A Novel Design in Nuclear Fusion

Jeff Brown
|
Sep 26, 2025
|
The Bleeding Edge
|
12 min read


The CLARITY Act is going to light a fire under crypto unlike anything we’ve seen in the industry before. (More on that in today’s AMA…)

The entire financial system is going to be equipped with blockchain technology. A new wave of asset tokenization is coming. Innovation on a scale that’s almost hard to fathom.

And the CLARITY Act will be the catalyst.

That’s why I hosted my AI Supercharged event earlier this week.

With so much happening, it can be difficult to know where to focus our attention. My team and I are particularly interested in what I call “AI coins” – a unique subset of cryptocurrencies that marry crypto and blockchain innovation with the power of artificial intelligence.

Each of these tokens is designed to benefit from the shift from traditional finance to decentralized financial solutions, integration of it into the global financial system, and the expansion of the Web3 agentic economy.

They are all connected to projects uniquely designed to either actively progress or benefit from the establishment of the agentic economy, the advancement of AI agents, and the further buildout of the on chain financial system.

It’s an incredible thing to see happening… and that resonated, it seems, with a lot of Brownstone readers.

We had a great time on Wednesday discussing the CLARITY Act, how the Trump administration has been working tirelessly to establish the U.S. as a global leader in crypto and artificial intelligence, and some of our favorite crypto initiatives benefiting from the “all-in” mentality we’ve been seeing all year.

The CLARITY Act will be the second major piece of crypto legislation passed into law this year – not long behind the GENIUS Act passed just this past July.

We have an incredible opportunity to get in at the foundation of the next great financial shift… the first truly substantial change to our financial system we’ve had in decades.

Now is the time to prepare and position ourselves ahead of the shift… which is why we’ve made a replay of Wednesday’s AI Supercharged strategy session available.

We want to be ready when CLARITY comes. You can go here to watch the replay.

Then, read on for this week’s AMA…

Ammonia Cracking for Hydrogen Fuel

Dear Jeff and Crew,

Please identify companies that are likely to be successful in bringing ammonia-cracking technology to the hydrogen fuel industry. Thanks.

 – Patricia H.

Hi Patricia,

Ammonia cracking has actually been around for decades and has been used at an industrial scale since the 1970s. It is an exothermic process, which means that it absorbs heat in order for the reaction to take place.

In the “cracking” process, ammonia is heated to very high temperatures. A catalyst is used that causes the ammonia to “crack” or split into nitrogen gas (N2) and hydrogen gas (H2). The hydrogen gas can then be used in hydrogen fuel cells.

Several well-known public companies have ammonia-cracking technology:

  • Air Liquide (Euronext: AI)
  • Johnson Matthey (London: JMAT)
  • KBR (KBR)
  • Linde (LIN)
  • Mitsubishi Heavy Industries (Tokyo: 7011)
  • Siemens Energy (Germany: ENR)
  • Thyssenkrupp (Germany: TKA)

Companies that have developed this technology typically have a long history of working with industrial gases. And to avoid any misunderstanding, I am not recommending or endorsing any of these companies. I’m simply providing them as a reference for you and your question.

In addition to the public companies, there are a couple of private companies working in this space that I actually do like… Topsoe, which is a Danish company, and Amogy, which was founded by four MIT PhDs in 2020.

The bigger question, however, is whether or not hydrogen as a fuel makes any sense at all.

Ammonia is manufactured for industry through the heavily energy-intensive Haber-Bosch process, which uses natural gas to produce hydrogen, and then, under high pressure and temperature, hydrogen and nitrogen are combined to form ammonia. The ammonia is then liquified for transportation.

Hydrogen is a very pernicious gas due to the tiny size of the hydrogen molecules, so transporting hydrogen in ammonia is a way to avoid leaks. Hydrogen gas is volatile and very flammable, much more so than natural gas. And hydrogen has no odor, so we can’t smell a leak. For this reason, it is very dangerous.

And 95% of all hydrogen produced is made using fossil fuels in very energy-intensive processes. It is not clean in any way. Hydrogen fuel cells are also far less efficient than using batteries. And that means that it is substantially more expensive than electricity or gasoline.

Most don’t know this, but to produce a megawatt hour of hydrogen as a fuel, it also consumes about 5,000 liters of water.

The use of hydrogen is not ecological, it is not environmental, it is not “green.”  The only benefit is that it has no emissions used as fuel (i.e., the emissions were already displaced in producing hydrogen in the first place).

But that doesn’t mean that it won’t be adopted. If enough politicians and lobbyists spend enough money on spin and bribes to get large taxpayer-funded spending programs in place, a lot of money can be extracted from public resources, sadly.

 If you would like some more context on this subject, I recommend reading The Bleeding Edge – Hydrogen as a Fuel (scroll down to the last question), The Bleeding Edge – Why I’m Skeptical of Hydrogen as a Fuel, and definitely read Outer Limits – The Reality of Hydrogen Fuel Cells.

I hope this information is useful.

How Can We Trust a “Corporate” Blockchain?

A question for your AMA:
I found this article introducing Figure Technology Solutions very informative.

It says Figure tokenizes documents on its proprietary Provenance blockchain. Please elaborate on how this proprietary blockchain works. Who hosts this proprietary blockchain and maintains its trustless integrity?

A blockchain is typically hosted by numerous independent parties coordinating to operate the blockchain with consensus voting to keep each other honest. For this effort, they’re paid in the blockchain’s coins. (If this description is inaccurate, then please explain how non-proprietary blockchains are typically hosted.)

How does hosting work for a proprietary blockchain? Who besides Figure is performing this function for Provenance? How do they keep Figure honest to maintain Provenance‘s integrity, preventing Figure from altering the blocks on Provenance? What coins does Provenance use for payment? Or if not for payment, why do entities other than Figure perform this function?

Thanks for explaining further.

Bobby W.

Hello Bobby,

This is an interesting question because it highlights some of the issues or concerns a private company using blockchain technology, and whether or not that centralized entity (i.e., the company) can be trusted. And Figure Technology’s approach is an interesting example of how this can be addressed.

Figure tokenizes its documents and creates loans on the Provenance blockchain. Provenance, however, is not a proprietary permissioned blockchain. It’s a public, permissionless blockchain that uses the Cosmos software development kit (SDK).

Cosmos SDK is a toolkit for builders to create their own application-specific blockchains.

The Cosmos ecosystem was built around the idea that some use cases would want independence from Ethereum and other Layer-1 blockchains and their development roadmap. This is referred to as a modular solution to blockchains.

Cosmos positions itself as the internet of blockchains. That’s because each blockchain built on the Cosmos SDK is able to easily communicate with any other blockchain that was built on the SDK.

The chain is not permissioned in terms of a walled garden or enterprise-like solution. Instead, there’s a network of 100 validators that verify the transactions and keep building blocks.

These validators include companies such as Blockdaemon, Cosmostation, and Citadel. One among many others that are not associated directly with Figure. These validators all have to delegate native tokens – in Provenance’s case, the HASH token – to the chain to provide validation services.

This makes it decentralized and permissionless.

There are also blockchain explorers to track the activity of the chain, such as provenance.io/pulse and zonescan.io, where you can see the blocks getting formed in real time. This provides public transparency of all transactions.

For example, here’s a snapshot of what we can see in real time on Provenance concerning loan statistics:

Provenance Loan Statistics | Source: Provenance Blockchain

There isn’t a single lender in traditional finance that makes this kind of information available on a real-time basis, let alone the publicly available transactions of anything that takes place on the Provenance blockchain.

This design and implementation ensure that Figure isn’t altering blocks on Provenance and keeps them honest with their activity.

First Light Fusion’s Novel Design

Hi Jeff,

I’ve been following the Bleeding Edge newsletters for several years now and love reading them. They keep me up to date on the exciting news without me needing to dig through huge amounts of negativity to find it. They’re also fascinating, and I love how you manage to break down complex subjects in ways a layperson can understand. Please keep up the good work!

I’m writing in to let you know about a recent (today) whitepaper release from a fusion company in the UK. I have been working as a Software Engineer at First Light Fusion in the UK for the best part of three years now.

We’ve recently undergone a significant change in approach that has dramatically improved morale in the company and appears (at least to us) technically feasible. I would love for you to have a read and to hear your thoughts on the whitepaper.
Best,

 – Ciaran W.

Hi Ciaran,

Thanks for the feedback. I appreciate you reaching out. What a fantastic time to be working in nuclear fusion!

Your timing is great as I just heard of the release of the new whitepaper from First Light Fusion (FLF) yesterday.

This is an exciting pivot. I have to say that I had almost written FLF off early this year when I read the news about focusing on commercial partnerships around its amplifier technology, and other, less interesting applications. That wasn’t looking good.

I took a fast pass through the white paper. This is typically something that I’d prefer to read and think on for a week before responding. There is a lot there, but I understand what FLF is proposing.

And to your point, conceptually, there should be no reason that this won’t work. Technically feasible, definitely. These are engineering problems to be solved. And with all complex systems, like what FLF has proposed, it’s all about execution and optimization of the design.

There are some clear advantages with the lower power requirements, pulsed power for compression, cylindrical targets for stability, and what should be lower cost for reactor construction and operation.

But the part of the design that I liked the most is the lithium pool reactor design. I had actually been wondering why no inertial confinement fusion (ICF) designs had adopted such an approach.

For everyone’s benefit, here is a rendering of what the reactor design would look like…

Lithium Pool Reactor Design | Source: First Light Fusion

We can think of this like the reactor being submerged in a large liquid lithium pool. It’s like a liquid lithium “blanket” that surrounds the reactor and efficiently captures the heat produced by the fusion reaction (FLF estimates 99.9% efficiency with this design), captures the neutrons, facilitates the breeding of tritium (fuel for future fusion plasmas), and also helps to mitigate structural damage to reactor encasements.

Solid lithium blankets do the same thing. However, they are leaky. They don’t capture all of the neutrons and simply aren’t as efficient as a lithium pool.

The reason I was surprised that no fusion company had tried this before is that similar technology is used widely in data centers.

It is very common to immerse GPU servers in a dielectric fluid – this is usually referred to as liquid cooling.

Doing so is far more efficient at dissipating the incredible heat produced by the GPUs, which reduces the thermal stress on the chips and electronic circuit boards. And that does two things: it improves computational performance and extends the lifetime of the servers. And energy costs for cooling can drop by 30–50% with liquid cooling systems.

While it’s not exactly an apples-to-apples comparison, there are some strong parallels to the lithium pool approach in reducing structural damage, improving heat capture, and improving overall efficiency of the system.

Very exciting approach.

Ciaran, from my perspective, this is a viable path forward, one that should be pursued. At the end of the day, the industry needs to build prototypes of a wide range of approaches to nuclear fusion in order to figure out which approaches work most efficiently and cost-effectively to support grid-scale clean energy power generation.

Objectively, there is one additional thing that First Light Fusion needs to be successful – a whole lot more capital. FLF is behind, and there is an incredible sense of urgency in the industry right now. The sooner FLF can build a prototype and demonstrate proof of concept, the better.

I wish you and your team well on your journey and hope you’ll be successful.

What Drives Up the Value of a Cryptocurrency?

Good Evening,

Financial markets are going on-chain. How does this make investments in crypto go parabolic? Going on-chain doesn’t mean buying crypto; it means using the technology. Which causes no price change in altcoins. I don’t want to be left behind, but this makes no sense. Thanks.

 – Chad S.

Hi Chad,

I’m glad you wrote in with this topic. It really speaks to the importance of what is happening right now as traditional finance migrates to widespread use and employment of blockchain technology.

In order to interact with blockchains, we need to do so with digital assets (i.e., cryptocurrencies) to facilitate transactions on blockchains. In most cases, the native token for any given blockchain is the “currency” used to transact on that blockchain.

Sometimes, transactions can be facilitated by paying “gas fees,” which might be made using a specific token, and for some blockchains, gas fees can be paid using U.S. dollar stablecoins.

Regardless of the application/blockchain, transactions are facilitated using digital assets. And that means that as traditional finance migrates to using blockchain technology for financial services, it will need to transact in digital assets.

The first major wave, which is happening right now, will see trillions of dollars migrate into blockchains as U.S. dollar stablecoins. The price of a U.S. dollar stablecoin will not change, of course, as a dollar is still a dollar.

However, blockchain projects that support the stablecoin ecosystem will see radically higher utilization of their blockchains, which will drive an increase in their token price. This has already been happening, as evidenced in our Permissionless Investor model portfolio.

The second major wave is that once financial institutions have converted their fiat U.S. dollar into U.S. dollar stablecoins, they are now easily able to transact and convert those U.S. dollar stablecoins into other digital assets to facilitate financial transactions on other blockchains.

The key point is that increased utility and utilization of any digital asset – assuming a stable monetary policy of the underlying blockchain – will drive the price of that native digital asset higher. Higher utilization means more buying of the digital asset for transacting.

Some digital assets are required for transacting on their respective blockchains, while other digital assets, like gas tokens, are designed to facilitate transactions. Some digital assets represent exposure to increased activity on a blockchain. As fees increase on a blockchain, the value of the representative token increases in value.

Better yet, the direction of the industry is to tokenize a much wider range of assets. For example, equity has already been tokenized, so the digital asset trades like a stock. Future cash flows can be tokenized. Intellectual property can be tokenized, for example, the tokenization of future royalty streams from a patent or a song/album.

Blockchain technology provides near-instantaneous settlement times, dramatically reduced transaction costs, transparency, immutability, reduced friction, and enables smart contract functions that automate transactions.

This environment is also optimized for the use of agentic AI because protocols are well defined, and an AI can have the agency required to transact on any blockchain and abstract away any complexity in doing so.

This is exactly what I discussed earlier this week at my AI Supercharged strategy session. There’s so much happening right now, and I’m keen to help as many people as I can understand the weight of the opportunity that we’re looking at here.

And Chad, one final point. The concern you raise is valid. There are altcoins out there that provide little to no utility. They serve no purpose and are destined to fail.

And there are also some incredible altcoins and digital assets out there that provide immense utility that will experience exponential growth in the utilization of the underlying blockchain. A big part of what my team and I do here at Brownstone Research is identify which projects are worth pursuing and which ones are just capitalizing on the movement without anything substantial to support them.

Trillions of dollars of capital moving into digital assets will result in the best quality blockchain projects that provide a great utility increase in value.

I hope you don’t miss out

That’s all for this week. As always, anyone with a question or concern can reach us right here. Just keep in mind I can’t offer individual investment advice.

I hope everyone has a great weekend.

Jeff



Want more stories like this one?

The Bleeding Edge is the only free newsletter that delivers daily insights and information from the high-tech world as well as topics and trends relevant to investments.