What Happened ESS Tech has signed a letter of intent with Alsym Energy to add 8.5 GWh of sodium-ion battery cells and modules to its product portfolio. Alsym’s Na-Series batteries are non-flammable, developed using an AI-assisted platform, and sourced entirely from materials outside foreign entity of concern supply chains. The deal marks ESS’s first move into short and medium-duration storage, beyond its existing iron flow battery business.
Why It Matters Lithium-ion dominates grid storage today, but thermal runaway risk means installations require fire suppression systems and active cooling which adds cost and complexity. Sodium-ion sidesteps those requirements entirely. For utilities, data centres, and commercial buyers, that translates to lower infrastructure overhead and simpler site requirements. For ESS, it means being able to serve customers across multiple storage durations from a single portfolio.
Industry Context Energy storage demand is growing faster than lithium supply chains can safely scale. Sodium-ion offers a credible alternative — it uses abundant, widely available materials and carries no combustion risk, though it trades some energy density against lithium in the process. Chinese manufacturers currently lead sodium-ion production volumes, but Western supply chains are developing quickly, driven partly by domestic sourcing requirements attached to grid-scale project funding. Alsym’s use of non-foreign entity of concern materials directly addresses that procurement consideration.
Our Take ESS is repositioning from a long-duration specialist into a broader storage provider. The strategy is logical; serving multiple project types from one portfolio is a real competitive advantage as storage procurement scales up. The underlying bet is that buyers will value supply chain security and non-flammability enough to accept sodium-ion’s energy density tradeoff in applications where both chemistries would otherwise qualify. That’s a reasonable assumption in the current procurement environment, but it still needs to be proved out at scale.
