There are multiple ways to visualize information about pretty much anything. Hydrogen is no different. There are entire color wheels devoted to the industrial feedstock at present, which presumably are informative to people who are less color blind than I am. But one of the more useful hydrogen infographics over the past few years has been Michael Liebreich’s Hydrogen Ladder, and now he’s updated it after a couple of years of stability.
I’ve discussed the ladder numerous times, both with Liebreich himself, but also with members of the Hydrogen Science Coalition such as Paul Martin, and with policy makers and clients. It’s a useful framework to consider.
Liebreich is, of course, the founder of what is now Bloomberg New Energy Finance, and has been assessing where the big money is in half-trillion dollar chunks for the past 20 years or so. He’s been investing personally in the transition. He’s usually been right.
I have my own way of visualizing hydrogen and its future, one I’ve iterated a few times as well. I most recently updated mine earlier this year when I stepped through the full scope of iron and steel globally, the technologies which were already commercialized or proven commercially at small scale such as HYBRIT, and projected demand, solutions, and decarbonization through 2100.
The differences between the info dumps, both of which have copious thinking, analysis and detail behind them, make for fruitful ground to consider. By the way, different people use different numbers for the current hydrogen demand. I worked mine up and am happy with 120, but others use 100 or 110. And, of course, error bars on my scenario are large, and it is a single scenario of many I could project, just the one I think most likely.
One of the first things that is obvious in contrasting the two is that my projection is through time while Liebreich’s is a point in time perspective by him of the relative competitiveness of hydrogen in different domains. When he and I discussed our perspectives earlier this year in London, he’s slightly less bearish on hydrogen demand than I am, thinking that total demand will be in the 200 or perhaps 250 million ton per year range, while I project a decline to the 90 million ton range from the current 120 million ton.
I suspect Liebreich’s is a shorter term perspective, perhaps through 2030 or even 2040, but it’s unstated. Certainly I was surprised to see an upward blip between 2040 and 2050 in my own projection when I integrated the iron new demand projection. And in my aviation projection, demand for biofuels goes up before it goes down as electrification takes over more and more of the space, so take everything like this with a grain of salt and use it as a springboard for thinking and discussion.
Unsurprisingly, at least to me, the bottom row of completely uncompetitive proposed uses for hydrogen has become a double stack. Liebreich is far from a hydrogen hopium addict, quite the opposite, but even very bullish hydrogen for energy types like former chief scientist of Australia Alan Finkel are realizing that it’s just not fit for a lot of uses, such as cars. I’ve been watching hydrogen for energy plays fall apart for the past few years as spreadsheet jockeys who had to use real numbers got involved and found what’s obvious from napkin math, that hydrogen is a very expensive carrier of energy and that electrolyzers and fuel cells don’t come as giveaways with a tank of gas.
The bottom row is now replete with eight use cases, including cars, taxis, metro transit, agricultural and mining equipment, bulk delivery of fuels between countries, heating under 200° Celsius, domestic heating, and burning hydrogen that’s just been created to make electricity. Personally, I think all of the hydrogen for energy use cases above this will collapse into the completely non-competitive row over time as Liebreich continues to iterate this. The only growth area in energy for hydrogen in my projection is related to hydrotreating biofuels in much the same manner as hydrogen is used with crude oil today, but the volume is much smaller.
Having done my global assessment of aviation and maritime shipping demand and solution projections through 2100, along with a lot of work on battery chemistries and biofuels, I’m quite comfortable that batteries and biofuels will dominate, while Liebreich is more bullish on biologically sourced carbon with electrolyzed hydrogen for those use cases. He asserts that there is limited feedstock availability, which I don’t think is accurate.
My projections — remember, big error bars, opinion, not carved in granite — show that the demand for burnable fuels for transportation is in the low hundreds of millions of tons. Current waste food is 2.5 billion tons of the 7.5 billion tons we manufacture annually around the world. Every ton of dried waste biomass (less than the 2.5 billion tons obviously) turns into 0.4 tons of biofuel. Just the waste food stream, ignoring the waste agricultural stalks and leaves, waste livestock dung, and waste timber residue is in the same order of magnitude as all demand. European livestock dung alone is in the order of 1.5 billion tons. My assessment of aviation fuel requirements found that current waste stalks were more than enough.
Burning stuff for energy is deeply inefficient, and as we won’t be just digging the stuff up and processing it relatively lightly, but manufacturing it at great expense, I see no reason to think we will perpetuate the practice where there is the alternative of electrification. There’s lots of waste biomass from human industry.
On that note, there’s a lot of yellow on the chart (according to people who aren’t color blind). That’s all the stuff where the alternative is more directly using electricity, either from the grid or via the magic of electrochemistry, aka batteries. Except for steel on the second row, where I project a growth of potentially 30 million tons for the HYBRIT green steel process and similar ones, which use 55 or so kilograms of hydrogen per ton of new steel, I expect all of those to collapse into row G over time.
There will be short term exceptions, but they will be short term. Hydrogen fans were delighted that India was buying electric locomotives, for example, because they didn’t look at the rails they would be used on or compare to electrification. India is committed to 100% heavy rail electrification by 2025, and is approaching 90% of the way there. The hydrogen engines were for ancient, narrow, windy, decorative tourist trains running through scenic regions, as I understand it. They didn’t want to spoil the view with overhead lines and it would have been expensive, too. That’s going to become accessible to emerging battery energy densities, so the hydrogen engines will become superfluous and too expensive.
Lower Saxony’s experience is the reality of hydrogen rail trials. They stop them when what was obvious before the trials started becomes too obvious to ignore, that hydrogen for transportation is at minimum three times as expensive as grid-ties and batteries.
Liebreich is still considering high-temperature industrial heat as a place where hydrogen can play. I don’t think it will, although I think anthropogenic biomethane will play in the space until it’s worth scrapping legacy infrastructure for much more efficient and cost effective electrified infrastructure. Some very expensive hydrogen for heat plays will be tried, but I think it’s deeply unlikely that they will be competitive.
Let’s explore why this is a little bit. Let’s look at blue and green hydrogen as the two major pathways.
Blue hydrogen is interesting. It starts with natural gas, aka fossil methane, aka CH4, aka one carbon atom and four hydrogen atoms bonded together by millions of years of pressure and heat acting on long dead plants, with the occasional dinosaur mixed in.
Natural gas is a pretty useful source of heat. It burns relatively cleanly compared to coal and it burns hot compared to alcohols like ethanol and methanol, 900° Celsius to 1,000° Celsius hotter. It’s also dirt cheap a lot of the time with the advent of shale oil and fracking, so the heat hasn’t been expensive and its cost was stable from the early 2000s until recently.
Remember that it’s made up of carbon and hydrogen. Well, 45% of the energy from burning it comes from that one carbon atom. Only 55% of the energy is from the four hydrogen atoms. And stripping that carbon atom off by bonding it with oxygen from the air to form carbon dioxide, then capturing that carbon dioxide, shipping it to a sequestration site and shoving it underground takes energy too. When we make ‘blue’ hydrogen we are actually using energy to throw away 45% of the energy in natural gas, as well as adding a bunch of capital and operational costs.
That means that blue hydrogen, even if it’s made from natural gas at the exact place where the heat is required, will be more than twice as expensive per unit of heat as burning the natural gas. There aren’t any ways around that. The basics of chemistry and physics don’t bend to wishful thinking by finance types.
Green hydrogen is also expensive. It takes a lot of electricity to turn water into hydrogen, about 60 MWh per ton with the balance of plant. That’s inefficient because when the hydrogen is used, once again even if it is manufactured exactly at point of use, it would return a maximum of 70% of the energy with a bunch of additional capital and operational costs. Electricity per unit of heat is more expensive than dirt cheap natural gas, so green hydrogen is even more expensive.
That comment about ‘at point of use’ is important. The vast majority of hydrogen is manufactured at the point where it is plugged into a chemical process like manufacturing methanol or hydrocracking crude oil. That’s because it’s one of the most expensive molecules to transport. It has to be compressed massively or chilled to 20° above absolute zero, and even then, its density is lower than alternatives. It’s just expensive to store and move.
This isn’t to extol the virtues of natural gas, by any means. It’s to point out that hydrogen is a really dumb idea for industrial heat in different ways than natural gas or coal are. It’s always going to be very expensive regardless of how we make it, and electric heat sources are going to be cheaper to operate. As long as we are trying to actually solve climate change, not treat the atmosphere as an open sewer and not destroy our economy, using electricity directly for heat will always be cheaper and lower emissions.
And in my work digging through all of the various use cases and technology for all aspects of heating, I haven’t found a single use case that we didn’t have electric technologies that were fit for purpose. What I did find were a lot of people who framed the question as replacing their burning gases or liquids with clean gases or liquids instead of asking where the energy would come from and how to apply it. That misframing has a lot of reasons behind it, many of them venal.
As always with Liebreich’s Hydrogen Ladder, I like to remind people that the top row isn’t a statement of continued or increasing demand of high volumes of hydrogen. Hydrocracking and desulphurisation (and hydrotreating) are almost entirely things done to crude oil in refineries. While there may be no alternatives to hydrogen for those use cases, those use cases are going to be in serious decline over the coming decades. Peak oil demand is this decade, and the heavy, sour crude which demands the most hydrogen to refine is going to be off the market first due to economic challenges.
Methanol is a problem of a different order, and having reassessed my position on it, more likely to diminish as a demand source for hydrogen, not increase, something I’ll address in my next hydrogen projection. Liebreich points out that biomass / biogas may provide the feedstock for it, but the process still goes through steam reformation of methane into hydrogen. As hydrogen must decarbonize it will get more expensive, and alternatives for the industrial solvent and feedstock will become more competitive. While it’s being touted as a maritime fuel alternative, I’m on record as noting the bait and switch nature of the effort. In other words, economics will likely drive a reduction in methanol demand despite the industry’s lobbying to make it multiply instead.
Fertilizer is a different concern as well. When we say fertilizer in the context of hydrogen, we’re really talking ammonia, aka a nitrogen atom and three hydrogen atoms. The past 60 years of amazing crop productivity increases has been due in large part to manufacturing ammonia from fossil fuels and putting it on fields. The fields don’t care about the hydrogen, that’s added to make a molecule that plants can do something with. The plants care about the nitrogen. Up to 5% of a plant’s dry mass is nitrogen.
Every ton of ammonia put on fields results in about 14 tons more of crops. That’s why we’re using less land to grow a lot more food than we used to. Since the start of the current Green Revolution in agriculture, growth in ammonia fertilizers has not kept pace with growth in population or GDP. It’s been a good market growth, but we’ve been using it more and more efficiently. With the recent price spikes in natural gas and shutting down Russian exports — Russia is a big manufacturer and exporter of ammonia for fertilizer — agribusinesses have been looking for alternatives.
We’re not going to eliminate ammonia fertilizers. But they are a big climate problem coming and going. Every ton of ammonia comes with about six tons of greenhouse gases from manufacturing and distributing it, and when applied to fields a bunch turns into nitrous oxides which are the equivalent of about three tons of greenhouse gases per ton of ammonia. Both ends of that are priced in carbon pricing schemes, including the EU’s carbon border adjustment mechanism as of 2026, so fertilizer costs to farmers are going up, not down.
Their use can be minimized without anyone starving. Agrigenetics like the nitrogen fixing microbes from Pivot Bio are already reducing ammonia requirements by 25% on millions of acres of corn in the USA, and they have a target of 100% for all three major grain crops by 2030. Precision spraying with electrically powered megadrones is grounding the helicopters and light aircraft that used to overspray fields, and cutting product masses by 30% or more with the same crop yield. And low-tillage agriculture has advantages in this regard as well.
I project a reduction in ammonia fertilizer demand, if not elimination of it, while Liebreich asserts that there’s no real alternative for it. Once again, timeframes for that question matter.
Finally, there’s long duration grid balancing on the second row. I really don’t see anything about seasonal electricity storage for 10 or 100 year problems that means the characteristics of green hydrogen are the best ones. I think it’s more likely that diverting the massive amounts of anthropogenically created methane to salt caverns makes more economic and climate sense than manufacturing low-carbon hydrogen and shoving it underground where it will try really hard to leak. It is the Houdini of molecules, after all.
I also think that a 10 or 100 year problem is a very different beast economically than everything else we are talking about. It’s more in the category of the Svalbard Global Seed Vault, a serious rainy day concern. For the next few decades, putting fossil methane away for a rainy day is good enough, as if we have to burn it only a couple of weeks every few years, we’re way ahead of the climate game.
None of this is to say negative things about the Ladder. It’s a very useful tool for discussing the comparative merits of the different use cases for the Swiss Army knife molecule. Like the tiny corkscrew, blade, and toothpick on the Swiss Army knife, hydrogen just isn’t the go-to molecule if there are alternatives, and there are almost always alternatives. And in hydrogen’s case, a lot of the current use cases are going to diminish due to economic and climate imperatives, not grow.
Have a tip for CleanTechnica? Want to advertise? Want to suggest a guest for our CleanTech Talk podcast? Contact us here.
EV Obsession Daily!
https://www.youtube.com/watch?v=videoseries
I don’t like paywalls. You don’t like paywalls. Who likes paywalls? Here at CleanTechnica, we implemented a limited paywall for a while, but it always felt wrong — and it was always tough to decide what we should put behind there. In theory, your most exclusive and best content goes behind a paywall. But then fewer people read it!! So, we’ve decided to completely nix paywalls here at CleanTechnica. But…
Â
Â
Thank you!
Tesla Sales in 2023, 2024, and 2030
CleanTechnica uses affiliate links. See our policy here.