SK on Locks in $1.1bn US Storage Deal as Korean Battery Makers Pivot

A South Korean battery maker has just locked in a deal worth roughly $1.1 billion to supply US energy storage systems, at the very moment its electric vehicle business is running cold.

That company is SK On, and its new agreement with NeoVolta Power is not an isolated bet. It reads as evidence of a fast-moving reorientation across the industry, where Korea’s three largest battery manufacturers are converting EV production lines to energy storage output on US soil. EV orders have softened enough to make idle capacity a problem, and grid demand has grown enough to make it an opportunity.

The timing is notable: this agreement surfaced the same day as the Argus Media report detailing it, making this among the first published accounts of the arrangement.

What the deal reveals matters more than the deal itself. It shows where battery manufacturing investment and energy storage supply chain activity is concentrating in the US over the next five years, which companies are moving, where the capacity is being built, and what structural forces are pushing the shift to happen now rather than later.

SK On and NeoVolta lock in a $1.1 billion LFP supply deal with an 18 GWh ceiling

The starting point is a firm, multi-year supply contract. SK On will supply 9 GWh of lithium iron phosphate (LFP) pouch-type cells, a battery chemistry that uses iron and phosphate rather than nickel and cobalt and is favoured in stationary storage for its lower cost and longer cycle life.

LFP battery chemistry uses iron and phosphate rather than nickel or cobalt, giving stationary storage applications a lower cost base, longer cycle life, and a more stable thermal profile than the nickel-rich chemistries originally designed for electric vehicles.

Here are the core terms of the initial agreement:

• Volume: 9 GWh of LFP pouch-type cells
• Cell type: Lithium iron phosphate, pouch format
• Delivery window: 2027 to 2031
• Estimated value: approximately KRW 1.5 trillion, or roughly $1.1 billion
• Follow-on framework: a second 9 GWh agreement anticipated in 2026

The second tranche is where the structure gets interesting. Under the planned follow-on, expected to be formalised during 2026, NeoVolta would assemble energy storage packs for SK On to buy back, taking the combined cooperation volume up to 18 GWh.

The valuation The initial 9 GWh contract is estimated at approximately KRW 1.5 trillion, roughly $1.1 billion, covering deliveries from 2027 to 2031.

Set that against SK On’s stated target of securing 20 GWh of energy storage orders during 2026, and the significance sharpens. The combined 18 GWh framework, if finalised, would fill nearly the company’s entire 2026 order target from a single US partner.

That tells you SK On is chasing concentrated, large-anchor relationships rather than spreading its bets across many smaller customers. The buyback arrangement in the second tranche pushes the partnership past a standard supply contract into mutual dependence. For anyone tracking how the US storage supply chain is forming, this is a named, valued, and staged commitment rather than a vague memorandum.

Georgia becomes the manufacturing spine of the deal

The physical geography of this deal is where the strategy becomes tangible, and it all sits in one state.

The LFP cells will be manufactured at SK On’s existing operations in Commerce, Georgia, where the company’s EV battery cell sites carry a combined production capacity of 22 GWh. These are the plants originally built for electric vehicles, now feeding a storage contract.

The Georgia plant restructuring that preceded this supply pivot involved significant workforce reductions at SK Battery America earlier in 2026, a detail that contextualises how abruptly the company has shifted from EV output to storage-oriented production at the same facilities.

There is more Korean capacity arriving alongside them. SK On’s joint venture EV battery plant with Hyundai Motor, also in Georgia and rated at 35 GWh per year, commenced commercial production in early June 2026. That adds substantial domestic manufacturing weight within the same state.

The assembly side of the deal is equally local. NeoVolta’s production facility sits in Pendergrass, Georgia, currently running at 2 GWh of annual capacity, with a ramp-up scheduled before the end of Q3 2026 and a target of scaling to as much as 8 GWh per year by 2028.

Because both the cell supplier and the pack assembler operate inside Georgia, this is a genuinely localised US supply chain node rather than a cross-continental arrangement. That matters for three practical reasons: compliance with Inflation Reduction Act (IRA) sourcing rules, lower logistics cost, and the speed at which the 2027 delivery window becomes executable.

For anyone mapping US battery manufacturing geography, the read here is clear. Georgia is emerging as a storage supply chain hub, with multiple large-capacity facilities operating or ramping in the same state within the same two-year window.

LG Energy Solution and Samsung SDI are running the same playbook

SK On is not acting alone, and that is the point. Its two largest Korean rivals are executing near-identical moves, which turns a company strategy into something closer to industry policy.

LG Energy Solution (LGES) converted EV battery lines at its Holland, Michigan plant to produce LFP storage cells, beginning mass production in June 2025. The company reported an energy storage order backlog of roughly 120 GWh at the end of Q3 2025 and has set a target of securing more than 90 GWh of new orders in 2026.

The backlog signal LGES reported an energy storage order backlog of approximately 120 GWh at the end of Q3 2025, a figure that reflects demand already contracted rather than projected.

Its named US contracts show the scale in concrete terms:

1. DTE Energy: 6 GWh valued at approximately $1.6 billion over roughly two years
2. Terra-Gen: an 8 GWh order with deliveries from 2026 to 2029
3. Hanwha Qcells US: 5 GWh of Michigan-produced LFP cells, delivery 2028 to 2030

Samsung SDI is pursuing the same conversion logic. The company plans to expand US storage capacity to approximately 30 GWh by the end of 2026, primarily by converting EV battery lines at the StarPlus Energy joint venture plant in Indiana. It is positioned as the only non-Chinese supplier of prismatic LFP cells, with US LFP storage cell production scheduled to begin in October 2026.

The pattern is what carries the weight here. When three of the world’s largest battery makers convert the same asset class, EV lines, to the same product category, LFP storage cells, in the same geography, the US, the conclusion is hard to avoid.

Korean manufacturers have collectively decided the US storage market is now their primary growth vector, not a defensive hedge against soft EV sales. That tells you supply chain formation is moving faster than most headline coverage suggests, with multiple gigawatt-scale facilities shifting output inside the 2025 to 2026 window.

What is driving the pivot, and where the risks sit

The commercial logic is honest and worth stating plainly before the caveats arrive.

Three forces are pulling capacity toward storage. Slowing EV adoption in North America has left EV joint-venture plants underutilised. The power requirements of AI data centres and grid expansion have created surging demand for stationary storage. And US trade and tax policy has made domestically manufactured Korean cells competitive against Chinese rivals.

The profitability case is visible in Tesla’s numbers. According to Tesla’s reported figures, its energy segment generated gross margins of approximately 28.7% in H1 2026, well above its automotive segment’s 18.7%. Storage is structurally attractive on margin, not merely a way to fill idle lines.

The policy tailwind keeping Korean cells competitive

Two mechanisms create the current Korean advantage. US tariffs on Chinese LFP storage batteries are cited in the 45% to 58% range, while Korean imports face roughly 15% under the KORUS free trade agreement.

The tariff differential US tariffs on Chinese LFP storage batteries are cited at 45% to 58%, compared with roughly 15% on Korean imports under the KORUS FTA. That gap is what makes Korean cells competitive today.

Alongside tariffs, the IRA’s Advanced Manufacturing Production Credits are cited at up to $45 per kWh for domestically produced cells. The IRA’s Foreign Entity of Concern (FEOC) rules, which restrict tax credits for components sourced from entities with significant Chinese ownership, act as a further demand signal by pushing US developers away from Chinese-affiliated supply chains and toward Korean suppliers.

Three structural risks the pivot cannot ignore

The foundation is less solid than the headline growth suggests. Watch these three risks:

• Domestic oversupply. US battery manufacturing capacity is forecast to jump from 338 GWh at the end of 2025 to 601 GWh by the end of 2026. With LGES, SK On, Samsung SDI, and others converting simultaneously, a domestic LFP glut is a real possibility.
• Chinese raw material dependence. South Korea imports roughly 70% of its cathode materials from China and depends on China for an estimated 97% to 98% of its graphite. True supply chain independence remains distant.
• Tariff volatility. The competitive position is a policy artefact, not a cost structure reality. Analysts have flagged that Korean makers carry a structural cost disadvantage against China’s vertically integrated supply chains, so any shift in trade policy resets the entire equation.

The tariff arithmetic makes Korean cells competitive today despite that underlying cost gap. Read carefully, that means the commercial logic rests on US industrial policy holding its current shape, and FEOC exposure remains live if supply chain restructuring lags the contract commitments already signed.

Chinese battery export policy is itself in transition: Beijing’s removal of export tax rebates for batteries by 2027 will raise the landed cost of Chinese cells in global markets, a dynamic that partially offsets the structural cost advantage Korean producers are currently racing to close.

What the pivot will be tested on over the next 12 to 24 months

The combined ambition is now substantial. SK On is targeting 20 GWh of 2026 orders, LGES more than 90 GWh, and Samsung SDI roughly 30 GWh of US capacity. Stacked together, that is a large, coordinated Korean bet on a single market.

The demand backdrop supports it. According to Wood Mackenzie and the American Clean Power Association, the US installed 18.9 GW and 51 GWh of battery storage in 2025, a 52% increase in GW over 2024. Wood Mackenzie projects cumulative US storage capacity reaching 200 GW and 655 GWh by 2031, with 146 GW and 499 GWh of new installations expected between 2026 and 2031.

The U.S. Energy Storage Monitor, a joint publication by the American Clean Power Association and Wood Mackenzie, recorded 18.9 GW and 51 GWh of battery storage installations in 2025 and projects cumulative US capacity reaching 200 GW and 655 GWh by 2031, the demand base underpinning every Korean supply-side commitment now being formalised.

The demand anchor Wood Mackenzie projects cumulative US battery storage reaching 200 GW and 655 GWh by 2031, the demand base against which every Korean supply-side bet is being placed.

Two variables will most visibly test the pivot in the next 12 to 24 months:

• Whether US tariff policy holds at the current differentials that make Korean cells competitive
• Whether Korean supply chains can cut Chinese raw material dependence fast enough to preserve IRA and FEOC compliance

The combined scale of Korean ambition now looks larger than the currently contracted demand base can absorb without competition tightening. The question is not only whether these companies win orders, but whether the margin structure holds as more Korean capacity comes online at once. That gives you a concrete framework: three named companies, specific capacity targets, a clear 2026 to 2028 execution window, and two policy variables that will decide whether the financial logic survives contact with reality.

For readers wanting to model how margin structures evolve as more LFP capacity comes online, our dedicated guide to battery storage cost trajectories examines how falling cell costs interact with installation pricing and what historical margin compression patterns suggest about the 2027-2031 window.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions.

Past performance does not guarantee future results. Financial projections are subject to market conditions and various risk factors. Forward-looking statements are speculative and subject to change based on market developments and policy revisions.