Mainspring Energy on hydrogen’s role in long duration energy storage

In this edition of LDES Voices, Craig Gordon, Head of Business Development and Policy at LDES Council member Mainspring Energy, shares Mainspring’s perspective on the role of hydrogen in the long duration energy storage conversation. Drawing on the company’s project with Verdagy in Moss Landing, California, he explores the potential of hydrogen-to-power for LDES and the policy changes Mainspring believes could support its deployment.

About the author:

Craig Gordon is Head of Business Development and Policy at Mainspring Energy, bringing more than 20 years of energy industry experience. This includes twelve years at developer Invenergy, where he led domestic and international government and regulatory affairs and helped form the American Clean Power Association. Earlier in his career, he spent eight years at a Midwestern utility working in credit risk, wholesale power trading, generation dispatch, and renewable energy policy, and he holds a B.S. in geology from Wheaton College and an MBA from Washington University in St. Louis. 

Long duration energy storage is having a moment. For those who follow the LDES market closely, this may seem like an inevitable development. From the growing influx of intermittent renewable generation to skyrocketing load growth and constrained transmission systems, the pressing need for LDES has been clear for some time.

That understanding is increasingly being translated into new investments, commercial deals, and regulatory activity. According to a recent article in Latitude Media citing data from the energy intelligence platform Currence, the sector is experiencing “a step change” in deployment, with hundreds of megawatts of LDES expected to come online this year compared to just 33 megawatts in 2025.

The technologies being built this year are familiar to those with even a passing knowledge of the electricity sector. Currence’s data shows that about two-thirds of that 2026 capacity will be lithium-ion. A variety of other storage technologies, including iron-air and vanadium-flow batteries, carbon-dioxide storage, and compressed air storage, are also progressing toward commercialization.

Hydrogen is absent in the LDES conversation

Conspicuously absent from that list is hydrogen. In the early 2020s, hydrogen attracted enormous investment and widespread policy support because it was viewed as a versatile tool for tackling hard-to-decarbonize emissions from heavy industry and long-distance transportation. According to the International Energy Agency, 60 nations had developed national hydrogen strategies by 2024. The Hydrogen Council estimated that announced investment in hydrogen projects through 2030 had reached $680 billion.

For a variety of reasons including high production costs and infrastructure gaps, hydrogen has struggled to fully translate that support and momentum into actual projects. Yet there is a compelling case for hydrogen’s role in LDES, one that requires expanding traditional notions of what constitutes long duration energy storage. Hydrogen LDES will generally have lower roundtrip efficiency than other storage technologies. But because adding storage hours means adding tank capacity rather than more conversion equipment, its costs rise slowly with duration. A 2023 Idaho National Laboratory analysis found hydrogen-to-power systems had lower levelized costs than lithium-ion batteries for durations beyond roughly 7 to 12 hours.

Understanding why hydrogen can and should qualify as LDES also helps explain why it is often overlooked. Most energy storage technologies store and release energy in the same device and at the same location. By contrast, hydrogen typically becomes a long duration energy storage resource when three things happen in sequence. First, electricity powers an electrolyzer to produce hydrogen. Next, the hydrogen is stored in tanks, underground caverns, or pipelines. Finally, when power is needed, the stored hydrogen is converted back to electricity in a fuel cell or generator. Like a battery, the system stores energy in chemical form and releases it as electricity on demand. The difference is that each step happens in separate equipment, sometimes at separate locations.

Until recently, few hydrogen-to-power options combined the dispatchability, efficiency, and low emissions we expect from energy storage. Fuel cells are efficient and clean. But solid oxide fuel cells, the most widely deployed type for stationary power, have traditionally operated as steady baseload generators, and their operating limits constrain how quickly they can change output. PEM fuel cells respond faster but have mainly been used for backup power. And turbines designed for natural gas often require combustor modifications to run on high hydrogen blends, and they generate higher NOx emissions without costly abatement equipment.

However, linear generators are showing why they deserve to be part of the LDES conversation. Because they use a flameless, low-temperature reaction rather than conventional combustion, linear generators produce near-zero NOx emissions while remaining efficient, fully dispatchable, and well positioned to help states meet their LDES and clean energy goals.

Moss Landing makes the case for hydrogen-to-power

A collaborative R&D project in Moss Landing, California demonstrates the value hydrogen can deliver in the power system. Starting in 2025, electrolyzer manufacturer Verdagy began fueling a Mainspring linear generator onsite with hydrogen produced at the facility.

The project has provided valuable technical and business case lessons about co-locating hydrogen production with a linear generator. For example, Verdagy had a specific rationale for the combination: it wanted to capture the full value of the hydrogen produced by its electrolyzers. This is more challenging than it may seem because converting hydrogen to electricity only makes sense if the generation technology operates efficiently and can dispatch power on demand even when the fuel supply fluctuates.

From a technical and operational perspective, the Moss Landing project demonstrates that joining an electrolyzer and a linear generator works in a real-world industrial environment. The generator has already logged thousands of hours of runtime and generated over 100 megawatt-hours of clean power. The project is forecast to deliver Verdagy more than $100,000 in annual savings compared to electricity purchased from the grid.

The project also provides a business model worth replicating. At a time when rising demand is straining grid capacity and interconnection queues, an onsite generator paired with an electrolyzer can be up and running in months, far faster than the multi-year wait for many new grid connections. The combination also ensures Verdagy can maximize the value of the hydrogen it produces. The company produces hydrogen for multiple uses, and the linear generator provides an additional value stream by converting what might otherwise be unused hydrogen into valuable electricity. Co-locating the two assets also eliminates hydrogen transportation costs and simplifies operations.

A straightforward fix to unlock more LDES

With onsite hydrogen storage added, a project like Moss Landing could serve 12-hour, 24-hour, or even 100-hour applications, which utilities, data centers and other buyers increasingly want, provided state programs recognize hydrogen-to-power as storage. To fully unlock hydrogen-to-power as an LDES resource at scale across regions, three policy steps are needed.

First, policymakers and regulators must explicitly recognize hydrogen as a qualifying LDES resource. Progress across the country now is uneven. Because hydrogen can be produced through different methods, states may specify which types qualify and if there are any limitations on when it can be produced. That could mean all hydrogen qualifies, or only hydrogen produced using renewable or low-carbon energy. Since other storage types can charge anytime from the grid, we recommend that hydrogen production from grid power shouldn’t face new, added barriers.

Second, states should clarify which generation technologies qualify alongside hydrogen. One strength of California’s AB1373, signed into law in October 2023, is explicit guidance qualifying “non-combustion” technologies only. That draws a clear line between conventional engines and non-combustion technologies such as fuel cells. Linear generators, which use a flameless, low-temperature reaction, should be recognized on the same side of that line as fuel cells. Making that distinction explicit removes ambiguity for developers and investors evaluating whether to move forward.

Third, procurement agencies and utilities need a clear mandate to act. Under AB1373, the CPUC determined in 2024 that the Department of Water Resources should procure up to 2 gigawatts of LDES and formally asked DWR to begin in early 2025. As of mid-2026, however, DWR was still setting up its central procurement function, and no LDES solicitation had been issued. Further, AB1373 limits central procurement to resources with development timelines of more than five years, which excludes technologies that can be built quickly, such as Mainspring’s. States borrowing AB1373’s technology standards should leave that restriction behind.

More productively, a February 2026 CPUC order requires California’s load-serving entities, including investor-owned utilities and community choice aggregators, to procure 6,000 megawatts of new capacity by 2032. At least 1,500 megawatts must come from clean firm resources or long duration storage of eight hours or more. In 2026, Virginia enacted legislation directing its two largest utilities to seek approval for roughly 4.5 gigawatts of storage lasting 10 hours or more by 2045, and New York lawmakers introduced a bill that would set a 3 gigawatt LDES target by 2030. These frameworks should explicitly include hydrogen and the generation technologies capable of optimally converting it to power.

Verdagy’s Moss Landing project demonstrates that the technology and business case for hydrogen-to-power are real. The policy frameworks that would allow it to compete are within reach. Acting on them is a straightforward step that would unlock clean, dispatchable, long duration storage options that states are actively trying to incentivize.