Energy Storage:
Beyond Batteries
Lithium-ion gets the headlines, but the renewable grid needs everything: iron-air, flow batteries, compressed air, pumped hydro, gravity, thermal, and hydrogen. Here's the full picture.
Data: EIA, BNEF, Wood Mackenzie · Updated March 2026
Storage Technologies Compared
There's no single silver bullet for energy storage. A fully renewable grid will need a portfolio of technologies, from short-duration for daily peaks, long-duration for multi-day events, and seasonal storage for the deepest gaps. Here's every major technology in play.
How it works: Lithium ions shuttle between a cathode (typically LFP. lithium iron phosphate for grid applications) and a graphite anode through an electrolyte. Charging moves ions one way, discharging moves them back, releasing stored electrical energy. Grid-scale systems stack thousands of cells into container-sized Megapacks or equivalent units.
Li-ion dominates grid storage with 90%+ market share. Tesla delivered 31.4 GWh in 2024 alone, holding 15% global BESS integrator market share. Pack prices dropped to $70/kWh in 2025, a 45% year-over-year decline. But the 4-hour duration ceiling is a fundamental limitation for a grid that needs multi-day resilience.
- Extremely fast response (<1 second)
- Proven at massive scale (GW+)
- Rapidly declining costs
- High round-trip efficiency
- Modular and containerized
- Limited to ~4 hours economically
- Degrades over time (capacity fade)
- Fire risk (thermal runaway)
- Critical mineral supply chains
- Not viable for seasonal storage
How it works: The iron anode "rusts" (oxidizes) during discharge, absorbing oxygen from the air and releasing electrons. During charging, the process reverses: electricity removes the oxygen, converting rust back to iron. It's essentially a reversible rusting battery. The materials are dirt cheap: iron pellets and air.
Form Energy shipped its first commercial batteries in 2025 from its 550,000 sq ft factory in Weirton, West Virginia: built on the site of a former steel mill. The first deployment: a multi-day storage pilot with Great River Energy in Minnesota. Form has ~200 MW under contract and has produced 100,000+ electrodes. The factory employs 300+ workers with plans to expand to 1 million sq ft by 2028. The AI-driven demand boom has accelerated their timeline dramatically.
- 100-hour (multi-day) duration
- Extremely cheap materials (iron + air)
- No critical mineral dependency
- Made in America (WV factory)
- Negligible fire risk
- Low round-trip efficiency (~45%)
- First commercial deployment just started
- Slow response time vs. Li-ion
- Large physical footprint
- Manufacturing still scaling
How it works: Two liquid electrolytes are stored in separate tanks and pumped through a cell stack where they exchange ions across a membrane, generating electricity. Power is determined by cell stack size; energy capacity by tank volume. This decoupling means you can independently scale power and duration by adding more tanks for more hours.
ESS Inc. uses an iron-based saltwater electrolyte (no vanadium, no exotic materials). In October 2025, Salt River Project and ESS announced a 5 MW / 50 MWh (10-hour) pilot, "Project New Horizon," at SRP's Copper Crossing Energy Center in Arizona, expected operational by December 2027. Vanadium flow batteries from companies like Invinity are also commercial, particularly in China where several GWh-scale projects exist.
- Duration scales with tank size
- Minimal degradation (20,000+ cycles)
- Non-flammable electrolytes
- Long lifespan (20+ years)
- Electrolyte is recyclable
- Higher upfront cost than Li-ion
- Lower energy density (large footprint)
- Pumping systems add complexity
- Vanadium supply chain concerns
- Limited deployments at GW scale
How it works: When excess electricity is available, water is pumped from a lower reservoir to an upper reservoir. When power is needed, the water flows back downhill through turbines, generating electricity. It's gravity-powered energy storage, conceptually simple, massively scalable, and extremely long-lived.
Pumped hydro accounts for ~95% of all global energy storage capacity with nearly 200 GW installed worldwide. The US has about 22 GW of pumped hydro across ~40 facilities, most built in the 1970s and 1980s. New projects are extremely difficult to build due to geography requirements, environmental permitting (often 10+ years), and high upfront capital. But existing facilities are invaluable, and many will operate for another 50+ years.
- Massive scale (GW per facility)
- Extremely long lifespan (50–100 years)
- Proven technology (decades of operation)
- High efficiency for mechanical storage
- No degradation over time
- Requires specific geography (elevation)
- 10+ year permitting timelines
- Enormous upfront capital ($1B+)
- Environmental / land use concerns
- Almost no new US projects in decades
How it works: Excess electricity powers compressors that push air into underground caverns (salt domes, depleted mines, or purpose-built chambers). When power is needed, the compressed air is released through an expansion turbine to generate electricity. Advanced (adiabatic) CAES captures and stores the heat of compression, recovering it on expansion for higher efficiency.
Hydrostor, a Canadian company, received final permitting in December 2025 for its 500 MW / 4 GWh Willow Rock Energy Storage Center in Kern County, California, their first grid-scale deployment. Cost: ~$3,000/kW for an 8-hour system, with a path to 20% cost reduction. (CEC permit, Dec 2025; Utility Dive) Hydrostor secured $200M in early 2025 for its A-CAES projects. Two legacy diabatic CAES plants have operated for decades: McIntosh (Alabama, 1991) and Huntorf (Germany, 1978).
- Very long duration (8–24+ hours)
- Large scale (hundreds of MW)
- Long equipment lifespan
- Low per-kWh cost at scale
- Uses proven turbine technology
- Requires underground geology
- High upfront capital cost
- Lower efficiency than batteries
- Few commercial A-CAES deployments
- Long construction timelines
How it works: Electric motors lift heavy composite blocks (35 tons each) to the top of a tall structure. When electricity is needed, the blocks descend, spinning generators via cables and regenerative braking systems. It's the same principle as pumped hydro, but using solid masses instead of water, with no geography constraints.
Energy Vault operates a commercial 25 MWh demonstration system in Switzerland and has signed agreements for GWh-scale deployments. Their newer designs use modified building structures rather than open cranes, improving energy density and weather resilience. Still early: the technology needs to prove cost-competitiveness against rapidly cheapening lithium-ion for 4–8 hour durations.
- No chemical degradation
- No exotic materials needed
- Siting flexibility (no geology needs)
- Very long system lifespan
- Recyclable/reusable components
- Unproven at utility scale
- Low energy density
- Competing with cheapening Li-ion
- Complex mechanical systems
- Limited track record
How it works: Excess electricity heats a storage medium, molten salt, sand, crusite blocks, or solid carbon, to very high temperatures (500–1,500°C). When power is needed, the heat drives a steam turbine or thermophotovoltaic (TPV) cells to generate electricity. Some systems deliver heat directly for industrial processes, bypassing the electricity conversion entirely for even higher efficiency.
Antora Energy heats carbon blocks to 1,500°C and converts heat to electricity via TPV panels, no moving parts. Malta Inc. (Alphabet spin-off) uses electro-thermal storage with hot and cold molten salt tanks. Rondo Energy stores heat in brick for industrial decarbonization. Concentrated solar plants have used molten salt storage for years, proven at the Crescent Dunes (110 MW) and Gemasolar plants.
- Very cheap storage medium
- Long duration capability
- Can serve industrial heat directly
- Low self-discharge
- No exotic materials
- Lower electricity round-trip efficiency
- Heat-to-electricity conversion losses
- Large physical footprint
- Slower response time
- Complex thermal management
How it works: Electrolyzers split water into hydrogen and oxygen using excess electricity. The hydrogen is compressed or liquefied and stored in tanks, underground salt caverns, or existing pipeline infrastructure. When power is needed, the hydrogen runs through fuel cells or gas turbines to generate electricity. Can also be stored as ammonia for easier transport.
Hydrogen is the only viable technology for true seasonal storage: storing summer solar excess for winter demand peaks. The DOE's Hydrogen Shot targets $1/kg green hydrogen by 2031 (currently ~$5-7/kg). Salt caverns in the Gulf Coast already store hydrogen industrially. Projects like the Sauk Valley Green Hydrogen Plant in Illinois are demonstrating co-located solar-to-hydrogen production. The seven DOE-funded Regional Clean Hydrogen Hubs ($7B) are building the infrastructure backbone.
- Seasonal-scale storage (months)
- Can use existing gas infrastructure
- Zero-emission fuel cell generation
- Versatile (power, transport, industry)
- Abundant feedstock (water)
- Very low round-trip efficiency (30-45%)
- Expensive electrolyzers
- Storage/transport challenges
- Green hydrogen still costly
- Infrastructure buildout needed
US Storage Growth: Exponential
US battery storage has gone from niche to mainstream in under five years. The 2024 total was more than double 2023, and 2025 is on pace to top that again. The five-year forecast: 93 GW of new capacity by 2029.
The Duration Problem
Nearly all grid batteries installed today are 4-hour lithium-ion systems. They're well-suited for daily peaks, shifting solar from afternoon to evening. But a renewable grid needs storage for multi-day weather events, seasonal variations, and prolonged outages. 4 hours isn't enough.
Largest US Storage Projects
The scale of individual projects keeps growing. Multi-GWh installations are becoming routine in Texas and California, while new technology pilots push the frontier of what's possible.
| Project | Location | Capacity | Type | Status |
|---|---|---|---|---|
| Moss Landing | Monterey County, CA | 750 MW / 3,000 MWh | Li-ion (Vistra) | Operational |
| Edwards & Sanborn | Kern County, CA | 875 MW / 3,500 MWh | Li-ion + Solar (Terra-Gen) | Operational |
| Eleven Mile Solar | Maricopa County, AZ | 600 MW / 2,400 MWh | Li-ion + Solar | Operational |
| Sonoran Solar Energy Center | Buckeye, AZ | 550 MW / 2,200 MWh | Li-ion + Solar | Operational |
| Crimson Storage | Riverside County, CA | 350 MW / 1,400 MWh | Li-ion (Recurrent) | Operational |
| Willow Rock (Hydrostor) | Kern County, CA | 500 MW / 4,000 MWh | A-CAES | Permitted |
| Bath County | Bath County, VA | 3,003 MW | Pumped Hydro | Operational (1985) |
| Ludington | Mason County, MI | 1,872 MW | Pumped Hydro | Operational (1973) |
| Form Energy / Great River | Cambridge, MN | 10 MW / 1,000 MWh | Iron-Air (100-hr) | Under Construction |
| SRP / ESS New Horizon | Florence, AZ | 5 MW / 50 MWh | Iron Flow (10-hr) | Planned (2027) |
| Energy Dome / Alliant | Wisconsin | 20 MW / 200 MWh | CO₂ Battery (10-hr) | Approved |
State Leaders in Storage Deployment
Battery storage is following solar's geographic pattern, with sun-rich states with high electricity demand lead, but the buildout is spreading fast. Texas exploded onto the scene driven by ERCOT merchant opportunities.
Cost Trends: The Price Collapse
Battery storage costs have plummeted. Li-ion pack prices for stationary storage hit $70/kWh in 2025, a staggering 45% decline in a single year, and the steepest drop of any battery use case. The question for long-duration technologies: can they follow a similar cost curve?
Stay on Top of Storage
Get energy storage analysis and breaking news delivered to your inbox, free.
Subscribe →- Wood Mackenzie / ACP. US Energy Storage Monitor Q4 2025
- Canary Media: US to Shatter Grid Battery Records (2025)
- BloombergNEF. Li-Ion Battery Price Survey 2025
- Latitude Media: Form's First 100-Hour Batteries Hit the Grid
- IHA / Wikipedia: Pumped-Storage Hydroelectricity
- ESS Inc.. SRP 50 MWh Iron Flow Pilot
- Solar Power World: Tesla BESS Global Dominance
- Utility Dive: Hydrostor Willow Rock 500 MW CAES Permit
- US EIA. Battery Storage Capacity Data
- Governor Newsom: California Storage Record Growth (COP30)