road that use expensive materials or components. To be commercially viable this component cost has to decrease. This project can aid in the decrease of overall fuel cell system costs by lowering balance of plant material costs. Furthermore GM has shown in their fuel cell systems that significant degradation is possible when low cost materials are used as system components. And reducing the performance and the durability losses related to contamination to a few or zero millivolts would result in high impact.

Therefore, proper selection of materials is critical for balancing cost and fuel cell systems performance and durability. Contaminants can come from the parent material or from the additives that are in the automotive plastic. Some of the common additives are listed here such as glass fibers for structural materials.

[Next slide]

I don't know who are all in our audience but just to give a quick definition for those who don't know what a fuel cell is, a fuel cell is an electrochemical device that efficiently converts chemical energy into electricity through an electrochemical process without burning.

It represents an alternative to batteries, generators, and internal combustion engines to provide power for portable electronics, stationary, and transportation applications. You can see here this is a picture of a GM fuel cell stack. And when you take the stack, which is represented here in the middle and you put the balance of plant together we call that the fuel cell system. This schematic here shows the fuel cell stack along with different balance of plant components such as the air compressor, the recirculation pump, etc. You can see there are many potential sources of system-derived contaminants.

[Next slide]

So our project focuses on system contaminants that originate from the system itself, contaminants that come from system components such as hoses, structural materials, air compressors, fuels, gases, adhesives, and lubricants have shown to have an impact on performance and durability of fuel cell systems.

[Next slide]

Now I will talk a little bit about material selection criteria.

[Next slide]

I want to emphasize that the materials that we've selected for the screening study were not made specifically for fuel cell applications. The materials were selected based on the criteria listed here. As you can see we wanted to make sure that materials based on their physical properties can operate under the fuel cell operating conditions. But they are commercially available so that we can publish the work that we do. We wanted to cover a range of different costs. The material has to be both accessible for manufacturing purposes and we also received inputs from OEMs and fuel cell system manufacturers like those listed here.

Our goal for the screening experiments was to identify and quantify what contaminants leach out of these balance of plant materials and understand where the contaminants come from. Do they come from the parent materials? Do they come from a solvent? Or do they come from the additives that we're adding in there? And determine what the impact to the fuel cell components are. So we find that some of the additives that leached out are not useful for fuel cell applications. Perhaps that will help the fuel cell developers and material suppliers have a conversation about that and design balance of plant materials that are specifically for fuel cell applications.

The materials that are in red here are the ones that we chose for this study. You can see we did not look at—excuse me, there’s a message on the screen for some reason but I think it's the LCD itself—all right.

[Next slide]

So this an example of the different balance of plant materials that were selected based on physical property and cost, so typical structural plastic materials that may be used in fuel cell applications are listed here. The materials are shown as a function of cost on a non linear basis. And the least expensive materials reside in the nylon family and will be used for large components while the more expensive materials may be used in more specialized components. The numbers in the brackets here are the numbers of materials in that material class that we used in our study.

[Next slide]

We screened about 60 balance of plant materials as listed here. There are three categories of materials. There are structural plastics; the assembly aids materials, and then hoses. You can see that listed here based on some of the manufacturers, the chemical descriptions, and different grades. Again our project was focused on commercially available polymeric materials. We [inaudible] the study and we need a systematic way of screening the materials for their potential impact on fuel cell performance and durability.

[Next slide]

[Next slide]

This slide here will show you the material screening approach and the experimental method. On the left hand side here is our leaching protocol. We took the material and then we soaked it in DI water at 90 degrees Celsius for one or six weeks, and then we removed the material for analysis. Note that we chose these conditions as accelerated conditions so that we can get a high enough concentration of contaminants so that we can detect the contaminants using our analytical techniques.

One of the first quick, easy methods we used is called the TOC or the total organic carbon. We quantify the amount of organics and ions present using these two methods. We also used advanced analytical techniques like GCMS, ICP, and IC to identify the organic and ionic contaminants that are present. We also used ex-situ techniques to determine the impact of these contaminants on membrane conductivity and electrochemical testing to determine its effect on catalysts.

And finally we also carried out in-situ fuel cell infusion testing to determine the overall impact on fuel cell performance. And you see some plots of this based on using voltage loss as a criteria.

[Next slide]

On our website, on our interactive Web tool, you can select a different material based on these different criteria. You can determine whether you want to find information based on the material type or class or manufacturer. And what you get—the output of the data—will be listed here at this side: the TOC and the organic species that are present or the leaching index and what's the performance impact on fuel cells.

These data will be shown graphically for quick determination of the impact and comparison with other materials in the same material class or type. However, the website will not provide you with interpretation of the data. Therefore in the next few slides I will be showing some screening result examples to help you interpret some of our data and also to share with you some of the significant implications and learning that arose through this work.

[Next slide]

[Next slide]

These two plots here are the solution conductivity versus TOC. It gives you an overview of the range of the amount of ions that are present and the amount of organics that are present. You can see for all the assembly aids materials that we studied there's quite a range of ions, from Krytox materials which has a very low amount of ions and organics, to other materials that have very high concentrations of those. The same thing for the structural materials that were studied.

One of the first things I want you to know is that there is a wide range of concentrations of contaminants from the materials we studied. The other thing you might take from here is that potential contaminants consist of a mixture of species: organics, inorganics, and ions. You can see that the likely target balance of plant materials used in these systems would be where they have low TOC and low solution conductivity. And as you can see materials such as this here are the higher cost, non-commodity materials that leach out less ions and organic contaminants. So as they say, you get what you pay for.

[Next slide]

In this slide I want to show that GM has screened and categorized 34 plastic materials and they're classified into groups based on their basic polymer resin and brands. As you can see from these plots the leaching index shows that in general the higher the leaching index the higher the cell voltage loss and the lower the material cost. Again the message is you get what you pay for. The implication is that fuel cell developers can do a quick screening of balance of plant material candidates by carrying out leaching experiments only and measure just the TOC or the solution conductivity. And then these measurements are quick and easy to do. And if you find some good candidates then you can carry out some further testing for just electrochemistry, or membrane conductivity, or in-situ fuel cell testing, which require a lot more time and capability to do. So this is a nice quick way to screen your materials.

[Next slide]

So we used ICP, that's inductively-coupled plasma technique, to identify what different elements are present. We looked at 29 different elements as listed here in this table. We only list on this slide some of the major components—the different elements that had the highest concentration found. We did elemental analysis by using this technique and then through our knowledge of the materials that were used, such as additives and fillers used in the polymeric formulation and datasheets, we were able then to identify that the elements that were leached out were likely linked to common additives.

An example is shown here where the aluminum and calcium are likely flame retardants, and others could be fillers or flame retardants and catalysts as shown here.

[Next slide]

We used the IC, ion chromatography, technique here to identify the different anions that are present as shown here. These are the different ones that we detect. And again these observed species can be attributed to fillers and additives.

[Next slide]

We used the gas chromatography mass spectrometry to identify what organic species leached out of these materials. You can see here that the organic polymers come from the polymer resins, additives, residual solvents, and by-products of incomplete polymerization, and that the more expensive materials didn't leach out any organics and tend to be quite clean.

[Next slide]

And finally, we did some in-situ fuel cell testing to determine the impact of these materials on fuel cell performance. Here are some examples of a few assembly aid materials that we screened. These were selected to demonstrate the range of impacts that these contaminants have on fuel cell performance. The Krytox materials shown here belong to the PFAE/PTFE family of materials. And their extracts had very low TOCs and solution conductivities. Essentially no contaminants were identified in these materials. And you can see from this in-situ fuel cell screening plot that shows there's no impact on voltage loss at 0.2 A/cm2. So what you're doing is comparing the colored lines here versus the black, which is the baseline where no contaminants were added. This set of materials would be classified as clean.

In this example you can see that these urethane materials had an impact on fuel cell performance. They showed about 100–150 mV voltage loss but there was some reversibility of the fuel cell. At this point we stop infusing the fuel cell with the contaminant and you can see that there's some reversibility of the voltage here. This material is classified as contaminating but it can partially recover.

And the third example demonstrates that there's a very large voltage loss here resulting from the contaminants that came out of this material and that even when we stopped infusing the contaminant there was no recovery of the fuel cell voltage at all. This type of material we consider to be contaminating and it's not recoverable. As you can see some materials can have adverse effects on fuel cell performance, but the effect is complex and some contaminants are recoverable when others are not. Again keep in mind that the concentrations that are used for these screening methods are relatively high because we use accelerating methods—leaching conditions—to get these contaminants.

[Next slide]

Overall I wanted to summarize our learnings here. You can see that the resin type, the cost, and additives do have an impact and they need to be considered when selecting balance of plant plastic materials. In our other work we've also found that contamination impact depends on the operating conditions such as current density and concentration, RH, platinum loading, and so on. You need to consider those as well and those operating conditions that can cause more contact time between the liquid and the plastic need to be considered in development of a fuel cell system.

Through our project and our learnings there are some suggested mitigation strategies. In order to minimize the amount of extract solution concentration that comes out and gets to the fuel cell and may affect the fuel cell performance, you need to minimize the contact time of the plastic materials with the water in the fuel cell, to minimize exposure of plastic materials to high temperature. If you could, you could use the RH as a way to flush out the contaminants and perhaps even help you do some potential cycling as an ex-situ way of recovering the fuel cell performance.

There are probably easier ways to choose clean balance materials, but then they're usually more expensive. Or else you can work with the plastic manufacturers to modify the commercial plastic materials to minimize these contaminants. Then perhaps there are other types of mitigation that can be used, such as filters for consideration.

[Next slide]

I'd now like to go to the live demo to give you a little flavor of what the website can do for you.

[Website demonstration]

Can you see this? [inaudible comment and laughter] Okay so this is our main website. You can see that it gives you the general information of our project. And there are different tabs here. This overview tab really just gives you a project overview, why we do what we do, and when the project started, and so on. When you click over to the materials tab you can get the same table that I've shown you before, the different materials that we studied and how many in each of those categories. You can also click on these links to get the table that I showed earlier as well.

When you click on the methods tab you can see that we've listed the different methods that we used in the screening of the different balance of plant materials. And our flow chart here shows what we did there. Again if you want to know what organic model compounds that we studied further you can click on that to get the chemical structure and name of those model compounds. The partners tab lists the different partners that we collaborated with.

And there is also a publications tab. If you click on this link you see a list of the different publications and presentations that are related to this particular project. And then the data tool is where our interactive Web tool lies. There are three places that you can click here. You can click in the big button here, here, or here. I'm just going to use the one that's up here because that's the most easily visible on that particular website. You can see that when you click on that it brings you to a different page.

Like I said earlier you can choose from the assembly aids materials, structural materials, or hose materials. I chose structural and you can see that the material class has changed, with the different manufacturer choices and trade names and grades. I'm going to pick maybe this one. It's a PA material. It's a nylon material from EMS and there are different grades here that you can select from.

One of the first plots that you'll see here is total IC. That's the total ions—anions you have versus the total ions in general. The big green dot is the data for the material that you just selected. When you hover over it you can see that there's a number that shows up for IC total and ICP total. This here compares all the data—all the materials that are in the structural materials class. And the list goes on. If you click on here we show the rest of them. We also put in the data for the aged DI water as a reference.

On this side is a plot you saw earlier in my presentation that shows the conductivity versus TOC. If I change any of these things you can see that green dot will change. It has gone from here to here. This is the material that we just chose compared to a set of materials in the same class—all the nylon materials from the same manufacturer shown here. Perhaps this is the one that you want to choose and you would do it from up here. I don't actually know which one that is but you can see that it changes when you select the different materials.

Our next set of plots—as you can see if you scroll down this part up here is fixed so that you can continue to choose the different materials you want to look at and compare them. This plot here comes from the ion chromatography experiment. It tells you what ions were present in that particular material and what the concentrations are. If I select a different material you can see that these plots change. And if you hover over them again you can also get the data—the exact number for those.

This is IC data and this is ICP data. You can this one has a lot of potassium and sodium that leached out, versus this as well, this one seems to have other types of elements that leached out.

The third row of plots shown here is the effect of the balance of plant contaminants on the voltage loss itself. This is a static plot, that's why it is well-labeled here so that you know that you can correlate what materials with which one that you selected. And you can see that there's a range of impact of materials here. This one seems to have a lot more higher impact than the other one. And this choice plots a PPA material instead. You can see that that plot changed for you. And here you can really see that the 30 percent GF—stands for glass fiber, that's an additive that's being added to the material to give it structural integrity—you can see that this has an effect. This additive has an effect on the amount of contaminants that leached out and its performance effect. So this is the voltage loss data compared to the baseline. And this is the HFR, or high frequency resistance data. And again black is the baseline.

On this quadrant here is the GCMS and LCMS result. You can see that we listed how many peaks were detected and the number of weeks they were soaked. All of the structural materials were soaked for six weeks. And here is a list of the two organic species that were identified in the extract. We only listed the organics that we have high confidence in based on the hits with the NIST spectral library reported here. And these were liquid injection GCMS method.

The fourth set of data that's shown here is what you saw earlier—the average leaching index plotted against the different material classes. We plotted them with increasing leaching index. And again if you hover over them then the data shows up. We would like to improve these two plots by really color coding them so that when you choose a different material it will highlight the bar. And I chose not a very good material. For example, if that was the material we just chose it would highlight this particular bar and show you that's the material we just chose. Right now we're not at that point yet.

This plot is the voltage loss versus material class. And you can see there are similar trends between the amount of voltage loss and the leaching index.

I think that's all I have. I’m going to go back to our presentation.

[Next slide]

I want to summarize by saying that we have a relatively large material database on the website now and it's publicly accessible to everyone. I've gotten feedback from folks who have used it that thought it was very useful because all of the information can be found in one place.

The purpose of doing this is really to help the fuel cell industry select appropriate balance of plant materials and in cost-benefit analysis. We also hope that the fuel cell developers and the material suppliers will have a conversation so that they can develop clean materials for fuel cell applications since the materials that we studied were just commercially available materials; they're not specifically for fuel cell application.

We've designed this web tool to be interactive, informative, and easy to use. And I hope that is the case. We'll continue to improve the website so feedback is welcome. Please look for our publications for more in-depth analysis of these types of results.

[Next slide]

I'd like to acknowledge the Department of Energy EERE Fuel Cell Technologies Office for project funding. Dave Peterson and Kathi Epping were our technology development managers. I'd like to thank all of the companies listed here for providing materials and for giving inputs on the material selection. I'd like to thank the Granville Group at Colorado School of Mines for the ICP-OES and IC measurements, all of our NREL team members listed here, a special shout out for NREL website developer team, Chris Ainscough, Sara Havig, Shauna, and Michael. And of course all of our collaborators from General Motors, University of South Carolina, University of Hawaii, Colorado School of Mines, Los Alamos, and 3M.

[Next slide]

Here are my contacts if you need to contact me and have some questions.

[Next slide]

We're ready for questions and answers.

Dave Peterson:
All right, this is Dave Peterson again. Thanks Huyen. Thanks for the presentation. We now have some time for questions. Please submit your questions according to the process shown on the slide. With that we'll go ahead and get into the first question, which is, there are many different types of chemistry additives that can occur with polymers. What steps were implemented to ensure an apples to apples comparison?

Huyen Dinh:
What we've done was to keep the leaching conditions all the same. So when we leached—we soaked the materials we kept the surface area to the solution constant. We leached them with DI water at 90 degrees Celsius, the same way. That part we can control and that was done the same for all of the materials that we studied. So whatever leached out at these conditions we will be able to identify using all the different analytical techniques that I talked about and figure out where the origin may have come from.

I just said earlier some of them came from the parent materials, the resin itself. Some of them were additives. If they leached out into the solution and we can detect it, we can identify what they are and quantify them and also look at the overall fuel cell performance affect. Does that answer your question? I guess I can't ask that. [laughter]

Dave Peterson:
Thanks Huyen. The next question is: is there any understanding mechanistically how these attacks occur?

Huyen Dinh:
How these what? Can you repeat the question?

Dave Peterson:
Sure. Is there any understanding mechanistically how these attacks occur—what's being attacked on the polymer and what damage is being done?

Chris Ainscough:
And also we had a request to put the contact information for the team back up if you can do that.

Huyen Dinh:
This one or this one?

Chris Ainscough:
Probably the other one. Yes.

Huyen Dinh:
Yeah they can play on the website right now. Yes we've done some work to understand what the degradation mechanisms are. We've done that with the extract solution itself where there's a combination of different contaminants that are present. Like I said when we leach these out it's really a soup, a pot of different things: ions, organics, and inorganics. And then we feed the whole thing into the fuel cell to look at its impact on fuel cells. We also worked on identifying what those organics could be and focus on the study of specific organics.

Because when you look at this whole soup of them you don't really know whether it's the organics that's having an effect on fuel cell performance. Or is it the inorganics, or the ions? And which one of those has the impact or the most dominating effect? Therefore we had a series of work looking at model compounds. We looked at a range of different model compounds. That's listed here [return to slide 30]. We injected—looking at one specific model compound at a time and looked at its effect on catalysis.

We found that there are some general trends with some of these organics that has an impact on the catalysis itself. And we also found that these organics themselves have an impact on membrane conductivity. We did an ex-situ membrane conductivity test. Some of them absorbed onto the membrane. Some of them actually ion exchanged with the proton. We have some of these organics that actually just loosely, physically adsorb onto the catalyst and so they're easily removed.

And we have others that have irreversible effect on the catalyst and even get oxidized or reduced on the catalyst to form something else. And then we also looked at the overall effect of the fuel cell performance of individual compounds as well as the mixtures of compounds to see if there was interaction between them. And there were. Another message there is that we looked at them individually and there were effects. But there is also interaction between the different components that are present in the mixture.

Dave Peterson:
Thanks Huyen. The next question: have you further identified the contaminant species that cause irreversible loss from epoxy as an example?

Huyen Dinh:
I could talk about this one specifically yes [points cursor to 4-methyl-benzenesulfonamide on slide 30]. This particular organic compound we found to have an irreversible effect on the catalyst. I don't think I have the CVs or anything on here. But I injected this particular organic compound into an RDE setup specifically and looking at its effect on the electrochemical surface area and the effect on the ORR—the oxygen reduction reaction. It definitely decreased the ECA by quite a bit—a substantial amount. And it also affected the ORR as well.

And then when you took that electrode out of the contaminated solution and transferred into a clean solution with everything clean you can see that the effect is not reversible at all. That's another way that we look at both ex-situ and in-situ tests to determine whether an organic compound or species has reversible or an irreversible effect [returns to slide 21]. You saw earlier on the in-situ testing when we stopped the contamination, you can see for this species, there's some recovery. And this species there isn't. And that's an in-situ one.

I hope that answers your question.

Dave Peterson:
Okay, yes, next question: has there been a statistical analysis to determine the strength of the correlation between the leaching index and in-situ voltage loss?

Huyen Dinh:
[Returns to slide 17] I don't know the exact answer to this. GM did this analysis and I think they did do some statistical analysis on it. I've done some work on doing a similar plot for the assembly aids materials and they seem to show a general trend as well. Again this is a general trend, not an exact science. We can see that there are exceptions like this one here. The PPSU material shows a very low leaching index. But here the voltage loss is higher than the PSU. So there are exceptions.

And the same thing for this one here, but it's just a nice general trend for you to use if you wanted a quick way to screen your materials using just two measurements: the TOC measurement and the solution conductivity, which are easy to use, and quick.

Dave Peterson:
The next question: how many different fuel cells were tested to determine how generally the in-situ measurements can be applied?

Huyen Dinh:
Can you repeat that question? [Returns to slide 21]

Dave Peterson:
Sure. How many different fuel cells were tested to determine how generally the in-situ measurements can be applied?

Huyen Dinh:
Okay. So we took a lot of care at the beginning of the project to ensure that the test is repeatable—the in-situ fuel cell testing was repeatable—not only within the different organizations, for example GM or NREL, but that our baseline experiments were all matched with each other within statistical error from lab to lab. We did a round robin as well using the same fuel cell hardware, using the same type of MEA and we compared the data where the MEAs were all built at GM and then sent out to other labs, whether that’s NREL or the other universities—project partners in our project.

And then we tested them all and we all got very, very similar data. That's in an AMR presentation if you want to see the comparison between the different fuel cell testing results for the baseline. We also compared data by assembling our own fuel cell using the same materials and hardware. And we also got very similar results from lab to lab. We've done things to make sure that our baselines are similar for what we do repetitively.

And many times—it's not just one, I forget exactly how many different tests we did. I know that GM did quite a few and did a statistical analysis for the baseline itself to make sure that our methodology for in-situ testing is repeatable. And we do a baseline at the beginning of the in-situ testing as you can see too. This is a general baseline that we did. But then we also do at the beginning of each experiment for a contaminant, we just do a baseline where we don't add any contaminants in and then we add in the contaminants and then we don't add contaminants in. So they always seem to line up with our other baselines that we've done in the past.

Dave Peterson:
Thank you.

Huyen Dinh:
So we're very confident that we have a very robust method that is repeatable by everyone on the team.

Dave Peterson:
Great. Next question: can you please describe how the infusion test is performed? Ionic contamination needs to be carried through a liquid phase and GDL/MPL constitutes a barrier to this type of transport. So for the most part just describe how the infusion test is performed.

Huyen Dinh:
Yes. GM spent a lot of time developing this method. And then we all applied it. So we were all able to do it as well. We used a—what did we call it? I can't recall the name of it. But we used a pump to pump the solution and we used a special nozzle to infuse the gas phase and the solid phase of the contaminant solution into the fuel cell. And we controlled that very carefully to know what the relative humidity is and concentration and everything.

So it comes in the same way. It's an aqueous gas phase—a gas phase of the contaminants that goes into the fuel cell and penetrates through and gets to the MEA for the reaction. I think that's what they're trying to get at. So we start with just the water—not water—we start the experiment without any contaminants present to make sure that we have the same baseline for several hours. And then we inject in the contaminant itself to see what the voltage loss is and the HFR effect is for some hours.

And we want to make sure that hopefully it will reach some steady state before we stop the infusion of the contaminants and just perhaps humidified air and hydrogen and see what the recovery looks like. So we do beginning of test, beginning of life types of diagnostics at the beginning to determine what, for example the pol curves look like in the beginning, what the ECA looks like in the beginning. And we also do it at the end so we can compare what they lost, what ECA is, and what the effect is over a range of current density by looking at the pol curve data.

It's a very long and comprehensive infusion testing. And we did this for almost all 60 materials. That took a long time to do.

Dave Peterson:
We've got a couple more questions here. If anyone has any last minute questions please submit them now. The next question: is this contaminant research effort relevant to fuel cell-powered passenger vehicle hydrogen fueling dispensing equipment? Is there concern about contaminants at this stage of hydrogen—I'm sorry I paused there. Is there a concern about contaminants at this stage of hydrogen distribution as these fueling locations expand throughout the country?

Huyen Dinh:
First of all I’d like to say that yes, even though this study was focused on PEM fuel cell effects, I think the ex-situ data could be useful for other types of fuel cells as well if these types of materials are used in the systems or even for electrolysis and not just fuel cells. For every system that has a balance of plant associated with it and it uses similar materials, then yes it's widely applicable to other types of fuel cells as well as electrolysis. And it may even be applicable to the delivery of the materials if these materials were used. [Returns to slide 7]

[Returns to slide 6] You can see in a system you have a compressor, a coolant, a radiator, and you may have structural materials. You may use hoses and you may use these general types of materials. And if they are in your system then they will be applicable. You can look at our data to get some general guidelines on what type of contaminants can come out of some of these materials. As for hose delivery or the fueling infrastructure it may or not be relevant because the way we do our leaching experiment, we soaked it in water. And unless the material gets exposed to liquid water or humidified gas like the way we did the experiment here, then the effect may be very minimum, like in a fueling infrastructure where I think it's usually dry gas that passes through different things or there is a hose or nozzle before it gets to the fuel cell. So it's more of a stretch to correlate that to concern for the hydrogen fueling infrastructure and the hose delivery because of the way that we leached the materials and extract out the contaminants in this particular work.

Dave Peterson:
The final question we have for now is: what is the plan to expand the materials list, e.g., different types or new materials in the online data tool? Are you open to information of new materials leaching from other collaborators?

Huyen Dinh:
Currently there is no plan to expand on the list of materials that we studied. That stage of our project is over. We are focused more on the detailed contamination mechanisms of the work. We are funded by DOE so we do what we need to do for what DOE is interested in. But we are open to adding to the database. If you have information that is relevant to this type of work we're open to adding that. Because I think it's important to have one site for all these types of contamination information in one place.

I know that people said that they looked everywhere, and it's spotty information from this paper or that paper. So it's really nice to have all this information in one place. Perhaps we'll put air contaminant type of data on here as well or fuel contaminants or even other system contaminants data. I've talked to other material suppliers who have done some contaminants work and they're discussing internally whether they would be able to share that information. And if they could perhaps we can either link to it or add it to this database.

Furthermore NREL could also work with specific companies or universities on the side, meaning differently from the follow up with DOE. It could be like a TSA, a technical service agreement where we can use our expertise to help analyze and figure out whether a material leaches out certain types of contaminants. And if the company does not have that kind of capability themselves and may be interested in working with us we're happy to do that.

Dave Peterson:
Okay and we had a final follow-on question for the infusion test. The question is asking if you could provide a reference to the infusion test procedure.

Huyen Dinh:
I think some of them may be already in some of the papers that we have. But yes we also hope to put together procedures and then publish them on the website as well. So the general procedures, the methodologies of how we do that are in some of the published ECS Transactions or papers that we already have. But we want to put together a more detailed one and hopefully be able to publish that online as well. But that's future work. You can go to our list of publications and find those papers relatively quickly.

Dave Peterson:
Okay, thank you Huyen, and with that I will turn it back over to Alli to wrap things up.

Alli Aman:
Thank you so much Huyen for taking time out of your busy schedule to present. I just wanted to remind everybody that the PowerPoint slides as well as a recording of the webinar will be up on our website in roughly 10 days. I will send an email out once those have posted. And again I just encourage you to check back to our website and sign up for our newsletter so you can be in the know of what's going on in the fuel cell technologies world here at the DOE. Thank you Huyen, thank you Dave, and until next time.

Huyen Dinh:
Thank you. Bye-bye.