If validated, the parties have outlined a roadmap to scale up to 2 MW across NTS facilities by 2028
Decision Focus
Inlyte Energy, a California-based startup, announced plans in early 2026 to run two concurrent pilots of its iron-sodium battery technology: one inside a Tier IV colocation facility in Bern, Switzerland operated by NTS Colocation, and a second at a utility energy storage test site in Wilsonville, Alabama in collaboration with Southern Company. The operational signal for global data center energy teams is specific: a battery chemistry claiming thermal-runaway immunity and unusually high deployment density is entering live mission-critical and grid validation environments for the first time this year.
90-Second Brief
As the week closes, inlyte Energy is commissioning a 600 kWh iron-sodium battery installation at NTS Colocation’s Tier IV Bern facility, targeted for commissioning by end-2026. If validated, the parties have outlined a roadmap to scale up to 2 MW across NTS facilities by 2028. Simultaneously, Inlyte is running a utility-scale demonstration with Southern Company in Alabama for a minimum one-year evaluation. The technology is positioned as an alternative to both diesel backup generators and conventional lithium-ion BESS, directly targeting the safety constraints and community-opposition dynamics that now complicate diesel infrastructure approvals.
What Is Really Happening?
Diesel generators remain the dominant backup power solution across global data center portfolios, and their continued use is increasingly difficult to defend in European and dense urban markets. Community opposition to on-site fuel storage and standby exhaust emissions is intensifying. Regulators in several jurisdictions are tightening emissions requirements for backup generation, and carbon reporting obligations make diesel a visible liability even when it runs infrequently. Meanwhile, lithium-ion BESS has scaled rapidly for grid applications but carries thermal runaway risk that constrains deployment density in mission-critical environments—fire suppression requirements and building-code spacing rules add significant capital and footprint overhead.
Iron-sodium chemistry occupies a different position on the tradeoff curve. According to Inlyte’s own claims, its design produces no fire or thermal runaway risk, and modules can continue operating through internal cell failures. These are manufacturer claims, not independently verified field outcomes, but they are the design basis for the density figure the company cites: up to a gigawatt-hour of storage capacity per acre. If that specification holds under operational conditions, it would represent a materially different land-use and siting calculus than lithium-ion installations at comparable scales.
The Swiss deployment context raises the validation stakes further. NTS operates on 100% renewable energy with waste heat recovery. That profile removes any carbon offset argument for retaining diesel and increases the operational pressure to demonstrate that an iron-sodium system can meet Tier IV uptime requirements cleanly. The Alabama utility test adds a second validation axis: grid-scale performance under utility evaluation criteria rather than data center SLA conditions.
Why It Matters for Global Heads of Data Center Energy
The diesel question is not resolved by this pilot, but the pilot creates a reference point that energy procurement teams should formally track. For facilities in European markets where diesel permits face tightening and community relations affect site approvals, an iron-sodium system that clears Tier IV validation would constitute a viable procurement option within a medium-term planning window—not an immediate replacement program, but an alternative that moves from concept to field-tested within a 12-to-24-month horizon.
The density claim warrants specific attention. If the GWh-per-acre specification is confirmed operationally, iron-sodium storage would fit behind-the-meter into urban campus footprints where lithium-ion is already constrained by fire-risk spacing and local building codes. That dynamic changes the economics of behind-the-meter storage in constrained geographies—Frankfurt, London, Singapore, Tokyo—where both land and power availability are binding constraints. It could also affect how energy teams size battery commitments in colocation environments where landlord and tenant fire-safety rules limit lithium-ion expansion.
The dual-pilot structure signals a second ambition beyond backup replacement: grid services. An asset that satisfies both Tier IV backup requirements and utility evaluation criteria is a candidate for VPP and demand-response participation. For energy procurement teams already building grid-services positions, a single battery chemistry that crosses both qualification thresholds would simplify the BESS strategy stack considerably.
Forward View
Three fronts are worth monitoring. First, operational data from the Bern installation, expected by end-2026, will be the first field dataset on iron-sodium performance under Tier IV SLAs—covering cycle behavior, response latency, and failure-mode outcomes under real load. Second, the Southern Company evaluation, running for at least one year, will generate utility-grade metrics on round-trip efficiency and degradation that no public source currently provides. Third, track whether the 2 MW NTS roadmap through 2028 holds or accelerates; an early contract expansion would signal that internal validation milestones are being cleared ahead of schedule and would warrant a closer procurement review.
What Is Still Uncertain
The evidence base at this point is narrow and largely sourced from company announcements. Installed cost per kWh is undisclosed, and any cost-competitiveness claim against lithium-ion—whose pricing has continued to decline—requires independent benchmarking that does not yet exist in the public record. Round-trip efficiency figures are not confirmed. The GWh-per-acre density specification is a design-basis claim, not a validated operational outcome. The Southern Company collaboration appears in secondary sources without a primary press release, and its evaluation criteria and contractual structure are unspecified. Critically, the technology’s discharge performance under the high cycling frequency and rapid response conditions that govern actual data center backup utility has not been publicly reported.
One Question for Your Team
If the Bern Tier IV installation delivers verified performance data by year-end 2026, does your current backup generation strategy include a defined evaluation framework for qualifying an alternative battery chemistry into your medium-term site procurement roadmap—or does iron-sodium remain outside your tracking horizon until it clears a larger reference deployment?
Sources
- Indexbox — Sodium Battery Pilots for Data Centers and Grid Storage in 2026 (Link)
