>> 29 August 2021 – Updated to Version 4.1: renaming Off-Road Vehicles as Non-Road Mobile Machinery; added Generators at Level D; clarified that Creative Commons license applies to the accompanying images, not to this article.
The main changes from the last version are to clean up some of the terminology. It’s hard to summarise the full splendour of a use case in just a couple of words, and that has led to some confusion. I have downgraded the aviation use cases a tad since the last version, and upgraded off-road vehicles. Of which, a bit more below, where I give a thumbnail of my reasoning for each use case. Oh, and for the avoidance of doubt, by hydrogen I also include ammonia and e-fuels or synfuels made via hydrogen (don’t @ me!).
Many thanks to all those who have contributed to my education over the past few months, in particular the brilliant Paul Martin, from whom I think I borrowed the Swiss Army Knife imagery and whose generously-shared knowledge of chemistry far exceeds my own, and to Lord Bamford of JCB. Thanks too to Adrien Hiel of Energy Cities, who gave me the initial idea for a ranking based on the EU’s A to H energy efficiency scale.
For those who want to use and reproduce the Clean Hydrogen Ladder, there is a link to a high-definition set of images at the end of this article. I have made it available under a Creative Commons license, which means you can use it as long as you credit me, and you can even modify it, as long as you make it clear that any modifications are not approved by me. Enjoy!
For those who want more background, let’s start with the standard view of the hydrogen economy. I don’t mean Jeremy Rifkin’s bonkers view (“the Worldwide Energy Web and the Redistribution of Power on Earth”), I mean the view of hydrogen as the Swiss Army Knife of the future global economy, able to do pretty much any job that needs doing.
The problem is, just like a Swiss Army Knife, you won’t use hydrogen for everything you could theoretically do with it. Clean hydrogen will have to win its way into the economy, use case by use case. It could do so on its merits, or it could do so because of supportive policy (including carbon prices). But it will have to do so in competition with every other clean technology that could solve the same problem. And that is where the dreams of the hydrogen economy hit reality: in almost all use cases there is a good reason why hydrogen is not currently used – because other solutions are cheaper, simpler, safer or more convenient.
What the ladder does is summarise in a single, simple graphic, my view of where clean hydrogen is sure to be part of a net zero future – starting with where we currently use grey, or polluting hydrogen – and where there are almost certainly other and better solutions – generally direct electrification and batteries.
Some people have asked whether the ladder is based on peer-reviewed research. The answer is yes, lots of it, all by other people. But, importantly, what it is trying to do is bring together all the different factors that will decide success or failure of a clean hydrogen solution, including thermodynamics, micro- and macro-economics, safety, human behaviour, resilience and geopolitics. If anyone knows how to do that, without building a model of Integrated Assessment Model complexity (and lack of credibility) I would love to meet them. And yes, there will be regional differences – based on availability of resources, need for heating and so on – but the ladder at this point presents a single global view.
Let’s take a quick look at what the ladder says, by sector.
What you’ll immediately see that the top of the ladder, level A, is all about replacing the use of grey hydrogen in the economy.
Current uses of hydrogen – principally for fertiliser, oil refining and petrochemicals production – currently accounts for around 2% of global CO2 emissions. Clean hydrogen has to win here, as there is no alternative.
Clean hydrogen offers a very promising way of decarbonising steel, but its not 100% sure, because there are alternatives like molten oxide electrolysis that could out-compete it. As clean hydrogen gets cheaper, and if cheap CO2 becomes available via Direct Air Capture (unlikely) or as a byproduct of biogas production (more likely), we could start to see it used to make a range of chemical feedstocks.
In the power system, you won’t routinely use hydrogen to generate power because the cycle losses – going from power to green hydrogen, storing it, moving it around and then using it to generate electricity – are simply too big. The standout use for clean hydrogen here is for long-term storage.
The economy of the future, which is going to be vastly more deeply electrified than today, needs long-term storage. It’s not just about providing back-up for when there is no wind or sun, it is also going to be about providing deep resilience in the case of weather disasters, cyber or physical attacks, neighbouring countries shutting off interconnectors and the like. Hydrogen can be stored in salt caverns, depleted gas fields or as compressed gas or liquified at various strategic points. It can be converted back into electricity centrally at 60% or more electrical efficiency via fuel cells [disclosure, I’m an investor in Ceres Power], providing as high a level of grid resilience as you want to pay for.
Even long-term storage is not a slam dunk for hydrogen: there are scalable alternatives (in addition to things like pumped hydro, demand response, batteries, etc, etc, which won’t get to the right scale or are geographically limited). It is possible that compressed air storage might be cheaper than hydrogen (Professor David Cebon is a big fan of this approach). Or we might just use unabated natural gas and offset a few percent of residual emissions.
It is possible to see a role for hydrogen in things like island grids and uninterruptable power supplies (UPS). In each case you might need more days of resilient supply than can be cheaply provided by batteries, and in the case of non-grid-connected islands where renewable power can be generated, making and storing hydrogen might be less absurdly expensive than alternatives. They make row E.
The other thing on row E is clean power imports. Hydrogen pipelines are a cheap way of importing energy, as long as what you have in Country A is hydrogen, and what you can use in Country B is hydrogen. Similarly for ammonia/fertilisers. The problem is that these are edge cases. In the net zero economy of the future, what you are going to have a lot of is clean electricity in Country A and a need for clean electricity in Country B. While you certainly can import power by converting it into hydrogen, compressing or liquefying it, transporting it, removing and adding back heat, storing it, and using it to regenerate power, the round trip losses are gargantuan. HVDC connections look more likely to win [disclosure, I’m an investor in XLinks, though my views drove the investment, not the other way round].
One thing we won’t be doing with clean hydrogen is using it for short-term grid services like balancing and inertia. Software (demand response) and batteries will simply be waaaay cheaper and simpler than hydrogen.
In aviation and shipping, the opportunities for clean hydrogen range from shipping – where clean ammonia based on clean hydrogen looks promising – down to short-haul and light aviation, where battery electric aircraft look like they are going to win as battery energy densities increase and costs come down.
Medium- and long-haul aviation is going to be impossible to electrify due to fundamental limitations in battery chemistry/physics. Hydrogen, however, also suffers fundamental limitations, not because of its gravimetric energy density (which is excellent) but because of its volumetric energy density (which is terrible). My colleague in Ecopragma Capital, Henry Lawson likes to say you can fly London-New York on a hydrogen plane as long as you have two other planes in tow carrying the hydrogen. However, I have medium and long-haul aviation fairly high up, levels D and C respectively, because it is possible to see them use some volume of synthetic fuels based on hydrogen, which counts in my book. Otherwise, if we want to keep flying, it’s biofuels, and there just isn’t enough of it for all the uses it will need to be put to.
Local ferries, routes up to a few hundred kilometers, are most likely to go electric, as described in the report Liebreich Associates wrote last year for the IADB. However, coastal and river vessels, working routes of a few hundred to a thousand or so, look like a very promising market for hydrogen. These lengths of routes can’t be served by batteries, but marine vessels don’t have the energy density challenges of aviation, so hydrogen might work, either in a fuel cell or just burned in an internal combustion engine which can benefit from the existing marine engine supply chain (more on combustion engines below).
OK, this is the controversial one. I’ve written and tweeted endlessly about this, so it shouldn’t come as any surprise that I am not bullish at all about any use of hydrogen in any but niche settings.
By now everyone should have managed to get their head around the fundamental inefficiency of turning electricity into hydrogen, compressing it, storing it, moving it around and then converting it back into power on board a vehicle. Somewhere between half and three quarters of the input power is wasted, and this is not going to change much – there are fundamental thermodynamic constraints. Not only that, but the H2FC vehicles are also much more complicated so they have higher maintenance costs, and although hydrogen can be safely handled, you really don’t want it in every garage and workshop. You can forget use cases like urban delivery, two and three wheelers, metro trains and buses.
It’s not just about efficiency. We do plenty of inefficient things in life, like drive SUVs and sports cars, cook on barbecues and go on holiday. The fundamental problem with H2FC cars is that they are worse vehicles on all the dimensions bar one which people use to choose their next car: worse acceleration (because BEV’s are astonishing), less seating and cargo space (because they are full of hydrogen tanks), higher maintenance (because their drive trains are so complex) and you can’t refuel them conveniently at home at the office, at the mall, at a nearby lamp-post and so on.
The one area in which hydrogen cars win is that you can refuel them quickly during long trips. Is this a killer app? I don’t see it. Most of us do so few trips over 250 miles each year that the disadvantage of the odd enforced charging stop will pale into insignificance by comparison to the enforced weekly trip to a hydrogen fuelling station. By the time EV charging is as ubiquitous as internet bandwidth, those who do drive over 250 miles at a go a bunch of times per year (and I am one of them) will get used to doing slightly longer natural breaks, which will also be good for road safety.
I could have put hydrogen cars on level F, because it is conceivable there might be some use cases and drivers who really need to drive 400 miles in any direction at the drop of a hat, in freezing conditions. But let’s keep things simple – Level G, hydrogen cars are not going to be a thing.
Local trains and buses simply don’t cover enough miles for hydrogen to have any advantages. All you get is a vehicle with higher maintenance costs and a safety problem in your depot. Yes, investment is needed in charging infrastructure, but once that is done, you have a lower cost of ownership, a more comfortable vehicle for driver and passengers, and no problem charging during vehicle down time at night or during the day away from rush hours. Level G.
What about slightly longer train routes, particularly rural ones? Of course they can work using hydrogen fuel cells, but why would you bother? Just electrify the track already if it is a busy segment or, if it is not, stick a battery on the train. Simple, low cost of ownership, job done. So that’s an F.
Regional trucks are an F too. Most trucks do not thunder across continents – 85% of them do less than 500km per day. That’s the sort of distance that lends itself perfectly well to battery-electric vehicles. Analysis by my great friend Auke Hoekstra of Technical University Eindhoven shows that by the time you’ve taken out the internal combustion drive train, a 40-tonne battery truck will weigh barely more.
Long-distance trucks are a D. Wait, you cry, everyone has been saying that hydrogen is the only solution here. Sure, you could build a whole hydrogen infrastructure just for the 15% of trucks that do thunder across continents, but equally you could just invest in a bunch of high-capacity charging. Once you realise that all local trucks are going to be BEVs, along with anything doing up to 500km a day, the infrastructure for electric long distance trucks is much easier to envisage. Intermittent catenary or contactless charging in the road surface would combine elegantly with platooning autonomous operation on remote highway stretches. Long-distance coaches will go the same way as trucks.
That leaves remote trains – I’m thinking the sort of trains a mile long carting minerals on tracks used by no one else – where hydrogen might beat batteries, assuming there really way a policy requirement to go clean. So that’s a C, along with vintage vehicles, which could use synthetic fuel made via hydrogen, and damn the cost.
There is a lot of lobbying around e-fuels right now, particularly in Germany, but that really looks more like a delaying tactic than a serious attempt at a net zero transport solution. If you don’t believe me, read about Astongate, the sock-puppet scandal with Robert Bosch at its beating heart. Bulk use of e-fuels is a G. If we went to H, bulk e-fuels would be an H. Or a Z.
Yes, you can make a synthetic liquid fuel from hydrogen, but far too high a cost to be used for general land transportation. It’s not just that each stage in the production process involves energetic losses (it does), but in addition to your hydrogen you have to find a source of carbon. If all you are doing it taking out of flue gases of a natural gas plant, you might as well just save money and drive dirty using CNG.
That leaves off-road vehicles. We are talking things like graders used in mining and road works, cranes and construction machinery, forestry vehicles. These are vehicles that operate in environments where there are generally not great grid connections. So if you go electric, you will either need to drive out to recharge, or bring in batteries to swap, or invest substantially in local charging infrastructure. Hydrogen looks like an attractive option here. I very much enjoyed a trip recently to see the JCB plant near Uttoxeter to see a digger using a hydrogen-burning internal combustion engine. This was fascinating – JCB have developed an engine based on a series model of diesel engine, which means it uses an existing supply chain and vehicle design – unlike a fuel cell solution. Compressed hydrogen can be supplied via a trailer, it looks like a simple and elegant solution, though I did not ask if the vehicles can carry enough fuel to last a shift, because gaseous hydrogen is so much less volumetrically dense than diesel.
We’ll finish our tour of sectors with heating. Right now there is a mammoth lobbying campaign going on to place hydrogen at the heart of the country’s future space heating needs. It’s obvious, say the owners of existing gas distribution networks or boiler-related businesses. You just swap out a natural gas condensing boiler, for a hydrogen-ready duel fuel one, and then in due course just switch over to hydrogen. Sorry, I’m not really seeing it.
Heat pumps multiply, hydrogen divides. If you are looking at using the UK’s primary winter source of zero-carbon energy, offshore wind, heat pumps are not just a bit more efficient, they are about six times more efficient. Using wind power to generate hydrogen, and then using that for heat, would have a system efficiency of around 50%. Using the same power via a heat pump would have an average coefficient of performance of 3.
Even the most committed hydrogen boosters seem to be admitting that when it comes to new homes, heat pumps win. But they are spreading all sorts of misinformation about heat pump retrofits, claiming they require replacing all your radiators, re-insulating your home, replacing your grid connection. The fact is that modern heat pumps work at 50C or above, just like properly-installed boilers (most UK boilers are not properly installed, they are set to excessive circulating temperatures so they are hugely inefficient). Octopus Energy is looking to install heat pumps for £5,500 a time. Keep your heat pump chugging along in the background, responding to peaks and troughs in supply and demand, you won’t overload the local grid and you’ll help smooth demand for power.
Ground source shared loop heat pumps look particularly interesting for community heating projects – just as we now have a gas main, we could well end up with most homes connected to a low-temperature heat main.
And that’s before you even get started on where all the hydrogen is meant to come from. If it’s green hydrogen, the volume is so colossal, it would pose huge challenges to the supply chain for renewable energies and the land or sea area available. If it’s blue hydrogen, fine then the calculus might change, as long as it can be produced without fugitive emission and at nearly 100% CO2 capture rates, otherwise there goes net zero. Then there’s the question of the extent to which hydrogen will leak from the distribution pipes (it’s greenhouse gas, not many people know that), whether householders will need to invest in new venting and valves outside their properties for safety reasons, and what the impact on NOx will be of large-scale hydrogen combustion for heating.
Of course there may be some buildings that really can’t be converted to heat pumps – perhaps listed ones, very remote from the grid, irremediably draughty, or whatever. So I do have domestic heating as an F, not a downright G. Commercial heating is an E: it is just possible we might use hybrid heating in commercial properties, using electricity most of the time but switching to hydrogen to relieve the grid during the Beast from the East.
Industrial heat is another interesting one. Low-and mid-temperature industrial heat – up to 160C and perhaps higher – will be much more cheaply and precisely supplied by heat pumps. Many people seem to think that high-temperature industrial heat has to be delivered by gas. That is simply not true, there are lots of ways of delivering high-temperature heat electrically. It’s going to be a street fight for low carbon high-temperature heat between hydrogen and electricity, process by process, plant by plant, so that’s a D.
Oh yes, what about blending hydrogen into the gas network? Pointless. 20% hydrogen blend by volume only delivers a 6% reduction in CO2 because of its lack of density and the increased work required to shift it through the pipes. And for what? To burn it in power stations regenerating power? Puh-lease. Or for use in heating, when our current fleet of boilers are so badly set up the majority don’t condense and fail prematurely? Anyway, blending doesn’t even appear on the ladder because it’s not a use case for hydrogen: it’s just a way of storing and moving it – perhaps the topic for another epic graphic one day.
One final look at the ladder. This one is pure Paul Martin – thanks Paul. Here I’ve tried to indicate which is the leading contender which hydrogen has to beat if it wants to win in any use case.
Broadly speaking, the landscape divides into those where there is no real alternative so clean hydrogen is clearly going to win, use cases where it is in competition with biofuels, so it has a good chance of winning, and use cases where it is in competition with battery-electric solutions, in many of which hydrogen is going to find it hard to win.
If you have enjoyed this piece and want more detail on how I came to these conclusions, you might want to read part II of my blog on hydrogen for BloombergNEF, entitled Separating Hype from Hydrogen, or listen to Audioblog Episode 3 of Cleaning Up, which is based on it:
Alternatively, if you want to know my thoughts on the supply side of clean hydrogen, read part I of that blog or listen to the Audioblog Episode 4 of Cleaning Up, which is based on it:
Thanks to everyone who has helped me refine my thinking on the Clean Hydrogen Ladder over the past months – on Twitter, here on Linked In at various events where I have presented the ladder for discussion, colleagues from BloombergNEF, Liebreich Associates and Ecopragma Capital. Please keep your comments, critiques and suggestions coming, and I will try to reflect them in Version 5.0, which will no doubt follow in due course.
And finally, I promised you a high-resolution version for download. Here you go!
To Read the Linkedin article CLICK HERE