Green hydrogen has a seductive appeal. Done right, this zero-emissions energy source has the potential to penetrate many corners of our economies and be instrumental in the fight against climate change.
It can be transported over long distances, stored for lengthy periods and some existing fossil fuel infrastructure such as gas pipelines can be adapted to handle it. These attributes help explain the rush of excitement around the gas, also referred to as “clean” or renewable hydrogen.
Unlike its most common form of production, known as grey hydrogen, which is extracted from natural gases in a carbon-intensive process, the green version relies on renewable energy, such as solar or wind power, to split water into hydrogen and oxygen. It creates no carbon during a production process called electrolysis and emits only water when it is burnt.
Some 1,000 new projects globally have been announced to date, requiring total investment of $320bn, according to the Hydrogen Council, an industry body whose members include oil companies such as BP and carmakers like BMW Group. Would-be developers, however, have only committed $29bn so far.
But spending not only lags behind announcements, it lags behind requirements. Lex calculates that a net zero energy system might require 500mn tonnes of hydrogen annually, which would entail some $20tn of investment by 2050. That means we are only about 0.15 per cent of the way there.
The industry has so far been held up by its high costs. Slowly, however, governments, particularly in the US and Europe, are now moving to support it and kick-start the ascent of hydrogen.
“Green hydrogen will be essential to reach net zero,” says Lord Adair Turner, chair of the Energy Transition Commission think-tank. “But we are still at the starting gate. Without strong policy support, it will not scale up in the timescale required.”
Sizing up the hydrogen economy
The first question that anyone considering hydrogen as a fuel asks is: why is it needed at all?
Yes, climate change means that we need to stop using fossil fuels, which account for 80 per cent of global energy usage. The answer, critics say, is to use renewable and low-carbon electricity to power electric vehicles or heat pumps directly. That would be a better bet than going through the rigmarole of using renewable electricity to split water and generate “green” hydrogen, that can then be burnt in boilers.
For the most part, that is an accurate representation of the energy transition. Hydrogen is abysmally inefficient. Consider EVs, for example. Even when factoring in the 5 per cent lost in transport and 10 per cent as batteries charge and discharge, EVs can be up to 80 per cent efficient. In hydrogen vehicles, between 30 and 40 per cent of the starting renewable electricity is lost in making the fuel and a further 40 per cent in the fuel cell.
“Hydrogen’s poor efficiency means that it holds a compelling case as a decarbonisation solution only where direct electrification is not feasible — in industrial processes that require a chemical reaction, for instance,” says Mark Meldrum, of climate consultancy Systemiq.
That still leaves a substantial slice of the energy pie. The International Energy Agency reckons that by 2050 we will use electricity for about 50 per cent of the energy we need, up from 20 per cent today. An estimate from the ETC predicts almost 70 per cent. By those calculations, that leaves 30-50 per cent for other fuels to play for.
Industries that use non-renewable hydrogen at present, such as fertiliser production, are sure bets for the green version of the fuel — around 50mn tonnes a year by 2050, according to the ETC.
Another big opportunity for hydrogen is in industries that use fossil fuels in their processes today. Hydrogen can be swapped in as a feedstock in chemicals production, for example, and can replace the coal used to heat steelmaking furnaces. The ETC pencils in 120mn tonnes of hydrogen a year for such purposes.
Long-range transport is also likely to be powered by hydrogen in the future. That is because aircraft, ships and lorries on long journeys need to store a lot of energy on board and batteries are heavy compared with the amount of energy they can hold.
Hydrogen, by contrast, is very light. Its issue is that it is difficult to compress and would require unfeasibly large tanks in aircraft and ships. The solution is to combine the hydrogen with nitrogen to form ammonia, which is more energy dense and can be burnt in the internal combustion engines of ships if found to be safe.
For long-haul air travel, the idea is to make “green” jet fuel by taking renewable hydrogen and adding in carbon captured from the air. This synthetic hydrocarbon would emit CO₂ when burnt, but because the carbon comes from the atmosphere in the first place, it would be carbon neutral over its lifecycle.
Lastly, green hydrogen will be required for energy storage. Batteries do a good job of that, but can go flat if left unused. Long-duration storage — storing summer sun for winter heat, for example — is more difficult. Using surplus renewable energy to make hydrogen and then using that to make electricity is convoluted, but there are not many alternatives for seasonal storage. Pulling this off, however, will depend on how other storage systems develop.
Add all of these use cases together and you get the 500mn tonnes of hydrogen needed by 2050 — accounting for more than 10 per cent of the energy mix. That is a ballpark estimate, of course.
The cost of being carbon-free
Making and transporting 500mn tonnes of hydrogen will obviously require a lot of capital and exact numbers are hard to come by as the cost of technologies is expected to fall over time.
For an idea of the size of this capital project, however, it is worth dividing hydrogen capex three ways; the cost of renewable electricity needed to make the gas, expenditure on electrolysers used to split water into oxygen and hydrogen and, lastly, the infrastructure — pipelines, ships and storage sites — required to take the hydrogen where it is needed.
Generating this amount of hydrogen will need almost 25,000TWh of renewable electricity a year, about 100 times the UK’s current electricity demand. On the assumption that solar panels and wind turbines are placed in sunny and blustery areas, we would need 10TW of infrastructure, Lex calculates. At an average cost of $800 per kW, the investment required would be about $8tn.
The second bucket is the electrolysers. Today, inflation has pushed up the price to about $1,500 per kW but it could be as low as $250 per kW by 2050. Using a midpoint of $875 per kW implies a capex requirement of $7tn to achieve the desired goal.
When it comes to infrastructure, the expenditure needed for transport and storage depends on the specific technology. The cheapest option involves pushing hydrogen through repurposed gas pipelines and using repurposed storage sites. Supply chains involving shipping hydrogen transformed into ammonia will be more expensive. Overall, capex may hover at about $5tn.
By these calculations, the total outlay would be around $20tn — a hefty number especially if one considers that investment might end up being concentrated in the 2030s and early 2040s. To put it in context, the energy transition as a whole is expected to require about $100tn of capital, according to the ETC.
These large sums point to the scale of the challenge. But the real question for investors, policymakers and consumers is how much the hydrogen generated through this investment might cost — in absolute terms and compared with the fossil fuels that it would replace. The difference is the amount that would need to be filled by subsidies.
A simple way of calculating the average cost of hydrogen is to divide the capex by how much hydrogen the kit it buys might produce over its 20-year lifespan. By that reckoning, the average cost for the hydrogen would work out at about $62 per MWh.
This really is a rough number. Both investments and energy flows stretch into the future and do not allow for the time value of money. The calculation assumes no operating costs and, most importantly, does not include any return for those putting up the capital.
The second leg of this calculation is no easier considering the price of the fossil fuels hydrogen is set to replace will continue to gyrate. For example, using the high natural gas prices of last year, hydrogen would already be cheaper than fossil fuels with no subsidies required.
But gas-crunch prices may be a poor guide. Assuming that natural gas will stabilise at a more reasonable $50 per MWh, that would suggest every unit of hydrogen needs a $12 per MWh subsidy on average. Multiplying that for the whole of the hydrogen produced, we are looking at about $4tn in subsidies.
Not all of this would need to come in the form of additional support. In Europe and the UK, carbon pricing is already in place. The EU emissions trading system means that those companies using natural gas already pay an additional $20 per MWh for the CO₂ they emit — and that number is expected to rise. It follows that, in these regions, hydrogen will be cheaper than natural gas and the cost of carbon emissions combined.
Such back-of-the-envelope maths would suggest the industry should be able to get going without subsidies. Yet that is not the case. As a report by the Hydrogen Council makes clear, the new projects being announced are not matched by a commitment of capital. If hydrogen is both necessary and, over the next 30 years, not much more expensive than fossil fuels, why are things not moving faster?
A serious push for subsidies
Hydrogen’s problem, today, is threefold.
Renewables are not being built at the rate needed to decarbonise electricity, let alone make hydrogen, and their cost has ticked up. Moreover, the few hydrogen projects that do exist are small-scale and piecemeal, representing less than 1 per cent of total hydrogen production over the past three years. That makes the cost of infrastructure, which becomes tolerable when there is a bigger demand, high on a per-unit basis.
As a result, the hydrogen produced today is still very expensive. A look at Platt’s hydrogen price wall, which shows the cost of hydrogen produced in different regions, suggests that, while some projects manage to come in at $50-$100 per MWh, the cheapest hydrogen in Europe today costs more than $150 per MWh without transport and storage. European natural gas meanwhile is below $32 per MWh.
This means that a serious subsidy push is needed if hydrogen is going to reach the scale required to break even with existing energy sources.
Europe and the US are both trying to push the market to its tipping point, in radically different ways. But neither approach is, so far, sufficient to get the ball rolling.
The US Inflation Reduction Act throws money at the problem. It offers producers of green hydrogen a tax credit of up to $3 per kg. At this level, the fuel is a bit more expensive than natural gas, but much cheaper than so-called grey hydrogen. For industries already using grey hydrogen, such as refining or ammonia, the switch is a no-brainer.
“The IRA has kick-started the switch from grey hydrogen in North America,” says Markus Wilthaner, partner at McKinsey who co-leads its global hydrogen team. “The US now accounts for 70 per cent of committed clean hydrogen production globally.”
But the trouble with the IRA is that the production credits only last for 10 years. That makes it harder for businesses to have confidence about what the market will look like long-term, which is important where upfront investment is needed. Steelmakers phasing out coal need to build whole new plants to support hydrogen, for example. And a sensible hydrogen system also needs centralised production facilities and transport and storage infrastructure.
Europe is taking a diametrically opposed approach. Rather than throwing money at production, it is making some hydrogen consumption compulsory. It wants 42 per cent of the hydrogen used in industry to be renewable by 2030. The existing obligations add up to a couple of million tonnes of demand by 2030. Hydrogen Europe, an industry body, expects a further 1.5mn tonnes of demand for steelmaking and heating, leading to a total of just over 4mn tonnes — a sizeable quantity, although still falling short of the EU’s stated 20mn tonne ambition.
The trouble here is that the obligation imposes additional costs on industry by forcing them to switch from cheaper natural gas to more expensive hydrogen. On top of that, the EU has applied a restrictive definition of what constitutes renewable hydrogen. Plugging an electrolyser into the mains will not suffice, because grid electricity may be generated by nuclear or fossil fuels. To ensure the hydrogen is truly “green”, it must be produced off-grid during the limited periods when there is an excess of renewable electricity. Spreading production across fewer operating hours further raises the cost of hydrogen at the beginning of its development.
There is encouraging news on the horizon. That so-called green premium will narrow as the cost of emitting carbon rises and covers more industries. The EU is also seeking to cover the gap with its own subsidies and has recently launched a new European Hydrogen Bank, which will run auctions to finance the most competitive hydrogen production.
On paper, all bases are covered, but this cumbersome and uncertain process makes it hard for developers to borrow on the promise of future subsidies.
But with hydrogen, while it is easy to focus on the negatives of the gargantuan task ahead, progress has been made. For one thing, it has a widely accepted role in the energy transition after once being largely dismissed.
There is now an ecosystem of companies ramping up to serve this industry. Electrolyser manufacturers exist. German industrial group Thyssenkrupp is set to list its own hydrogen equipment maker, Nucera, for a reported $4bn valuation.
Tentative policy is coming together. If Europe and the US deliver on their hydrogen pledges, that will help reduce costs for others. Hydrogen has been late to the low-carbon party, but given the right incentives, it has potential to fuel the revelry.
Data visualisation by Graham Parrish and Bob Haslett