Webinar: Increasing Renewable Energy with Hydrogen Storage and Fuel Cell Technologies

You are here

Increasing Renewable Energy with Hydrogen Storage and Fuel Cell Technologies

Below is the text version of the webinar titled "Increasing Renewable Energy with Hydrogen Storage and Fuel Cell Technologies," originally presented on August 19, 2014. In addition to this text version of the audio, you can access the presentation slides.

Alli Aman:
—technical glitches, which I'm sure we're all very familiar with. I'm 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 business days; however, I will send an email out once those are posted to the website. So feel free to check back to the website, but know that you'll be getting an email from me once they are posted.

Everyone is on mute, so if we can advance to the next slide, there is a way that you can submit questions via the Question box.

[Slide 2]

So you can see here on the example if you just want to submit your questions while today's speaker is presenting, we will cover those at the end of the presentation. So we leave about 10 or 15 minutes at the end of presentations to go over questions. So definitely submit those as the speaker is presenting.

I also encourage you to check back to our website for future webinars. We do host these monthly. Each month it's a different topic, so I encourage you to check back to our website for future webinars. I also encourage you to sign up for our monthly newsletter. That'll keep you in the know of what we're doing at the DOE and also of future webinars.

So on that note, I'm going to turn it over to Dr. Monterey Gardiner. He is a fuel cell technology manager for the Fuel Cell Technologies Office. He's developing a roadmap for hydrogen energy storage, and is the office liaison to the Office of Energy Efficiency's grid modernization effort. And on that note, Monterey, I'll turn it over to you.

[Slide 3]

Monterey Gardiner:
Well, thank you Alli. So good morning everyone. My name is Monterey Gardiner. I'm in the Office of Fuel Cell Technologies here at DOE. And for the last couple years, I've had the honor of working with Josh Eichman in exploring how hydrogen for energy storage might be competitive with traditional energy storage options. So this is going beyond just looking at hydrogen for fuel cell vehicles, which we have done the majority of that in the past.

So Josh joined NREL's hydrogen systems analysis team in January of 2013 as an EERE postdoctoral fellow, and he's very focused on the integration of advanced technologies with the grid. He completed his Ph.D. with a master's in mechanical and aerospace engineering from the University of California, Irvine, with a focus in renewable integration.

He also has a degree with a fuel cell—sorry, a bachelor's degree in mechanical engineering from Clemson University. So it's been a pleasure to work with Josh over these last couple of years and really exploring how can hydrogen compete as an energy storage mechanism, and he was instrumental in standing up an energy storage workshop we had in May that he'll cover in this talk.

So with that, Josh, I'll turn it over to you.

Josh Eichman:
Thanks Monterey. And really excited to be here and have the opportunity to present some of the work that I've done on hydrogen energy storage looking at how we can integrate with the renewables and transportation, and to do this we've performed some experimental analysis as well as modeling, so I'd like to go over both of those topics with you today.

[Slide 4]

So by way of motivation for why we're interested in hydrogen energy storage, I think I see really two main drivers, the first one being we recognize that additional renewables are going to bring challenges to the grid operation, and hydrogen technologies can interact well with the grid as I'll show in the next few slides. But as far as renewables go, their challenge is with over-generation, ramping concerns, additional reserve requirements that really can change how we typically have operated the grid and make it, again, more challenging, and hydrogen has a role to alleviate that. Additionally, there's environmental regulations and mandates, and hydrogen has the opportunity as a potentially clean fuel to mitigate those.

So what I would posit below is that hydrogen can be made both dispatchably and that way can be made from electricity or a clean fuel, and be made renewably also, either clean electricity again, or some type of biofuel. So in that way, it's pretty flexible. And then additionally, it can enable these multi-sector interactions, which I'll get into a lot more detail about later, with the potential to reduce greenhouse gas emissions and criteria pollutants, particularly for the transportation sector, but we also see opportunities in the electric sector.

[Slide 5]

So just want to give you a spoiler alert. I'm about to go through some of what I think are the findings from this so that you can, as we're going through you know sort of what to look for, and then I'll reiterate it at the end. So, hopefully, drive the points home. So the first one is I think through today's presentation I'll show that hydrogen can be operated flexibly and in a variety of configurations, and that is to support either the electric sector or a transport sector by enabling these multiple interactions. So the electric, transport, heating fuel, and then even industrial supply, which is the largest consumer for hydrogen right now.

And then the last one is I want to show that hydrogen can participate in the electricity markets, and has the potential to improve competitiveness and further enable renewables.

[Slide 6]

So as I mentioned, I'll go through some of the system configurations and then I want to give kind of an electricity 101, so what does it take to operate the grid, and then how can hydrogen technologies fit into that. And then show some experimental flexibility tests we did, and the ones I'll show are for electrolyzers that was done here at NREL. And then in terms of some of the results, I'm going to perform a techno-economic comparison, so conventional storage technologies, and then how hydrogen compares to all those cases. And then also perform an energy capacity sensitivity, so essentially look at how much storage is valuable and from a market perspective would more storage be more valuable.

And then the next two bullets I don't really have enough time to talk about, but I think are really important, so I included in the backup slides. For those I'd look at how increased renewables are going to impact the value for hydrogen technologies moving into the future, and then also looking at impacts on the larger grid system in the event that you have hundreds of megawatts or even gigawatts of capacity installed for hydrogen.

And then the last one, again, Monterey mentioned briefly the workshop that we held in May, so just give you some of the highlights from that, and a few of the results.

[Slide 7]

So as I had mentioned, hydrogen has this ability to integrate multiple sectors, and how does it do that? So what you see here is the electric grid, basically hydrogen in the form of some kind of storage, and then the natural gas grid. So technology, like an electrolyzer, is going to be able to take electricity and water, split that into hydrogen and oxygen, and we can use that hydrogen— as for oxygen, there are opportunities for its use, but it isn't typically used at this point. And then when you're looking at integrating the natural gas grid, kind of the corollary for natural gas would be a steam methane reformer. So, again, you take natural gas and water and through a series of reforming steps, you're able to get purified hydrogen. And then in reverse, for the electric sector, you can take a fuel cell, turbine, internal combustion engine, take hydrogen, and then turn that into electricity.

Similarly, a corollary for natural gas would be pipeline injection, so you can take hydrogen that you've made through any means and put that into a natural gas pipeline, either through direct injection, which would be putting hydrogen molecules directly into the gas, or methanation, so you take hydrogen and you make it into methane, and then you put it into the natural gas system.

So in addition to—those are just two of the outputs. Another output is for fuel cell vehicles, so enabling that transport link. And that can be hydrogen from any source, similarly chemical and industrial processes. So I think that is pretty well understood, and I feel that what I really want to focus on today is this ability to provide grid services, interacting with the electric sector, providing additional revenue for those systems, and then looking at it in the context of all these different configurations. So you could have an electrolyzer that puts hydrogen directly into the pipeline, or an electrolyzer that uses some of its hydrogen for a fuel cell and some for a vehicle, so in that way it can be pretty flexible.

And there's sort of a growing interest in both literature and demonstration projects in this idea of power to gas, so I just wanted to briefly describe it a little bit. It's really the idea of taking power in the form of electricity, converting that to gas in the form of hydrogen that can be used for a variety of these processes—typically considered taking electricity and putting it into a natural gas pipeline. But I think the definition can extend as wide as providing fuel for transport or chemical, as long as that fuel's used in its gas form.

[Slide 8]

So now that I think we have an understanding of sort of how these configurations are set up, is anyone doing this? How many of these projects exist, and where are they? So there was a paper released in 2013—you can see the first bullet point there—that it looked at different hydrogen projects, and it said that there were 41 realized projects which include laboratory, include demonstration projects, as well as both systems that are commissioned and out of commission. So that includes kind of the energy storage-type projects that would have either an electrolyzer or an electrolyzer and fuel cell combination. And you can see that as a group, the highest is Germany, and then USA and Canada.

So then if we drill down a bit, the next bullet, so Germany had a paper that was released that said they had 22 green hydrogen power-to-gas projects, so you can see on the right-hand side there's a figure and those projects include mobile applications as well as the power-to-gas as I mentioned before, also green hydrogen. And then if you look in the northeast corner of Germany, there's the sort of medium blue that takes wind, and they have a lot offshore wind, and then converts that into hydrogen, so it includes those energy storage projects as well.

And then just announced, there was a two-megawatt power-to-gas project planned for Ontario, Canada, and for this project, they want to use a system to provide energy storage for grid management as well as grid services, and in this case regulation. And I'll describe a little bit more about what these kind of grid services are next, and what the grid needs to operate.

[Slide 9]

So the first thing, probably the most important, about the grid is that that you have to balance the demand with the supply or production, and that has to be done on a real-time basis. So we don't know exactly what our electricity demand is going to be so we have these day ahead projections. So if you have a day ahead projection, as the hour comes around, we have a better projection. And the difference between those, there's a lot of different names for it, but we'll just call it hour adjustment here.

Similarly, once the five minutes before that time period when we know we have to provide some amount of electricity comes on, we have a better projection for how that is. So the difference between those we're calling load following here. Down on the chart you see LF, that's load-following adjustment, and that ends up being less rapid changes, a little bit more coarse, but it can be a significant amount of energy and difference between hour ahead and 5 minute.

And then actual and real-time, the difference between those is what we call regulation, and this is used to correct any of the energy imbalances on the very rapid time scale, and keep the grid's frequency maintained, in the U.S. around 60 Hertz and Europe around 50 Hertz.

So in addition to being the energy bounds, they do that by providing  bulk energy as well as some of these ancillary services. So the first two I'd already mentioned, load following and the coarse adjustment, regulation is a finer adjustment, but also you want to be able to manage in case there's a plant outage. So we have spinning, non-spinning, and other reserves, and so those first five are the ones I'll really talk about. I'm not going to get into voltage support or black start today.

[Slide 10]

But in the figure, you see—again, you want to keep that frequency around 60 Hertz in the case for North America, and do you see, OK, we're moving along this timeframe around the 60 Hertz, and then all of a sudden, we lose 2,600 megawatts of generation on the grid, which causes the frequency to drop because there's not enough generation on the system, so what happens—if you look at the top half of that chart, the first thing is the regulation, or frequency response is called up. So all the capacity available is called up, and it tries to correct that imbalance. You see a little blip down at the bottom where it's started to call the frequency, and then as soon as possible, you call the spinning and non-spinning reserve, which is essentially generators that are either off or already spinning with additional capacity, and they pick up and replace the regulation so that it can go back and provide its normal operation.

And then a few minutes later, about 25 minutes, and then approaching that 30-minute mark, you have other reserves, or operating reserves, that then replace spinning and non-spinning reserves. So that's sort of how the flow goes. The frequency response is first, then spinning and non-spinning, and finally the other reserves are operating. And then that enables you to experience another grid outage, in case another plant goes down, you have the additional capacity in spinning and non-spinning waiting in case there's another failure.

[Slide 11]

So then the last market we'll look at is what's called a capacity market. So this one's a little more abstract, and I'm going to give you kind of the general idea of this one. But at the bottom there's a reference if you want more details. I also have some other references if there's interest. But the idea behind a capacity market is this is what they use to make sure that you have sufficient capacity on a longer term. So you can't build a plant overnight—some kind of power plant overnight, so you have to have the capacity market as a mechanism to enable new generation to be built.

And typically what this capacity does, capacity in the case of payment, is that they'll pay you to install your generator, and it basically should offset your capital costs. So in terms of the work that I'll present for the economic competitiveness, we assume that we're getting a capacity payment, and we're assuming that value is based on the cost of new market entry. So this is what it would cost to install the next megawatt of power, and we're assuming that's $150 per kilowatt year.

So those are the kind of three things we're looking at, the energy, the ancillary service, and the capacity market. So one thing to note about these is, again, each of their values vary, and they vary both in time and then throughout the year. And what you see in the figure to the right is historical prices for California's independent system operator in 2012—and I'll actually use these in the analysis later—and what you can see is the average price for electricity is around $25, denoted by the green bar in the plot furthest to the left, and then there's also a range around that, so the prices fluctuate.

Similarly, with the reserve products, their prices fluctuate, and then they're in sort of decreasing value. So the resource that needs to respond the fastest has the highest value, and then those go down. So regulation is the highest paid resource, then spinning reserve, and then non-spinning reserve.

[Slide 12]

So now I feel like I've established this idea of what are the grid services, how can you provide them, and then the idea of difference in value. So now I want to take a look at hydrogen technology specifically, and how it can relate.

[Slide 13]

So at NREL, as a cooperation between Xcel and the DOE, we built the wind-to-hydrogen project. I think that was around 2009. And what this project does is it combines renewables for wind and solar and then those can be used in electrolyzers, so that's wind and solar, so the left and the upper left of the figure, and in the center you see there's different electrolyzers, which, again, takes that electricity and turns it into hydrogen. And then that is compressed and we can use that in different forms. So we can either use the hydrogen directly to generate electricity through a fuel cell or an internal combustion engine, or we can compress it further and use it for transport.

So again, it's this idea of integrating renewables, integrating these responsive loads in the form of electrolyzers as demand response devices, because that's essentially what they are. And if you look at this entire figure, the most important part in this is going to be those electrolyzers. So what they're going to do is they're going to be able to see fluctuations coming from the renewables, or even from the grid if you connect them to the grid, and they're going to be able to respond to that.

[Slide 14]

And so how fast are they going to be able to respond? We performed some electrolyzer flexibility tests to see both start-up/shut-down, turn-down response time, ramp rate, and frequency response. And we used 40-kilowatt units, so you're not talking megawatt scale yet, but still it's a decent size.

[Slide 15]

And so now what you see is one of the tests, I think this is one of the more interesting ones, that involves ramping the units from low power to high power, either 25, 50, or 75 percent power to 100 percent, or vice versa ramping down. And this will show you how fast can you respond and based on what the market requires, are you going to be able to compete in the market. So that's what we want to answer toward the end of this.

OK. So what you see is there's—basically you're staying at whatever output you started, then there's a trigger where you're supposed to ramp up to whatever the set point is. And we want to look at what's the initial response time. So how long does it take for an appreciable difference, in this case, plus or minus one percent max current, which is noted by the vertical lines that are different shades of blue in each of the figures. And that happens on the order of milliseconds, so we'd begin to see a response on the order of milliseconds at the stack level. And then how long does it take to settle to the final set point? That takes on the order of seconds.

So what we're saying is electrolyzers can rapidly change their load point in response to grid needs. So we also want to see if they can respond fast enough to be able to stabilize the grid. So in the previous one we saw that they can respond very quickly and maybe they can provide some of the slower-moving services, but regulation is the fastest service that they'll need to respond to, and how can they respond to that.

[Slide 16]

So as part of work that was done by Kevin Harrison—and you can see the link down there to the left, he established a microgrid using a synchronous generator and a load simulator at the Wind Technology Center, and then used electrolyzers to then put those on the system and see if when the load simulator is changed, so a change that either causes an increase or a decrease in the power requirements of that microgrid, and then that's going to cause a frequency imbalance, and then see if the electrolyzer can respond.

So I think it's easier to describe by looking at the figure, and what you see is the green line is sort of the base case, so the electrolyzers aren't responding to anything. And this is the natural response from the synchronous generators. So there's a change in the load simulator, so it's going from zero to ten kilowatts, and that causes a frequency sag, and then you're going out of the bounds, which we're setting at 0.2 Hertz, and then eventually you come back, on the order of one second.

But if you're able to then use the electrolyzers and dispatch those in response to that—in this case under-frequency, then you see that the frequency can be—it recovers much faster so that instead of being one second, it's 0.2 or 0.3 seconds response. So we're saying that electrolyzers can accelerate frequency recovery, and in this way, they can provide regulation.

[Slide 17]

So now if we compare the test that we just did to the requirements for the grid, we can see how electrolyzers stack up, basically if they can compete. And the figure you see here is a list of a lot of the different resources. So the top three are contingency in case there's an outage; then regulation and load following are provided during normal operations, that's the blue bars. And then I won't talk about voltage control today, but the left side of the bar represents how fast you have to be able to respond. Again it's that response time, which we said was on the order of seconds for electrolyzers. So you're exceeding all of the bars that are there, so you're able to provide all those services from that point of view.

The right side of the bar represents the duration that you'll need to provide that service. So in the case of load following, you need to be able to respond in five minutes, but you need to be available for 100 minutes or more. And because electrolyzers are a demand/response device in that they're a load side device that you're able to turn down to provide equivalent generation, and you can turn those units down for however long is necessary. And then that way, again, I'm saying that those devices can meet all the services required.

And then lastly, this is supporting for non-spinning reserve the electrolyzer shutdown time is on the order of minutes, 1 to 6 minutes, so we're saying that indeed, electrolyzers can respond fast enough and for sufficient duration to participate in all the electricity markets shown here.

[Slide 18]

So it turns out it's not just being able to provide those services, but you have to be the correct capacity as well. So we've just established what the services are, how electrolyzers could respond to those, but now it's a question of how large are the units. So there's a minimum capacity for bidding in the market, and as of 2006 in parts of Germany this is 50 megawatts and 30 megawatts for the minute reserve, and 10 megawatts for the primary and secondary markets.

But I think the real take away from this one is that we're starting to see this minimum capacity requirement come down, and at the same time, a lot of the times this capacity can be aggregated so you can take multiple small units and put them together and bid them as one unit. So I think with those two facts, combined with the fact that manufacturers are scaling up electrolyzer sizes, then you're really seeing sort of this nexus where the grid capacity requirements are falling while electrolyzer manufacturing is scaling up, and they're both sort of meeting around that megawatt scale right now. So electrolyzers are 1 to 2 megawatts plus, and the grid requirements are coming down one megawatt, half a megawatt better. So now I think we've established both the ability to, and electrolyzers in this case can meet the capacity requirements as well.

[Slide 19]

So now I want to move into the techno-economic analysis I had mentioned before, and we use three techniques. I'll step through each one, and then the one I'll show more results for is the Price-Taker. So what exactly is a Price-Taker model? It takes historical or modeled energy prices, reserve prices, hydrogen price, and operational parameters, that's what goes into the model, and then it performs time-resolved co-optimization for both energy and ancillary service. And what comes out is the maximum revenue that you can get from arbitrage, which would be this idea of buying low/selling high, as well as ancillary service in the sale of hydrogen.

So the assumptions we're making in this model is that there's sufficient capacity in all markets, so essentially you're not going to bottom out the regulation market. You don't have so much capacity that you're going to take all of the regulation market for yourself, which is I think a reasonable assumption particularly as we're starting to roll out the technology. And then second, objects don't impact market outcome. So you're not large enough to impact the prices, which is again I think reasonable now. But in the future, that may be not be reasonable, so we're also interested in looking at that.

[Slide 20]

So the second type of modeling approach I looked at was this production cost model. So for this model, it's a large-scale grid simulation that takes all the generators and transmission lines from a certain area. And you can see down on the left, there's an example of California where there are several thousand generators and similarly, several thousand transmission lines. All that goes into model, along with what are the load requirements and reliability, which would be all the ancillary service requirements, any other systems constraints. And then, again, it performs a co-optimization for energy and ancillary service. And in this case, it does so to minimize the system production cost. So you want to minimize the cost of energy coming out.

And what you get is how the generator should operate, when they're going to use fuel, what are the emissions, the load served. And then I think the last two bullets on the right are pretty important, this energy prices and reserve prices come out of that. So I'll link that in the results section, so we'll come back to that, but, again, I just want reiterate, I didn't have enough time to present on the results from this one, but I did put the slides in backup so if you wanted to look at them afterwards, and, also, they're available in my Annual Merit Review presentation that I gave earlier this year.

So the positive side of the production cost model is that you have this flexibility to look at a lot of different things in detail, both temporal and spatially. One of the down sides though is it takes a long time to run through it, it can be days or weeks of potential run time depending on the size of your system. So that's where the Price-Taker has a benefit because it can run very quickly.

[Slide 21]

So now I want to look at a hybrid of those two. So if you can run the production cost model for more generic cases, for instance, look at increased renewables, or look at what happens if you change the gas price, or what if the market design on the electricity side changes.

And you don't want to run many, many scenarios in the production cost model. You can run a few of the basic ones, and then you can take the resulting prices from that, along with the hydrogen price and whatever your technology's configuration is, and put that in the Price-Taker, which runs very quickly. So in this way we can form this kind of hybrid approach. And the example that I, again, don't have time to discuss today but it is in the backup slides, is one where we look at in the case of California if we were to install many times the renewable penetration that's now. I think we scale all the way up to around the 60 percent level, and then see how that impacts the energy and reserve prices, and then that filters down to the Price-Taker. So again, we can look at how hydrogen technologies compare in both today's grid as well as in the future grid by using this hybrid modeling approach.

[Slide 22]

So now I want to get back to the first method I talked about, which was the Price-Taker model, and we use in this case historical prices for California in 2012, and we're assuming that all the units that we're looking at are one megawatt. We're going to compare conventional technologies of pumped hydro, as well as the lead acid battery, with the hydrogen technologies of a stationary fuel cell—in this case, we're looking at a PEM—and a PEM electrolyzer as well, and then the kind of competitor to producing hydrogen in this case, the steam methane reformer, or the technology that's being used the most right now.

So I won't go through all this. This is really for reference purposes, but I do want to show a few things. So the capital cost in fixed O&M, you see there's a range there. So what we're doing is representing on the low side, that's a more forward-looking value, future value, optimistic, and then on the high side, that's more of a current pessimistic-type value. So I'll show that range in the next few slides, and when you see the range, you can know that, OK, depending on how you think the technology's going to be affected, you can move to wherever you want on that part.

[Slide 23]

So here are the results. It's going to be a complicated figure in the end, so I broke it into little pieces. And really what I'm going to show is the yearly revenue, so that's the output of the Price Taker—we took all these energy prices, reserve prices, and hydrogen prices, optimized that to see what the maximum revenue you could get was based on your technology configuration, and then want to compare that to the cost. So we annualize the cost for that system over its lifetime, and then compare the two. Obviously you want revenue to be greater than the cost, so in the figure, you want the blue bar to be greater than the red bar.

And as I mentioned, the red bar in this case represents the fixed O&M, as well as capital costs and an installation multiplier, and has the values ranging from low to high. In terms of technologies, what you see down at the bottom, there's a key to the right, but HYPS is pumped hydro, and the services that it can provide are designated Eonly here, which is energy arbitrage only. So you're only engaging in energy markets, just shifting when you're buying electricity, when you're selling electricity, without providing any ancillary service.

And then the next column, the HYPS All, represents a case where you're actually selling ancillary service. So if we compare the revenue from those, which is the blue bars, the revenue from providing all services is greater than just providing energy, which is a theme throughout all these results. So if you're able to integrate more tightly with the grid, you're going to achieve more revenue.

[Slide 24]

Now I want to expand beyond just the conventional technology—oh, let me first say when you think of a technology that would be competitive, you look at pumped hydro. There's a lot of that installed. It may be geographically constrained at this point, but it's a technology that's generally competitive, so you see its revenue is on the order of its cost. But for the battery, when you look at a lead acid battery, they're not installing them necessarily because they're competitive at this point. They're just in very specific applications.

So what we're showing is that the revenue doesn't quite meet the cost, which I think is another kind of sanity check here. And now if we expand it to include hydrogen technologies—so all these first eight technologies do not include the sale of hydrogen, so you're strictly electric storage devices. Electricity in/electricity out, you're able to in this case integrate the electric sector with hydrogen and provide grid services.

And then a little later on I'll look at integrating some other sectors, but what you see is for the last four technologies, FC-EY, which is fuel cell-electrolyzer, again Eonly if you're engaging only in energy markets, FC-EY All, all services, and then the last two are FC-SMR, so fuel cell coupled with a steam methane reformer. So we're saying steam methane reformer is inflexible, you're not going to be able to provide as much grid services, but the cost might be a little bit lower. I think independent of that, though, the revenue that you're able to achieve from strictly providing electric services isn't going to be sufficient to pay down your capital in this case.

[Slide 25]

So what if we look at a case where you actually do sell hydrogen? So these last eight bars represent cases where you sell 80 percent of the potential generation of your hydrogen, or every one's a megawatt plant, so we're selling around 400 kilograms per day of the capacity. The first four on the right side of eight are the same. You have the fuel cell electrolyzer that can engage in energy markets, and then all, and you see this revenue goes up a lot. So what does the blue bar represent? It's a range of potential prices that you can sell your hydrogen at. So we're saying if you think in the future you're going to be able to get $3.00 per kilogram sold, then that's the bottom end of that bar. If you think you can get up to $10.00 per kilogram for your hydrogen, then that's the top end of the bar.

So this is another one. Both the cost you can decide how optimistic you are and as well as the revenue you can decide how optimistic you are about the cheap sale price. And what we see for those devices that can provide electricity storage, so those first four, as well as selling their hydrogen, being able to sell the hydrogen greatly improves the economics for your cases here. So if you're going to increase your revenue significantly while the cost stays the same, you're able to get much higher revenue. And I think some of the most interesting cases are the last four.

So if you look at the SMR baseload, that's conventional technology, again another sanity check. It's the technology we use the most and its revenue is significantly greater than the cost. But if we look at the electrolyzer baseload, and the last three technologies, you see that, again, tighter integration with the grid is going to provide you more revenue. Being able to provide all services is better than energy only, which is better than your typical baseload operation.

And, really, those last three are pretty promising, particularly for the case when you're looking at increased renewables. But those particularly when you're being able to provide all services, you're going to have more revenue than your cost, again. So that's another one of the potential positives about this system configuration.

So I want to provide one more set of results, and then we'll go into workshop findings. And what it—so now as I mentioned before, I'm going to do an energy capacity sensitivity. So, basically, how does the amount of storage you can provide—or the amount of storage you have—affect the economics of your system?

[Slide 26]

So from a cost point of view we're using aboveground storage so the cost will increase linearly as you increase storage capacity. So how's the revenue affected? In current energy and ancillary service markets, there's no mechanism to achieve revenue from energy that you're going to sell [break in audio] now, right? So there's only a slight increase in the achievable revenue from that, so what we're saying is that more storage is not necessarily more competitive in current energy and ancillary service markets. If those markets change, there might be opportunities in the future, particularly in the capacity market area, but as of right now, it doesn't appear that there's that much more value.

So then you wonder what's the optimal storage. Well, it's the minimum to get whatever you're trying to do across. So maybe around the eight-hour mark it'll give you a little bit of extra revenue while not increasing your cost significantly, so you can integrate well with the grid while also providing hydrogen for whatever your final products are.

[Slide 27]

So now I want to move away from the results and go into the Hydrogen Energy Storage Workshop. This was held in Sacramento, California, in May, and it was convened by the U.S. DOE and Industry Canada. And really, we brought together groups of stakeholders to identify challenges, benefits, and opportunities for hydrogen energy storage applications. And some of those would include grid support, intermittent renewable integration, hydrogen vehicles.

[Slide 28]

So who attended? There were 65 participants, and really a pretty diverse set. We had both federal government from the United States and Canada, as well as state government, and then utilities, academic, I mean you see the split there. One important one is the industry. It's good to have their feedback on all this as well.

[Slide 29]

So in terms of some preliminary findings, to sort of whet your appetite, this report is in review right now, but will be released in the next few weeks. And some of the example findings from that is we wanted to isolate what are the criteria and barriers that are hindering hydrogen energy storage systems. And what the participants recognized is that they both want to understand better the technical and economic viability, and see if that is a barrier. So I think that was both a level of necessary understanding there, as well as concern that it might be a barrier. And then similarly, multiple end uses. So we recognize that it could be used for multiple end uses, but better describing how those could be used and systems configuration type of ideas I think was important.

So it wasn't just technical. It was also policy, and one of the important results from the policy was that everyone thought that hydrogen energy storage should receive equal treatment and credit in the market. So one example was the low-carbon fuel standard. But it's also hard when you're engaging these multiple sectors to use credits from each of the sectors and combine them. Sometimes that's not allowed. Sometimes there's no mechanism for it. And then finally, we also wanted to look at what are the next steps. What do we need to do going forward? So the participants recognized that the technology is there, that each of the components are there, but there was a need for demonstration of pilot projects to sort of prove out some of these business cases that both I've presented today and then even other ones that we can talk more about in the report.

[Slide 30]

So the conclusions from the analysis if you look at the flexibility for hydrogen equipment, we focused on electrolyzers in this report, but electrolyzers can respond sufficiently fast and for long enough duration to participate in electricity markets. And then from the economic viability point of view, we really recognize that there's a need to sell hydrogen, that this strictly providing electricity in/electricity out isn't going to be a good business case, that you're going to need to sell hydrogen. It's more valuable than electricity, so if you produce it, it needs to be sold. Similarly, tighter integration with the grid is going to increase your revenue, so being able to provide ancillary service is better than being able to provide just energy services. It's better than the traditional baseload operation.

Also, electrolyzers operating as a demand response device are very favorable prospects. So they showed a potential positive revenue, and it's a technology that's well developed, and you can also provide grid services for renewable integration in that case. And then lastly, more storage is not necessarily more competitive than the current energy and ancillary service market.

So that's what I wanted to go through. If there are any questions I'll take those now.

Monterey Gardiner:
Hello, Josh. This is Monterey Gardiner. Do you have the question list up or is there somebody there that can show the questions to you?

Alli Aman:
Amit, you'll have to read those questions to him. So you have access to be able to read the questions.

Monterey Gardiner:
OK. So we have a handful of questions here, Josh. Probably the first is about the round-trip efficiency for electrolysis and then back to electricity. And I can take a stab at it, or you can answer.

Josh Eichman:
I'll take the first cut, and then if you want to talk about anything else. I think we recognize that when you combine the—both the fuel cell and electrolyzer, if you're going electricity in to electricity out, it's not a very favorable proposition, particularly in comparison to some battery technology or even the pumped hydro. So the round-trip for that, we were showing maybe it's a little bit low. We were showing a 40 percent fuel cell efficiency, and again for the electrolyzer 70 percent, which gives you about 28 percent round-trip, which isn't very favorable compared to 80 or 90 percent for the battery.

But I think the important takeaway there is, yeah, you don't necessarily want to do this electricity in/electricity out. If you're using electrolyzer-produced hydrogen and it's already more valuable than electricity, you should sell it as that, and then use the demand/response capabilities of the electrolyzer to provide grid services. And that way, again, looking at these multi-sector interactions, you're not staying in just electricity. You're able to expand to the transport or to the industrial. So hopefully that answers the question.

Monterey Gardiner:
Thanks, Josh. That was great. Just the second half of that was, any ways to increase efficiency? And so I think I'll just make two quick points. If you look at high-temperature electrolysis that has an opportunity to increase the efficiency on that side by about 30 percent or so. So either just high temperature, whether it's nuclear reactions or something else, but 1,000 degrees C. The other option is looking at solid oxide electrolyzers. So we have solid oxide fuel cells. There's some research looking at doing that in reverse as an electrolyzer for higher efficiency.

And then on the generation side, so if you look at a fuel cell, these are based on just oxygen from the air. So you have parasitic losses from pumps and blowers, but if we were to capture that oxygen, that could also increase life of the fuel cell and increase efficiencies at some cost for storing that hydrogen.

And then the last item is if we're going to scale up, we're considering or there is some interest in looking at hydrogen and oxygen combustion turbines for zero-emission, very large-scale power production. And then the final part as far as efficiency, we're really looking at curtailed renewables where you have so much renewable energy that this would otherwise be demand or transmission constraint. So if you have renewables being produced when it can't be used if it's too much in the middle—solar in the middle of the day or too much wind at night, if you're not using that electricity anyway, the efficiency is not really the question. It's a question of economics.

One other question, Josh. Let me just kind of scan through here. There was a longer question about just the interaction of electricity and natural gas, right? So you can burn natural gas to produce electricity, and you bypass hydrogen, and then if you're looking at low-cost natural gas or CO2 emissions regulations coming up, how does that affect hydrogen as an intermediate. So I think this question somewhat bypasses the reverse of that, of natural gas to electricity, but looking at electricity into natural gas. But Josh, you want to say anything about that interaction?

Josh Eichman:
Well, yes. I'd say that's sort of the conventional way, going from natural gas directly into electricity, and providing service at that point. But I would just say for the hydrogen you're going to enable, in addition to the transport area, that there are natural gas-fueled vehicles. They're just not as prevalent as some people would like, I suppose. So being able to create this fuel as well, and then having the flexibility to go across these—providing energy to multiple sectors I think is a really interesting one. So it sounded a bit like a comment from their point of view. You think I addressed it there?

Monterey Gardiner:
Yeah, definitely. And I think this question gets into the results of our workshop, which was really these cross-sector policy supports. So whether you have a renewable portfolio standard or you have some type of a carbon tax or some policy that's looking to reduce carbon, hydrogen fits in between those two, and so it doesn't necessarily take advantage of these policy support items. So it takes more explanation of what the value of hydrogen is, this intermediate transfer is.

There was a question about PEM versus alkaline electrolyzers and response times.

Josh Eichman:
Yeah. So we had tested both, and as far as being able to participate in electricity markets, it looked like alkaline were sufficiently fast. We recognize that they might be a little bit slower than the PEM technologies, at least in terms of their response time, and there may be some other options. We didn't do a turndown test, so we didn't look at how far you could turn down an alkaline unit. But—so I think generally what we're finding was that PEM was slightly faster, but in terms of looking at being able to participate in electricity markets, there wasn't really a difference between those two. I think they both would be able to.

Monterey Gardiner:
OK, thank you. There was a question on Slide 24—I don't know if you can flip back to that slide—and why the revenue from batteries was higher than everything else.

Josh Eichman:
Let's see. The revenue from batteries?

Monterey Gardiner:
Battery all is higher than—

Josh Eichman:
Oh, yeah, so that one's strictly a round-trip efficiency question. You're able to purchase electricity at the same price as someone else, but you're able to purchase less electricity and then sell it for the same amount. So that's why the revenue is higher. And then for the fuel cell technologies, that's why it's lower because you have the round-trip efficiency.

Monterey Gardiner:
There was a question about the efficiencies on your assumptions table for the fuel cells, the 40 percent lower heating value, and I don't believe that included co-generation. That was electricity production only. Is that right?

Josh Eichman:
Right. Yeah. So there's some concern there. You do get waste heat, so you might be able to do some kind of a bottoming cycle, but as far as an actual co-generation goes, I'm not sure how that would affect the flexibility of that system. It's the same—I mean we—PEM versus SOFC, so there's questions of flexibility across even the different fuel cell technologies. That's why we're saying we're modeling it after a PEM. It could have been an alkaline, like I said, I think those might be able to respond fast enough.

Monterey Gardiner:
OK, here's an easy question. Can you give some examples of electrolyzer manufacturers?

Josh Eichman:
Yeah, who sent this question in? So the two-megawatt power-to-gas project in Ontario is being done by Hydrogenics, and they also have a number of projects in Europe, and then I've talked a little bit with ITM Power and Proton and Giner, so those are the four that I'll talk about—that I would basically mention today. Hopefully no one feels offended if I didn't include them, but I guess Versa Power, too. They also have a solid oxide fuel cell electrolyzer, reversible, which is pretty interesting and might prove a little higher efficiency. So that's my shameless plug.

Monterey Gardiner:
OK, sounds good. So there's quite a few questions. I'm scanning through them as we come down to our last five minutes or so.

Here's an interesting question about emerging economies in other countries. How well do you think grid storage balancing will work in highly unstable grids, such as is typical in these emerging economies, and does it make sense for these countries to invest in grid services, I guess presumably around hydrogen?

Josh Eichman:
Right. So that is an interesting question because especially in the case when you're building up your infrastructure. And the cases we looked at in California, you're competing against stuff that's already there. They already have relatively good reliability. They already have a system that has sufficient capacity. You're not needed a lot of times. But in a place that's a developing market or trying to improve their reliability, it might present some opportunities to actually come onto the scene and be able to achieve either maybe a higher capacity value—I mean there are concerns if your grid is unstable and it goes down often that you're going to have some process issues if you want to produce hydrogen all the time. But I think if you're moving into this area and you're able to provide services that aren't already provided by someone else, there's a nice opportunity there. So, yeah, I would say it would be an interesting thing to do in those areas.

Again, because—so the electrolyzer is demand response, I want to get that across. So you already have to have generation available to meet your demand, and then in the case of an electrolyzer you could turn down to whatever power point and essentially shed load, but you still need your generation. There might be a complication there. But yeah, I think you can provide services from it in principle for any grid.

Monterey Gardiner:
Thank you, Josh. So I'm scanning through this. There's a question about looking at hydrogen mixed with natural gas for kind of lower-emission combustion. I think we covered that briefly just in relation to the high capital cost of the fuel cell. Let's see.

Josh Eichman:
So I have seen a little bit of work on that sort of mixing hydrogen in for emissions, and then that would kind of play into the blending pipeline injection. So yeah, I think it's interesting. It's something that I haven't looked at. But again, I think it would be interesting for some more information about that to come out.

Monterey Gardiner:
OK. Here was a question about kind of how those cost points were chosen for the $3.00 to $10.00, and whether or not there's like a break point for what the sale price would have to be. So I guess that's basically our distributed generation cost analysis. But if you want to say anything about the model and trying to work back to price targets.

Josh Eichman:
Right. Yeah. So that's one of the challenges from looking at strictly the production side in this case, and at the distribution side. There are nice opportunities for, for instance, one of those electrolyzers to be on site and not have a distribution cost. But there's a lot of details with that. There's already a lot of details that have to be rolled in to get the stuff that I'm presenting here on the figure on 25. So that's why we sort of side-stepped that by saying there's this $3.00 to $10.00 value and that represents a range, and you can decide whatever you want from that. So I think looking at some of the distribution stuff is interesting. I think it needs to be kind of parallel analysis in this case.

Monterey Gardiner:
OK. That sounds good. So for the last two questions, maybe we can combine them. What are the technical or policy steps that would make this longer duration or increased storage more competitive, and then the companion question to that maybe is FERC recently, or rather a federal court struck down FERC's requirement for demand response. I haven't followed that much, but anything about kind of policies that would encourage increased storage or longer duration kind of power generation.

Josh Eichman:
Right, and I think that would actually require some adjustment in the market because right now, the markets set along the hour. You have some day-ahead capacity, but if you're looking at shifting, that's just barely in the range. That's like 24 hours of capacity. You need more of like a week-ahead kind of thing to be able to monetize that. And I know pretty much the provider of this longer-term storage right now is pumped hydro. So they can work out specific bilateral agreements with you, depending on which regions you're in. So there are some opportunities there, but more generally in a regulated market, like in California or New York, or one of those, I think that you might actually need some other mechanism to really prove out that value.

And as I mentioned maybe the capacity market is a venue. So if you can prove that you not only can provide capacity, but you could provide it for a week, then maybe they can use that capacity market mechanism to get you some additional value and to incentivize you installing that. Again, it's a question of if it's necessary as well. If the grid is able to operate without it, and it's at a lower cost, it begs the question why would you install more storage and operate at higher cost just so you can then use it at a later time.

But I think there's a few questions that need to be answered. But in terms of monetizing it, I think the capacity market would be a nice opportunity.

Monterey Gardiner:
All right. Thank you very much, Josh. I guess we'll take this point to go ahead and close out the webinar. We had a lot of really good questions, so please go ahead and forward those questions to Josh or myself. I hope you found this informative. This month, we're going to be releasing a draft report from that energy storage workshop, and by the end of the year, we'll have something that can be shared publicly. So I want to thank everyone for their time, and looking forward to your questions.

Josh, you want to say a few final words?

Josh Eichman:
Yeah. Just thanks everyone for attending and, again, thanks for the questions. I hope you found it informative and as Monterey said, I would be available for any additional questions or comments. Just feel free to contact me if there's anything you're interested in knowing more about. And thanks, again.

Monterey Gardiner:
All right. Thank you Alli, and thank you Amit for the support in getting this webinar scheduled and set up. It was a great opportunity.

Alli Aman:
Great. Thank you guys so much. And just a reminder, slides and the recording will be on our website in roughly about 10 business days. So thank you guys so much.

Monterey Gardiner:

Alli Aman:

Monterey Gardiner:
Everybody enjoy the rest of their week. Goodbye.

Alli Aman:

Josh Eichman: