This Week in Voltage
The NRC just handed nuclear another historic milestone — a construction permit for TerraPower's Natrium advanced reactor in Kemmerer, Wyoming, the first ever issued for a commercial non-light-water reactor. That's the headline. But the deeper story this week is what's happening in the storage sector, because the economics that will make the Natrium reactor — and every other intermittency-dependent grid asset — actually work at scale just locked in.
Lithium-ion battery costs have found a floor. The question now is whether the grid can absorb what's coming.
Deep Charge: The $25K Floor and What It Actually Means
Here's the number that matters: lithium carbonate spot prices have stabilized near $25,000 per metric ton as of May 2026. Not crashing. Not spiking. Stabilizing — and that distinction is everything for grid-scale storage economics.
For years, battery storage investment was hostage to commodity whiplash. Prices surged to irrational highs in the early 2020s, then cratered during the supply gluts of 2024. Project developers couldn't underwrite long-term power purchase agreements when their primary input cost could swing 60% in eighteen months. The result: capital sat on the sidelines, interconnection queues backed up, and the grid buildout lagged the renewable deployment it was supposed to support.
That dynamic is changing. The structural driver isn't EV demand anymore — it's stationary storage. Demand for lithium in storage applications jumped 71% year-over-year in 2025, and the BESS sector is projected to expand another 55% this year. Unlike consumer EV sales, which wobble with interest rates and subsidy calendars, utility-scale storage projects are locked to long-term infrastructure mandates. That's a fundamentally different demand signal — steadier, less cyclical, and increasingly divorced from the quarterly delivery reports of passenger car manufacturers.
The market has noticed. The U.S. added a record 57.6 GWh of new battery energy storage capacity in 2025, bringing total deployed capacity to 166.1 GWh. Developers are now planning to add 24 GW of utility-scale storage in 2026 — a massive jump from the 15 GW added last year, per EIA data. By end of Q1 2027, the EIA projects total U.S. battery storage capacity will exceed 67 GW, up from 44.6 GW today.
This is what a cost floor enables: long-duration planning. When developers can model input costs with reasonable confidence, they can sign PPAs, secure financing, and break ground. The $25K floor isn't a ceiling on ambition — it's the foundation that makes the math work.
The Bottleneck Hasn't Moved
None of this means the path is clear. Battery storage firms are seeing surging interest from AI data centers — behind-the-meter systems that manage demand spikes, cover outages, and reduce dependence on diesel backup — but the interconnection queue problem is metastasizing. Connecting to the grid can take three to seven years in parts of the U.S., per Reuters reporting. Data centers can be built in 18 to 24 months. The math on that gap is brutal.
Meanwhile, Goldman Sachs Research projects U.S. data center power demand climbing from 31 GW in 2025 to 66 GW by 2027 — with the Mid-Atlantic, Texas, and Mid-Continent markets each individually scheduled to add more capacity in 2027 than the entire nation added in 2025. That demand is real and accelerating. The storage economics to serve it are now viable. The wire connecting the two remains the wall.
By the Numbers
- 57.6 GWh — Record U.S. battery storage capacity added in 2025 (Reuters)
- 24 GW — Utility-scale storage planned for U.S. grid addition in 2026 (EIA via pv magazine)
- 110 GWh — Projected annual U.S. battery storage deployments by 2030 (Reuters)
- $25,000/ton — Lithium carbonate spot price floor as of May 2026 (Skillings)
- 66 GW — Projected U.S. data center power demand by 2027 (Goldman Sachs)
What We're Fighting For
A stable lithium price floor is not a commodity story. It's a civilizational infrastructure story. Every gigawatt-hour of grid-scale storage we deploy is a ratchet click upward — more renewable capacity that can actually dispatch on demand, more data centers that can run without diesel backup, more of the grid that operates on electrons instead of combustion.
The NRC's Natrium permit and the storage buildout are the same story told from different ends of the timeline. Nuclear gives us the baseload spine; batteries give us the flexibility layer. Together, they're the architecture of a grid that doesn't apologize for the weather.
Watch the FERC interconnection queue reform timeline. That's the next constraint that needs to break — and when it does, the $25K floor becomes the starting gun.
