Eric Parker, Hydrogen and Fuel Cell Technologies Office: Hello, everyone, and welcome to the first H2IQ hour of 2021, part of our monthly educational webinar series that seeks to highlight research and development activities funded by the U.S. Department of Energy's Hydrogen Fuel Cell Technologies Office, or HFTO, within the Office of Energy Efficiency and renewable energy. My name is Eric Parker and I'm the HFTO webinar lead. As always, we'll be announcing more topics like these soon, as well. So, stayed tuned.
So, really quickly, I'd like to go over some housekeeping items. This WebEx call is being recorded and maybe posted to DOE's website or used internally. If you do not wish to have your voice recorded, please do not speak during the call or disconnect now. If you do not wish to have your image recorded, please turn off your camera. If you speak during the call or use a video connection, you are presumed to consent to recording. But that being said, all attendees will be on mute throughout the webinar. So, please, submit your questions in the Q&A box you should see in the bottom right of your WebEx panel now. You can submit those at any point during the webinar today, and we'll cover those during the Q&A session at the end of the presentation. With that, I'm going to turn it over to our super DOE host, Neha Rustagi, to introduce today's topic and speaker. Thanks, Neha.
Neha Rustagi, Hydrogen and Fuel Cell Technologies Office: Thank you, Eric. I'm happy to introduce Mark Ruth, from the National Renewable Energy Lab. Mark sits within the Strategic Energy Analysis Center where he manages the industrial systems and fuels group. He has been at NREL for 27 years where he's led many projects for DOE on nuclear renewable hybrid energy systems and, most recently, in 2017, he led the multi-lab effort to characterize the technical and economic potential of H2@Scale. This work was published towards the end of 2020, and it will be the focus of his presentation today. So, with that, I'll kick it over to Mark.
Mark Ruth, National Renewable Energy Lab: Thank you, Neha. Thank you everybody for joining us today. I'm excited to be able to talk to you a little bit about our analysis on the technical, demand, and economic potential of H2@Scale or hydrogen as an intermediate within the United States. Before I begin, as Neha mentioned, a couple of points at the bottom of this slide, you can see the URLs for where you can find the full report. In 45 minutes today, I'm only going to have time to be able to go through some highlights and some pieces around it. So, I won't be able to go into the level of detail that the report covers. Please look up for more information at those URLs. The full report's at the NREL site, and then there's a detailed analysis of the demand that's available at the Greet site at Argonne National Laboratory.
Also, before I begin, I'd like to recognize and acknowledge the others who worked on this project. As you can see by the author line here, it is quite the multi-laboratory effort. Paige Jadun worked with me at NREL and did much of the work within the effort. Amgad Elgowainy at Arbonne led the detailed demand analysis. Richard Boardman at INL led a lot of work in terms of high-temperature steam electrolysis. Then we had contributions from Nick Gilroy, Elizabeth Connelly, A.J. Simon, Jarett Zuboy in terms of developing data, analyzing data, doing a lot of spatial analysis, and helping us write up the full report.
So, we talk about H2@Scale, and I'm sure many of you have seen this figure or a previous version of it, but when I think about H2@Scale, what I really think about is the opportunity for hydrogen to provide an alternative infrastructure that carries energy and carries molecules for molecular use. So, in other words, let's really focus on being able to make hydrogen from a lot of different opportunities, whether that's from electricity from the grid or from other energy sources more directly like steam-methane and reforming of natural gas potentially with carbon-catching utilization.
You can see that the whole left side of this figure is focused on being able to make the hydrogen from those many sources, being able to do the most cost-effective and efficient way to be able to make that hydrogen, and then being able to go back to electricity is a low-value opportunity for seasonal or long-duration hydrogen storage to be able to utilize that hydrogen. However, I believe to be able to make that hydrogen utilization for seasonal storage valuable, there's a lot we have to think, not just about it, but across the mini potential applications for hydrogen. In the upper right, in blue, you can see three of those applications in terms of the transportation sector. Right now, we do a lot of use of hydrogen for upgrading oil and cracking and desulphurization of oil, which you can see there. But there's also the potential to produce synthetic fuels using hydrogen plus CO2 or direct use of hydrogen in fuel cell vehicles for transportation.
In the purple dots, you can see another opportunities for use of hydrogen in terms of its chemical properties, for producing ammonia and fertilizer products from it, for the metals production, which I'll talk about on a later slide, and in terms of providing feedstock to the chemical and industrial processes, which we'll talk a bit more about later. Likewise, it has the opportunity – it's one of the few ways we'll be able to provide high-temperature heat for industrial processes and then distributed by heat and power as well. So, hydrogen has a lot of different opportunities and a lot of different capabilities to use, and it can come from a lot of different sources.
Today, we actually use quite a bit of hydrogen. You can see here, this green square in the middle of the Sankey diagram that talks about how much energy is used to produce hydrogen today and then where we use it today. This Sankey diagram is the overall energy use in the U.S. in 2014, and we used about 98 quads of energy. As you can see, about 40 percent of that went to electricity with the other 60 percent – well, with about 30 percent going to transportation and then the remaining 30 percent being used for the industrial sector with some for natural gas for residences and commercial buildings.
Hydrogen uses about 2 percent of the total energy today to be able to produce the 10 million metric tons of hydrogen that we use for oil refining and for ammonia production today. So, even though I talk about hydrogen net scale in the future, hydrogen's already at scale today. We've solved some of those at-scale problems. However, there's a lot of new opportunities that we see as potential in the future.
So, how might we make that hydrogen for the future? One way, and I focused a lot in the H2@Scale figure on electricity production. One of their challenges with producing hydrogen from electricity is the way we've thought about it historically. This figure, the bar on the right shows the cost of producing hydrogen from steam-methane reforming, the all-in cost which includes not just the cost of the natural gas and the cost of operating but – blue – the cost of the newer ties capital cost for that steam-methane reforming. You can see a total cost of $1.95. Today you might be able get natural gas for a slightly lower price than that, as natural gas prices have been low for the last couple of years. But prior to that, that was about the cost with everything in on that.
On the left, you see a bar with electrolysis, the way we originally thought about. In that bar, we used a very high-capacity vector, 97 percent capacity vector for electrolysis, an industrial rate of electricity of $0.066 per kilowatt-hour and a capital cost for the electrolyzer, $400.00 per kilowatt. That is not the capital cost you could get the electrolyzer for today. That's closer to $900.00 or $1,000.00 per kilowatt. But it's a number that we believe we would be at if we had a fully built-out supply chain for the electrolyzer manufacturer.
As you can see in that bar on the left, the total levelized cost of hydrogen production is $4.20 per kilogram. The vast majority of that, $3.46, is the electricity cost for producing that hydrogen. However, we're seeing the opportunities today due to low-cost wind and solar electricity generation for low-cost electricity, say $0.02 per kilowatt hour or $0.01 per kilowatt hour. Even though that's not available 100 percent of the time, it's available somewhere around 40 percent of the year in certain parts of the country like the Texas panhandle, like some locations near large nuclear power plants that don't ramp down very easily, and other locations.
That second bar from the left shows that the opportunity using that low-cost electricity might allow us to really rethink the way the economics of electrolysis. You can see, in the left hand, half of that $0.02 per kilowatt hour, and right-hand, $0.01 per kilowatt hour, the share of electricity goes down from 346 to $1.05 or $0.52, depending upon the cost of electricity. On the other hand, the cost of the capital goes up pretty dramatically because the annuitized capital goes down. In other words, we're only producing about 40 percent of the hydrogen over the course of the year. You still have the same capital expenses. You still have to pay that back in 20 years or so.
Hence, instead of being $0.53, it's now $1.22 or $1.23. That cost is still too high at 224 to 277. But now, we're looking at something that R&D can make improvements to. For example, if you were to be able to reduce the cost of capital from $400.00 per kilowatt to $100.00 per kilowatt, even if you take a hit on efficiency by 10 percent effectively, going from 66 percent efficient to 60 percent efficient, suddenly we're moving to the third column on the right, which has a total cost of $1.14 to $1.73 per kilogram, which is very competitive with steam-methane reforming either at the slightly higher natural gas prices of $1.95 that we had a few years ago or even at current natural gas prices.
So, this really excites us about what the potential for hydrogen is and really led to this analysis. You can see here the objectives of the analysis were to try to quantify that potential of the H2@Scale vision. In other words, I was trying to understand how big that vision might be in terms of production of hydrogen, the application of hydrogen and its uses, and what the real opportunities are.
The report's broken down into two key areas. The first one involves the serviceable consumption potential and the resource technical potential. I'm not going to cover that today in this presentation in the interest of time, but in the report, you can find some details about what we consider the serviceable consumption potential which is kind of the top end or the maximum market size that one would think about for hydrogen because it's constrained by the services for which society currently uses energy as well as real-world geography and system performance but not constrained by economics. So, in other words, it's looking at it if hydrogen were very cheap, very available, how big might those markets be. It gets into what the size of those markets might be in different categories, each one of the ones at the right-hand in the bubble figure that you saw a couple slides ago.
It then compares the total amount of that serviceable consumption potential to the resources that are available to be able to show that there's sufficient resources in the U.S. to be able to meet that potential hydrogen demand from wind, solar, biomass, hydro and other renewable technologies, as well as natural gas, coal, and uranium through nuclear-type technologies. So, it really tried to compare and show that there are sufficient resources because that's one of the key considerations today.
Instead of focusing on that first half of the report, I'm going to focus, in this presentation, on the second half, which is the economic potential. The economic potential is the quantity of price of hydrogen at which suppliers are willing to sell and consumers are willing to buy that hydrogen, assuming there's market and technology advancement scenarios. It's really looking at, "Well, what is the actual potential for hydrogen under the economics in the economics scenario that we have today?" The method that we use to be able to estimate that economic potential you can see here. We use a market equilibrium methodology which, for those of you who might remember economics 101 from college might remember this, you've got a demand curve which goes down. As quantities go up, the prices go down, meaning that as prices get lower, more and more consumers are willing to buy it or, in other words, fewer consumers are willing to buy their quantity at higher prices. You can see on the PQ or the price quantity curve on the left, it goes down as you go across.
The flip side is the supply curve. The supply curve shows that as the price goes up, more and more producers are willing and interested in producing the product – in this case, hydrogen – for consumers to be able to buy. Then, economic equilibrium is where supply and demand cross. You can see that price quantity. From that, you can estimate a quantity or a market size for that scenario. Note that the quantities where, at that point, we assume no excess supplier demand and the market pushes it towards the equilibrium. We're very – just like in chemical situations, we're very seldom actually at equilibrium. We're always moving towards equilibrium and things are always changing.
You can think about that analogously for those of you who are chemists, temperature changes. That moves the equilibrium. Things change in the economy. That moves the equilibrium, and we're moving towards it in a lot of ways. However, it gives us a good idea of what the opportunity might be.
Of course, with this methodology, there's a lot of limitations and caveats. We tried to think about what the markets might look like and what the economy might look like in 2050 and what that might be. However, we did not get into issues such as stock turnover and whether or not we could actually provide towards that market by 2050. So, in other words, we really focused on what the market might look like in 2050 if we didn't have to worry about buildings changing or vehicles getting purchased in that timeframe and technologies available for them, those types of issues. So, that is outside the scope of this analysis.
Likewise, a major caveat of this analysis is that new policy drivers, especially those like emission policies, like 100 percent renewable incentives or carbon taxes or others are not included for either hydrogen or the grid within this analysis. Of course, there's a lot of assumptions about adjacent technologies. We tried our best to fairly consider how adjacent technologies might develop. However, that is very difficult to be able to capture, especially for those of us who are expects more in the hydrogen area than in those adjacent technologies. So, there's some questions about whether or not we were able to make that happen.
The demand analysis is limited to sectors such as those that we could foresee in the future. So, as you'll see, in a few slides, which markets we incorporated and how we estimated those, that's based on what our insight and on our crystal balls. Of course, there's a lot of assumptions around that. Then, estimates of delivery costs were standardized without getting into a lot of location specificity. Obviously, hydrogen is going to be very regional, very localized in many ways. We tried our best to standardize those, to simplify this analysis and to allow it to be completed. We also didn't include some long-term production technologies such as photoelectrochemical options. We didn't consider economic feedback such as rebound effects on natural gas prices and on other things.
So, having said all of that, what I'm going to go through in the next set of slides and kind of as the meat of this presentation are the exports and results for the five national scenarios. The five national scenarios you can see labeled across each one of the columns here, going for the ones that look the most like today, that Reference scenario on the left, to the one that looks the most aggressive on the right, which is what we call Lowest-Cost Electrolysis, with some changes in between. You can see here some of the major differences between them. The natural gas prices for the Reference scenario and the R&D Advances in Infrastructure scenario use the Annual Energy Outlook's Reference scenario. The other three scenarios that we use use what the Annual Energy Outlook calls the low oil and gas resource, which has higher natural gas prices for its scenario. So, you can see that we assume the low-cost natural gas for the first two and higher cost natural gas for the next three scenarios.
For high temperature electrolyzers or high temperature steam electrolyzers, we use the current costs which are pretty high because of low market sizes in the Reference scenario and then we assumed improvements down to around $380.00 per kilowatt for the other scenarios. Low temperature electrolyzers we also assumed improvements. We use the current cost of around $900.00 per kilowatt. In the Reference scenario, we reduced that to the $400.00 per kilowatt, which I mentioned previously is kind of the trajectory we're on with built-out supply chains for R&D Advances and Low Natural Gas Resource/High Natural Gas Price scenarios, and then we assumed further improvements; $200.00 per kilowatt in the High Natural Gas Price R&D scenario and $100.00 per kilowatt purchase cost in the Lowest-Cost Electrolysis scenario.
The electricity prices available for producing hydrogen with those natural gas prices is what we call low-cost dispatch-constrained electricity, which is really based on some supply curves and in very interesting analysis in terms of developing supply curves of how electricity might be available. Those supply curves were developed using our ReEDS capacity expansion model at NREL and putting in price floors on them. It's a pretty extensive analysis and one that I'd love to talk further upon at a later date, but we don't have time to talk through today within this analysis. However, what it did lead to was these supply curves for electricity that we could use. However, those supply curves are for selling electricity into the grid and there's, essentially, a $20.00 per megawatt adder to be able to buy electricity from the grid. So, we assume for our first three scenarios, we would include that $20.00 per megawatt adder.
For the aggressive analysis R&D scenario, we assumed that the adder would only be $10.00 per megawatt adder because the electrolyzer, being a dispatchable load, could provide some services such as demand response and frequency and voltage control-type services to the grid. So, that would reduce that adder cost. In the lowest cost electrolysis scenario, we reduced it all the way down to a zero adder, or the wholesale price to sell into the grid because we were assuming that the dispatchable load from the electrolyzer could provider all of those services and, hence, reduce the overall cost.
A few other key parameters. Distribution for fuel cell electric vehicles, we used the current cost which is around $8.00 per kilogram to be able to do distribution in dispensing the vehicles for the Reference scenario. For the other four scenarios, we assumed that the HFTO cost target of $2.00 per kilogram in 2016 costs was met, and it's about $2.20 in our dollar values.
The metals demand, we assumed market competition with Coke in the Reference scenario, but we assumed that there was a willingness to pay a premium for hydrogen in the other four scenarios. In other words, the desires to reshore metals and steel production is higher and exists and, therefore, is involved in the other four scenarios. We're seeing something very similar to that in terms of the tariffs that we currently see on steel imports in the U.S. today.
So, with all of that, this side starts to go through the nine different applications for hydrogen that we use, showing the threshold price, the demand of the threshold price and if there's a couple of steps of the threshold essentially what we had. So, as I mentioned, we looked at nine different applications within this study. The first one is refining and chemical processing, kind of the current industry as it currently exists. That one, they're willing to – they do not pay today, but would be willing to pay a high threshold price for hydrogen because there's really no other option when you think about hydrocracking or hydrodesulfurization or other options. We assumed a little bit of growth in that market as primarily based on additional exports of refined products. So, therefore, at a high-threshold price, we assumed a little bit of growth up to 7.5 million metric tons per year.
We also looked at the metals market. Metals is primarily steel and, as you can see there, a hydrogen price of $1.70 per kilogram which is based on EIA's Low Oil and Gas Resource Scenario price where hydrogen would be competitive. In that case, what's happening is we're assuming that up to 4 million metric tons of hydrogen would be used for producing steel and the pig iron necessary for steel within the U.S. today. That 4 million metric tons per year essentially replaces all the pig iron for steel that would be needed in the U.S. in 2050 but still has imports. That number could go all the way up to 12 if the price of hydrogen were lower and you used hydrogen for heat as well as for its chemical properties in the direct reduction of iron that you might see in the first couple columns, the 4 million metric tons per year.
Ammonia production in the U.S. today is projected to grow up to – requires about 2 million metric tons per year of hydrogen. It's projected to grow up to about 2.5 million metric tons per year based on growth in the agricultural sector. They have very few options so they will expectively use hydrogen for that. But, however, if the price of hydrogen is at $2.00 per kilogram or lower, we assumed the growth will be even higher as we offset other imports.
Biofuels production is based on a 50 percent market share in the aviation market in 2050 and replacing hydrogen that is produced via catalytic conversion of the biomass within a pyrolysis process to be able to produce the hydrogen. In other words, it's essentially producing hydrogen from biomass. If we could replace that with hydrogen from our other sources that we considered in this analysis, then there would be a high value for it and that would lead to a higher-carbon yield for that biofuel and a better chance of being able to meet that aviation market.
Synthetic hydrocarbons are based on a methanol market and being able to compete within a methanol market at 6 million tons per – we assumed this market size is 6 million metric tons per year at a threshold price of $1.73 per kilogram based on today's technology of taking hydrogen plus CO2 and competing with today's methanol prices in the market. Natural gas supplementation requires $1.40 per kilogram hydrogen to be able to compete on an energy basis for natural gas. So, in other words, if we're assuming that if hydrogen goes in a natural gas system, it's only value is its combustion value and it's got an equivalent value as – the particularly natural gas price at $1.40 per kilogram of hydrogen. The market size there is set by achieving 20 percent by volume in the natural gas system which is only about 4.5 percent by energy. That 20 percent by volume is based on an assumption that that is what would be reasonable in natural gas markets – sorry, what would be reasonable in the natural gas infrastructure and used by natural gas applications, like furnaces, combustion turbines, heat, et cetera.
Seasonal energy storage for the grid is based on competing with natural gas to be able to provide that last peaking power that is necessary within our grid analyses. You can see there, hydrogen price would have to be $1.10 per kilogram to be able to compete with natural gas. The size is only 14 million metric tons per year for that.
Then you get down to the two vehicles markets. $2.20 at the terminal is equivalent to about $5.00 at the pump, assuming success in the distribution and dispensing costs. We assumed based on vehicle choice modeling that we could achieve about 20 percent of either the light-duty market and/or the medium- and heavy-duty market, and that 20 percent market size is 12 million metric tons for the light-duty market and 5.2 million metric tons of hydrogen annually for the medium- and heavy-duty market.
So, those are all of our hydrogen applications and hydrogen demands. We did not include this list that you can see here, which are either smaller or harder to imagine without key carbon emission reduction targets or policies. We took those that I mentioned in the previous one, and then we put them into these demand curves here. The one that I went through for you was really focused on the gray curve there. We also have some that are in the orange curve and the blue curve that are less aggressive in terms of their hydrogen demands, primarily based on assumptions of lower natural gas prices and assumptions of a couple other changes within policy things. So, those are the blue and orange ones. Those were used for the first two scenarios, the Reference scenario and the R&D Advances and Infrastructure scenario.
We also develop supply curves. I'm not going to go into nearly as much detail in terms of what we did on supply curves however, here, you can see that the five supply curves that we used. On the left, you can see the Reference scenario in blue, the R&D Advances and Infrastructure supply curve in orange, and then on the right, you can see the other three scenarios in gray, green, and blue respectively. Each one of the scenarios incorporates hydrogen production using steam-methane reforming in the solid lines, low-temperature electrolysis from low-cost dispatched-constrained electricity in the short-dashed lines, and high-temperature electrolysis coupled with nuclear power plants where those nuclear power plants are economically challenged in the solid lines.
Essentially what you see, as the supply curve's going down, certain costs become cheaper. The Reference scenario in blue is based on low natural gas prices and the opportunity to be able to convert more natural gas to hydrogen there. The R&D Advances and Infrastructure scenario allows for an additional amount of nuclear with high-temperature electrolysis prices and at very large hydrogen markets. We're starting to get into the opportunity to use low-cost dispatched constrained electricity and low-temperature electrolyzers. As you move to the right, you go to the gray and green, and the Lowest-Cost Electrolysis scenario is primarily made up of using that low-cost electricity, which would be available due to additional PV and wind generation, which we see as potentially being built based on their cost and willingness for hydrogen to pay at these different prices.
We then took those supply and demand scenarios and used them to be able to develop our markets, as I mentioned to the supply and demand curves and the market equilibrium methodology. Here, you can see for the R&D Advances and Infrastructure scenario, the supply curve in orange and demand curve in blue. There you can see the crossover point that these two black lines point to showing a market size of somewhere around 30 million metric tons per year at a hydrogen price of about $2.20 per kilogram. For each one of those, we then broke out, "Well, what are the supplies and demands?" Here, you can see the supply is primarily from brown which is steam-methane reforming, in this scenario, with a little bit from high-temperature electrolysis in the mustard-colored yellow. That makes up the supply for that 30 million or so metric tons per year.
The demand for this one is made up of refineries, in gray, on the left, plus some ammonia, plus some biofuels. Recall that all of those were set because of high hydrogen prices because of a lack of alternatives to be able to meet the markets that they have. There's also a demand for metals, in this scenario, which is shown in gray, and then some demand for light-duty vehicles and medium- and heavy-duty vehicles for fuel cell electric vehicles in turquoise and purple, respectively. Those demands are based on market competitiveness at this $2.20 per kilogram of hydrogen at the terminal price. So, you can see kind of how we get the supply and demand curves.
To be able to remind you, we did this for all five scenarios ranging from the Reference scenario, on the left, which is closest to the one we have today, to the Lowest-Cost Electrolysis scenario, which is the one with the most optimistic assumptions. I showed you this slide before, so I'm not going to dwell on it. But the real focus is kind of, "What are these different scenarios here?" With supply and demand curves for each one of the scenarios which you can see on this slide across the top with the supply curves being in black, the demand curves being in more that grayish color, you can see the supply and demand curves for each one of them. Then you can see the stacked bar charts breaking out the supply and demand for each scenario.
I will go through each one of those supply and demands for each one of those five scenarios. So, I just took the supply and demand curves that you saw at the bottom of the previous slide, stacked them separately, so now you can see supply on the right side of the stacked bar charts and the demands on the left side of the stacked bar charts, and the five scenarios going from reference, on the top, down to the Lowest-Cost Electrolysis on the bottom.
As you might recall, today's hydrogen market size is around 10 million metric tons per year. So, this analysis shows that the economic potential of hydrogen demand is two to four times that. You can see here, our lowest scenario is right around 20 million metric tons per year, or twice as big, and our Lowest-Cost Electrolysis scenario is about 40 million metric tons per year, or four times the total market size today.
In the Reference scenario, we get up to about 22 million metric tons per year. The market size, here, you can see, is based on refineries in gray and ammonia in orange demand that are based on today's growing markets. It stays at about that 10 million metric ton per year size, which is where we are today. So, essentially, assuming very little market size growth in those. However, it identified some new demands. Biofuels and using biofuels for aviation based on the opportunity to reduce their costs by having exogenous hydrogen in them. That shows up in blue. Then in kind of the tan color, you can see methanol production where methanol is competitive with other technologies.
As you can see, the supply side is fully made up or steam-methane reformed hydrogen. So, because the natural gas prices are low, the prices of high- and low-temperature electrolysis are high, the only way you're able to meet that demand at that price is through reforming natural gas through the steam-methane reforming process.
Moving into a scenario with some R&D Advances and Infrastructure. We still have low natural gas prices. However, now we've got reduced cost for delivery and dispensing of that hydrogen so the market size gross. We also have competition within the metals sector due to R&D there and policies that are pushing towards onshoring steel production within that. So, you can see growth on the left to a little bit over 30 million metric tons per year based on those additional market demands. To be able to meet that supply, we're using even more natural gas through steam-methane reforming. However, we are now starting to see pushing the natural gas prices up to a point where we're starting to be slightly supply limited there and hence, we're getting into some low-cost nuclear high-temperature electrolysis opportunities.
So, in other words, instead of building new natural gas, it's economically more favorable to use some nuclear power plants with high-temperature electrolysis that are – with those nuclear power plants being economically challenged today. In other words, they'll willing to sell electricity at the equivalent of $25.00 per megawatt hour because they can't get that price within the electric sector because of all the competition from natural gas and wind and solar within that sector.
The third scenario, the Low Natural Gas Resource and High Natural Gas Price used all the same parameters as that second scenario. However, it switched from the Annual Energy Outlook's Reference scenario to its Low Oil and Gas Resource scenario which has higher natural gas prices. Therefore, you can see the price of hydrogen goes up, goes to a point where the transportation fuel cell vehicles are pushed out of the market. It's no longer economically competitive for them to exist within the market. You can see that there's a lot more – that nuclear high-temperature electrolysis is a lot more competitive. So, we see a lot less natural gas production with the hydrogen market being met by a greater share of nuclear high-temperature electrolysis.
In fact, in this scenario, we're seeing almost 60 percent of the nuclear power plants in the U.S. today being converted to hydrogen production because even at $40.00 per megawatt hour, they make more money selling hydrogen than they would be able to within the – at $40.00 per megawatt hour electricity equivalent prices, they would make more money selling hydrogen than they would producing electricity and selling it just to the grid.
The Aggressive Electrolysis R&D scenario, you may recall, it really involves a change where the low-temperature electrolyzer purchase cost is reduced from $400.00 per kilowatt, which we had in the second and third scenarios, to $200.00 per kilowatt, and the electricity price adder is reduced from $20.00 per megawatt hour to $10.00 per megawatt hour, making it more economically favorable to be able to buy that electricity. You can see the demands are the same within this scenario as they were in the Low Natural Gas Resource scenario. However, now, with low temperature electrolysis production of hydrogen using low-cost dispatched-constrained electricity is now more competitive. So, you can see that little bit of that green bar at about – what is that – about 7 or 8 million metric tons per year, competing both with natural gas and with nuclear high-temperature electrolysis within it.
Further reductions of the low-temperature electrolysis price down to $100.00 per kilowatt and taking out the price adder on low-cost dispatched-constrained electricity, in other words buying electricity at the same price it would be sold into the grid for because there are some cost benefits of being able to use the electrolyzer as a dispatchable load leads to the much larger market size of 41 million metric tons per year. You can see that there's a larger amount of ammonia production because you're cost competitive and competing with importing ammonia. Fuel cell electric vehicles are now much more competitive with battery electric vehicles and other vehicle technologies within our vehicle choice model because the price of hydrogen is a bit lower in this scenario. So, we're getting up to about 40 million metric tons per year.
So, where would the electricity come from to be able to produce that additional hydrogen? Here, you can see the growth in generation that we show across our five scenarios for PV solar electricity on the left and wind on the right. We broke it down into three different colors; blue is generation to serve the electric load, green is hydrogen production, and then the yellow color is still curtailed. As you can see here, going to higher-priced natural gas as we did between the R&D Advances and Infrastructure scenario and the Low Natural Gas Resource scenario, increases the amount of PV and wind serving load on the grid. However, what we also see, and especially in the wind case, is, as we start to use – as we move to the Lowest-Cost Electrolysis scenario, we get a lot more production of hydrogen. In other words, the green bar gets larger.
Note, on the wind side that as you move from that Low Natural Gas Resource scenario to the Lowest-Cost Electrolysis scenario, the blue bar also gets larger. So, in other words, what's happening there is new wind is being built. Between those two scenarios, you've got additional wind being built in the Lowest-Cost Electrolysis scenario, a small fraction of that wind generation is being sold to the grid with the remainder being sold to hydrogen.
In other words, buying – say you've got a wind turbine that you would not build because of the coincidental nature of wind generation. You wouldn't build because you could only sell electricity to the grid, say, 20 percent of the time that your wind farm is producing electricity. However, if hydrogen is willing to buy the other 80 percent of the time, even at a slightly lower cost than what you had hoped to sell into the grid for, you might build that wind farm to be able to sell into the grid 20 percent of the time and sell it to the hydrogen market 80 percent of the time and be economically competitive doing that and have a sufficient return on the investment to be able to do that. So, essentially, what you're seeing here is a doubling of the size of the wind generation from just electricity to thinking about hydrogen and its benefits, not just for making hydrogen but its benefits to the grid as well.
Within the analysis, we also really looked at what the spatial aspects that the economic potential scenarios are. As I noted before, we had to simplify our delivery costs. We did not have the resources to be able to think through all the potential delivery vectors and delivery costs as a whole. So, what we did was, as a way to try to provide information for subsequent analyses that might want to be able to use this, you can see where the demands are in the top figure, broke it out by the same colors so you can see what the different demands are and what the supply is on the lower left. This is for the Lowest-Cost Electrolysis scenario, as you might imagine. You see that there's a lot of wind in the wind belt through North Dakota, South Dakota, Nebraska, Kansas, and Texas.
Then there's some high-temperature electrolysis at currently existing nuclear power plants that you can see in the orange throughout it. You can see much of the demand is near the population and industrial centers. So, the eastern seaboard, California, and the Gulf Coast, with a lot of demand, though, spread out throughout the country, not just because of vehicle demands but because we do such things as produce ammonia in other parts of the country or could produce ammonia for farming closer to the farms and do other things like that. I just want to note that we do have these spatial aspects of all five economic potential scenarios within the reports. You can see those.
We also looked across, "Well, what is the impact on the overall energy sector?" So, if you remember the Sankey diagram, I talked about using about two percent of the total energy in the U.S. today is used to be able to produce hydrogen. Here, you can see that that percentage would go up. In fact, you see that we would use about eight quads of energy to produce that energy. However, a big chunk of that, maybe over 90 percent of that would be from electricity, which has a factor going to the way that they estimate the fossil fuel equivalents on the left side of these types of diagrams.
So, it's actually ends up being about – about 20 percent of the total market of the total energy use is hydrogen and the impact of being able to use hydrogen – this isn't shown in this slide but is reported in the report – is about a 20 percent reduction in greenhouse gas emissions above and beyond the improvements to the electricity grid alone. So, as you can see in this scenario, the Lowest-Cost Electrolysis scenario, the electric grid is pretty clean. We're only using a very small amount of coal and natural gas for some dispatchable electricity generation or needed electricity generation because of location in the country. So, the electricity sector is already cleaned up very much. In addition to that, there's still a 20 percent reduction by being able to use hydrogen for all the purposes that I spoke about previously in the slide deck.
So, in summary, you can see some of our key conclusions from this analysis. The economic potential of hydrogen demand in the U.S. is two to four times the current consumption within this analysis. At those market sizes, hydrogen production is 4 to 17 percent a primary energy use. I think I said 20 percent on the previous slide, rounding that 17 percent up. The range across the five scenarios was developed using a variety of economic and R&D success assumptions. So, anywhere in that range is depending upon your assumptions around what happens to the economy and what happens in terms of success in terms of the R&D. Total petroleum use could decline by up to 15 percent below a scenario with high renewable penetration on the grid, primarily due to the penetration of hydrogen for fuel cell electric vehicles.
We show that an increase hydrogen market size can be realized even if low-cost, low temperature electrolysis is not available as long as other hydrogen productions options are available. In other words, in this analysis, low-cost steam methane reforming and/or low-cost, high-temperature electrolysis using nuclear generated heat. However, to be able to achieve our highest market sizes, I would say low-cost, low-temperature electrolysis or an equivalent technology – think something like photoelectrochemical hydrogen production at low-cost – are essential to really achieve the high market sizes and the high impacts on the energy sector as a whole.
One of the other benefits of doing low-temperature electrolysis is, as I mentioned, grid integrated electrolysis can increase the wind power generation by more than 60 percent because it monetizes the additional opportunity to be able to generate that wind. As I noted previously, that not only has the potential to benefit the wind generator but it could also benefit the grid and reduce some of those dispatchable generation requirements that we saw from natural gas and coal on the safety diagram slide that I just showed previously.
Up to about 60 percent of the new current nuclear power plants could potentially improve their profitability under these scenarios by producing hydrogen. That's not to say that they should switch over today, but if we see improvements in high-temperature electrolysis technologies, we see prices on the grid as we're seeing them today, that is a potential for those nuclear power plants to switch over instead of shutting down because they're no longer cost competitive on the grid. We showed that the potential for up to 20 percent reductions in CO2 equivalent emissions over just the electricity grid improvements and reductions due to those improvements on the grid and using more renewables on the grid. Higher reductions may be feasible given policy drivers and development of additional demand sectors.
With that, I'm going to end the formal part of this presentation and open it up to a Q&A session. I'm hoping you've been typing your questions into the Q&A box, and I will trust that Eric and Neha will be able to dig through those and be able to pick out some of the high – the questions that are of high interest. I'd like to thank you for your time and thank the team that put together this analysis. It was a really valuable analysis and I think it's really helpful for us in terms of our thinking about the potential for H2@Scale into the future. So, thank you.
Neha: Thanks Mark. We have loads of questions. I'm going to start listing them out and any that we don't get to, we'll try to respond offline. So, one of the first questions we got was, "Can you speak to how the COVID-19 pandemic might affect the ability of these projections, for example increase in natural gas prices? Would they affect some of your conclusions?"
Mark: Yeah. So, as I mentioned in the caveats, one of the things that we did not do within this analysis was we did not consider the potential impacts on the macroeconomy as a whole. I think with the COVID-19 crisis that we've been going through, I think we're starting to see and we have seen points where we've had lower costs for resources such as natural gas and oil because of the reduced demand globally there. I believe we're going to rebound in those areas and that those will probably go back up and we're probably going to see some issues still in those spaces. So, I would say COVID-19 is going to have a slightly lower effect on these than natural growth in transportation markets and industrial markets and some of those types of issues. I think it's a really important question to consider and one that is definitely at the cutting edge of how we think about the economy and how the economy works.
Neha: Thanks Mark. So, another question around Slide 17 – and it might be good to just pivot back to it – but, "Can you explain the supply and demand intersection again? I think it's a little counterintuitive. I comes across as a little as the price goes down, demand goes down.
Mark: Right. So, essentially, what's happening here is that as – if we – maybe it's easier to go back to this slide here. So, as – if you think about the hydrogen – if you want to think about price as an independent variable, which is not an unfair way to think about these types of curves. As the price goes up, the supply goes up. So, in other words, as the price goes up, more and more actors are willing to be able to provide in this case hydrogen but any type of commodity into the market. As the price goes up, you'll get more and more people who are willing to provide it. Likewise, as the price goes down, you've got bigger and bigger demand.
So, here, you can see at $3.00 per kilogram, our demand is less than 20. It's about 18 million metric tons per year. But if we were at a price of $1.00, then we would be at about 45 million metric tons per year. You can't really look at these independently though. So, you really have to look at the intersection between them. So, as you can see, the intersection is what's really important here. It's hard to think about the multiple players within each one of these. There really are fewer independent variables and there's more interactions between them. So, this is why economists have developed this kind of supply and demand curve methodology to be able to think about what's happening.
Neha: Thank you. Another question that's specifically about the biofuel sector. So, within the biofuel sector, did we assume 100 percent use of biofuels in aviation or a mix?
Mark: So, we assumed that 50 percent of the aviation market would be met by biofuels that are producing using a pyrolysis method and exogenous hydrogen or hydrogen that's not produced from that biomass within that pyrolysis process to be able to meet that market size. So, I don't know if that means 50 percent of aviation will be biofuels and 50 percent will be on traditional jet fuel or if all fuel for that aviation is 50 percent biofuel and 50 jet fuel or somewhere in between. It just breaks down into that kind of market size. So, when I say I don't know, I think that means that it really doesn't matter how we did it. It's just that that was our assumption was 50 percent.
Neha: Thank you. Another question around the third and fourth economic central scenarios. Can you speak to why the fuel cell vehicle demand went away in those scenarios or went down in those scenarios?
Mark: Oh, yeah. So, let me move to this figure here. So, the fuel cell vehicle demand is this long stair-step at about $2.20 per kilogram of hydrogen. What happens in the third and fourth scenario compared to the second scenario is that the supply curve goes up because we assumed higher natural gas prices in the third, fourth, and fifth scenarios than we did in the first and second scenarios. With those higher natural gas prices, the supply curve goes up. You end up with hydrogen prices at the intersection that are greater than $2.20 per kilogram. At that point, it is more expensive or it's not competitive to go to the light-, medium-, and heavy-duty vehicle markets.
One of the shortcomings of this analysis is we have that one long step for light-, medium-, and heavy-duty vehicles. It would be ideal to be able to say, "Well, this is $2.20 at the terminal. What happens if it were $2.30 at the terminal? What happens if it were $2.40 at the terminal and you're able to break that one long step into more discreet chunks?" However, due to resources, we weren’t able to do that many vehicle choice modeling analyses. Hence, we end up with the one long step. It probably gives you a good bookends and good ideas, but more fidelity would be possible with additional resources and additional thinking into the analysis.
Neha: Thank you. Sort of a related question, a two-part question. If you can speak to the premium added to the terminal prices for fueling station. So, in your vehicle modeling what the actual experience cost _____ to the driver was and then also, just in general, how you modeled infrastructure, the pipelines or trucking, and approximate cost?
Mark: Mm-hmm. So, all the prices, we tried to do on an industrial equivalent basis. So, we think about that as what is the price at the terminal, kind of a large scale for hydrogen. Obviously, as the questioner sees, that is not the price at the pump because people don't go to industrial sites that are massive. They go to local corner fueling station pumps. So, there's an additional cost for delivery and dispensing of the hydrogen. We assume for the R&D Advances and Infrastructure, Low Natural Gas Resource, Aggressive Electrolysis R&D and Lowest Cost Electrolysis scenario – so for the four on the right of this figure – $2.20 for the cost of delivery and dispensing which is meeting the DUE HFTO delivery and dispensing target of $2.00 per kilogram by converting it to equivalent year dollars. So, at $2.20 which is where that long bar is on the demand curves for the price at the terminal plus $2.20 for delivery and dispensing, plus $0.50 in taxes or so are overall costs at the pump, to the consumer at the pump or the assumption for the overall cost to the consumer at the pump is $5.00 per kilogram.
On the left, the assumption was much higher than that. I don't remember the exact number because it is, as you can see, that long step is down here at about $0.80 per kilogram. So, I think we were assuming something like $4.00 or something like that for delivery and dispensing. But it was high enough that it required a much lower and unreasonable kind of cost for hydrogen production.
Neha: Thank you. So, another question was around aside from reductions in the cost of electrolysis, were there other R&D improvements that you considered, for example, in happy industry?
Mark: Yes. So, probably easiest to talk about here. We did assume R&D improvements depending upon – actually let me go back to our prices – depending upon the technologies. So, for metals, we assumed development and engineering for direct reduction of iron to become competitive at $1.70. Right now, the TRL of that technology at a manufacturing readiness level, that technology is much lower. So, it would require more competitive prices than $1.70 per kilogram of hydrogen to be able to do it today. So, it is definitely an R&D advance.
I spoke a couple of questions ago about biofuels. The pyrolysis process to produce aviation fuels is kind of at a pilot scale. Some would even argue it's at less than a pilot scale today. So, there's obviously R&D advances that would be necessary for that technology to be competitive. Synthetic hydrocarbons, likewise, are at a bench scale or maybe a pilot scale to be able to produce methanol from CO2 plus hydrogen. We know it can be done but the electrochemical conversion of the CO2 to CO, the use of CO or the potential competing thermochemical catalytic processes would all require R&D advances there as well. So, those are kind of R&D advances in those spaces.
Then, obviously, the fuel cell electric vehicles are to be competitive at the $5.00 per kilogram price, we see some cost reductions from what those vehicles are costing today. You would have to talk to Toyota and others to be able to get exactly what those reductions are, what the R&D needs for those are.
Neha: Thank you. Another question around, "What fraction of hydrogen is produced by water electrolysis today and what type of electrolysis is it; high temperature or low temperature?"
Mark: In the U.S. today, the fraction is in the very, very small percent. One percent, two percent, maybe three percent of the total hydrogen production today is from water electrolysis. That fraction is essentially all, to the best of my knowledge, essentially all low temperature electrolysis, and it's built for very small-scale purposes. Essentially, today, it's cheapest at a large scale to use steam-methane reforming. So, if you've got a refinery or an ammonia production, that's how you want to do it. At the smallest scale, so if you've got a very small lab that doesn't much hydrogen, you just buy doers or compressed cylinders of hydrogen gas.
But there's an in-between point where you might want to use a little bit more hydrogen and it would take you tens or hundreds potentially compressed gas cylinders, but you aren't quite big enough to actually – so you're big enough to be able to get beyond that kind of price point, which is using steam-methane reforming to produce that hydrogen but you're not big enough to have your own steam-methane reformer. So, you start buying hydrogen – you a hydrogen electrolyzer for that. So, this is some large laboratories. This is some electronic production. This is cooling within some turbines at some coal and nuclear power plants, those types of demands. But it's a very, very small percentage today.
Neha: Thank you. This might be, I think, the last question we have time for. Did you characterize the impact of these scenarios on water consumption or water risk?
Mark: I think that's a really interesting question. We did not go into much detail about water consumption. But the thing I think I want to note about that is that in our scenarios, we had competing – we went through a capacity expansion model given the electric grid. As you can see here, a significant amount of our grid is wind plus solar, really pushing out natural gas plus coal. By pushing out natural gas plus coal to wind and solar, a lot less water is needed to be able to produce the electricity. Because of that, there's a lot more water that would be available for other options. Hydrogen, as a whole, does not require a lot of water for low temperature electrolysis. In fact, the total amount of water going into, say, a kilogram of hydrogen is essentially one flush of a toilet. So, it's a very, very small amount of water that goes into there.
There's a lot more water that's required for cooling for both electricity generation, if you're producing it from nuclear natural gas or coal and for even cooling the electrolyzers. So, that's where the water demand is. It's not for producing the hydrogen itself. That said, water is really important. We didn't get into a lot of details about that. I think there's a lot of value getting into that. Amgad Elgowainy, who's one of the co-authors on this, has published a few really nice papers on that. So, I recommend looking at that Greet site that I showed the demand report for, and look for some water papers within there. You can learn a lot more about water for hydrogen use using those papers.
Eric: Okay, I think that does conclude our Q&A session. We are at the top of the hour. So, thank you, again, Mark and Neha. I believe we have way more questions than we could have gotten to today. So, thank you for submitting those. We'll make sure to capture all of them. If you want to reach out to Mark directly – I think he had his email address on one of the prior slides – as you'd like to continue the conversation. But with that, that concludes today's scheduled H2IQ hour. So, thanks again to everyone for joining. Reminder, this will be posted online, the recording and the slides, on the HFTO website in around a week. So, be on the lookout for that. We'll announce some more topics in the near future. So, with that, have a great rest of your week, everyone, and goodbye.
Mark: Thank you.
[End of Audio]