Chemical Engineer Paul Martin Reflects On Liebreich’s Hydrogen Ladder & #Hopium — Part 1
In this edition of CleanTech Talk, Paul Martin and I discuss Michael Liebreich’s hydrogen ladder. Paul is a working chemical process engineer, and has spent his career building prototypes of biofuel, hydrogen, and chemical processing plants as part of scaling them to full, modularized production systems for clients. Paul’s piece in CleanTechnica on why hydrogen is not suitable as a replacement for natural gas in buildings is a must read.
Liebreich is an entrepreneur, founder of what has become Bloomberg New Energy Finance (BNEF), chairman on multiple boards, has engineering and business degrees, and represented the UK on their skiing team in 1992. He’s had a rich and interesting life, but for the purposes of this pair of podcasts and attendant articles, it’s his iteratively improving hydrogen ladder Paul Martin and I are focusing on.
Regular readers of CleanTechnica will know that I have been assessing hydrogen’s place in the decarbonized economy in the areas of transportation, oil refining, and industry, among others. Paul and I share a strong opinion that “blue” hydrogen, which is sourced from fossil fuels with 10-30 times the mass of CO2 which is theoretically going to be sequestered or used, is a fossil-fuel industry lobbying effort and not a viable climate solution.
Listeners are recommended to keep the hydrogen ladder in front of them as Paul and I talk through aspects of it.
We start with a discussion of one of Paul’s frequently used hashtags, #hopium, which he defines as the drug that is made out of our own hope to overcome our faculties and divert government money to things which aren’t useful. We agree that the fossil fuel industry are masters of PR when it comes to giving false hope to governments and individuals that we can just vacuum CO2 out of the air or out of smokestacks after emitting it, rather than the reality that we leave most fossil fuels unburned and unused.
Paul steps through existing hydrogen production, pointing out that of the 120 million tons used annually today, less than 0.1% could be considered green hydrogen, intentionally cracked from water using renewably generated electricity. All hydrogen today is actually black, at least 30% blacker per unit of energy than the fossil fuel it was made from. For coal, up to 30 kg of CO2 is created for every kg of hydrogen, with one data point suggesting a proposal in Australia to make hydrogen from low-grade coal with 35 kg of CO2 for each kg of hydrogen. For natural gas, it’s up to 10 kg, but there is also methane leakage with its 86x worse than CO2 on 20 years global warming potential. Creation of hydrogen from natural includes an almost equal amount of GHGs in methane leakage, which is typically not counted in the emissions.
We continue with a discussion of ground transportation, where there is no place for hydrogen, in our opinion. Paul draws out the efficiency versus effectiveness argument first. Gasoline isn’t efficient, as perhaps 15% turns into useful energy, but it is effective due to being cheap, easily poured into gas tanks, and easily transported.
Hydrogen is neither efficient or effective for ground transportation. The misleading truths that are used for #hopium are that it’s the most common element in the universe and has excellent energy density for its mass.
The first truth is not helpful, as all hydrogen available to us is tightly chemically coupled with other substances, whether that is fossil fuels or water. It takes a lot of energy to break those bonds.
The second truth is not helpful either. Hydrogen, as the lightest element and lightest gas, has very poor energy density by volume, regardless of whether you compress it to 700 atmospheres, a little over 10,000 pounds per square inch, or chill it to 24 degrees above absolute zero to liquify it. As a gas, it has less than a third the energy density by volume of methane, and as a superchilled liquid, its energy density by volume is only 75% better.
Paul points out that the Toyota Mirai vs Tesla Model 3, otherwise comparable cars, is illustrative in that the Mirai weighs as much as the Tesla, even though it only carries 5.6 kilograms of hydrogen. The tanks weigh hundreds of kilograms. A standard hydrogen cylinder weighs 65 kg and only delivers 0.6 kg of hydrogen, a problem that transportation uses have to overcome with expensive thin-walled aluminum tanks wrapped in carbon fiber. It’s also worth noting that hydrogen cars have less interior and luggage room due to the hydrogen storage and fuel cell component space requirements.
Paul points out the lost mechanical energy of compression. He calculated once that the energy used to compress 5 kg of hydrogen to 700 atmospheres was equivalent to the kinetic potential energy of suspending the car 500 meters in the air, ready to drop. That energy is lost. If superchilled hydrogen were used instead, 40% of the energy in the hydrogen would have to be used to chill it.
The final devil in the details is thermal management. Hydrogen is an interesting gas in that unlike many other gases, it gets warmer as it expands. Anyone used to compressed air cans know that the jet of air comes out cold, but an equivalent jet of hydrogen would come out hot. Even though compressed hydrogen isn’t liquified, in other words, it has to be chilled in its tanks before being pumped into cars, another loss of energy.
This all leads to the common myth that hydrogen cars are quick and convenient to refuel. The reality is shown by Toyota’s entry in the 24-hour enduro Super Taikyu Series in Japan’s Shizuoka Prefecture. They prepped a racing Corolla with a hydrogen combustion engine. It had four huge carbon-fiber tanks in the area where you would normally have back seats. They brought four tractor trailers full of equipment to fuel the car. The car had to spend four hours of the 24 hours of the race refueling. Ineffective, inefficient, and with startling infrastructure requirements.
As Paul says, the devil isn’t hiding in the details, he’s waving his pitchfork in plain sight of anyone willing to see him.
We move on to agreeing in general that hydrogen might have a direct play in long-haul shipping, or at least hasn’t proven itself uncompetitive in that space. I recently assessed Maersk’s methanol drivetrain dual-fuel ships announcement, and 40-day journeys with thousands of tons of fuel are a very hard problem to crack. Maersk has proposed a green methanol manufacturing facility capable of producing enough synthetic green methanol annually to cover half of one trip for one of the eight ships.
For the rest of the first half of the podcast, aviation is in our sights. Paul and I agree that short- and medium-haul aviation — basically all air trips within the boundaries of most continents — are going to be battery electric. Hydrogen has no advantages for those ranges.
And we agree that long-haul aviation is another hard problem. I went deep on long-haul aviation’s global warming contributions and challenges recently, so had the concerns at top of mind. First was the problem of direct carbon dioxide emissions of course, but aviation also has contrail and nitrous oxides emissions problems.
Contrails are water vapor, effectively clouds. Due to the altitude of especially night-flying high-altitude planes, they keep more heat in than they reflect. That’s something that can partially be managed by changing operations, reducing altitude and night-time operations, but there are economic reasons why planes fly high and at night that need to be addressed with economic incentives.
Nitrous oxides are trickier. Any fuel burned in oxygen produces nitrous oxides with a bunch of the nitrogen from the air, which is, after all, 78% nitrogen. Nitrogen combined with oxygen in the form of N20, nitrous oxide or laughing gas, has a global warming potential of 265 times that of CO2, and persists in the atmosphere a long time.
Another form of nitrous oxide, NO2 or nitrous dioxide, is the chemical precursor to smog, causing asthma and other heart lung problems. For those following along, yes, if you have a natural gas stove or furnace in your home, it’s also putting NO2 into your home’s air along with carbon monoxide, which you need a detector for if you don’t have it. All the more reason to electrify to induction stove tops and heat pumps as your appliances age out.
Paul’s perspective is that hydrogen for long-haul aviation has multiple problems. The first is that it can’t be stored as a pressurized gas in airplanes due to the increasing loss of atmospheric pressure and bulk as planes ascend to 30,000 ft. The second is that even chilled, it’s much less dense by volume than kerosene, so it would have to be stored in the fuselage. The third is that fuel cells are bulky for energy output of sufficient electricity, so would also have to be within the fuselage, and fuel cells give off a lot of heat. So that means either jets lose a fair amount of passenger and luggage storage, or get a lot bigger and heavier, even before the cooling and venting requirements for the fuel cell heat. That makes the economics of jet travel problematic, which might be just fine, as it arguably should be more expensive than it is.
However, this means that it would be hydrogen jet engines that would be used if hydrogen were to be used directly as a fuel. And burning hydrogen in a jet engine will produce a lot of water vapor, hence the same contrails, and nitrous oxides, hence the high global warming potential. Hydrogen would only deal with two-thirds of the problem.
Paul and I agree that biofuels for hard-to-service transportation modes such as long-haul shipping and aviation, along with operational changes and reduced use, are likely the best we can do until we achieve a battery as much better than lithium-ion as lithium-ion is than lead acid, and that took a century.
But we’ve had biofuels certified for aviation use since 2011, and they just aren’t being used. They are more expensive, despite being much lower CO2 emissions cradle-to-grave than kerosene. Once again, negative externalities have to be priced.
The next half of the podcast discussion gets into places where hydrogen actually has a place in the sun, but makes it clear that hydrogen is actually a decarbonization problem, not a decarbonization solution.
Appreciate CleanTechnica’s originality? Consider becoming a CleanTechnica Member, Supporter, Technician, or Ambassador — or a patron on Patreon.
Environment
Ford and SK On kill massive $11.4B US battery joint venture, split factories between them
In a massive shakeup for the US electric vehicle supply chain, Ford and South Korean battery giant SK On announced today that they are ending their massive “BlueOval SK” joint venture.
The two companies will effectively split the custody of the three massive battery factories they were building together, with Ford taking the Kentucky site and SK On taking the Tennessee site.
Back in 2021, Ford and SK Innovation (SK On’s parent) announced a massive $11.4 billion investment to build three battery gigafactories in the US: two in Kentucky and one in Tennessee. It was, at the time, the single largest manufacturing investment in Ford’s 118-year history.
The idea was to create a vertically integrated battery supply chain for Ford’s next-generation electric trucks and SUVs.
But today, SK confirmed reports that they are dissolving the JV structure entirely.
According to the announcement, Ford will take full ownership of the two battery plants in Kentucky. One of these plants had already begun initial operations earlier this year, while the second is still under construction.
SK On, meanwhile, will take full ownership of the BlueOval City battery plant in Tennessee.
Ford acknowledged the announcement from SK, but it refused to comment:
We are aware of SK’s disclosure and we have nothing further to share at this time.
The South Korean battery maker was surprisingly candid about the reasoning, stating that the split allows them to “supply batteries for both electric vehicles and energy storage systems not only to Ford but to a wider range of customers.”
This confirms that SK On didn’t want to be tied exclusively to Ford’s production volumes, which have been fluctuating wildly over the last year as the automaker adjusts its EV strategy. SK On explicitly mentioned expanding into the Energy Storage System (ESS) market,a sector that is booming right now, using the capacity at the Tennessee plant.
The deal is expected to close in the first quarter of 2026.
This comes amid narratives of “slowing EV demand” from legacy automakers and a changing political landscape in the US that has introduced uncertainty following the end of federal incentives.
Electrek’s Take
This is a huge deal, and frankly, it smells like trouble for Ford’s original volume targets.
When you form a multi-billion-dollar JV, you do it to lock in supply and share the massive capital burden. Breaking it up less than four years later, before all the plants are even running, suggests that the original plan is effectively dead.
In short, it would suggest Ford does not want the battery capacity from the factory it gave to SK.
This is not exactly surprising considering Ford’s pullback of EV plans, such as the F-150 Lightning, recently.
For SK On, this looks like a smart pivot. If Ford isn’t buying enough cells to fill the lines because they are delaying EV models or pivoting to hybrids, SK needs the freedom to sell those batteries to Hyundai, VW, or into the energy grid storage market, which is insatiable right now.
They basically just freed up 45 GWh of capacity in Tennessee to sell to whoever writes the biggest check.
FTC: We use income earning auto affiliate links. More.
Another EV is reportedly on the chopping block from Hyundai’s US lineup. Here’s what we know so far.
Is Hyundai pulling the IONIQ 6 EV from the US?
The sporty “N” model may be the only IONIQ 6 trim Hyundai offers in the US for the 2026 model year. Hyundai confirmed the 2026 IONIQ 6 N will be “extremely limited” in the US, but has not offered any updates on the standard version.
Hyundai gave the electric fastback a major refresh for 2026, complete with a sharp new look, a bigger battery for more range, a refined interior, and more.
That sounds nice and all, but will we ever get to see it in the US? According to a report from Jalopnik, the IONIQ 6 N may be the only trim available next year.
When asked whether the updated IONIQ 6 would be coming to the US, Hyundai said no updates were available at the moment. On Hyundai’s website, the 2026 IONIQ 6 is listed with “more details to come.”
Although the Hyundai brand has not scrapped any EVs from its US lineup, its luxury Genesis brand dropped the Electrified G80 and is shifting from initial plans for an all-electric lineup to launching its first hybrid and extended-range electric vehicles (EREVs).
Hyundai’s sister company, Kia, has indefinitely delayed the EV4 electric sedan and the high-performance EV9 GT for the US market.
Is the IONIQ 6 next? A Hyundai spokesperson told TheKoreanCarBlog there are currently no new updates on plans to bring the updated IONIQ 6 to the US. The high-performance N variant remains the only 2026 IONIQ N trim confirmed to launch.
Hyundai has yet to reveal prices for the 641 hp electric sports car, but given the 2025 IONIQ 5 N starts at $66,200, the IONIQ 6 N will likely be slightly more expensive.
With the 2026 models arriving at dealerships, Hyundai is offering a few deals worth checking out. The IONIQ 5 remains one of the most affordable EVs in the US, with leases starting at just $189 a month. You can use the links below to find available models by you.
FTC: We use income earning auto affiliate links. More.
Environment
Rivian AI & Autonomy Day: In-house silicon chip, next-gen AI platform, LiDAR for Level 4 self-driving [Video]
![Rivian AI & Autonomy Day: In-house silicon chip, next-gen AI platform, LiDAR for Level 4 self-driving [Video]](https://i0.wp.com/electrek.co/wp-content/uploads/sites/3/2025/12/Rivian-autonomy-R2-LiDAR-front.jpg?resize=1200,628&quality=82&strip=all&ssl=1)
As promised, we are here in Palo Alto, California, live at Rivian’s inaugural AI and Autonomy day, which the American automaker has been teasing for a month now. The event was filled with numerous exciting updates, including new in-house technology, a new AI assistant, the addition of LiDAR, and a uniquely wrapped Rivian R2.
Today’s AI and Autonomy event has been on our radar since Rivian released its Q3 2025 financial report, back in early November. At the time, the American automaker shared it had recently founded a new AI-focused company called Mind Robotics, joining its recently launched e-bike brand, ALSO.
When Rivian’s comms team shared the news about Mind Robotics, it said not to bother asking any additional questions, as the company wasn’t sharing any other details at the time. Instead, more would be revealed at an upcoming event called Rivian’s AI and Autonomy Day, slotted for December 11, 2026.
Here we are, and if you’re reading this, Rivian’s livestream of the inaugural event is either happening or has recently taken place. In addition to AI news, Rivian shared a slew of exciting updates, including some new component integrations as the automaker joins the race to achieve full-fledged Level 4 autonomy someday.
There’s a lot to unpack here, so let’s dig in.
Rivian to deliver new AI and autonomy tech in early 2026
Since today’s event was partially an AI Day, let’s start there. To begin, it revealed a new Rivian Autonomy Processor (RAP1) and Gen 3 Autonomy Computer, powered by a new in-house silicon chip designed for vision-centric AI.
According to Rivian, RAP1 will power its Gen 3 Autonomy computer, called the Autonomy Compute Module 3, or “ACM3.” Here are some key specs:
- 1,600 sparse INT8 TOPS (Trillion Operations Per Second)
- Processing power of 5 billion pixels per second
- RAP1 features RivLink, a low-latency interconnect technology allowing chips to be connected and multiply processing power
- RAP1 is enabled by an AI compiler and platform software developed in-house by Rivian
Additionally, AI plays a key role in Rivian’s business infrastructure and user experience through a shared, multi-modal, and multi-LLM data foundation it calls “Rivian Unified Intelligence” (RUI).
Rivian says RUI will integrate AI into diagnostics to assist techs and quickly identify “complex issues.”Rivian states that the AI platform will also aid in the development of powerful new features, enhance service infrastructure, and facilitate predictive maintenance.
A notable addition that stood out to me in this new architecture is the Rivian Assistant (pictured above), which we’ve learned will launch in early 2026 on Gen 1 and Gen 2 R1 vehicles. Here are some highlights, per Rivian:
- Built on Rivian’s edge models to understand your vehicle, your digital life, and the world around you
- Connects vehicle systems with third-party apps using an in-house agentic framework, with Google Calendar named as the first integration
- Augmented by frontier large language models for grounded data, natural conversation and
- powerful reasoning.
Now for the second part of Rivian’s special event – autonomy. This morning, Rivian shared details of its new Autonomy Platform as well as an end-to-end data loop used to train it. This Large Driving Model (LDM) is trained similarly to a Large Language Model (LLM), but programmed to achieve foundational autonomy, with the ultimate aim of reaching Level 4 self-driving capabilities.
The new LDM will gather ideal driving strategies from massive datasets using Group-Relative Policy Optimization (GRPO) and will then incorporate those techniques into a compatible Rivian vehicle. According to Rivian, those autonomy software enhancements will be rolled out to Gen 2 R1 models in the near future, unlocking Universal Hands-Free (UHF) driving capabilities for over 3.5 million miles of roads across the US and Canada. Check it out:
To access this feature, Rivian has introduced an autonomy subscription called Autonomy+, launching in early 2026 for a one-time fee of $2,500 or a monthly payment of $49.99 to access the feature and its ongoing improvements.
Rivian hinted at some of those autonomy-centric improvements in the works for the Gen 2 R1 and R2 EVs, which include point-to-point, eyes-off, and personal Level 4 driving capabilities. Rivian founder and CEO, RJ Scaringe, spoke to the new tech debuting at AI and Autonomy Day:
I couldn’t be more excited for the work our teams are driving in autonomy and AI. Our updated hardware platform, which includes our in-house 1600 sparse TOPS inference chip, will enable us to achieve dramatic progress in self-driving to ultimately deliver on our goal of delivering L4. This represents an inflection point for the ownership experience – ultimately being able to give customers their time back when in the car.
Last but not least, the upcoming R2 made an appearance, cleverly “dressed” as R2D2 from Star Wars. What was more interesting, however, was that the R2 was equipped with a new piece of autonomy tech new to Rivian EVs – LiDAR.
That’s right; today, Rivian confirmed that it plans to integrate LiDAR into future R2 models, alongside the previously mentioned ACM3. Per Rivian:
LiDAR will augment the company’s multi-modal sensor strategy, providing detailed, three-dimensional spatial data and redundant sensing, and improving real-time detection for the edge cases of driving.
Here’s a peek at how it works:
How are we feeling? I know I just threw a lot of technical jargon at you, but here’s the gist – Rivian is continuing to develop, introduce, and integrate its own in-house technologies. Today’s was a purpose-built silicon chip and software-centric autonomy platform bolstered by AI.
LiDAR is likely to be introduced in R2 next year, and future models beyond that, and new autonomy capabilities are on the way (for a fee). That will include more hands-free-friendly routes around the US and Canada. Exciting stuff all around from Palo Alto.
Electrek’s take
While we will still need to see Rivian actually roll out these AI and autonomy technologies to owners for real-world use (and critique), it is exciting to see the American automaker not only deploying technologies like software-first autonomy platforms and silicon chips, but also developing them in-house.
Rivian is by no means the first to do so (Tesla introduced its FSD silicon chip years ago, and several Chinese automakers have followed suit). However, it’s still quite encouraging to see Rivian put a pony in the race to true, Level 4 self-driving.
I also find it interesting (while not surprising) that Rivian has chosen to integrate LiDAR into the R2 to support autonomy as part of its multi-modal approach discussed above. LiDAR remains a hot topic as Tesla still swears by vision cameras and has been joined by other recent converts, such as XPeng Motors. Meanwhile, most other OEMs still see LiDAR as the long-term best option, especially when paired with other sensors.
Rivian has opted for that same route to autonomy, and I can’t disagree with the decision. From a business standpoint, owning IP is a significant advantage, and it’s exciting to see Rivian continuing to develop such technological advancements internally, similar to other American EV brands like Tesla and Lucid.
Their focus on EVs not only promotes a sustainable future but also gives these companies more freedom to invest more heavily in one form of technology (BEVs) compared to legacy automakers, who must spread their R&D budgets across combustion, hybrid, PHEV, and BEV models.
Automakers like Rivian, Lucid, and Tesla have a few proprietary aces up their sleeves that they could one day wield toward other OEMs, whether that’s building and selling that hardware or software to them, or even licensing it. Take, for example, Rivian’s agreement with the Volkswagen Group, which includes EVs from Scout Motors.
Right now, I’m sure those brands are utilizing most of their technology in their own models, but developing their own motors, inverters, chips, and software platforms does have its advantages, whether that’s in Rivian EVs or other makes and models.
I’m personally excited to see Autonomy+ roll out to my Gen 2 R1S and will be sure to test the expanded Hands Free routes when I can. That’s all for now; thank you for reading.
FTC: We use income earning auto affiliate links. More.
-
Sports2 years agoStory injured on diving stop, exits Red Sox game
-
Sports3 years ago‘Storybook stuff’: Inside the night Bryce Harper sent the Phillies to the World Series
-
Sports2 years agoGame 1 of WS least-watched in recorded history
-
Sports3 years agoButton battles heat exhaustion in NASCAR debut
-
Sports3 years agoMLB Rank 2023: Ranking baseball’s top 100 players
-
Sports4 years ago
Team Europe easily wins 4th straight Laver Cup
-
Environment3 years agoJapan and South Korea have a lot at stake in a free and open South China Sea
-
Environment1 year agoHere are the best electric bikes you can buy at every price level in October 2024