Beyond the Hydrogen Mirage: A Candid Conversation with Joe Romm – CleanTechnica

---

---

---

Recently, I had the opportunity to sit down again with Dr. Joseph Romm to discuss his then about to be released book, The Hype About Hydrogen, available now on Amazon. This is the second half of our conversation, lightly edited.

Michael Barnard [MB]: Welcome back to Redefining Energy — Tech. I’m your host, Michael Barnard. My guest today is Dr. Joseph Romm, senior research fellow at the University of Pennsylvania center for Science, Sustainability and the Media, working with Michael Mann. His work focuses on the sustainability, scalability and scientific underpinnings of major climate solutions. The 20th anniversary version of his book The Hype about Hydrogen dropped on Earth Day, and we are here to talk about it. This is the second half of our conversation.

Joe Romm [JR]: Let’s be honest. Part of the resurgence of interest in oil and gas companies is because they’re the ones who know how to use hydrogen. They’re the ones who know how to move it around. I’ve always felt the reason they pushed it so hard is that they never believed green hydrogen would be cost-effective. They assumed people would eventually come running back to them to make it from methane—with promises to capture some carbon along the way.

And they were right. Now we’re seeing all these apologists saying, “Okay, well, green hydrogen may not be cost-effective for a while, so in the meantime, we’ll make it from methane. We promise we’ll capture the carbon.” But as we’ve seen with regular carbon capture, almost no one delivers. Everyone claims they’ll hit 90 or 95 percent, but hardly anyone captures anything close to that.

[MB]: I used to point to Sleipner’s North Sea facility as probably the best-case scenario. And even then, it was still a bit odd. For those who don’t know, it’s an offshore natural gas platform. They extract gas from beneath the seabed, but it contains too much carbon dioxide—about 8%, if I remember correctly. So they separate out the CO₂ and get massive tax credits from the Norwegian government to inject it back underground. And they actually do it.

I used to think, at least it was Norwegian engineering—efficient, reliable. But then last year we found out they had been underperforming for five years. They’d pumped far less CO₂ underground than they claimed. Even the Norwegians can’t get it right.

[JR]: I have a section in the book on Sleipner because there’s a common misconception in this country about carbon capture and storage. The people pushing it are mostly oil companies, and most of the time they use the captured CO₂ to extract more oil from the ground. Occidental’s acquisition of Carbon Engineering was clearly for that purpose. I hope we all understand that capturing CO₂—from a power plant or from the air—and then using it to extract more oil is not a sustainable solution. It doesn’t solve climate change.

The reality is that effective carbon storage requires a lot of money for monitoring and verification. Sleipner is a good example: the CO₂ is injected underwater, beneath the ocean floor, into a formation they claim is geologically sealed. But to know it’s truly sealed—and that the CO₂ isn’t migrating—you need continuous, expensive monitoring. CO₂ spreads. It can find old cracks you didn’t know were there, or create new ones over time.

In the book, I discuss two case studies: Sleipner and the In Salah project in Algeria. In both cases, long-term monitoring revealed that the CO₂ didn’t just stay where they put it. It moved. This matters. Especially now, when the literature is clear—and we saw this emphasized at COP 29 in Azerbaijan with two major studies—that if you want to genuinely displace fossil fuel emissions, you need to store CO₂ permanently. CO₂ stays in the atmosphere for a long time. So if you’re going to remove it, you need to lock it away for centuries. If it leaks in 100 years, you haven’t really solved anything. You’ve just delayed the problem.

This is why measurement, reporting, and verification (MRV) are so important—but no one wants to pay for them in this country. Oil companies say, “Give us the CO₂, pay us a tax credit, and trust us—we’ll bury it and it won’t come back.” But they don’t want liability. They want immunity in case something goes wrong. If a CO₂ plume resurfaces in a decade and harms people, they don’t want to be held responsible.

That’s the definition of a moral hazard. No accountability, no real incentive to get it right. If you truly want to do carbon capture and storage responsibly, you have to invest in long-term monitoring and verification. Otherwise, it’s just another illusion.

[MB]: Well, the good news about Northern Lights—the Norwegian carbon storage project—is that the ships are finally going to start moving this year. And I say it’s good news not because it makes any real sense, but because it will soon become painfully obvious to everyone that it doesn’t.

Norway paid for roughly 80% of the capital cost using money from its sovereign wealth fund, so it’s already pulled a huge amount of value out of fossil fuels to fund this. On top of that, they’re subsidizing BECCS plants to send CO₂ to Northern Lights. The only facility that even approaches fiscal sanity is Yara’s dockside ammonia plant, which produces a relatively pure stream of CO₂.

But even then, Yara has to buffer, compress, and liquefy that CO₂ at great expense, while waiting for one of the Northern Lights ships to arrive. Then the ship travels 700 kilometers round-trip to the injection site. And that site, while technically on land, is reached via a 100-kilometer undersea pipeline that dives 2 kilometers down to a storage formation supposedly sealed by impermeable shale that will hold the gas forever.

It’s an astonishing amount of engineering and money. They’ve gone so far as to equip the ships with Flettner rotors to gain an extra 3% efficiency. They’re also using air lubrication systems under the hulls, slow steaming—tactics we don’t typically apply on standard cargo vessels barring the slow steaming—all to reduce the CO₂ emissions from the maritime fuel powering the ships. When the illusion breaks and people start adding up the real costs, it’s going to be eye-opening.

[JR]: It’s important for people to understand that when you capture CO₂, it’s a gas—but to store it, you need to convert it into supercritical CO₂. That’s a state where it’s neither a true gas nor a true liquid. It has about half the density of water, and it’s kept at around 1,000 pounds per square inch. In that state, it behaves as a solvent—supercritical CO₂ is actually used in industry for exactly that purpose.

So when you inject it underground, you’re injecting a high-pressure solvent into geological formations. This isn’t a simple “fire and forget” process. It requires serious engineering, long-term oversight, and a deep understanding of subsurface behavior. The first time I saw the equation for this, it really hit me—this is far more complex and risky than most people realize.

Vaclav Smil did a calculation where he pointed out that if you want to capture and move around 3 billion tons of CO₂—whether it’s from power plants or any other source—you’re dealing with a logistical burden equivalent in volume to more than 90 million barrels of oil per day. That’s roughly the same scale as the entire global oil production and delivery system, which took a century to build. If you think you’re going to recreate that kind of infrastructure in a generation, you might want to think again.

And that’s just for 3 billion tons. Total global greenhouse gas emissions are 50 billion tons annually. Even if you’re only aiming for a 6% solution, you’re still talking about building an entire global petroleum-scale infrastructure just to bury waste—and it better stay buried. If it leaks out over the next hundred years, you haven’t solved the problem.
The point isn’t that carbon capture or hydrogen are completely worthless. The point, as I emphasize in the book, is that we need to focus on technologies that are scalable now and capable of driving emissions down rapidly. We’ve been increasing emissions for over 30 years. We’re at COP29 now. In a TEDx talk, I pointed out that there have been over 30 annual global climate meetings—including one we missed during COVID—and emissions have kept rising the entire time.

So unless we start cutting emissions sharply and soon, we’re in serious trouble. That’s what I posted about recently, and that’s what the financial sector seems to be acknowledging quietly. Instead of screaming for immediate action, they’re hedging—investing in air conditioning, insurance, and adaptation. That tells you something.

The real trick is to spend as much money as possible on the things that are likely to work—and as little as possible on things that probably won’t. I’m a physicist, and I ran a billion-dollar R&D office. I’d never say a problem can never be solved, but the thing about hydrogen is, it’s not solving just one problem.

People talk about “gold hydrogen”—naturally occurring hydrogen underground—as if just discovering it solves everything. But, as I argue in the book, there are at least five major challenges. Twenty years ago, I used to say you needed three or four miracles to make hydrogen viable. And usually, it only takes one fatal flaw to kill an idea. But over time, I realized something deeper: if you’re willing to believe in one miracle, you’ll believe in four. It’s like infinity—whether it’s one or four, it’s still an endless leap of faith.

So, saying “we just need to make green hydrogen” isn’t enough. That still doesn’t get hydrogen to end users. It still leaks. It’s still one of the most dangerous substances known to humankind. And no one wants to talk about the safety issues. So no, I’m not saying we should abandon all hydrogen. We will, at some point, need to replace the dirty hydrogen we currently produce. Right now, we make about 100 million tons of it a year, and production keeps growing by about 5% annually.

But hydrogen accounts for only about 2% of global greenhouse gas emissions. So yes, it’s important—but not urgent. There are hard-to-decarbonize sectors, like international air travel, that contribute 2–3% of global emissions. We all agree they’re difficult and expensive to fix right now. So maybe let’s not focus on them first.

What we need is the kind of cost-curve thinking that McKinsey and others used to do. Let’s go after the relatively easy 80%. Let’s focus R&D on the difficult 20%, like hydrogen, without prematurely scaling up expensive, risky technologies for marginal gains. We need to stop chasing shiny distractions and focus on what actually gets emissions down—fast.

[MB]: The thing about hydrogen is that around 40% of global production is used for refining oil—and that 40% is overwhelmingly tied to heavy, high-sulfur crude from places like Alberta, Mexico, and Venezuela. I actually did the math and the workup on this, and folks at Schlumberger looked at it and said, “Yeah, that checks out.” And they would know.

It works out to about 7.7 kilograms of hydrogen per barrel for Alberta’s crude. By contrast, for light, sweet crude—like some of the best from Brent or Saudi Arabia—it’s only about 1.2 kilograms per barrel. So when you look at that, it becomes clear: if hydrogen becomes more expensive, and if oil demand declines, hydrogen demand is going to decline as well.

The same logic applies to ammonia-based fertilizers. If hydrogen becomes more costly, we’ll stop overusing them. Alternatives like agrigenetics and precision agriculture become more competitive, and in many cases, more cost-effective. There’s a real economic argument there.

I had a conversation recently with Michael Liebreich where he admitted he’d gotten the price point for hydrogen wrong when doing the first version of the hydrogen ladder. He had other reasons for thinking hydrogen wouldn’t be a big deal, but he said the hydrogen ladder would have looked different if he’d had the right price assumptions. I got lucky—I did the cost workups and the modeling before I put out my hydrogen projections. I keep saying this: I don’t think I’m right. I just think I’m less wrong than most. And in this case, I got lucky. I could have been just as embarrassed as a lot of other people are today.

But there’s something we haven’t really talked about: hydrogen leakage. There are two major concerns. First, if hydrogen accumulates in an enclosed space and ignites, it’s extremely dangerous. But the second issue is more subtle and often ignored.

You’ve smelled natural gas before—it stinks. That’s because we add odorants so that leaks can be detected and people can evacuate. But you can’t do that with hydrogen. The odorants that work for other gases destroy fuel cells. So if you want to use hydrogen for both electricity and heating, you’d need two entirely separate distribution systems: one for clean hydrogen feeding fuel cells, and another with odorants for safety in buildings.

Oddly, this seems to be completely overlooked by many hydrogen proponents. I find that strange. Do they just not know? Are they refusing to deal with it? Or is this just one more miracle they assume will somehow be solved later

[JR]: The safety issue around hydrogen is often casually hand-waved away by people who say, “Well, it’s used safely.” And sure, that’s true—under very strict conditions. But let’s look at what countries like India actually do to use it safely. Their regulations require a 100-foot setback between any building that produces or stores hydrogen and the nearest structure. That’s because the fire risk is so high. You also need massive ventilation in any enclosed space where hydrogen might accumulate. Otherwise, you risk a gas bubble forming—and hydrogen, as we know, burns.

But it’s worse than that. Hydrogen is odorless, and as you pointed out, it burns invisibly. That’s why, in NASA safety handbooks, you’ll find guidance like this: if you’re entering a room where there might be a hydrogen fire, carry a broom. Because the broom will ignite before you do. That’s not a joke—it’s a workaround for the fact that hydrogen flame detectors aren’t very good. Maybe people are working on better sensors, but hydrogen is the tiniest molecule in the universe. It leaks through seals, gaskets, joints—materials that easily contain other gases.

And that leakiness matters. In any facility where hydrogen might be present, workers have to wear static-free clothing. Why? Because hydrogen has one-twentieth the ignition energy of gasoline. It’s so combustible that a static discharge—or even a lightning storm miles away—could set it off. It also burns at a much higher velocity than natural gas, increasing the blast risk.

There’s another crucial difference. Natural gas only ignites in air at a fairly narrow concentration—something like 5% to 15%. Hydrogen, on the other hand, can ignite in air across a massive range—from roughly 4% all the way up to 75% or 80%, depending on conditions. That means it’s far more likely to find an ignition point.

The bottom line is, you have to treat hydrogen with extreme care. And that kind of care costs money—money people don’t want to spend. That’s also one reason it makes little sense to put hydrogen anywhere near a nuclear reactor. In fact, nuclear engineers have studied hydrogen in detail because of what happened at Three Mile Island. During that disaster, a hydrogen bubble formed inside the reactor containment vessel. It shocked the public. No one had expected it, and there was real fear it could explode and breach the containment structure.

So yes, hydrogen can be used safely—but only with serious precautions. And most of those precautions make it too complex and costly for broad, distributed use.

[MB]: That’s actually what happened at Fukushima—it was hydrogen that exploded. The reactors generated hydrogen, which accumulated and eventually ignited, causing the blasts that destroyed parts of the facility.

But I’ll point out something interesting: hydrogen is also used in a very controlled way at nuclear plants. It’s used to lubricate the bearings on large turbines because it’s an excellent coolant and lubricant in those high-speed environments. There’s actually one small-scale nuclear-hydrogen use case that I thought made a lot of sense. A plant installed a small electrolyzer onsite specifically to replace the gray hydrogen they had previously trucked in for turbine lubrication. Instead, they used a tiny amount of auxiliary “vampire” power—around 0.003% of total output—to produce all the hydrogen they needed.

That’s a genuinely good use case. But it was small, and crucially, it wasn’t about using hydrogen as a fuel. That’s an important distinction I want to emphasize: everything we’re talking about here—hydrogen’s safety, leakage, infrastructure challenges—it’s all in the context of hydrogen for energy. That’s where the problems lie.

Joe and I are both very supportive of green hydrogen when it’s used as an industrial feedstock. In that role, it makes sense. It has real use cases. It’s hydrogen for energy that remains fundamentally flawed.

[JR]: Making ammonia cleanly is possible—it’s just expensive.

[MB]: Making hydrogen to burn it or run it through a fuel cell is a bad idea—plain and simple.

[JR]: Right. And I know we often try to avoid getting into ethics, but it’s worth stating the basics. Fossil fuels are hydrocarbons. When you burn them, you oxidize the hydrogen into water and the carbon into CO₂. Both reactions release heat, which we’ve historically valued. But water and CO₂ are the end products of combustion. That’s the end of the thermodynamic road.

So when people try to reverse that—when they talk about turning water back into hydrogen and pulling CO₂ from the air, where it’s present at just 420 parts per million—and then combining them to make synthetic fuels, they’re trying to reverse entropy. And thermodynamics tells us very clearly: if you attempt to reverse entropy, you’re going to pay a massive efficiency penalty. That’s the second law—the famous concept of exergy. If it’s bad for hydrogen, it’s worse for direct air capture.

And if you’re foolish enough to say, “I’m going to take hydrogen from water and CO₂ from air and run them through a Fischer-Tropsch process to make a synthetic fuel, just to burn it again”—well, maybe consider that it would be better not to burn anything in the first place. The literature is clear: that pathway is 10 to 20 times less efficient than direct electrification.

And people forget—or conveniently ignore—that it’s not just the electrolyzer that has to run on 100% clean electricity. That electricity has to be new, local, and hourly matched. And you have to power the direct air capture system with that same clean energy. And the Fischer-Tropsch plant too. The total renewable energy requirement is staggering.
So then the question becomes: where are you going to put this thing? We’ve already used most of the easily accessible, high-quality renewables. Are we going to build this massive synthetic fuel complex in the middle of the Sahara Desert? Is that really the signal?

That’s the kind of logic we’re seeing from Germany, for example. I was talking to a Bloomberg reporter who mentioned a story about plans to use solar in Namibia to make hydrogen for export to Germany. I said: so instead of using that African solar power to build up the local economy, you’re going to make hydrogen, find some way to ship it north in some costly and inefficient form, and then burn it in a steel plant in Europe?

That’s your plan? You’re going to build a steel plant that depends on imported hydrogen from an African desert? And what’s truly hard for you and me is trying to talk about this with a straight face—because these are smart people. Serious people. And they’re seriously talking about investing billions into something that depends on multiple miracles to even function.

[MB]: Yeah, a few years ago I did a major study of the Maghreb region and North Africa—Morocco, Algeria, and Egypt—and the European plans to build green hydrogen programs there for export to Europe. I spoke about it at a conference in Tunisia, where I was on a panel, and I said quite plainly: this is all going to fail.

But while the Europeans are being foolish and spending a lot of money, the opportunity for these countries is to leverage that investment. Build out wind, solar, transmission, and storage infrastructure. Use it to decarbonize your own economies. Because whether or not the hydrogen export plans succeed, you’re still going to be affected by the EU’s Carbon Border Adjustment Mechanism (CBAM). Everything you currently export to Europe will face increasing carbon tariffs. The way to avoid that? Decarbonize domestically.

But what struck me—and I’ll try to say this politely—is the degree to which Europe still behaves as if it doesn’t have a colonial legacy. It does. And it’s often blind to that fact. The rest of the world isn’t.

There’s a powerful moment captured on video: a German minister—possibly even a chancellor—is speaking to an African leader, laying out climate or energy expectations. And the African leader just blasts them. He says, in effect: you don’t have the moral authority to tell us how to live well. And he’s right.

[JR]: I try to come up with analogies in the book to help people understand this. For me, the best analogy is this: imagine you want to ship water somewhere. So instead of just sending water, you convert it into champagne, ship the champagne, and then distill it back into water at the destination. That’s the plan. And somehow we’re supposed to think that makes sense.

Yes, it’s true that hydrogen can be used for direct energy applications. But is it the only way to do those things? No—not even close.

In the book, I interviewed one of the senior leaders of the International Energy Agency’s hydrogen program, and I quote him at length in the conclusion. One of the reasons people are still so positive about hydrogen is that the IEA’s Net Zero by 2050 roadmap includes it. Hydrogen is in the model because for some sectors, there’s no other obvious pathway. So it becomes a placeholder.

But what he told me was striking. He said, basically, all the major technological advances of the past decade have made hydrogen less plausible, not more. Every big step has been pro-electric: advances in batteries, in heat pumps, in electric vehicles. All of it points to electrification as the cheaper, more efficient, more scalable path.

[MB]: Molten oxide electrolysis is now being developed in labs around the world. Then there’s China’s new green steel process, which is reportedly based on their existing copper production method. Neither of these approaches—molten oxide electrolysis or China’s new process—uses hydrogen at all.

I’m still hearing rumblings, and I haven’t had time to fully dig into them—one person, two eyeballs—but some early indications suggest that molten oxide electrolysis may be using less electricity end-to-end than other decarbonized steelmaking methods. And if it consumes less energy and avoids the complications of hydrogen entirely, it’s probably going to be cheaper too.

[JR]: Right. That’s exactly the point—anything you can do directly with electricity, you’re never going to do more efficiently with hydrogen. And even if electricity has some limitations, they’re nowhere near as severe as the challenges that come with hydrogen.

Here’s what I’d say to the steel industry: let’s list the sectors that are hard to decarbonize but that we don’t have to rush right now. We don’t need to replace all the dirty hydrogen immediately—it only accounts for about 2% of global emissions and comes with high costs. We don’t need to fully decarbonize long-distance air travel yet. We don’t have to replace all international shipping. And we don’t have to fully decarbonize steel today. Those are four of the hardest problems. Let’s give them some time.

Because the choice right now is this: are you going to spend billions building a hydrogen-based steel plant today, even though there’s no green hydrogen available and likely won’t be at scale for years—if ever? Or could we invest in R&D on alternative steelmaking technologies that don’t depend on hydrogen at all? Some of those are already emerging.
Yes, they might not be ready tomorrow. But until we’ve achieved the relatively easy 80 to 90 percent of emissions reductions—through electrification, renewables, efficiency, and grid upgrades—we shouldn’t be spending huge sums to chase technologies that end up costing $500 or more per ton of CO₂ reduced. That’s not climate strategy—that’s waste.

[MB]: I’m a broad-spectrum nerd—I just need to know how things work. And then I leave a breadcrumb trail of what I’ve figured out. Most of the time, I’m not terribly wrong. I get great corrections from people, and that helps refine things. When it comes to steel, I actually see a really encouraging story—with or without hydrogen.
China produces half of the world’s steel, and it’s at the end of its infrastructure boom. It stopped permitting new blast furnaces last year and is pivoting toward electric arc furnaces (EAFs) to make use of its 260 to 280 million tons of domestic scrap. That’s a big shift.

Meanwhile, Europe and the UK are sitting at just 20 to 40% scrap utilization. They’re still exporting tens of millions of tons of scrap each year instead of turning it into new steel, and they’re still running blast furnaces. It’s just baffling.
The United States—despite my various critiques, both historical and current—has been running EAFs for about 70% of its steel demand since around 2000. They’re actually the global leader in electric arc furnace deployment. Yes, they still use natural gas for preheating and could electrify further, but the foundation is already there.

Between the global shift toward electric arc furnaces and a likely reduction in total steel demand, we’re going to see major changes in the steel sector’s carbon footprint. This is one of the few bright spots.

And yes, we did talk about leakage. I mentioned wanting to go in two directions with that. Because, 20 or 25 years ago, hydrogen was hyped as the clean solution—it burns cleanly, and when used in a fuel cell, the only byproduct is water. That was the narrative. But the more we’ve learned about leakage, infrastructure costs, and real-world implementation, the less convincing that story has become.

It’s presented as a climate solution. Yes, we know it leaks—but somehow that’s brushed aside as just a safety issue. And for some reason, people feel comfortable discounting it. Why? I don’t know. But I’m guessing you’ve been following the emerging research on the global warming potential of hydrogen.

[JR]: As it turns out, hydrogen isn’t a greenhouse gas in the traditional sense—it doesn’t directly trap heat. But it is an indirect greenhouse gas, because it extends the atmospheric lifetime of other heat-trapping gases, most notably methane.

Over the past five to seven years, scientists have revisited the numbers. Our understanding of atmospheric chemistry has improved, our models have gotten better, and—frankly—I don’t think anyone ten years ago imagined we’d still be seriously entertaining a hydrogen economy. But once interest resurged, the scientific community took another look. And what they found is concerning.

The 20-year global warming potential (GWP) of hydrogen is now estimated to be around 35, give or take. That’s much higher than we previously thought—and it’s a serious problem.

Historically, the focus was on the 100-year GWP, which is why we didn’t worry too much about natural gas. Carbon dioxide lasts a long time in the atmosphere, so it dominates the hundred-year frame. But now, with growing awareness of short-lived climate forcers, we’re looking at the 20-year impact more closely—because we urgently need to limit warming in the near term to buy time for deeper, long-term solutions.

That’s why methane has come under such scrutiny. Over 20 years, methane has a GWP of about 80. And we now know there’s widespread methane leakage across the economy. Robert Howarth at Cornell was heavily criticized for raising this early on, but he’s since been vindicated. His research showed that you only need 2–3% methane leakage before natural gas is no better than coal. And as it turns out, hydrogen leaks far more easily than methane.

This brings us to the infrastructure problem. How do we transport hydrogen? Ideally, through pipelines—but those require a guaranteed buyer and seller before they’re built. That’s the classic chicken-and-egg problem. If you don’t have established hydrogen demand, no one builds the pipelines. But without the pipelines, no one builds hydrogen-using facilities. So no one goes first. That problem was identified over 20 years ago—and it still hasn’t been solved.

In practice, most hydrogen is likely to be moved by truck, either compressed to very high pressures—up to 10,000 psi—or liquefied. Liquefaction allows for much greater energy density, so you can transport more hydrogen per trip. But it comes with huge energy penalties. And in certain cases—like tunnels—liquid hydrogen poses additional safety concerns that compressed gas might not.

So between its indirect warming effects, its high leakage rate, and the unsolved logistics of safe and efficient distribution, hydrogen as a climate solution looks far less promising than proponents would like us to believe.

[MB]: Right—you’re not allowed to take liquid hydrogen through tunnels. The safety risks are just too high.

[JR]: There are always complications. One of them is that canisters can’t actually dispense all the hydrogen they hold—the pressure dynamics prevent it. These are the kinds of practical realities that get brushed aside in the magical thinking that often surrounds hydrogen.

When people imagine hydrogen-powered trucks, they often talk about using liquid hydrogen—because if you try to cram compressed hydrogen onboard at 10,000 psi, you don’t end up with much fuel. You need specialized, rigid, non-moldable tanks, which limits how you design the vehicle. And every fueling station would need to be equipped with 12,000 psi overpressure pumps just to refill those tanks.

That adds massive complexity and cost. And here’s the kicker: all of that infrastructure is completely worthless if the hydrogen economy doesn’t materialize. If you build 1,000 hydrogen fueling stations with ultra-high-pressure pumps and the market doesn’t take off, you’re left with stranded assets—facilities no one can repurpose and no one wants to maintain.

There are just so many points of failure in this vision, and that’s why no one’s writing the check. The risk is too high, the return too uncertain, and the alternatives—electrification in particular—are simpler, cheaper, and already scaling.

[MB]: And they’re vastly more expensive and far less modular or manufacturable than megawatt-scale charging infrastructure.

[JR]: But if you want to produce green hydrogen locally at each fueling station, then every station needs to be located near a massive renewable energy source. Otherwise, you’re just pulling electricity from the grid—which likely includes fossil generation—and that defeats the whole purpose. You’re not solving the emissions problem; you’re just shifting it around.

[MB]: Let’s face it—even if we power battery-electric trucks with today’s grid electricity, they’re still not as clean as they could be. But they’re vastly better than hydrogen-powered trucks. Hydrogen has about one-third the efficiency of direct electrification for road freight. So if you’re using electricity to make hydrogen, you’re effectively multiplying any CO₂ emissions from that electricity by three.

But let’s get back to the core point—you’re going to a specific place with this, because we’re talking about global warming potential. And that changes how we evaluate all of this.

[JR]: Leakage is a major issue, especially given the pressures involved. That’s why a lot of people suggest switching to liquid hydrogen instead. I keep seeing proposals: liquid hydrogen for planes, liquid hydrogen trucks, trucks powered by liquid hydrogen, or trucks delivering liquid hydrogen. It’s all over the place. But the assumption seems to be that using liquid form somehow solves the storage and transport problem—when in reality, it just introduces a whole new set of challenges.

[MB]: Daimler is heavily invested in this. They’ve even got a member of their board of directors acting as a vocal spokesperson for hydrogen, especially in transport.

[JR]: This is one of the craziest ideas out there. First, liquefying hydrogen consumes about 40% of its energy content—you have to cool it down to near absolute zero. We’re talking much colder than liquid nitrogen or liquid CO₂. The energy inefficiency of that process is staggering.

But it doesn’t stop there. Once the liquid hydrogen is in the tank, it starts to warm up. It sloshes around during transport—and yes, there are actual studies on the sloshing effect. As it warms, it begins to re-gasify, creating pressure inside the tank. And right now, the standard way to deal with that pressure? You vent it. You just let the hydrogen escape into the atmosphere.

[MB]: I will say that Air Liquide actually captures boil-off in Europe—because they’re required to by regulation.

[JR]: And sure, you can pay to do that—capture the boil-off—but in the U.S., I don’t think there’s a single truck doing it. To make that possible, you’d have to scrap the existing fleet and install entirely new technology. And remember, you’re not just capturing the vented hydrogen—you also have to re-cool it.

So somehow this truck that’s already transporting liquid hydrogen would also need to carry the power and equipment to keep it cold enough to prevent boil-off. That’s a huge ask. It means you can’t transport it very far. And that’s the point—I’ve been looking at this, and it just doesn’t add up.

[MB]: Hydrogen leaks everywhere. Every time you do anything with it—every transfer point, every touch point—you’re looking at at least 1% leakage. That’s what the data consistently shows. In California, there was one hydrogen fueling station with 35% leakage. After years of remediation, they managed to bring it down to just under 10%.

In South Korea, when they inspected hydrogen cars and buses, 15% were leaking. An electrolyzer station in Northern Europe—engineered to high standards—still showed leakage rates between 1% and 4%. That’s just the reality.
And when you start multiplying those numbers across a full hydrogen supply chain, things get worse fast. If your value chain has seven or eight transfer points—and many do—you’re easily looking at 10% leakage end-to-end.

Multiply that by hydrogen’s 20-year global warming potential of 35, and you’ve got a significant warming impact. That’s not a climate solution. That’s a problem.

[JR]: It’s a lot of warming—full stop. And even setting aside hydrogen’s global warming potential, the inefficiency alone is reason enough to avoid losing any of it. It’s insane, really. What we’re saying is that our supposed solution to global warming is a gas that extends the lifetime and abundance of methane in the atmosphere.

And then I hear people say, “Well, we can’t do it all with renewables, so we’ll just make the hydrogen from natural gas.” Right—so we’re going to use a leaky fossil system to make hydrogen, which will then leak out itself, further extending the life of methane in the atmosphere. That’s not a solution; it’s a feedback loop. And as I say at the end of the book, the last thing you’d ever want to do in a world worried about near-term warming is expand the use of natural gas. And yet, that’s exactly what hydrogen does.

Even before Trump, there were real questions about whether oil and gas companies were serious about tackling methane emissions. And keep in mind—methane is valuable. You can sell methane. Hydrogen? Not so much. So if we haven’t gotten serious about containing methane, where there’s a profit motive, what makes us think we’ll do better with hydrogen?

Whatever framework you use—three or four miracles, or “turtles all the way down”—the point is the same: there is no foundational layer where this hydrogen economy actually makes sense. It’s built on a stack of wishful assumptions.
And I get it. The climate crisis is dire. Emissions keep rising. It feels like we’re not acting fast enough. But we’re optimistic people—we believe technology can solve problems. And it can. There are real technologies that are scaling today and delivering emissions reductions.

But people need to understand: hydrogen isn’t one of them. Not for energy. Hydrogen isn’t a solution that exists waiting for just one breakthrough to make it all work. It’s not like a “cure for cancer” situation where one discovery unlocks everything. It’s a complex problem that requires solving dozens of hard engineering, safety, infrastructure, and economic challenges—many of which don’t even overlap.

And that’s why the real answer—the practical, scalable, economic answer—is the electrification economy. That’s the future.

[MB]: So we’re at the top of the hour. Normally I’d leave it with an open-ended question, but you’ve got a book coming out in six days. So let people know where they can get it, what formats it’s available in—and if there’s some sketchy black market seller out there, give folks a heads-up to steer clear.

[JR]: Well, look—I get that some people don’t want to give money to Amazon. And I’m not here to defend Bezos. But the truth is, before he became whatever he is now, he did revolutionize book production and delivery. You can think of him a bit like Elon Musk: there’s a “before” and an “after.” The fact remains—Amazon built a remarkably efficient system for both paperback and digital books.

So yeah, if you want to feel conflicted and virtuous at the same time, buy it from Amazon. You really should. Even my publisher doesn’t recommend buying the ebook through other platforms because they can’t legally make it compatible. It’s not a true PDF, and it’s not a true Kindle file, so they’ve explicitly said: don’t buy it there.

This isn’t a book filled with figures or complex formatting, so the Kindle version works great. There will be an audiobook eventually, but for now, grab the paperback or the Kindle.

Personally, I recommend the Kindle. It’s more environmentally friendly, and honestly, it’s more useful to me as an author. I can see what readers are highlighting. And when a bunch of people underscore the same line, I think, okay, maybe that’s the part I should emphasize in a talk.

[MB]: Do you have a launch event or anything planned for the 22nd?

[JR]: No, I’ve been doing book talks, but we live in a world where they don’t really drive sales anymore. Podcasts are the modern book tour, I think..

[MB]: Well, I’m glad to be part of it.

[JR]: Well, it’s electronic, right? And it’s Earth Day—that’s the point. I really worked hard to get this out by Earth Day. So go to Amazon and buy the paperback.

[MB]: Excellent. This is Michael Barnard, the host of Redefining Energy – Tech. My guest today has been Dr. Joseph Romm, whose 20th anniversary edition of The Hype About Hydrogen is out in six days. As he said—buy it on Amazon. Joe, thank you so much for being on.

[JR]: My pleasure. Thanks for having me.