Energy / Grid storage · Deep dive

Peak Energy

Sodium-ion, passively cooled grid-storage systems for utilities and IPPs — a lithium-free BESS built by Tesla, Enovix and Northvolt veterans betting that salt plus a factory in Sacramento beats cheap Chinese LFP.

emerging

The question that decides it: Sodium-ion cells today cost more per kWh than Chinese LFP and store roughly 30% less energy per unit of volume, so Peak's whole case rests on three offsets stacking: passive cooling that strips out the failure-prone fans and pumps, chemistry that is safer and materials-abundant, and a domestic supply chain shielded by tariffs and production credits. Does that bundle actually beat delivered-cost LFP on a usable-kWh basis for a utility buyer — and can Peak stand up 4 GWh of Sacramento manufacturing and hit its 2027 Jupiter deliveries before the tariff-and-subsidy window that makes the math work narrows or the LFP price gap widens?

My take ★★ Very interesting ★ Interesting − Maybe ✗ Not interesting

Snapshot

Peak Energy is a Denver-based battery company building grid-scale storage systems on sodium-ion chemistry rather than lithium, and cooling them passively — no fans, pumps or vents. Founded in 2023 by veterans of Tesla, Enovix and Northvolt, it delivered what it calls America’s first grid-scale sodium-ion battery in mid-2025, has signed $1 billion-plus in commercial agreements with developers including Jupiter Power, RWE and Energy Vault, and in July 2026 announced a 4 GWh factory in Sacramento — the first U.S. plant dedicated to sodium-ion grid storage. It is raising an ~$80 million Series B at roughly $475 million pre-money (Axios, July 2026) on top of $65 million previously raised, and in June 2026 added a strategic investment and cell-development partnership with General Motors. It is genuinely pre-scale: ~170 employees, no product revenue, first shipments slated for Q1 2027. The thesis — abundant sodium plus passive cooling beats lithium on total cost and safety for stationary storage — is a bet the physics makes far from obvious.

Founding story

Peak was founded in June 2023 by Landon Mossburg and Cameron Dales, two operators who had each spent a career on the hard part of the battery business — building and selling at scale. Mossburg was an engineering director at Tesla and then President of North America and Chief Automation Officer at Northvolt, Europe’s marquee (and ultimately troubled) battery maker, standing up U.S. cell production. Dales came from silicon-anode maker Enovix, where as GM and CCO he built the commercial funnel and helped take it public in 2021. The name comes from the Colorado peaks behind their Denver base.

The insight was picking the right chemistry for a specific job, not inventing one. Grid storage does not need the energy density a car or phone demands — a battery in a field cares about cost per kWh, safety, cycle life and supply security, not weight. Sodium is abundant, needs no lithium, cobalt or nickel, and tolerates heat and full discharge better than lithium. Peak bet that what held sodium-ion back was not the cells but the willingness to build a purpose-designed U.S. grid system around them — and that people who had scaled Tesla, Northvolt and Enovix could do the manufacturing the science never finishes.

How it works

A sodium-ion cell shuttles sodium ions between electrodes just as a lithium cell shuttles lithium — same mechanism, cheaper and heavier ion. Peak uses a sodium iron phosphate pyrophosphate (NFPP) cathode; its July 2025 pilot was billed as the world’s largest NFPP system. Sodium’s penalty is density — the ion is ~3x heavier, so cells store about 30% less energy per unit volume and a system needs more space and material for the same MWh. Its payoff is thermal: the cells tolerate heat and abuse far better, with much lower fire and thermal-runaway risk.

That thermal margin enables Peak’s real differentiator — passive cooling. A conventional lithium BESS is stuffed with fans, pumps, liquid loops and HVAC; Peak says those active systems drive 85%+ of historical BESS failures and burn parasitic energy. Because sodium runs cooler, Peak removes them entirely — its 2025 pilot was, per the company, the first MWh-scale battery running purely on passive cooling. Fewer moving parts is a three-way pitch: lower upfront cost, higher uptime (Peak claims 99%+), and less parasitic energy loss.

Product and business overview

Peak sells a complete system, not bare cells — the sodium-ion packs, the passively cooled enclosure and the balance-of-system — to utilities and IPPs as a turnkey BESS for daily cycling (the 2-8 hour range that dominates U.S. procurement).

The June 2026 GM partnership reshaped the supply chain: GM develops and will manufacture the sodium cells in its Michigan labs and keeps exclusive manufacturing rights, while Peak builds the systems. A real trade — Peak offloads the hardest, most capital-intensive step to an automaker with manufacturing muscle and a GM Ventures check, but no longer fully owns its key input. The Sacramento factory announced in July 2026 handles system assembly and, per Peak’s roadmap, U.S. cell production — 183,000 sq ft, up to 4 GWh a year.

Business model and pricing

The model is system sales — capital equipment sold to utilities and IPPs, booked on delivery with service and warranty attached. Peak publishes no per-kWh price; it markets relative cost: a ~20% system-cost reduction versus conventional active-cooled LFP, from stripping out cooling hardware and parasitic load, plus lower lifecycle cost from higher uptime and safer, longer-lived cells.

The uncomfortable reality sits underneath: at the bare-cell level sodium is not yet cheaper than lithium — early-2026 figures (PatSnap) put sodium cells ~$70/kWh against Chinese LFP at ~$40-45/kWh. Peak’s cost argument therefore cannot be won on the cell; it has to be won at the system and lifecycle level (no cooling, more uptime, fewer failures) and on policy (domestic manufacturing shielded by tariffs and credits). A legitimate argument — but a total-cost story a buyer must be talked into, not a lower invoice.

Traction over time

Peak can ship a working unit and sign large forward contracts, but has not delivered at commercial scale. The $1B+ in “commercial agreements” is real but mostly forward: the Jupiter deal (up to 4.75 GWh, could exceed $500M) runs 2027-2030 with a ~720 MWh first phase; Energy Vault’s 1.5 GWh is a development agreement; RWE is a 3.1 MWh pilot. Headcount is ~168-177 (mid-2026), product revenue effectively zero. Every number is a promise with a delivery date, not a shipped result.

Market analysis

The demand backdrop is as strong as any in energy. Grand View Research pegged grid-scale battery storage at $10.69B in 2024, reaching ~$44B by 2030 at a 27% CAGR. BloombergNEF said the sector crossed into the “100-gigawatt era” in 2025 — ~112 GW added (up 48%), ~307 GWh of batteries deployed globally — and projects the global market growing ~15-fold across the decade, with the U.S. adding nearly 67 GW / ~284 GWh of utility-scale batteries over five years.

Peak does not sell into that whole TAM — only the slice willing to pay up for a non-lithium, domestically made, passively cooled system. That slice’s size turns on policy and the sodium-vs-LFP cost gap. If U.S. tariffs on Chinese batteries (Section 301 rates rising, LFP facing much higher stacked duties in 2026) and the IRA’s $35/kWh cell credit hold, domestic sodium carves out a defensible niche. If those soften or Chinese LFP keeps falling, the addressable slice shrinks toward zero however big the overall market grows.

Competitive intel

Peak is small and pre-scale among giants and cautionary tales (full profiles above). Fluence (~$2.3B FY2025 revenue) and Tesla Energy (Megapack 3, 50%+ annual growth) own the utility relationships and scale, run mostly on LFP, and could adopt sodium the instant it pencils — Peak’s tech lead is time-limited. CATL and BYD are the existential threat on the same chemistry, targeting cell costs (CATL’s Naxtra, claimed ~$19/kWh at scale) no U.S. startup can touch; the whole domestic-supply thesis exists to keep those cells out via tariffs. Form Energy attacks the same “cheap, domestic, non-lithium” story from the long-duration end with far more capital ($1.2B+, ~$3.42B valuation). And Natron Energy, the other U.S. sodium name, shut down in September 2025 after failing to raise — the clearest proof this path can hit a wall.

Where Peak wins: the only U.S. company with a shipped grid-scale sodium-ion system, passive cooling as a real differentiator, a team that has scaled battery manufacturing, and a GM deal that de-risks cell production. Where it is exposed: a rounding error next to Fluence, Tesla and the Chinese majors, with a product-and-policy edge, not yet a cost or scale moat.

History and evolution

• Jun-Oct 2023 — Founded in Denver; launches from stealth with $10M led by Eclipse and TDK Ventures.
• Jul 2024 — $55M Series A led by Temasek’s Xora Innovation.
• Jul 2025 — Ships the first grid-scale sodium-ion BESS in the U.S., a 3.5 MWh passively cooled system.
• Sep 2025 — System online at the SolarTAC facility in Watkins, Colorado, in a nine-party utility/IPP pilot. (Same month, rival Natron collapses.)
• Nov 2025 — Jupiter Power contract for up to 4.75 GWh through 2027-2030.
• Feb-Mar 2026 — 1.5 GWh Energy Vault development deal; 3.1 MWh RWE pilot in Wisconsin.
• Jun 2026 — GM partnership and investment: GM develops and manufactures the cells, Peak builds the systems.
• Jul 2026 — Selects Sacramento for a 4 GWh factory (~$71M, shipments Q1 2027); reported ~$80M Series B at ~$475M pre.

No public crises yet — but the company is barely two years old, and the hardest chapters (the factory, 2027 deliveries, holding cost against LFP) are all ahead.

What people say

The case for. Trade coverage (Electrek, pv magazine, Energy-Storage.news, 2025-2026) treats Peak as the credible U.S. sodium-ion story after Natron’s failure — it shipped a working, grid-scale, passively cooled system no other domestic sodium player has. Cutting the fans, pumps and HVAC behind most BESS failures is a genuine reliability argument. The customer roster is unusually strong for a pre-revenue company, and the GM partnership was read industry-wide (Utility Dive, Latitude Media, June 2026) as validation that a Tier-1 automaker will back sodium-ion. The founders’ Tesla/Northvolt/Enovix pedigree buys credibility on what storage startups most often fail at: scaling production.

The complaints. The skeptic case is mostly engineering and economics, and it is sharp. Lower energy density means more space, enclosure and material per usable kWh — a structural handicap passive cooling must overcome before parity. Worse, at the cell level sodium is currently more expensive than Chinese LFP (~$70 vs ~$40-45/kWh, early 2026), so the “cheaper than lithium” line rests on lifecycle math a buyer takes on faith plus tariffs and subsidies that could change. Analysts (C&EN, IEA, 2025-2026) note sodium has not undercut LFP in most applications because lithium prices simply are not high enough. Natron’s September 2025 shutdown — same chemistry, same domestic-factory ambition, $1.4B plant abandoned for lack of capital — is proof the scale-up can fail. The pointed bear case: U.S. sodium-ion may be a solution chasing a subsidy, viable only while tariffs and the IRA credit prop it up. And Peak carries execution risk — never manufactured at scale, does not fully own its cell (GM does), and has signed far more GWh than it can yet build.

Outlook: the open question

Peak works if the total-system and lifecycle economics of passive-cooled sodium — not the raw cell price — beat delivered LFP for a utility buyer, and if it builds the Sacramento factory and hits its 2027 Jupiter deliveries while tariffs and credits still tilt the field toward domestic supply. It fails if sodium cells stay costlier than LFP without a lifecycle edge to close the gap, if the scale-up stalls the way Natron’s did, or if the policy shield that makes ‘made in America’ sodium pencil erodes before Peak reaches volume. The bet is genuinely two-sided.

The bull case is coherent: grid storage is the fastest-growing slice of the energy transition, sodium’s abundance and safety are real, passive cooling is a defensible edge, and Peak has the manufacturing pedigree, offtake pipeline and GM partnership few storage startups can match. The bear case rests on hard numbers: sodium stores less and, today, costs more per cell than LFP; the cost story leans on tariffs and subsidies, not intrinsic economics; and the last U.S. company to try this exact thing is dead. What settles it for Peak: the Sacramento plant commissioning on schedule, the 2027 Jupiter phase delivering at a system cost that undercuts LFP-plus-cooling on a usable-kWh basis, and net-new offtake signed at those prices without a subsidy crutch. What settles it against: slipping timelines, sodium cell costs failing to fall toward LFP, or a policy shift that readmits cheap Chinese LFP to U.S. procurement. With ~$65M raised and an ~$80M round in market against rivals measured in billions, Peak cannot be wrong for long — the next 18 months of execution and delivered cost decide which company this is.

How a challenger would attack it

Sell the passive-cooling insight without the sodium tax. Peak’s most defensible claim is not the chemistry — it is that active cooling drives 85%+ of BESS failures. But that insight is separable from sodium: a challenger building simplified-thermal or immersion-cooled LFP systems captures most of the reliability pitch while keeping the $40-45/kWh Chinese cell against Peak’s ~$70 sodium cell and its 30% density penalty. Peak has to win a total-lifecycle argument the buyer takes on faith; the LFP challenger just shows a lower invoice. The second attack is the supply-chain seam Peak created for itself: GM develops and exclusively manufactures the cells, meaning Peak’s key input is hostage to an automaker’s shifting priorities — a competitor that vertically owns its cell line (or simply buys CATL Naxtra cells the moment tariff policy wobbles) has no such dependency to point at in a utility’s bankability review. Third, attack the order book: the $1B+ in agreements is almost entirely forward — Jupiter’s first phase is ~720 MWh starting 2027, Energy Vault is a development deal, RWE a 3.1 MWh pilot. A rival that ships commercial volume in 2026, before Sacramento is even commissioned, converts Peak’s customers’ delivery risk into its own pipeline — the exact anxiety Natron’s September 2025 collapse planted in every procurement team.

Same playbook, new buyer

Take heat-tolerant, passively cooled sodium where its physics is the headline, not the footnote. Peak sells into standard 2-8 hour U.S. utility procurement, where sodium’s density and cost penalties fight LFP head-on. The chemistry’s genuine edges — heat tolerance, no thermal runaway, full-discharge tolerance — are worth far more in markets Peak isn’t structured to serve. Hot-climate deployments (Gulf states, Indian and Australian grids, desert mining operations) punish active-cooled LFP with exactly the fan-and-HVAC failures Peak’s data cites, and parasitic cooling loads there are a real cost line, not a rounding error. Second: safety-constrained siting — urban substations, indoor commercial installations, sites near occupied buildings where fire codes and insurers penalize lithium — where a no-thermal-runaway system clears permitting LFP cannot. Third: military and critical-infrastructure procurement, where domestic supply chain is a requirement rather than a subsidy-dependent preference, and price sensitivity is lowest. Peak won’t pivot to these: its Sacramento capacity, Jupiter offtake and Series B story are committed to mainstream utility volume, GM’s exclusive cell rights constrain export and defense structuring, and a pre-revenue company racing a 2027 delivery deadline cannot requalify its product for three new certification regimes.

Sources and further reading

• Peak Energy Selects Sacramento to Build America’s First Sodium-Ion Grid Storage Factory (PR Newswire, July 8 2026)
• Peak Energy raising $80M as it plans battery factory in Sacramento (Axios, July 2026)
• Peak Energy to build 4 GWh grid-scale sodium-ion BESS factory in California (pv magazine USA, July 2026)
• Peak Energy and General Motors Partner to Scale Next-Generation Energy Storage (PR Newswire, June 2026)
• What does the GM-Peak Energy partnership mean for sodium-ion? (Latitude Media, June 2026)
• Peak Energy just shipped the US’s first grid-scale sodium-ion battery (Electrek, July 2025)
• Peak Energy to supply up to 4.75 GWh of sodium-ion batteries to Jupiter Power (ESS News, November 2025)
• Temasek’s Xora Innovation leads $55M Series A in Peak Energy (FinSMEs, July 2024)
• Sodium-ion batteries: Should we believe the hype? (C&EN, November 2025)
• Sodium-ion battery maker Natron Energy shuts down, halts $1.4B factory plans (Manufacturing Dive, September 2025)
• Grid-scale Battery Storage Market Size, Industry Report 2030 (Grand View Research, 2025)