CatCost: An Estimation Tool to Aid Commercialization and R&D Decisions for Catalytic Materials—Video Text Version

This is the text version for the CatCost: An Estimation Tool to Aid Commercialization and R&D Decisions for Catalytic Materials webinar.

>>Fred Baddour: Good afternoon, everyone, and thank you for joining us to learn more about the CatCost tool. My name is Fred Baddour, and I'm the PI of the project that developed CatCost. I'm going to begin by introducing some of the background of the project and describing what CatCost is, and then I'll hand it over to Kurt Van Allsburg who is the lead developer on the project, and he'll go through an in-depth walkthrough of the tool.

So the problem that we sought to address when we started this project was the high commercialization risk associated with biomass conversion technologies. Analysis work out of our laboratory has shown that in a number of biomass conversion technologies modeled, catalyst cost can constitute up to 10 percent of the total capital cost. In addition, the uncertainty surrounding the cost of potential breakthrough catalysts drives uncertainty in the minimum fuel selling price of fuels produced during these processes. To address this uncertainty and help mitigate risk, we have developed a spreadsheet and web-based catalyst cost estimation tool called CatCost that is easy to use and requires no outside process design experience.

Our CatCost tool aims to enable researchers to build an understanding of economics of catalyst production and make more informed R&D decisions early in the catalyst development process. The release of CatCost in the fall of 2018 is the culmination of a three-year collaborative effort between NREL and PNNL with the guidance of an industrial advisory board. Both the spreadsheet and web-based versions of the tool complete with in-depth documentation and user tutorials will be released in fall of 2018.

We built CatCost integrating it with industry standard estimation methods, and it's adaptable to a wide range of catalyst synthesis types and user experience levels. CatCost provides researchers with a number of ways to input their own cost data or built-in raw material libraries. Users are able to select from two modes of processing cost estimation, including a detailed full plant analysis or simple step-based approach.

Both methods are bundled with pre-populated templates to help users initiate an estimate for a wide variety of common catalysts, including metals on metal oxide, zeolites, and nanoparticles. In addition, CatCost captures the value remaining in discharge catalysts at the end of their usable life to understand the effective catalyst cost. And so we've effectively integrated some of the best industry standard methods into the cost estimation to make a powerful analysis tool.

And so researchers are able to use CatCost to perform detailed sensitivity analyses, compare multiple catalysts, and component cost analyses with our simple tool that has been verified with industrial market data. CatCost enables researchers to make better R&D decisions by providing them with actionable cost information and powerful interactive visualizations that make their cost analyses easy to understand. It has been built from the ground up with industry-verified analysis features, including the ability to scale any estimate to a desired production rate; scenario and sensitivity analysis options allow researchers to test what if scenarios rapidly and see what steps or materials contribute the greatest uncertainty.

And researchers can also enter and manipulate data in any desired unit, removing the need for tedious conversions and allowing users to focus on the content of the estimate and not syntax. Coupled to the estimation capabilities of the tool is a suite of powerful visualizations that allow researchers to view and manipulate their data graphically for ease of analysis and presentation. We have designed the CatCost tool to be flexible for a variety of use cases to empower researchers to make better R&D decisions.

So as I mentioned, this will be free and will be released in the fall of 2018. You can get it at catcost.chemcatbio.org where you'll be able to interact with the online web version of the tool, and you can download the spreadsheets and the documentation and some of the tutorial videos that we'll be putting out in the future.

And if you have any questions from here on out, you can contact us at [email protected]. And with that, I'm going to hand it over to Kurt Van Allsburg, who is going to give us a walk-through of the system. Kurt.

>>Kurt Van Allsburg: Okay, so let's start with the demo, and what I'm going to do here is take you through the process of completing a typical catalyst estimate using the CatCost tool, the Excel spreadsheet version.

And the idea here is that you get a sense of both kind of our design philosophy and some of the estimation methods that are included in the tool. And I've completed a lot of the inputs already, but some I've left open so that you can see a little bit of the process. For this, I'm going to do a platinum on titania catalyst, and so the first thing that I need to do is put in some broad overall inputs that go onto the tool, and you can see starting out in the CatCost Excel spreadsheet that things are laid out in these tabs at the bottom in a sequence of steps that you need to complete in order to finish your estimate. We try to make it very easy to navigate and use. So we have to start here first at the inputs tab. And I'm going to input some information like the catalyst name.

And the basis here for cost estimation —so this is the —basically when you're building the plant. So this uses built-in price escalation databases both from Chemical Engineering and U.S. Bureau of Labor Statistics. And then we want to select what units we're going to do our estimate in, so I'm going to select kilograms here. You can see that I have a lot of options, and all of these things are implemented in a way that makes it so unit conversions, price escalations, and so forth are all transparent and done for the user for your convenience so that you can focus on the really interesting parts of the catalyst costing.

So then over here, we have some inputs that basically size the production plant that we'd be making this catalyst in. Of course, this is important because if we're trying to scale information from lab scale to industrial scale, we need to know what industrial scale means in this context. Now as Fred mentioned, we have two approaches to processing costs, in CatCost. One is a very sort of simple step-based method where you have an all-in cost. That is here in this first section, step method. An all-in cost for each step, like a filter, and that includes all utilities cost, operator costs, capital, and operating, in a single hourly rate for that piece of equipment. This is based on contract manufacturer approaches to getting a price quote for manufacturing a catalyst at say a toller. The other approach is the CapEx and OpEx factors approach. This is what you'd be more likely to find in a traditional chemical engineering process design text, and this involves a list of equipment and operators, costing every individual item of OpEx and CapEx. So this is a little bit more detailed. I'm going to show you both of those approaches.

But you can see here that we have to enter some things for either of these approaches. If we were doing the step method, we'd say we want an order size of between 1 and 1,000 tons. And that determines some things like what size equipment —at the catalyst manufacturer, since they already have existing sizes of equipment —what size of equipment would be used. And then determine some things like the length of this synthesis campaign and your selling margin. So you could see if I were to go for an order size of 200 tons, that the selling margin would be substantially lower. We actually have a just-accepted paper on this in organic process research and development on that part.

Then again, on the CapEx and OpEx factors approach where we're doing everything in detail, then you just have to enter some information like how big is this plant in terms of annual production of catalysts, how many operating hours do you expect, how much maintenance downtime will there be. So these are all things that are pretty standard in design reports like those published by our TEA group at NREL and elsewhere, and also by companies like IHS. So let's then proceed from these basic inputs to the next step, which is materials cost. In materials cost, I've already completed a number of these things, but what we're trying to do here is capture both sort of the stoichiometry and materials balance considerations, and actually what materials are used in the synthesis.

And so the first thing is to enter information about how large this —you kind of put in context the synthesis that you're entering, so if you're entering a particular consumption of materials that has to be for particular production rate of a product —so in this case, the synthesis information I've entered is to produce 1 kilogram of catalyst, so I'm going to enter that here, and you can see that some calculations are already done based on these inputs here, and we also have the option of using a percent yield calculation or to enter active base mass instead, which might make more sense for a different application. We also assume something about some losses because inevitably when you're loading reactors and carrying materials around, you will lose some. So we've tried to capture little components like that.

So down here, I've already completed how much I need of my metal source, but what I'm going to do is select that from our materials library. This is actually just a subset of the materials that we offer to users. But I wanted to make it reasonable to scroll through here. You can see that selecting a material and entering it in the list is really quite easy. You just select from the drop down. And now that I have my metal precursor in here, the quantity and the units, I am getting a whole bunch of quantities calculated for me, notably the consumption in kilograms of this material per kilogram of finished catalyst. And the unit price, which is looked up from the library, and then ultimately the cost of that material per kilogram of finished catalyst. So what is its contribution to the final cost? Just to show you how easy this can be, I'm just going to add one more material here. I'm going to say let's do hydrochloric acid, which is right here. And maybe I want to enter it in volume units, so I'll say I want to use 25. I can either type milliliters or I can look in here, and you can see that if the density is entered appropriately in the back end for this in the materials library, then it will automatically do all these calculations to give me the amount of HCl that is required in kilograms, the units I selected earlier, and flow all the way through to a cost and the bottom line. Turns out HCL is quite cheap, even at this consumption rate.

So that is our materials, our materials cost estimation component, and the idea is by the use of convenient unit conversions and stoichiometry calculations, and the bundling of a materials library for which it's easy for users to add their own materials and their own library of common materials, including proprietary ones that you wouldn't find in our library, through all of that, the user is able to accurately get materials cost for this component of catalyst cost. Now let's move on to the first method for doing processing cost estimation, and that is the step method. This again is the all-in approach where you have just a single hourly cost number for each particular step of a procedure, like a spray dryer or a ball mill, and what I'm going to do here is introduce a concept that it also applies for the other more detailed approach.

And that concept is that what the user can do is either select from a list of pre-bundled processed templates —so this is a preexisting list of steps that we've developed for things like an FCC catalyst, a metal carbide, a PGM metal on carbon or metal oxide. For this one, I'd choose the PGM on metal oxide, even though I'm going to look at more detail here. As you can see at the top, it says this is not included in the final estimate. I wanted to demonstrate the step method for you, but I think I'm actually going to look in more detail at the alternate approach. So you'd not normally do both of these because then it would be sort of duplicating the same elements of cost.

But one thing to note is that it is possible to combine these two approaches, the step method and the detailed CapEx and OpEx factors method. The idea being that if you have this really detailed plant design and all these steps with operators and capital costs, operating costs, but then you want to change the process by adding just one additional step, it might be easier rather than trying to figure out how to size that piece of equipment just to use our built in step method and add a single step. So that's the kind of use case that we'd imagine that it would be useful for.  

Now as I mentioned, you can select from one of the process templates, or you can just enter a custom process. So in the custom process —oh, actually I have something entered here, but normally that would be blank. But you can see down here I have all these process templates. So I'm going to look at the one that we're actually using right now, which is metal PGM on metal oxide, and you can see that it has a number of process steps that are included. And these automatically scale to the correct order size that you have selected. And so basically the user could interact with this and modify these things, the power user could, or someone who is just looking for a quick estimate could just accept this information. They don't even have to look at those details. All they need to do is select from this drop down, the metal PGM on metal oxide. So now let's switch to the more detailed approach, the CapEx and OpEx factors approach, which covers sheets 3b through 3e here at the bottom. Again, this would be the alternate approach. You do not have to do both the step method and CapEx and OpEx factors.

So if we look at the equipment, we have the same choice, essentially. We have a set of process templates, but the information included in these process templates is a little bit different. Now we're looking at a lot of detail. What is the piece of equipment? What material of construction is it made out of? How many? What size are we requesting? And of course the size units are different for each piece of equipment. So for a bin, for a feeder, it would just be a volume in cubic feet. For a pump or a compressor, it might be horsepower. For a tank, naturally, it would be volume, and a few things to note here. The —there is a very —we have a very large list of pieces of equipment that we have available from various textbooks. This is I think the largest list of equipment cost correlations available anywhere that I'm aware of, at least in the open —in the public domain.

So this allows you to select from a very wide range of high-quality cost correlations, so what you do is if you want to add a piece of equipment, it's going to automatically select the default material construction, or you can override that. You can say whatever quantity you want, maybe you want two of those, and then the horsepower. I don't know what an appropriate horsepower is for this, so let's try 50. That's apparently too large, so this is giving you feedback here. One thing to note, these size limits here are giving you feedback on whether you maybe have chosen the right piece of equipment, but either way, even if you're outside the limits or if you're inside the limits at the scale that you entered, but then as you chose a different industrial production scale and the CatCost tool automatically scaled it up or down for you, it incorporates logic to say that if you go above the upper limit of a cost correlation, it is going to automatically double or triple that piece of equipment until you're back in the correct range, or if you go below the minimum size for a piece of equipment, it is going to automatically go to that minimum and price it at that, which is pretty standard chemical engineering process design practice.

So these feedback mechanisms right here are really just to let you know maybe you want to choose a different kind of compressor or something along those lines. So then let's scroll over here and see what it's doing with these inputs that the user has given. So the first thing is that CatCost is automatically going to scale the user inputs to the production scale in pounds per year, tons per year, whatever, that was entered on one input. So it'll take the entered quantity and the entered size, and scale it to an appropriate quantity and size at that design scale. And then using the cost correlations that are built into the tool, it will develop a purchase cost and an installed cost for that piece of equipment along with a labor factor. This labor factor is just the number of operators at any given time that are needed for that piece of equipment. So in this case where you see .1, that means one operator could split his or her time between 10 such pieces of equipment.

If it were one, that would mean this piece of equipment needs a dedicated operator, and this information stepping back, this information is then summed, and it is brought up —haven't selected one yet. Since we were doing platinum on titania, I'm going to do what impregnation of metal-on-metal oxide, but that information is then pulled from down there from whatever process template or the custom process you've selected, and is pulled up to here. And then that information will now flow through to the rest of the sheets.

The next step after doing the equipment is your utilities consumption. Again here, we also have process templates for each one of the equipment lists, we have a list of the utilities consumption associated with that process design, and this comes from Aspen modeling so that the user, if they want to use a process template, doesn't have to do that. So again, the wet impregnation metal-on-metal oxide is selected already, and what that is pulling is the consumptions. And then we have unit costs, and ultimately that gets us to our utilities cost for this component of catalyst cost. Then the next part is where we get to the factors part of CapEx and OpEx factors. Here, what it's going to do is it will pull that total for the purchased equipment and the installed equipment from the equipment list, and it's going to then apply a number of fixed factors to get to other components of CapEx and OpEx costs.

So purchased equipment here then is defined as 100 percent, but then we have factors for things like instrumentation and controls. I think electrical indirect components, like engineering costs and legal expenses. These are all going into building the plant in the first place, and then analogously, we have the same type of factors for components of operating costs in the OpEx sheet. So most of this information is based on the direct labor cost. So it'll round up the number of direct labor operators from the plant design. And then it is going to add things for say laboratory charges, supervisory labor, overheads, and distribution and so forth.

All of these factors here are just fixed percentages, and one thing that I didn't highlight already that I'll just quickly show now is that for each of these values, we've built in scenario or sensitivity analysis to this tool, so that if I click up here at the top, there's this little plus icon, it's going to show that I not only have a base cost —so in this example, I'm estimating supervisory and clerical labor at 18 percent of direct labor up here. But I also have scenarios where I have a low scenario of ten percent and a high scenario of 20 percent. So some of that information is built in, and of course the user can customize these numbers to maybe in-house values that they may have, or they can just accept them as is.

The last component of the actual catalyst estimate that I want to show you is the spent catalyst part. Of course in order to really accurately compare two different catalysts, it's essential to understand what is going to happen to each at the end of their life. If one of them is PGM catalyst and the other is not, and we're not considering the value of that platinum metal content, it's going to be a really grossly distorted comparison. So this too —this section of the tool relies on some robust libraries that we have on the back end thanks to our collaborators at PNNL and some consultants that have helped us with this. And the —all the user has to do is complete a set of inputs here. I just deleted a few of these so I could show you the process of entering those. We've already entered that we have platinum on titania that the active phase is 100 percent platinum. And the bulk density, which I have a lookup tool for over here so that if you aren't sure what the bulk density of your catalyst might be, you can look it up. Then I'm going to select that we're using a fixed bed, which will affect how much attrition losses we expect for that catalyst. We do not have any contaminating metals from this list, and if we needed to landfill the catalyst, it would fall under the category of PGM or noble metal catalyst.

So all of these inputs have now generated a series of calculations of how much metal will we lose during the course of running this reactor with the catalyst in it, how much of the support will we lose, and thus how much total catalyst solids will we get after unloading the reactor to send to a refiner potentially. With that information, then we can calculate based on the spot price for the metal —one thing I'll notice that's very important that if you want to be using this in the future that you're updating your spot prices because of course that does vary a lot —it's going to calculate how much recoverable metal value there is there, and then based on information that we have that has not been published in this form before, we also are able to estimate how much it's going to cost to do that recovery. So then you get a net recovery value, and this —the tool is automatically going to choose the best option here, which we can see that the value of the metal at 104 is quite a bit higher than the fees at $6, so it's definitely going to be worth it to recover, and that is what we see down here. In a different situation, if I maybe selected aluminum, we'd find that the best choice for this catalyst might just be to sell it, and then we'd be selecting this sale value instead.

Last thing to show is that we have sort of a range of complexities of outputs from the tool starting with a very simple approach, which is just to show kind of a financial summary style description of all these results. So if we were using a step method, we have a list of the step method outputs here. Step costs and raw materials cost. You can see that raw materials costs are really dominating here, which isn't surprising considering that it's a platinum catalyst. I should go back and change this from platinum to aluminum because it's distorting the results. Then down here for the CapEx and OpEx factors outputs, this is designed to sort of just be directly comparable to something like you might find in an IHS PDP report, formerly SRI. So we have a breakdown of our capital investment, both broken down in terms of its annual depreciation cost, and its cost per unit kilogram catalyst and so forth.

And so the idea here is the user can do what they want with this information, and they can drill down to greater detail. We have more detailed outputs also. I'm not going to go into more detail here because the last thing I want to do is show you the web tool that has been designed to mirror the functionality in this tool, and we even have, I should highlight, we even have thanks to a really great intern working for us, we have the ability to interconvert between the libraries that are found in the Excel tool and the web tool. And the last thing I'll say about that is even if you are using the web tool, all that information is stored locally on your browser, so there's no risk. There's no disclosure of your proprietary information to us. The whole idea here is that you own your library, you can use our tool, we hope it's useful, but it would not be something where you have to worry about where your information might go. So we now are able to use these two tools very closely in tandem. So I'm now going to swipe over and see whether we can —looks like this is working. I'm just going to go to the home page here. I think I have too many things open on my computer. This is the web tool version of CatCost, and as I mentioned, it's designed to mirror what we see in Excel. In Excel, I didn't show you the actual materials libraries and equipment libraries they were farther over in the list of sheets. But you can see that I have materials library and equipment library. I'm going too fast.

And the spent catalyst library all incorporated into this web tool, and that is what allows us to make estimates in here and have confidence that they'll use the same calculation approaches in the Excel version. So I'll open this up, and you can see that the same flow of information is here. We go from one inputs to five outputs. And you put in your basic inputs, materials costs, including adding materials to the list, add pieces of equipment, and we have the factors for capital and operating costs. One of the things that's great about the web tool is we have some really powerful and exciting visualizations, like pie charts where you can zoom in on each individual component. You can click on materials and then zoom in to see what are my metal sources and supports and other materials, and then zoom in from there. And great tornado plots that are customizable on the fly. So it's a really powerful tool, and I think it goes a little bit beyond the capabilities that we have in Excel. So with that, I am already a few minutes over the initial 30 minutes. I think we're ready to take some questions. Thanks very much for your attention.

>>Moderator: So we do have one question. How is the input data for the web-based tool managed? What information is retained on the CatCost server?

>>Kurt Van Allsburg: Great, yeah. So I already touched on this a little bit, but I'll give you a little bit clearer picture of how this works because I know that this is something that could cause concern, and we want people to feel comfortable using the web tool. So the way that this works is the application that you're doing the estimate in, and this is —if there —if a question comes up about some of the technical details, we can refer this to our web tool team. But the application is the only thing that's downloaded from the server, and nothing is ever uploaded to the server —unless we will have one capability to send data over to your personal site on the data hub, the ChemCat bio data hub. However, that would only be your site. It would not be public, and you'll know when you're doing that because it'll say upload to the data hub.

However, that's not —the way you'd normally be interacting with the tool is you go to the website, and that automatically downloads the application within your browser. Then when you create estimates or add a materials library from Excel or from one of the ones we provide, all of that information is stored locally in your browser, and this is important to note, too, because that means if you switched browsers, it would be —you wouldn't see that same information. Now this was important to us because this allows you to have confidence that your information is not out there for potential server vulnerabilities or data mismanagement or anything. So the inputs that you provide are solely stored on your machine, and so even though we call it a web tool, you might think of it sort of as a downloaded browser add on.

>>Moderator: Great, and then is there an ability to maintain multiple databases of cost estimates? For instance, spot prices of metals, etc., if we wanted to evaluate using internal pricing data? And then there's a second question, but you can go ahead.

>>Kurt Van Allsburg: Yeah, absolutely. So in the Excel spreadsheet, it's really quite easy. I think many of us are used to using Excel for this kind of thing. You can just have multiple versions of the Excel spreadsheet that are sort of like your master copy that has one set of raw materials prices that maybe you want to keep separate from a different set. In the web tool, it's also possible, and the reason for this is that let's say we go to the materials library here. What we can do is download the —it's a JSON file for the file format that contains all this information —and it's actually human readable, so that's a nice feature that you can go in there and see what's in there and if there were anything that concerned you, you know exactly what is in that file.

It might be quite large because as you can see, there's a lot of information here. But you can download this file, and then you could clear your library so that then you only have the information in there that you upload from a file. So you can keep control of those versions and those different materials libraries by having different files for different data sets, if that makes sense. So I think absolutely. That works. And if there are any questions or problems with getting that to work for you, please reach out to us. We have this new email address, [email protected], and we will definitely be monitoring that as people are starting to play with the tool.

>>Moderator: And then the second half to that question is are the supplied data periodically updated from the tool side, or is that a user responsibility?

>>Kurt Van Allsburg: Great question. Yeah, so what we've decided on, we were originally thinking —we were kind of debating about whether to have —if we update something on our end, is that information then going to be pushed through to everybody's individual local version of the tool? And what we decided on is giving the user the most control over that. So the way it works initially is if I want to open a materials library JSON file here, I can either drag a file from my desktop —that would be your personal library that you created separately, or you can import it from the data hub. That would be where you get our most up-to-date materials library or spent catalyst library or equipment library.

So the way —the idea here is that you always know when you're choosing to update from the —our version, and I think it's not maybe in there yet, but there will be something that asks you if there's an existing material in your library, whether you want to override it or whether you want to duplicate it, and so forth.

>>Moderator: Last question. How many custom estimates in Excel can I see within the document, or do I see multiple versions of Excel?

>>Kurt Van Allsburg: Great question. So that is one potential limitation of Excel. It is really sort of one estimate per sheet, unless you know some way to do this I'm not aware of. So the way that I recommend handling this is to have —if you're the sort of user who will be adding materials, a lot of materials, or even pieces of equipment you'd expect to use from estimate to estimate, then you might consider keeping a master version of the Excel spreadsheet for yourself, and then you create your estimates from that. I think we looked at other ways to do this, and we rapidly got very large files and things got sort of complicated. So I found this as a somewhat better solution.

So that is the way that is managed. So it's one estimate per file. The Excel —the web tool of course is a little bit more flexible because you have a library of estimates that you've completed. I think I only have one here, but you can just add another, and you can see —I now have these two here.

>>Moderator: So that was it for questions, and now I'm going to unmute –

>>Kurt Van Allsburg: Yeah, if you can do that, it would be great. I think we're ready for —what was the name?

>>Moderator: Siva.

>>Kurt Van Allsburg: Siva, I think we have your —we're ready for you to make your announcement.

>>Siva Sivasubramanian: Okay. Yeah, thank you. This is Siva. This is Siva from BETO. The reason I wanted to talk to you about is that we have an RFI that is currently open. The RFI is Understanding Catalyst Production and Development Needs at National Laboratories. And this RFI will be open until October 12, so if you have some time and take a look at the RFI and provide your responses, we'd appreciate that. And the RFI number is 1951. Again, the title of the RFI is Understanding Catalyst Production and Development Needs at National Laboratories. Thank you very much.

>>Kurt Van Allsburg: Great. And yeah, can they contact you if they need more info about that RFI, or did you say there was a website for that?

>>Siva Sivasubramanian: Yeah, it is available on the exchange, and the RFI is published there, so they should be able to get it.

>>Kurt Van Allsburg: Great, thanks.

>>Siva Sivasubramanian: Thank you.

>>Kurt Van Allsburg: If there are other questions, feel free to type them into the questions box in your GoToWebinar application. Are we unmuted now? So we had one question about the actual download link where this is going to be hosted. So it's not on the website in final form yet, but the web address that you would want to use is catcost.chemcatbio.org. So it's just catcost.chemcatbio, and you can also access it just by going to the ChemCatBio webpage directly, and it is under Capabilities there.

So we will probably be doing some email blasts. Inevitably, you put like a software tool together, especially with a team of over 10 people, and there's some things that come up at the last minute, so we have been wanting to make sure we know everything is buttoned up before we really push it out and advertise it to people, but that is the web address is catcost.chemcatbio.org, and the email address is [email protected].

>>Moderator: Will there be a publication that compares the estimates to actual production costs?

>>Kurt Van Allsburg: Great question. So we have for the step method approach, we actually have an accepted paper in organic process research and development, and that will also be linked from the CatCost website when it goes live. For the CapEx and OpEx factors approach to processing cost, we have compared three catalytic materials. Actually both approaches we found that we can get within 20 percent of some industrial cost data. And so I think for that second one, for detailed CapEx and OpEx factors, that might just go into the documentation that accompanies the tool. We've got a big PDF with all the information you'll need, and that is going to have a change log for each time we update the tool with, we changed this in the web version, we changed this in the Excel version. So just like any other software or Excel-based tool, you'll be able to track what's going on with the development.

So the last thing I'll say on publications is that we do intend to submit a publication that highlights overall this approach what CatCost delivers by doing some performance cost kind of comparisons between different catalytic materials and highlighting the benefit that considering cost alongside performance can provide to a research program. This is really important to us, and especially to advertise this tool and its usefulness and perhaps its accessibility to a more fundamental audience who might find that seeing a paper on it is the best way to convince them that it's worth using. So that is worth coming. And all that information I think will be clearly displayed on the CatCost website as it becomes available.

>>Moderator: Those are the questions.

>>Kurt Van Allsburg: Great. Well it looks like we don't have any more questions, so I think with that I'll say thank you very much for your attention, and it was my pleasure to present this really exciting tool and the work, which has involved a really big team. We have just at NREL we have four or five people that have worked with us on the web tool. We have a great team at PNNL that has developed the spent catalyst component, and the step method as well. In fact, all that work was originated by a guy named John Super, who some of you may know. Really knowledgeable and has been a huge help to us in this project. So big shout out to John. It has been a very fun and exciting team effort and we appreciate your interest and attention. We hope that when it's publicly released very soon that you'll use it and hopefully it's useful to you, and that you'll tell us what we can do to make it better. Thanks a lot.

[End of Audio]