Non‑lithium long‑duration storage is moving from niche to necessary. Here’s how zinc and iron flow batteries fit into the next wave of green, AI‑driven infrastructure.
Most of the new power demand in the US isn’t coming from homes or cars anymore. It’s coming from data centers, AI clusters, and round‑the‑clock digital infrastructure that can’t afford a blackout or a price spike.
Here’s the thing about that kind of load: you can’t run it on solar panels and a couple of short‑duration lithium battery packs. You need long‑duration energy storage (LDES) that can hold clean energy for 8–10 hours or more, cycle hard every day, and do it for decades. That’s where non‑lithium batteries step in—and why companies like Eos Energy Enterprises and ESS Inc. matter for anyone serious about green technology.
This post looks at what their latest financials really say about the future of long‑duration storage, why US policy is suddenly a massive factor, and how all of this connects to the broader shift toward AI‑powered, low‑carbon infrastructure.
Non‑Lithium Storage Is Finally Moving From Niche to Necessary
Non‑lithium grid batteries—zinc and iron flow in this case—aren’t a science project anymore. They’re becoming a required part of the clean energy stack for three big reasons:
- Duration: They can deliver 6–12 hours or more of storage, where most lithium projects still sit at 2–4 hours.
- Cycling and lifetime: Designed to be cycled heavily every day for 20+ years with less degradation.
- Supply chain and safety: Made from abundant materials like iron, salt, and zinc, with lower fire risk and fewer geopolitical headaches.
The Energy Storage News report on Eos (zinc hybrid cathode batteries) and ESS Inc. (iron electrolyte flow batteries) puts real numbers behind that story.
- Eos Q3 2025 revenue: US$30.5 million, up 100% quarter‑over‑quarter.
- Eos 2025 guidance: US$150–160 million.
- Eos backlog: ~US$644.4 million, around 2.5GWh of orders.
By contrast:
- ESS Inc. Q3 2025 revenue: Just US$214,000, down 91% from Q2.
- ESS Inc. adjusted EBITDA loss: US$7.1 million (slightly improved from Q2).
Same macro trends, very different short‑term trajectories. But both are essentially betting on the same macro thesis:
Long‑duration, non‑lithium storage will be indispensable as grids absorb more renewables and as heavy loads—especially AI and hyperscale data centers—demand 24/7 clean power.
From a green technology perspective, that’s the real story: this isn’t a gadget market; it’s critical infrastructure.
Eos: Zinc Batteries Chasing Scale—and Profitability
Eos Energy Enterprises is probably the clearer commercial story right now.
What Eos is actually selling
Eos builds a zinc hybrid cathode battery branded as its Znyth technology. It’s designed for 3–12 hour storage applications where lithium either gets too expensive or wears out too fast.
The company is now on its third‑generation Znyth module, tuned for:
- Long lifetime under heavy daily cycling
- Robust performance in harsh conditions
- Lower material and manufacturing costs over time
This is exactly the kind of tech utilities want for long‑duration projects that pair with wind, solar, or serve data centers needing overnight coverage.
The growth signals that matter
Eos’ numbers show a company early in scale‑up mode but moving in the right direction:
- Backlog: US$644.4 million (2.5GWh) in customer orders.
- Pipeline: ~US$22.6 billion in commercial opportunities, ~91GWh.
- Cash on hand (end of Q3 2025): US$126.8 million.
Neither backlog nor pipeline equals cash today, but together they tell you one thing: there’s real, concrete demand for non‑lithium LDES, especially when projects can qualify for US incentives.
Manufacturing and policy tailwinds
Eos is moving aggressively on domestic manufacturing, which is incredibly smart under current US policy:
- Building up to 8GWh per year of production capacity in Marshall Township, Pennsylvania.
- Relocating HQ from New Jersey to Pennsylvania, with a planned US$352.9 million investment in lines and facilities.
- Securing ~US$22 million in state support.
- Previously one of the few companies to benefit from the US Department of Energy Loan Programs Office for domestic battery manufacturing.
Why does this matter for green technology buyers and developers?
Because the US clean energy incentive regime rewards domestic content and penalizes “foreign entities of concern” (FEOC). Zinc batteries built largely from US-sourced materials can:
- Qualify more easily for production tax credits (PTCs).
- Avoid FEOC‑related restrictions on investment tax credits (ITCs) starting in 2026.
- Bypass some of the new and upcoming tariffs on Chinese battery components and materials.
I’ve found that the projects that pencil out best over a 20‑year horizon are the ones designed from day one to fit the incentive code, not retrofit it. Eos leans straight into that reality.
Eos still operates at a loss (EBITDA loss: US$52.7 million in Q3, slightly wider than Q2), but that’s typical of manufacturing scale‑up. The core question isn’t “Are they profitable this quarter?” but “Do they have a realistic path there?” With a deep backlog, clear factory plan, and policy alignment, Eos at least has a plausible line of sight to profitable growth.
ESS Inc.: Iron Flow Batteries in Reset Mode
While Eos is ramping, ESS Inc. is in reset—but that doesn’t make it irrelevant. It makes it a case study.
A tough quarter masking a strategic pivot
ESS builds iron electrolyte flow batteries, which are especially well‑suited for 10‑hour‑plus applications. Think:
- Massive utility‑scale LDES
- Large‑scale integration with renewables
- Industrial or data center campuses that want long, predictable discharge windows
In 2025, ESS effectively hit “pause” on its earlier products:
- Energy Center (larger‑scale solution) and Energy Warehouse (C&I and utility) have been discontinued.
- The company is now focused on a next‑generation product: Energy Base—designed for very large, 10‑hour duration projects where size and duration justify higher upfront engineering.
This pivot comes at a cost:
- Q3 2025 revenue dropped to US$214,000, down 91% from Q2.
- ESS warned it needs additional debt or equity financing to meet near‑term cash needs.
- Unrestricted cash and equivalents at quarter end: US$3.5 million.
That’s why management has flagged “substantial doubt” about its ability to continue as a going concern without new funding.
Why Energy Base still matters for green technology
From a pure green tech lens, Energy Base is the interesting part:
- 10‑hour duration opens the door to true firm clean power—covering overnight gaps in solar or long lulls in wind.
- Flow batteries inherently separate power (MW) from energy (MWh) via tank sizing, which is attractive at large scale.
- Iron, salt, and water inputs make it resource‑abundant and non‑flammable, aligning with safety and ESG priorities.
ESS has started to re‑anchor this bet around real projects, not just roadmaps. A standout is the 50MWh Energy Base pilot with Salt River Project (SRP) in Arizona. That’s not yet a full commercial fleet, but it’s a serious test bed with a credible utility.
To stabilize finances, ESS closed a US$40 million financing with an investment fund managed by Yorkville Advisors Global, while also undergoing leadership changes—former CEO Eric Dresselhuys stepped down and Kelly Goodman, previously VP of legal, is now interim CEO.
Is this messy? Absolutely. But it’s also what innovation on the grid actually looks like: risky, capital‑intensive, and bumpy. If you’re a developer or corporate buyer, you shouldn’t ignore a tech like this entirely—you just need a disciplined risk framework:
- Use pilot and demonstration projects first.
- Structure contracts with performance milestones.
- Blend portfolios: pair bankable lithium with strategic non‑lithium LDES.
Policy, Tariffs, and FEOC Rules: The Real Tailwind
Both Eos and ESS believe that US policy is about to give non‑lithium batteries a real boost, and I think they’re right.
Why non‑lithium matters under US rules
The US tax and trade framework now actively shapes storage decisions. Three levers stand out:
-
Foreign Entity of Concern (FEOC) rules
- From 2026, projects with significant content from FEOCs lose access to certain clean energy tax credits.
- Many lithium battery materials and components are still heavily concentrated in China.
- Non‑lithium chemistries using zinc, iron, salt, and steel sourced domestically or from allied countries can avoid these restrictions.
-
Tariffs on Chinese batteries and components
- New and proposed tariffs increase the cost of importing certain battery products and inputs.
- Domestic non‑lithium manufacturers can become more cost‑competitive simply because they’re onshore.
-
Production and investment tax credits
- US law increasingly rewards domestic manufacturing and high domestic content.
- Eos and ESS both plan to lean into production tax credits (PTCs) as they scale US plants.
The withdrawal of the sodium‑sulfur (NAS) battery from grid‑scale markets (after BASF stepped back from a planned gigafactory) only sharpens the picture: there’s now a vacuum in non‑lithium LDES at scale. Zinc and iron flow are stepping into that gap.
This matters because green technology isn’t just about physics and chemistry anymore. It’s about policy‑shaped economics. The winners will be the ones whose tech and supply chains line up with that reality.
Data Centers, AI, and the New Load Problem
The biggest wildcard—and the biggest opportunity—for long‑duration storage is load growth from data centers, especially AI‑driven ones.
Data centers today are:
- Growing at double‑digit annual rates in many US regions.
- Concentrating huge demand into specific grid nodes.
- Under pressure from customers and regulators to secure 24/7 clean energy, not just annual offsets.
Eos and ESS both argue that their technologies align well with this next wave of demand. I agree, for a few reasons:
- 10‑hour duration matches real‑world patterns. Solar peaks midday; AI clusters may be busiest at night. You need storage that can arbitrage that gap.
- Reliability trumps everything. Hyperscalers don’t want to babysit asset degradation curves every few years. Long‑life non‑lithium systems fit better into a 20‑year planning horizon.
- ESG scrutiny is intensifying. It’s easier to tell a compelling ESG story around abundant, non‑flammable, domestically sourced materials than around opaque mineral supply chains.
For businesses planning their green technology strategy, that leads to a very practical question:
How do you design an energy stack for a data‑intensive operation that balances cost, policy, and real‑world reliability?
A realistic answer usually includes:
- Short‑duration lithium for fast response and frequency services.
- Non‑lithium LDES (zinc, iron flow, thermal, etc.) for long‑duration shifting.
- Renewables plus firming (wind, solar, and possibly clean firm generation) to anchor the portfolio.
- AI‑driven optimization to orchestrate all of the above across price, carbon intensity, and reliability.
This is exactly where our broader Green Technology series sits: AI isn’t separate from clean energy—it’s the brain that keeps all these heterogeneous assets working together.
How to Act on This If You’re Planning Projects
If you’re a developer, utility, or corporate buyer trying to make sense of all this, here’s a pragmatic way to use what Eos and ESS are signaling.
1. Treat non‑lithium as a portfolio layer, not a silver bullet
Non‑lithium LDES shouldn’t replace lithium everywhere. It should complement it where duration, cycling, or policy justify it.
Good fits include:
- 6–12 hour storage paired with large solar or wind projects.
- Nodes with severe congestion or curtailment.
- Data centers or industrial campuses aiming for 24/7 clean power profiles.
2. Design around 2026 policy, not just 2025 pricing
FEOC rules and tariff structures are about to reshape the cost stack. When you’re modeling projects:
- Stress‑test revenue and capex under different policy scenarios.
- Ask explicitly how domestic content and sourcing choices affect tax credit eligibility.
- Consider non‑lithium options specifically for ITC/PTC‑sensitive projects.
3. Pilot first, scale fast if it works
For newer chemistries like iron flow or advanced zinc systems:
- Start with pilots tied to real operational use, not just R&D.
- Bake performance guarantees into contracts.
- Use pilots to build in‑house expertise, then roll out larger phases if results match the model.
4. Use AI and analytics to get real value from LDES
A long‑duration battery is only as smart as its dispatch strategy. This is where AI‑driven energy management is not optional anymore:
- Forecast prices, loads, and renewable output.
- Optimize when to charge, discharge, or stay idle.
- Align dispatch with both financial metrics and carbon intensity of the grid.
That’s the bridge between LDES as a cost center and LDES as a profit‑and‑sustainability engine.
Non‑lithium long‑duration storage is entering a sorting phase: some players will scale, others will consolidate or disappear. Eos looks like a company leaning hard into scale and policy alignment; ESS looks like a higher‑risk, higher‑upside bet on ultra‑long‑duration iron flow.
For anyone serious about green technology—especially those building or supplying energy‑hungry, AI‑driven infrastructure—the right move isn’t to wait for a “winner.” It’s to start designing portfolios, pilots, and procurement strategies that assume non‑lithium LDES will be part of the mix.
The grid of the 2030s won’t be powered by one battery chemistry. It’ll be an ecosystem—lithium for speed, non‑lithium for duration, AI for orchestration. The decisions you make on projects in the next 12–24 months will decide how smoothly you fit into that future.