When Ahmed Sheikh Yamani, Saudi oil minister during the oil shocks of the 1970s and 80s, famously said that “the Stone Age did not end for lack of stone, and the oil age will end long before the world runs out of oil”, he was not thinking of renewable energy and electric vehicles, he was thinking of hydrogen.
On the surface, the most common element in the universe seems like the answer to every energy question. It can be produced anywhere you have electricity and water. It can generate either heat or electricity. It can be produced, stored, transported and used without toxic pollution or CO2 emissions. It carries three times as much energy per unit weight as petrol, diesel or jet fuel. It can deliver power at 60% efficiency via a fuel cell which can also run in reverse. It can be pumped at similar transfer rates to liquid hydrocarbons. And it burns at a similar temperature to natural gas.
Sadly, hydrogen displays an equally impressive list of disadvantages. It does not occur in nature so it requires energy to separate. Its storage requires compression to 700 times atmospheric pressure, refrigeration to minus 253 degrees Celsius or combining with an organic chemical or metal hydride. It carries one quarter the energy per unit volume of natural gas, whether liquefied or as a gas at any given temperature and pressure. Fuel cells and other equipment designed to use hydrogen have many moving parts requiring maintenance. It can embrittle metal; it escapes through the tiniest leaks; and, yes, it really is explosive.
A bit of history
Despite these obvious disadvantages, hydrogen holds a vice-like grip over the imaginations of techno-optimists. This dates back at least to 1970, when Lawrence W Jones, a nuclear physicist at the University of Michigan presented a paper entitled Toward a Liquid Hydrogen Fuel Economy. In it, he noted: “The use of liquid hydrogen as a long-term replacement for hydrocarbon fuel for land and air transportation must be seriously considered as the logical replacement for hydrocarbons in the 21st century.”
In the mid-1970s, Japan listed hydrogen as one of five focus areas for its Sunshine Project, with a combined budget equivalent to $2.4 billion today, designed to identify ways of supplying the resource-poor country with energy in the aftermath of the first oil shock (the other four being solar power, geothermal, coal gasification/liquefaction, and general supporting research). Vestiges of the status afforded to hydrogen in Japan as a savior technology can be seen in the continuing support for its fuel cell car program.
When oil prices plummeted in the 1980s, hydrogen’s grip on the public imagination evaporated, only to come roaring back during the technology bubble around the turn of the millennium. Between September 1997 and March 2000, the share price of Ballard Power Systems, standard-bearer for hydrogen fuel cells, leapt by 1,000%. Futurist Jeremy Rifkin captured the zeitgeist with his book, The Hydrogen Economy, which he subtitled The Creation of the Worldwide Energy Web and the Redistribution of Power on Earth. He spun a heady cocktail of peak oil, climate change and dotcom mania into what he called the Worldwide Hydrogen Energy Web, in which “millions of end-users will connect their fuel cells into local, regional, and national hydrogen energy webs, and create a new decentralized form of energy production”.
This was not hydrogen as engineering solution so much as hydrogen as liberation theology. Even the Economist bought it: “Because hydrogen can be made in a geographically distributed fashion, by any producer anywhere, no OPEC cartel or would-be successor to it could ever manipulate the supplies or the price. There need never be another war over energy.”
Sadly, even by the time Rifkin’s book hit the stands in late 2002, the dreams had begun to sour – Ballard’s shares had shed 96% of their peak value. Soon the nightmares began: over the next decade Ballard shares would fall another 93%, before hitting their nadir in 2012.
Instead of a miraculous ability to redistribute power to the people, one of the main properties of hydrogen turned out to be relieving its backers of their wealth.
EU Green Deal and Hydrogen Strategy
And so here we are again. The EU’s Green Deal, which targets a 55% reduction in emissions relative to the 1990 baseline by 2030 and their complete elimination by 2050, and its Covid recovery plan, have at their heart two documents published simultaneously this July: the EU Strategy for Energy System Integration and the Hydrogen Strategy for a Climate-Neutral Europe.
These lay out in considerable detail a sweeping plan for the production and distribution of “green hydrogen”, in other words hydrogen electrolyzed from water using renewable energy.
The plan is built around an increase of electrolyzer capacity in the EU from the current 60MW to 6GW by 2024 and to 40GW by 2030, at a cost of between 24 and 42 billion euros ($28-50 billion). Keeping these electrolyzers fed with renewable energy will require the spending of 220-340 billion euros on 80-120GW of new solar and wind generation; then there is 11 billion euros to retrofit half of existing fossil-based H2 plants with carbon capture and storage; and 65 billion euros for hydrogen transport, distribution, storage and refueling stations.
The total comes to between 320 billion and 458 billion euros – $380 to $550 billion at current exchange rates.
Some of the funding will come from the Green Deal, to which the EU has promised 1 trillion euros from its own budget, two-thirds in the form of grants and one third in the form of loans. But this has to cover a lot more than just the hydrogen strategy, most notably energy efficiency. Member states and the private sector will be expected to pitch in too – and at the time of writing five countries had committed to a total of 26GW out of the 40GW electrolysis capacity target for 2030.
However it ends up being funded, throwing up to $550 billion of capex over the next decade at green hydrogen might seem, let us say, courageous, to use the terminology of Yes Minister (the U.K. TV series). But that is just the beginning. It only covers capital costs, and only for production and distribution; it does not include costs on the demand side, which we will look at in Part Two. And it only goes to 2030.
The working assumption for EU electrolyzer capacity in 2050, tucked away in footnote 35 of the strategy document, is 500GW. To put that in context, the maximum peak electrical load ever recorded for all of Europe was 546GW. The hydrogen strategy could drive up to a doubling of Europe’s power demand, a doubling of its power supply, a doubling of its power distribution capacity, and a Europe-wide hydrogen pipeline network. Goldman Sachs breathlessly predicts an addressable market for green hydrogen in Europe worth 2.2 trillion euros per year by 2050. Bank of America extrapolates total global hydrogen-related infrastructure investment of $11 trillion through to 2050. Big numbers.
Economics of green hydrogen
The EU Hydrogen Strategy is predicated on driving down the cost of producing green hydrogen in Europe, currently between 2.5 and 5.5 euros per kg, to between 1.1 and 2.4 euros per kg by 2030. Is that a reasonable target?
The cost of green hydrogen is driven by four main factors: the cost of renewable electricity; the capacity factor at which plants run; the cost of electrolyzers; and the cost of capital.
The cost of renewable electricity, of course, continues to plummet around the world. The best wind and solar plants in the best locations now generate power at around $15 per MWh, and by 2030 we are going to see this drop to $10/MWh, in my view. By 2030, there will be large parts of the world benefiting from $20/MWh wind or solar, around one-third the cost of power from any other source, and there is no reason to believe these sorts of costs will not be achieved in sunny and windy parts of Europe.
Electrolyzer costs too have been plummeting – with learning rates of just under 20% per doubling of capacity, similar to wind energy. There are still plenty of remaining pathways to reduce costs, and as the industry scales we will most certainly see electrolyzer costs come down. But there is a wrinkle. The EU Hydrogen Strategy wants to drive electrolyzers “from 900 euros per kW to 450 euros/kW or less in the period after 2030”. Leading Chinese manufacturers, however, are already supplying equipment at $200/kW – as revealed in BloombergNEF’s 2019 Economics of Hydrogen Production from Renewable Power (client links here web | terminal). What is going on?
Chinese producers benefit from cheaper raw materials and labor, but they have also focused on the more established alkaline electrolyzers. These, conventional wisdom used to say, don’t like to ramp up and down to follow peaks and troughs in electricity demand and supply. As a result, the EU, expecting electrolyzers to be powered by variable renewable energy, focused development for the better part of a decade on solid oxide and proton exchange membrane (PEM) technologies, the latter able to ramp up and down within tenths of a second. These are more expensive than alkaline and are still far behind in scale – and it turns out alkaline electrolyzers can also be designed to load-follow, albeit a little slower. Can the EU catch up with Chinese electrolyzer costs? Probably not. Can it hit its cost targets without doing so? Probably not.
The very idea of using surplus renewable energy to generate hydrogen will turn out to be, on the whole, a mirage. It might make sense for an island grid, but not when it comes to a highly connected, continent-scale energy system. Here, the only thing that matters is to produce the cheapest green hydrogen possible, or you will be outcompeted by producers using the lowest-cost renewable electricity at high capacity factors, delivering via pipeline.
Imagine, for the sake of argument, a future grid with such huge penetration of variable renewable generation that curtailment reaches 33%. It could happen: wind or solar power with levelized costs of $20/MWh could still profitably be sold at $30/MWh, half the price of any alternatives. However, it would be would be entirely uneconomic to run an electrolyzer on the curtailed power alone, even if it were free, because it would all be dumped onto the grid within a relatively limited number of hours each year.
What green hydrogen producers will do instead is, yes, take advantage of whatever free surplus power is available, but supplement it by buying renewable power at normal wholesale prices the rest of the time. By doing so, in the best locations in the world – the ones I called renewable energy superpowers in my piece at the end of last year – they will be able to achieve a combination of high capacity factors and low clean power prices. While individual renewables plants may operate at a capacity factor of between 15% (low end for solar) to 60% (high end for offshore wind), if they are co-located or linked via power lines, it will be possible to get very high capacity factors – perhaps 80% or more, given some judicious investment in ever-cheaper batteries – and produce very cheap green hydrogen.
By 2050, therefore, extrapolating long-standing trends in renewable power and electrolyzer costs, combining them with the likely ability to run at high capacity factors and with reductions in costs of capital as financiers master the technology and market risk, BloombergNEF estimates that green hydrogen will be available at between $0.8 and $1.0 per kilo. I would not be surprised to see it go below that.
Parts of Southern Europe boast the combination of low-cost solar and wind resources required to get there, in competition with renewable energy superpower regions of Australia, Morocco, the Gulf, Mexico, Chile, Brazil, southern U.S., China, India and so on, Northern Europe does not. Which is why Germany is planning to import both clean electricity and green hydrogen, and why Hydrogen Europe, the umbrella organization for hydrogen interests in that continent, has proposed a “2×40” plan – the EU strategy’s 40GW of electrolyzer capacity by 2030, plus 40GW more in North Africa and Ukraine.
It is also worth noting that green hydrogen based purely on offshore wind is unlikely to be competitive with production based on the combination of super-cheap solar and onshore wind. Even by 2050 it is likely to be at a levelized cost disadvantage, and offshore wind shows strong self-correlation, which means it won’t be able to deliver very high capacity factors unless hybridized with solar power.
What’s wrong with blue hydrogen?
Green hydrogen based on renewable energy is not the only possible source of zero-carbon hydrogen.
First off, there is ‘blue’ hydrogen – produced by reforming natural gas or gasifying coal, but with the CO2 emissions captured. The EU Hydrogen Strategy estimates the current cost of producing blue hydrogen at 2 euros per kg, but by 2030 there is no reason why it could not be produced at least as cheaply as the EU’s 2030 target for green hydrogen, at 1.50 euros per kg, given the extraordinary strengths in energy, carbon capture and chemicals of European industry. In the longer term, over a multi-decade period, because it benefits from a slower learning rate, blue hydrogen can be expected to fall behind green in the cost stakes.
The EU Hydrogen Strategy admits that blue hydrogen will play a role – as noted, it highlights 11 billion euros for retrofitting half of existing fossil-based plants. Separately, however, the EU has indicated it will not provide any of the required funds itself: that will be down to member states or the private sector. Germany is particularly hostile to blue hydrogen, declaring almost as an article of faith in its national strategy that “only hydrogen produced on the basis of renewable energies (‘green’ hydrogen) is sustainable in the long term”.
Two legitimate reasons are often cited for reservations about blue hydrogen, and one poor one. The first real concern is that generally only 90% of CO2 is captured; this can be increased, but only at additional cost – though you might think it sensible to devote some of the vast funds earmarked for electrolysis research to improving the process. The second real concern is the issue of fugitive emissions: wherever natural gas is extracted there is some loss to the atmosphere, and methane (the main constituent of natural gas) is a powerful greenhouse gas. Again, it seems odd to regard this as an intractable problem, particularly given the rapid strides in tracking leaks via satellite. Now that miscreants can so easily be caught, the oil and gas industry is rallying around efforts to choke off fugitive emissions. The third, poor reason for not supporting blue hydrogen is simply because its production will benefit oil and gas companies, something many activists find unacceptable. The idea that we can somehow effect a transition to net-zero energy without the involvement of the world’s largest energy companies is, of course, absurd.
Any green swans?
Truly zero-carbon hydrogen can also be produced by a pyrolysis process, by which natural gas is passed through a molten alkali or metal, producing carbon black as a by-product. This “turquoise hydrogen” process can be powered by clean energy and looks economically promising, though – like blue hydrogen – it is frowned on by purists because it does not eliminate the risk of fugitive emissions or the involvement of oil and gas companies.
Then there is nuclear power, also frowned on by purists, but potentially attractive in generating zero-carbon hydrogen nevertheless. I have written before on the ghastly economics of new nuclear power stations based on current-generation technology, but small modular reactors could perhaps turn this around. Nuclear power has two great advantages over renewables for industrial processes like hydrogen production: it likes to run 24/7, and it produces waste heat which can be used. Whether those advantages are enough to counter the likely persistent levelized cost disadvantage, as well as the risks that inevitably accompany nuclear power, remains to be tested.
One final zero-carbon hydrogen concept that is worth mentioning is the elegant “thermal hydrogen” approach proposed by Jared Moore. In most plans for hydrogen electrolysis, no value is ascribed to the by-product, oxygen. In this proposed system, an oxy-fuel mix is used to drive an Allam Cycle generator, producing power and heat and delivering a pure stream – after removal of water – of CO2, ready for use or storage, at far lower cost than separating CO2 from normal exhaust gas. Some proportion of the heat and electricity is in turn used for high-temperature electrolysis, generating hydrogen and the oxygen consumed in the process. I told you it was elegant – as long as the resulting hydrogen replaces the use of a fossil fuel.
Cost of transport
Distributing cheap green hydrogen around Europe, the other part of the EU Hydrogen Strategy, should add between 7 and 50 euro cents per kilogram, according to BloombergNEF’s most recent report on the economics of hydrogen transport and delivery (client links here web | terminal) – as long as it is done via a pipeline.
Start using trucks or ships – whether the hydrogen is gaseous, liquefied or combined with an organic vector like toluene – and the costs quickly soar to anything from 60 cents to $7 per kg of hydrogen, depending on volume and distance, rendering it uncompetitive against ‘brown’ hydrogen and against the target price of 1.1 to 2.4 euros per kg.
Where there are no pipelines, the most viable vector for bulk transportation of hydrogen over long distance looks like being ammonia. But even then, the total transportation cost, including conversion, storage and so on, could be three times the cost of producing the hydrogen in the first place.
In September, Saudi Aramco and Japan’s Institute of Energy Economics, in collaboration with Sabic, sent the world’s first shipment of blue ammonia – that is ammonia made from natural gas but with the CO2 captured and stored – from Saudi Arabia to Japan to be used for generating zero-carbon power. It was only a pilot, to test the different elements of the value chain, but it is hard to see this work economically. And, of course, if the captured CO2 is used for enhanced oil recovery, the whole thing becomes a self-licking ice cream in terms of emissions.
What can we learn from all this, other than the fact that a massive wall of money is about to break over the hydrogen supply side?
Will green hydrogen be competitive with blue hydrogen by 2030? Probably. Competitive with brown hydrogen in Europe? Not quite, but close. Will there be vast amounts of green hydrogen, cheaper than brown hydrogen by 2050? Absolutely, though in Europe much of it is likely to be imported from places where it can be made more cheaply. And even if it is made in Europe, it will probably not be made using European electrolyzers, and certainly not using exclusively European renewable power.
Next week, in Part II, we’ll be looking at the demand side. Where on earth is all this cheap green hydrogen meant to go?
Stay tuned, hydrogen fans!
Michael Liebreich is founder and senior contributor to BloombergNEF. He is on the international advisory board of Equinor.