OK, in fairness to everyone who has ever built a hamster wheel to generate electricity, it probably wasn’t about the electricity. It may have been about the merit badge, or your science grade, or just to satisfy your curiosity. But it almost certainly wasn’t about devising a way to make electricity that would really be useful in the world.
Sadly, some proponents of hydrogen are not similarly retrained in their aspirations. Reporters and politicians have been flocking to tiny Hannibal, Ohio on the banks of the Ohio River to lavish praise on the Long Ridge Power Station, where electricity is being made from hydrogen . . . kind of. I say, “kind of”, because initially Long Ridge’s combined cycle turbine is being powered by natural gas with just a little bit of hydrogen blended in – about 5% so far. But the hope is that, by 2030, with adjustments and refinements, they’ll be able to power the process with hydrogen alone. What’s not clear is why they hope to do so.
Sure, generating electricity with hydrogen avoids the release of climate polluting greenhouse gasses that result from natural gas-fired generation. At least that’s true if the hydrogen you’re using isn’t made in a way that merely accelerates the release of emissions during the manufacturing process. We’ll come back to that problem later. For the moment, the question is, even if you’re generating electricity with really clean hydrogen, is doing so worth the effort and, more importantly, the cost?
One perspective on the question is provided by Lazard, the consultancy that over the past 17 years has built its reputation on its widely accepted assessments of the comparative cost of power generating technologies. In Lazard’s recently released “Levelized Cost of Energy 2023”, we find that, whereas the mid-point cost of generating electricity with unabated natural gas is $62/MWh, the cost of doing so with hydrogen is $116/MWh. Other technologies, including renewables and natural gas-fired generation regularly dip down to the $20-$40/MWh range.
Except, the $116/MWh figure for hydrogen-fired generation isn’t exactly true because the scenario Lazard considers is one in which hydrogen makes up just 20% of the fuel mix with natural gas accounting for the rest. In other words, replacing just 20% of the natural gas feedstock with hydrogen causes the price of electricity to jump by a frightening 87%. And worse, emissions are reduced by just 7%.
This is why making electricity with hydrogen, even in a minor supporting role, is stupidly expensive. It’s also stupidly expensive as a means of reducing greenhouse gas emissions. Renewable resources combined with batteries, and other technologies reduce emissions at far less cost. Even gas-fired power plants retrofitted with carbon capture technology, which is itself stupidly expensive, are a comparative bargain. That’s why there are not now and never have been any hydrogen-burning power plants, nor are there any producers of clean hydrogen operating outside of pilot projects and laboratories.
Of course, the recently enacted Inflation Reduction Act, which provides generous subsidies for clean hydrogen, seeks to change all that. But will it? Can the IRA subsidies make a stupidly expensive technology not stupid, although perhaps still very expensive?
The goal of the US Department of Energy’s Hydrogen Shot project is to reduce the cost of clean hydrogen to $1/kg by the end of the decade. That represents a 29% reduction from the $1.40/kg figure that Lazard used in its modeling. And it might be achievable, both for blue hydrogen, derived from natural gas, and for green hydrogen produced by electronically splitting water molecules. But, even if that goal is achieved, will it be enough to make hydrogen price competitive in producing electricity? Honestly? No.
The problem is that the goal price of $1/kg for hydrogen equates to $8.71 per million British Thermal Units (mmbtu). But the price of natural gas, which is currently in the low $2s/mmbtu, is only expected to rise to about $3.50/mmbtu going well into the future. The reason for the price differential is that hydrogen has only about 35% the energy content as methane.
Consequently, hydrogen is two and a half to four times more expensive than natural gas as a power generating feedstock. In fact, for hydrogen to be cost-competitive with natural gas, its cost would have to be $.40/kg, which is less than half of the DOE goal. In the case of blue hydrogen, the cost just to acquire the natural gas and methane from which you hope to extract hydrogen is likely to be greater than $.40/kg. Meanwhile, renewable resources, including wind and solar are becoming even less expensive than both hydrogen and natural gas.
We also haven’t considered additional complicating factors. In addition to needing to be manufactured, hydrogen would have to be transported and stored, processes which would likely add $.15-$.30/kg to the cost. But perhaps a bigger problem with hydrogen is that its credentials as a “clean” resource are pretty shaky.
The manufacture of blue hydrogen, derived from methane, involves a number of steps, starting with the extraction of natural gas, which entail leakage of methane in non-trivial amounts. Even green hydrogen, derived from water by means of electrolysis, is only green if its production is powered by zero-emission renewable resources. If, on the other hand, power is supplied in part or entirely by electricity from the grid, then the resulting hydrogen is just as dirty as the grid itself. Finally, hydrogen is also susceptible to leaking and, while it is not a gas that traps heat in the atmosphere, it interacts with and compounds the effects of other gasses that do.
That’s why hydrogen, while less damaging than fossil fuels, is generally seen as a last resort for decarbonizing and is applicable only when all other options such as electrification or energy efficiency have either been exhausted or are not practical.
In short, hydrogen is never going to serve as a “base load” resource for generating electricity, nor is it likely to be sensible or affordable as a “peaking” resource or even as a means of balancing load in an increasingly renewables-based power system. The only energy system function in which hydrogen may play a role is as a back-up for periods of many days or weeks during which renewable resources may become unavailable and short-term storage resources are exhausted. And even that role will be up for a fight with other rapidly emerging long-duration storage technologies such as iron-air batteries, compressed air, pumped hydro, and demand response programs, which will become increasingly effective as reliance on renewable energy expands and homes and appliances become more deeply integrated with the grid.
So, yes, you can generate electricity with hydrogen just like you can build a car that runs on coal. You might do it, but not if you care about functionality, cost, and environmental consequences.