A reactor in Houston splits liquid ammonia into hydrogen that travels just a few toes right into a German piston engine and by no means will get saved anyplace alongside the way in which, and greater than 70 % of the world’s ammonia is made out of pure fuel to start with

Ask anybody constructing an information middle in Texas how they plan to energy the factor, and the reply has gotten boringly constant. Park a row of fuel engines out again, run them behind the meter, cease ready on the utility.

There is an effective purpose it grew to become the default. Interconnection in ERCOT and PJM is now a multi-year drawback, and no one spending billions on an AI buildout desires to clarify a two-year energy delay to a board.

So the attention-grabbing a part of what Amogy and 2G Energy pulled off at a facility in Houston isn’t that they constructed one other on-site generator. It’s what they poured into it.

The two corporations mentioned on August 3 that that they had accomplished an built-in ammonia-to-power check, based on their joint announcement on GlobeNewswire. A chemical reactor splits liquid ammonia into hydrogen. That hydrogen travels just a few toes right into a German piston engine. The engine makes electrical energy.

No hydrogen storage tank. No gasoline cell stack. And every time the operator desires, the identical engine goes again to burning methane out of the pipeline.

The reactor and the engine needed to study to speak to one another

The product has a reputation, AMMDrive, and a really particular {hardware} pairing. Amogy provides the ammonia reformer. 2G provides an Agenitor 412 reciprocating engine-generator set, which is a machine it has been promoting into European cogeneration for years with a fuel meter on the entrance as an alternative of a chemical reactor.

Data Center Dynamics reported that the reformer turned ammonia right into a hydrogen-rich stream that ran the engine and hit the numbers each corporations say a industrial product wants. Amogy CEO Seonghoon Woo referred to as the consequence “an important proof point for Amogy’s ammonia-to-power platform.”

What really obtained confirmed is the handshake. Fuel supply, management programs, two very totally different items of {hardware} agreeing on what they’re doing. That sounds unglamorous till you bear in mind it’s the step that kills most integration initiatives.

The corporations additionally say the engine might be configured to throw off usable warmth alongside the electrical energy, and that the structure is modular sufficient so as to add capability in levels reasonably than unexpectedly. Neither agency disclosed {the electrical} output of the unit that ran in Houston. For scale, 2G’s personal hydrogen product literature lists the Agenitor 412 operating on pure hydrogen at 360 kW electrical and 40.5 % electrical effectivity. Whether the ammonia-fed model lands in the identical neighborhood is precisely the type of factor a industrial deployment would settle.

Cracking ammonia is outdated chemistry aimed toward a brand new buyer

Ammonia is NH₃. Three hydrogen atoms hanging off one nitrogen. Push it by way of the appropriate catalyst on the proper temperature and it splits, or cracks, into hydrogen fuel and nitrogen fuel.

The hydrogen is the purpose. The nitrogen makes up about 78 % of the air you’re respiration proper now, so no one spends a lot time worrying about the place it finally ends up. Amogy has been promoting that trick for years, and largely it has pointed the ensuing hydrogen at a gasoline cell. The firm powered a drone, a farm tractor and a semi-truck, then in September 2024 sailed a retrofitted 1957 tugboat it named the NH3 Kraken on a Hudson River tributary north of New York City.

Houston is totally different as a result of the endpoint is a reciprocating inner combustion engine. Pistons, crankshaft, connecting rods, valves. The distinction issues greater than it sounds: a gasoline cell converts hydrogen electrochemically and a piston engine burns it, and people are two solely separate engineering issues with separate emissions profiles. Burning a hydrogen-rich stream in air at combustion temperature makes nitrogen oxides. A gasoline cell doesn’t, and that hole is what an air allow author in Texas will get paid to care about.

The gasoline cell model of this argument is already operating commercially. One Santa Clara knowledge middle has spent greater than two years off the utility grid making its personal electrical energy from hydrogen gasoline cells and cooling its racks with the water these cells produce. Amogy and 2G are betting a piston engine is the cheaper, extra acquainted path to the identical place.

2G’s engines already eat 4 totally different fuels

2G isn’t a reputation most Americans know. It was based in Heek, in North Rhine-Westphalia, in 1995, and it builds mixed warmth and energy models throughout a spread the corporate places at 20 kW to 4,500 kW.

Fuel flexibility has all the time been the enterprise mannequin. DCD reported in May that 2G’s containerized merchandise span 50 kW to 2.5 MW, and that the majority of them run on biogas, pure fuel, hydrogen or propane with out altering the block. Feeding one a hydrogen-rich stream from an ammonia cracker isn’t a philosophical leap for a corporation that already sells pure-hydrogen models. It’s the identical combustion chemistry with a distinct supply technique.

The American numbers are what make 2G price watching. On July 30 the corporate confirmed second-quarter order intake of €422.4 million, in opposition to €54.1 million a 12 months earlier. US knowledge facilities alone accounted for €350.3 million of that, up from €8.3 million.

First-half order consumption got here to €479.4 million, and administration held full-year income steerage at €490 million with an EBIT margin of 9.5 to 10.5 %. Pairing with Amogy offers that gross sales crew an ammonia story to inform on high of the fuel story it was already telling very efficiently.

Switching again to methane is the precise gross sales pitch

Here is the place the pitch will get intelligent and somewhat hedged on the similar time. Low-carbon ammonia at hyperscale quantity doesn’t exist but. So neither firm is asking an information middle operator to guess a multi-billion-dollar campus on a gasoline with no mature market behind it.

What they’re promoting is one bodily machine with two modes. Install it now, run methane right now as a result of that’s what the pipeline delivers, and pipe within the reformer later if the ammonia market matures or the sustainability reporting begins biting.

Strip the press-release language off that and also you get a reasonably trustworthy proposition: purchase the engine, maintain the choice. 2G CEO Pablo Hofelich framed the collaboration as proof its engine platform can deal with a broad set of gasoline pathways, ammonia-derived hydrogen included.

Whether a hyperscaler pays a premium for optionality it could by no means train is a distinct query. Optionality is simpler to promote to an engineering crew than to a CFO.

Reciprocating engines are quietly successful the pace race

Amogy and 2G aren’t chasing a distinct segment. They’re chasing the fastest-growing nook of American energy procurement.

Enverus Intelligence Research printed a forecast on August 11 placing 25.5 GW of new US industrial demand on behind-the-meter generation between 2026 and 2030. Data facilities account for 22.5 GW of that, or 88 %, which works out to roughly 36 % of all US knowledge middle capability added over the interval.

PJM and ERCOT collectively make up practically half of it, at 7.5 GW and 5.0 GW respectively. Serving all of that may take 31.6 GW of era, and 29.6 GW of it’s gas-fired. The expertise cut up is the half that ought to curiosity 2G. Piston engines, gasoline cells and generators within the small and medium body sizes take 61 % of that gas-fired whole: 7.1 GW, 4.8 GW and 6.2 GW respectively.

EIR’s acknowledged purpose is pace. Those machines normally attain industrial operation in beneath two years, the place a giant combined-cycle plant can take nearer to seven. EIR senior analyst Alex Nevokshonoff mentioned the faster-deploying choices carry “a significant timing advantage.”

Everyone with a clean-power pilot is chasing those self same consumers. FuelCell Energy mentioned in March that its development pipeline had grown 275 percent since February 2025, pushed overwhelmingly by knowledge facilities. The nuclear crowd is circling too, with shipping-container microreactors already on hyperscaler radar.

The fuel incumbents usually are not precisely quiet both. VoltaGrid has a 2.3 GW settlement with Oracle and a separate 1 GW cope with Vantage for containerized pure fuel programs, per DCD. That is the wall a container-scale ammonia unit is strolling as much as.

The ammonia itself continues to be the costly half

Here’s the awkward arithmetic. The IEA’s Ammonia Technology Roadmap places simply over 70 % of worldwide ammonia manufacturing on natural-gas steam reforming, with many of the the rest coming from coal gasification.

Cracking gas-derived ammonia again into hydrogen to keep away from burning fuel is a circle you possibly can draw however not shut. And the clear options value actual cash: the IEA pegs near-zero-emission ammonia routes at 10 to 100% costlier per metric ton than typical ones.

Supply is the opposite half. The company counted current and introduced near-zero-emission ammonia capability of practically 8 Mt due on-line by 2030, which amounted to roughly 3 % of what the world might make in 2020.

The Houston check didn’t contact any of that. It didn’t exhibit megawatt-scale steady operating, it didn’t repair a value per kilowatt-hour, and it didn’t conjure a low-carbon ammonia provide chain into existence.

What Amogy does have is a challenge checklist that has stopped wanting scattershot. In June it signed a supply agreement with Amun Energy, a GS Engineering & Construction subsidiary, for a 40 MW ammonia energy challenge in Pohang, South Korea, beginning with a 1 MW pilot and scaling by 2029.

A pilot with GreenHarvest in Taiwan is because of land at a big industrial energy person on the island across the flip of 2027. There’s a March memorandum of understanding with Hoku Infrastructure masking Japan and different Asian markets, and a December 2025 funding from Kinetics, a Karpowership initiative, aimed toward ammonia-powered floating crops that would feed floating AI knowledge facilities.

Floating knowledge facilities operating on cracked ammonia would have learn as satire in 2022. It now reveals up in precise press releases, alongside truck-mounted reactors pitched straight at server halls.

Running a fuel engine on hydrogen cut up out of ammonia just a few toes upstream is a legitimately intelligent piece of chemistry bolted to legitimately confirmed mechanical engineering. Selling it’s the tougher half. The engine works both approach, which is the strongest card Amogy holds, and likewise an admission that the gasoline it really desires to promote isn’t prepared for the shopper it really desires to land.