Below is the text version for the "U.S. Shale Revolution and Chemicals Industry Generates Low-Cost Large-Scale Supply of Hydrogen" webinar held on December 13, 2017.
Eric Parker, Fuel Cell Technologies Office
Good day everyone, and welcome to the U.S. Department of Energy’s Fuel Cell Technologies Office webinar. We’ve got a great presentation lined up this month from Neha Rustagi and Amgad Elgowainy from Argonne National Laboratory on the U.S. shale revolution and how the chemicals industry generates a low-cost large-scale supply of hydrogen. My name’s Eric Parker. I provide program support within the Fuel Cell Technologies Office, and I’m the organizer for the meeting.
We’ll begin in just a moment, but first I have a couple housekeeping items to tell you about. Today’s webinar is being recorded, and the recording, along with the slides you’ll see today, will eventually be posted in the coming weeks, and we’ll let you know. All attendees will be on mute throughout the webinar, so please submit your questions via the WebEx chat box that you should see on the WebEx panel. Here’s an example of it here on the slide. We will cover those questions during the Q&A period at the end of the presentation.
And with that, I’d like to introduce our webinar’s host, Neha Rustagi, who is here with me at DOE headquarters. Neha is the lead for hydrogen delivery technologies within the Fuel Cell Technologies Office. And it’s a pleasure to have you with us.
Neha Rustagi, Fuel Cell Technologies Office
Thank you, Eric. So, as Eric mentioned, our presentation will be from Dr. Amgad Elgowainy. Amgad is a principal energy systems analyst and the life cycles analysis team lead at Argonne National Laboratory. He conducts techno-economic and environmental analyses of alternative transportation fuels and advanced vehicle technologies, including hydrogen fuel cell electric vehicles. So, our presentation today will be on fuels that can be used for fuel cell vehicles and are being produced in abundance as a result of the shale boom. And with that, I’ll kick it off to Amgad.
Eric Parker
All right. Take it away Amgad.
Amgad Elgowainy, Argonne National Laboratory
Thank you, Neha, and thank you, Eric. Should I share my screen?
Eric Parker
Yes, go ahead.
Amgad Elgowainy
Can you see my screen now?
Eric Parker
Yep, go ahead.
Amgad Elgowainy
Okay, hello everyone. This is Amgad Elgowainy. Today my presentation is about near-term opportunity for low-cost hydrogen that could serve the early-market demands of hydrogen. In particular, I will focus on the emerging demand of fuel cell electric vehicles. So, and to put it in context, I am not sure if some of you have heard before about the H2 @ Scale initiative. Dr. Bryan Pivovar a year ago presented in this webinar, the Hydrogen @ Scale initiative, and the concept there.
I will provide a brief introduction here, but dive into the topic of the presentation. Hydrogen @ Scale, as you see in this graph, trying to provide synergy across sectors. On the left, you see the power sector, which would be in the future dominated by the renewables, but also nuclear power in the U.S. in particular is important. About 20 percent in the grid mix today.
Electrons produced by renewable and nuclear can be used in low-temperature or high-temperature electrolysis to produce the hydrogen molecules, which you can then serve a different application. And once in [inaudible] it could go back to serve the grid in a long-term scenario. Near-term scenario, it could be in a gas form injected into natural gas pipelines. And you see the transportation application in green here.
Hydrogen fuel cell vehicles would be one of the leading demands, both in the near term and long term. And this will be my focus today, but other applications, such as synthetic fuels, upgrading of conventional oil or bio-oils in addition to industry applications, such as ammonia production or metal refining could also be leveraged, production of renewable hydrogen.
So, my focus will be on hydrogen fuel cell vehicles. The demand is expected to grow in the near term, especially in California. And then the zero emission vehicle states in the Northeast of the United States. And as the vehicles ramp up, there will be growing demand for the molecule.
So, vehicles are being deployed today, and here I have brought a couple of vehicles that are in the market today. On the left you see the Toyota Mirai, rated at 66 miles per kilogram of hydrogen in the U.S. driving cycle. And on the right you’ll see the Honda Clarity, rated similarly at 67 miles per kilogram of hydrogen.
Typical driving distance in the U.S. annually is somewhere between 12,000 to 14,000 miles, and on the average, a driver will drive a vehicle roughly 34 miles a day. Now, looking at the fuel economy, 34 miles or 33 miles is roughly—require roughly half of a kilogram per day. So, if we understand the market penetration of fuel cell vehicles, we can calculate the market demand for a given vehicle scenario.
So, in California for example, you will see the projected increase in fuel cell electric vehicle deployment, and within five-year time frame, the projection is somewhere between 40,000 and 50,000 vehicles on the road. Considering a half kilogram per vehicle, then we are talking roughly about 20 to 25 tons per day of hydrogen will be required in California alone within a five-year time frame.
This will lead to an important question, where the hydrogen will come from. Twenty tons a day in a given market would need to be satisfied with certain conditions. So, it is important to understand where the hydrogen would come from, and to look at opportunities for a low-cost molecule that will serve that market in the near term. And the idea here is to look for some source of hydrogen that is near term. Of course we understand that the future is for renewable hydrogen.
As I mentioned earlier, from zero carbon sources such as solar and wind and nuclear, but until we get there and until the economics and the market demand grows enough to make these technologies economical, we need something near term, today, that can serve these markets.
So, are there opportunities that can help as a foundation as we ramp up the demand? So, this is the question we are trying to answer today. And how this might enable energy security within the United States, and promoting the hydrogen as an energy carrier, not only to serve the vehicle market, but to serve across sectors.
So, what will be the requirement of a near-term hydrogen supply source? First, we need the scale. We [inaudible] in California within five years, twenty tons a day. We need, also, the hydrogen to be produced in high-purity form. And in here I put greater than 80 percent. Eighty percent is what we get from an SMR before the PSA purification. So, if we get that high-purity stream, then it will be economic to put a PSA on it to purify the hydrogen and supply hydrogen purity consistent with the requirement of fuel cell electric vehicles.
We also need that in the near term we avoid large capital investment to minimize the risk of investment. And lower capital investment would result in lower cost molecules. And we would like to make sure that we can produce a molecule competitive with SMR costs today.
We also want the molecule to be near the demand sites, whether it is on the West in California or the Northeast. So, the proximity of supply and demand will minimize the transportation costs. And finally we need the molecule at least to be at par with hydrogen [inaudible] steam-methane reforming, or better, and therefore, we could leverage some of the incentives provided for a low-carbon hydrogen.
So, what are our options today to serve early markets? One option is to build a large-scale SMR plant in anticipation for a market growth. And this could be at a larger scale central or a smaller scale if the demand is low and localized.
Another option is to leverage some of the excess capacities among the capacity existing today that is not being utilized in existing merchant plants across the U.S. And we would shed some light about what is the capacity of these merchant plants, and what might the excess capacity amount to, and the potential for that merchant market. And then really the core of this presentation is to explore the stranded byproduct hydrogen in existing industrial and chemical plants. And I will provide some quantitative numbers about the amount and the location of these potential hydrogen supply sources.
So, let us first eliminate some of these options. In the near term, building a very large SMR plant, 20 to 100 or 200 tons per day, requires significant investments, hundreds of millions of dollars. And this will require to invest in such a project you would require significant demand and also you would like some certainty of the demand and the longer-term contracts which does not exist in the current market, in California or other states. These plants also require long lead time to build and operate, to justify the investment, to get permitting, to design and to construct and put in place.
So, in the near term, this is not a viable option. And in the near term here, I mean the next five years, five to 10 years. So, absent a large-scale plant, maybe a good option will be to build a small plant that satisfies a local demand, what we call distributed production or on-site production. At the scale consistent with the demand, maybe half a ton a day, maybe two tons a day, depending on the demand.
This may be appropriate for a fleet operation. For example, a bus fleet, a fuel cell bus fleet where you may have the consistent demand, and you have the land area, and you could have—you could depreciate your investment over the service of the buses provided over the years. However, this might not be as viable for a fuel cell light duty vehicle market, simply because you shift the burden of capital investment to the station operator who are usually small investors and risk averse.
Furthermore, you are adding capital, which would be under-utilized in the near term because these stations are not fully utilized as vehicles are being deployed gradually over the years. So, the economics will not work. In addition, you would need significant footprint. To install the on-site equipment, you would need purification, you would need, maybe, power upgrade. And all of this will deter that from being a viable option absent fuel application, as I mentioned, fleet operators. So, this is of limited viability in the near term as well.
So, how about hydrogen being produced today for refineries, I mean, and for the industry in general? And here I dissect the U.S. into three key regions. On the left you see California, and you see where are the different industries that use hydrogen. And you will see, also, the blue triangle there. These are the liquid merchant plants. There are two of these in California, one in the north, about six tons a day, and one at the south, about 20 tons a day; combined 26 tons per day. And they already have their markets.
In the Northeast you will see several of these industries that use hydrogen, and the hydrogen is being produced there with one large liquid plant in New York, with a capacity of 40 tons a day. However, the majority of the hydrogen is produced in the Gulf Coast, as you see in the middle: 7,000 metric tons per day, mainly to serve the U.S. refining industry. So, we have a total merchant hydrogen capacity of roughly 15,000 tons per day.
And out of these, a small portion, only 250 or 260 tons per day, are liquefied to supply the merchant market. If we consider the liquid hydrogen, which is really the dominant mode to supply hydrogen to several markets in California alone, actually, we will see that this is among refinery customers. And if we assume 10 percent excess capacity with combined liquid plants in New York and California, we are talking about a total of 60 to 70 tons per day.
If there is a 10 percent excess capacity, then we are talking about only six or seven tons a day, which is capable of serving roughly 15,000 vehicles. And we are talking about 40,000 to 50,000 vehicles, in California alone.
So, this has limited potential in the near term. So, what we are proposing here is that there’s significant production of hydrogen that is being stranded in industrial operation. And here I will show two examples. The first example is the chlorine plants. The chlorine plants produce roughly one million kilograms, or 1,000 tons a day of hydrogen. This, alone, can fuel two million fuel cell vehicles. And here you will see, actually, the chloralkali process here. Use electricity, mainly electrolysis, to separate the chlorine and to produce hydrogen. That hydrogen has several uses. Roughly half of it goes to the merchant market, which is the best use of the hydrogen molecule. In other cases, where there is no market, it is combusted for processes, and this is roughly 30 to 40 percent. And about 10 to 20 percent is being vented. So, if we consider a 10 percent vented hydrogen of the 1,000 tons per day, we are talking about 100 tons per day. This can serve 200,000 fuel cell vehicles in the very near term. And this is a very high purity hydrogen. We are talking about 98, 99 percent purity.
The other even bigger byproduct hydrogen comes from the cracker plant. And the cracker plants crack mainly ethane or butane or propane to provide the chemical feedstock ethylene or propylene or butylene. And in the process, you create a double bond and release hydrogen. That hydrogen is of high purity, too, somewhere between 75 to 90 percent purity, very similar to the purity that you would get from an SMR before the PSA purification. Unfortunately, that hydrogen is being burned and combusted again for processes, and is being supplemented by natural gas, as you can see, to promote the cracking process.
Now in these two examples, that hydrogen can be set free and the thermal value of it could be displaced by low-cost BTUs of natural gas. So, you can see, really, there is significant potential here. You have 1,000 tons per day from chlorine plants, you have 7,000 tons per day of hydrogen from cracker plants, combined eight million kilograms that can serve 16 million vehicles.
Here you can see where the chlorine plants are, and these are combined, again, produce 1,000 tons per day. You see they are fairly distributed, although a larger number of them is in the Gulf Coast, but you will see significant number of them in the Northeast, and some of them on the West Coast. This could produce very high purity hydrogen at very low cost. And if you consider hydrogen that potentially could be vented when it does not have use, this is a green hydrogen molecule.
This slide shows the cracker plants. On the top left you will see the distribution of the cracker plants across the U.S., and on the right you will see the concentration, really, in the Gulf Coast area. And there is roughly 51 ethylene production plants. This is just the clear ones. I am not talking about the propylene or the butylene ones. And these produce over 1.3 million tons of hydrogen a year, over 3.6 million kilograms a day of hydrogen. And this is, again, a high-purity hydrogen. It is being surrounded, it is being combusted. Hydrogen has much higher value than its thermal value. As we know, actually, you can compensate the thermal value at a low-cost, cheap natural gas. And the hydrogen can serve a bigger role in fuel cell vehicles where you can get the efficiency.
So, the proposal here is to make use of these stranded assets in a more energy efficient way. And as we have seen before, the fuel cell vehicles are at least twice as efficient as the closest counterpart in internal combustion engine vehicles.
So, this slide shows really what would happen if you free the hydrogen molecule in a cracker plant. So, if I set the hydrogen free, I need to compensate for its thermal value by supplementing the heating process with natural gas. And this is a very simple displacement. Natural gas is cheap, low-cost, $3 to $4 a therm. A therm is a million BTU. I mean, last year even that, below $3 a million BTU. If we translate that to displace the hydrogen, then you could really free the hydrogen at a very low cost. We are talking less than half a dollar a kilogram.
If you add to it the cost of purification through PSA, and with it the techno-economics for that and also vetted that through industry, you add a small fraction, maybe 10, 20 cents at the most. So, you are talking something in the order of half a dollar kilogram hydrogen produced from the PSA unit. And if it requires some compression, you may add another 30 cents or so.
So, all in all, we are talking at a high purity hydrogen molecule at less than a dollar a kilogram that is larger scale, that requires no initial large capital investment. Some investment in a PSA unit, maybe some compression, but this is relatively low investment compared to the scale that you can produce and the cost you could achieve with this stranded asset.
So, here it gives you a picture of the cracker plants, existing and the planned. In the next three to four years, the cracker plants will double. This is according to the EIA and the companies filed to build these plants. You will see, in particular, the ethylene plants will double, and again, the majority of these are in Texas, and the Gulf area, Louisiana, but you will see some are coming in the Midwest. You’ll see North Dakota, you will see Illinois. And then some will come in the Northeast. You see Pennsylvania, you see Ohio.
So, this will produce significant amount of hydrogen. We spoke with one of these cracker plants. We said, “Why do you not sell the hydrogen for a bigger value?” They said there is no market for it. Where these plants exist, they don’t see a big demand for hydrogen.
This chart will show combined, the cracker plants existing and the planned, and will show also the chlorine plants. Combined, they have the potential of 8,000 metric tons per day hydrogen. Again, this can serve millions of vehicles—over 15 million fuel cell vehicles could be served just by these stranded hydrogen assets. And, again, this is a large-scale, low-cost, requires small capital investment and is fairly distributed.
How about the environmental impact of that? We know SMR is, at best, at 70 percent today lower heating value base efficiency. So, for each hydrogen BTU value or heating value, you need to consume roughly one and a half BTUs of natural gas. Now, the cracker plants will just display a BTU is a BTU, as if it is 100 percent efficient. That efficiency gain will make the carbon footprint of hydrogen from the cracker plants to be 30 percent less than SMR. And remember that we crack fossil feedstock, in this case ethane or butane or propane.
But the carbon we bring from the ground, we do not release in the air. It is locked in the product, the chemical feedstock, and the hydrogen is fairly zero carbon. However, because we compensate for the hydrogen with burning natural gas, we need really to burn that to the hydrogen, the amount of natural gas for [inaudible] BTU. But all in all, it is lower carbon footprint compared to the SMR, at least 30 percent. In some cases it could be more.
For example, you will see cracker plants can provide 25 percent to 30 percent. It depends, really, on how much compression and how much purification you may need. But if the hydrogen is vented, and you make use of it, then it is a zero carbon molecule. And you may have a small penalty from compression, for example. And this is why you will see a large reduction if you could leverage and utilize some of the hydrogen that is being vented today.
This slide shows the value of the low carbon hydrogen. This is, I borrow from a CARB presentation back in 2016, last year, by Sam Wade. And some of these values are proposed, but calculated based on the carbon production potential. On the far right, here, you will see the SMR. Even a fossil source of hydrogen generates credits under the low carbon fuel standards in California. You will see $1.60 a kilogram credit for gaseous hydrogen.
You will see over $1 for liquid hydrogen. And this is all at the $100 per metric ton of CO2, which was really the average price in 2016 as you see in the table above. So, the credit prices kept going up from 2014 to 2015 to 2016. So, at the higher price, you will get higher credits. And if they say the market can generate $1.60, then that hydrogen from the cracker plant can produce $2, because it’s an even lower carbon.
And this is really $2 credit, while it costs you $1, then you could make money. Of course, the transportation cost is big, and the re-fueling cost is big, but you could offset some of the transportation costs with the credits you could generate in a market like California.
So, to conclude, the checkpoints here is that we have a large-scale production. We have a large-scale stranded hydrogen in these industrial plants, fairly high purity. May require some investment in PSA to purify, but this is relatively low cost. This source would require low capital investment. Again, maybe some purification equipment. It is a low-cost molecule that can compete with today’s SMR and it even provides a better environmental footprint, compared to the MR, and can generate better credit.
And these plants are distributed and the upcoming ones, especially in the Northeast, the cracker plants can serve the Northeast market as fuel cell vehicles are deployed there. So, with that I want to acknowledge my colleagues at Argonne who contributed to this presentation, Dr. D-Y Lee, Leah Talaber, Marianne Mintz, Michael McLamore, and Steve Folga. I want also to acknowledge DOE FCTO, the Fuel Cell Technologies Office, for their support of this work, and for the H2 @ Scale initiative in general. And in particular, I would like to thank Fred Joseck and Neha Rustagi. Thank you.
Neha Rustagi
Thank you, Amgad. It looks like we have a couple questions coming in. So, the first question regarded, so you discussed the costs of replacing hydrogen with natural gas at these chloralkali plants. Is there a cost associated with doing that to anything else, so to non-hydrogen products?
Amgad Elgowainy
So, I’m not sure about the question, but really, the idea here is simple. Rather than burning the hydrogen, burn natural gas and free the hydrogen. So, if I free a BTU of hydrogen, I consume a BTU of natural gas. So, the cost of operating a BTU of hydrogen, is the cost of a BTU of natural gas. So, this is really the point I was trying to make. So can you repeat the question? I am not sure I understood the question.
Neha Rustagi
I think it’s just to clarify that replacing the hydrogen with natural gas, doesn’t have any other impacts on the ethane or on the target product.
Amgad Elgowainy
Yeah, I mean, especially in cracker plants, you already have the infrastructure, you have already natural gas is being used to supplement the hydrogen energy for the cracking process. So, I imagine if there is any increase in demand for natural gas to offset the hydrogen, there might be some costs there. But if the infrastructure is there, then I imagine the cost will be low.
Neha Rustagi
Okay. Another question was on slide 19, so I think we can scroll back to that slide. What carbon emissions are you assuming supplies power to the chloralkali plants?
Amgad Elgowainy
So, the power, I mean, the chloralkali plants have different technologies for producing the chlorine, and both of them use electricity. However, what I did here, you have different methods to locate the energy and the emissions between the whole product. The most conservative one to give credit to the hydrogen is to say, well, after I produce the hydrogen, I use it for its thermal value. So, what if I displaced the hydrogen with the natural gas? Then I burden the hydrogen with the natural gas that is compensating for the hydrogen loss.
If I use a different life cycle analysis method, which really we do, then the carbon footprint would be much lower. So, if I split the electricity use between the chlorine and the [inaudible] and the hydrogen, then the burden of hydrogen will be much smaller, too. And so there are different methods, and how you allocate based on energy, based on mass, because roughly, for each ton of chlorine, 3% of that will be the mass of core product hydrogen.
So, when you allocate by mass or by energy, hydrogen will see very little burden. So, here I put the most conservative life cycle estimate of carbon emission associated with freeing the hydrogen and compensating it with natural gas. So, other methods will show even a much lower carbon emission than what you see on this slide.
Neha Rustagi
Okay, thank you. One other question came in related to delivery infrastructure. So, if you can speak to the types of infrastructure that could be used to transport this hydrogen, would it be carriers, like methanol, or what would be the infrastructure methods of interest?
Amgad Elgowainy
So, again, the thinking here is near term, how we can make things happen near term in a cost-effective way. And we are talking here about, like, next five years. So, in that sense it’s a traditional mode for transporting hydrogen absent a very large demand that can justify a pipeline, then we are only dealing with trucking options, either gaseous trucking or liquid trucking. And the gaseous trucking you could, today, you can carry a ton of payload with these composite tanks, lightweight composite tanks. You can carry a ton of hydrogen, and the liquid tankers can carry about four tons of hydrogen. So, depending on the distance and the transportation distance and the proximity of supply to demand, if you are within a couple of hundred miles, then gaseous delivery will be economic. If it is longer than 300 miles, if you truck, for example, from Texas to California, then it is over 1,000 miles, and liquid hydrogen will be more economical.
And, again, we are looking in the Northeast, perhaps, gaseous trucking will be more viable because of the proximity of supply and demand. And trucking is costing today about, depending on the distance, for a couple of hundred miles, the trucking may cost about $2 a kilogram. If you go, like, several hundred miles, it may cost more. So, the idea here is that if you transport to California, and you are able to leverage some of the credit, it will offset some of the cost of transportation. So, again, you may get $2 credit, GHG credit in California today, and this will offset some of the transportation costs.
Neha Rustagi
Okay. One more question was at what point will the demand, the hydrogen demand for fuel cell vehicles and lift trucks in North America begin to strain the existing hydrogen generation capacity?
Amgad Elgowainy
So, what we know is that today it is already starting to be a problem, as today I believe we have in California about 3,000 vehicles. That is expected to more than double, in the very short term. So, and then suddenly you will have a stress on the demand. So, today, a lot of it is being supplied by the merchant market, by the excess capacity here and there. But, this is very limited resource now.
And as we push forward, then, some companies starting to think about tapping into refinery hydrogen, and pipelining it or purifying it, trucking it. So, it is already a problem that has been mentioned in several forums, and in several conferences, and at the California Hydrogen Business Council, this has been echoed in many, many places that there is an urgent need to figure out where the next molecule will come from.
Neha Rustagi
Okay. The next question, this is, it regards, I guess, more for FCTO, on whether or not the office has an interest in use of hydrogen for stationary fuel cells. So, I can say while the majority of our funding is focused on fuel cells for vehicles, we have funded a lot of stationary fuel cell work in the past, and we do have some ongoing projects in that space. The next question for Amgad is what is the pressure that chloralkali or cracking plants are operated at, and whether or not those pressures would increase the cost burden to the PSA to purify the byproduct hydrogen.
Amgad Elgowainy
Yeah, this is a very good question, and I alluded to that. Hydrogen from the chloralkali plant will come at very low pressure, a couple of bars or so. And it will require really compression to transport and to supply to a hydrogen terminal, or a hydrogen liquefaction unit. But that cost is small, and I mean our estimate of that is something in the order of 25, 30 cents. So, in addition, actually, this is from the chloralkali plant. It is fairly pure, so it doesn’t require a PSA there.
For the cracker plant, it will require a PSA, and in fact, Neha shared with me yesterday, a link that Praxair reported last year, this was in the news in 2016, that they are putting a PSA, and this is the only cracker plant I know doing that, putting PSA to purify hydrogen and supplying it to Dow Chemical. So, it does really have a precedent here, and it appears that it did make a local economic sense to do that from Praxair. The cost of purification and compression combined is less than half a dollar a kilogram.
Eric Parker
Okay, thank you Amgad. Without any further questions, I think we will wrap up for today. So, I’d like to thank everyone for joining us today and the thoughtful questions, as well as a big thank you to Amgad and Neha for their help today. If you didn’t get a question, please feel free to email any of us on the slide you see. I also want to encourage everyone to sign up for the monthly newsletter. That will inform you on future webinars. And you can find the sign up for receiving the recorded webinars and slides on that website as well. Have a great rest of your week everyone, and goodbye.
[End of Audio]