Head-to-head: Long-duration energy storage (LDES) — comparing leading approaches on cost, performance, and deployment

The electricity grid needs storage that can discharge for 8 hours or longer to balance multi-day weather events and seasonal variability as renewable penetration climbs past 50%. BloombergNEF estimates that 1.5 TWh of long-duration energy storage (LDES) capacity must be deployed globally by 2040 to meet net-zero targets, representing a cumulative investment of $245 billion to $345 billion (BloombergNEF, 2025). Yet the LDES landscape includes at least six fundamentally different technology families, each with distinct cost structures, siting constraints, and maturity levels. For policymakers and procurement teams evaluating which approaches deserve support and contracts, the differences matter enormously.

Why It Matters

Lithium-ion batteries dominate the current storage market but are economically optimized for durations of 2 to 4 hours. As grids add renewable capacity, the frequency and duration of supply gaps increase: California's CAISO recorded 14 events exceeding 8 hours of net-load deficit in 2025, up from 3 in 2021 (CAISO, 2025). ERCOT in Texas experienced a 72-hour period during Winter Storm Elliott where wind generation dropped below 10% of nameplate capacity. These multi-day events require storage technologies that can sustain discharge for 10 to 100+ hours at costs well below what lithium-ion can deliver at those durations.

North American utilities, independent power producers, and state energy agencies are now issuing procurement solicitations specifically for LDES. The U.S. Department of Energy's Long Duration Energy Storage Shot initiative targets a levelized cost of storage (LCOS) of $0.05 per kWh for systems with 10+ hours of duration, a threshold that would make LDES competitive with natural gas peaker plants on a delivered-energy basis. Understanding which technologies can realistically approach this target, and when, is essential for sound policy design and investment allocation.

Key Concepts

LDES technologies are generally categorized by their energy storage medium: mechanical (pumped hydro, compressed air, gravity), electrochemical (flow batteries, metal-air batteries, iron-air batteries), thermal (molten salt, crushed rock, liquid air), chemical (hydrogen and derivatives), and geological (underground hydrogen storage). Each category has different relationships between power capacity (MW) and energy capacity (MWh), different degradation profiles, different siting requirements, and different supply chain dependencies.

The critical performance metrics for comparing LDES technologies include round-trip efficiency (the percentage of energy recovered after storage), cycle life (the number of charge-discharge cycles before significant degradation), energy density (MWh per cubic meter or per tonne), response time (how quickly the system can ramp from idle to full discharge), and LCOS, which integrates capital cost, operating cost, efficiency losses, and asset lifetime into a single dollar-per-kilowatt-hour figure.

Head-to-Head Comparison

Pumped Hydro Storage vs. Compressed Air Energy Storage

Pumped hydro storage (PHS) remains the most deployed LDES technology globally, accounting for more than 94% of installed storage capacity worldwide with approximately 170 GW in operation (International Hydropower Association, 2025). PHS facilities pump water to an upper reservoir during periods of excess generation and release it through turbines during periods of high demand. Round-trip efficiency ranges from 75 to 85%, cycle life is essentially unlimited (facilities built in the 1970s remain operational), and discharge durations of 8 to 24 hours are standard.

The limitations are geographic and temporal. PHS requires specific topography: two reservoirs at different elevations with sufficient water supply. Environmental permitting in North America now averages 7 to 12 years, and construction adds another 4 to 6 years. Capital costs range from $150 to $300 per kWh at 10-hour duration, competitive with any alternative, but the development timeline means projects initiated today will not deliver power until the mid-2030s.

Compressed air energy storage (CAES) uses excess electricity to compress air into underground salt caverns or purpose-built tanks, then expands the air through a turbine during discharge. The two operational utility-scale CAES facilities, the 321 MW Huntorf plant in Germany (operating since 1978) and the 110 MW McIntosh plant in Alabama (operating since 1991), use natural gas combustion during expansion, achieving round-trip efficiencies of only 42 to 54%. Advanced adiabatic CAES (A-CAES) systems that capture and reuse compression heat can reach 60 to 70% efficiency but have not yet been demonstrated at utility scale.

Hydrostor, a Canadian company, is developing A-CAES projects using purpose-built underground caverns rather than relying on existing salt formations, expanding potential siting locations. Their 500 MW Willow Rock project in Kern County, California, received its final environmental permit in 2025 and targets commissioning by 2029 with projected LCOS of $0.08 to $0.10 per kWh at 12-hour duration (Hydrostor, 2025). CAES offers faster permitting than PHS but requires either suitable geology or significant excavation costs that add $40 to $80 per kWh to capital expenses.

Iron-Air Batteries vs. Vanadium Redox Flow Batteries

Iron-air batteries represent a newer electrochemical approach using iron, oxygen, and water as primary inputs. Form Energy, the category leader, has designed a 100-hour iron-air battery system that "breathes" by reversibly rusting iron during discharge and de-rusting it during charge. The company's target LCOS is $0.02 to $0.03 per kWh at 100-hour duration, which if achieved would undercut every competing LDES technology. Capital costs are projected at $20 to $30 per kWh of energy capacity, roughly one-tenth the cost of lithium-ion on an energy basis.

Form Energy broke ground on its first commercial manufacturing facility in Weirton, West Virginia, in 2023 and began shipping battery modules in late 2025. Georgia Power's 15 MW / 1,500 MWh pilot installation is scheduled for commissioning in mid-2026, representing the first utility-scale iron-air deployment. The primary trade-off is round-trip efficiency: iron-air batteries achieve only 40 to 45% round-trip efficiency, meaning that more than half of the input energy is lost as heat. For applications where the marginal cost of renewable electricity is near zero during charging periods (midday solar surplus, nighttime wind surplus), this efficiency penalty is economically acceptable. For grids where curtailed renewable energy is scarce, the efficiency loss is costly.

Vanadium redox flow batteries (VRFBs) use liquid vanadium electrolyte stored in external tanks, allowing independent scaling of power (determined by cell stack size) and energy (determined by tank volume). Round-trip efficiency is 65 to 75%, substantially higher than iron-air. Cycle life exceeds 20,000 cycles with negligible degradation because the active material remains in solution rather than undergoing solid-state phase changes. Invinity Energy Systems and Rongke Power are the leading VRFB manufacturers, with Rongke operating the world's largest flow battery installation: a 100 MW / 400 MWh system in Dalian, China, commissioned in 2022.

The constraint for VRFBs is cost and supply chain. Vanadium pentoxide prices fluctuate significantly, ranging from $6 to $15 per pound over the past five years, creating capital cost uncertainty. At current pricing, VRFB systems cost $250 to $400 per kWh at 8-hour duration, declining to $150 to $250 per kWh at 12-hour duration as the fixed cell-stack cost is amortized over more energy. Several companies, including Largo Inc. and Australian Vanadium, are working to stabilize supply through dedicated mining operations and electrolyte leasing models that reduce upfront capital requirements.

Gravity Storage vs. Liquid Air Energy Storage

Gravity-based storage systems lift heavy masses during charging and lower them to generate electricity during discharge. Energy Vault, the most visible company in this category, uses a crane-based system to stack and unstack concrete composite blocks. The company's first commercial system, a 25 MW / 100 MWh facility in Rudong, China, began operations in 2024. Round-trip efficiency is reported at 75 to 80%, and the system uses no water, no rare materials, and no geologic constraints. Capital costs for the Rudong system were approximately $180 per kWh at 4-hour duration (Energy Vault, 2025).

However, gravity storage faces scaling challenges at longer durations. Because energy capacity is proportional to mass and height, storing 100+ hours of energy requires either enormous masses or extreme heights. The physics limit energy density to roughly 1 to 2 Wh per kilogram at practical lift heights (100 to 200 meters), meaning a 100 MWh facility requires lifting and lowering 50,000 to 100,000 tonnes of material. The land footprint and structural engineering requirements become formidable beyond 8 to 12 hours of duration.

Liquid air energy storage (LAES), also called cryogenic energy storage, compresses and cools air to liquid form at negative 196 degrees Celsius, stores it in insulated tanks, and then warms and expands the liquid air through a turbine during discharge. Highview Power, the category leader, operates a 50 MW / 250 MWh LAES facility under construction at Carrington, UK, with projected commissioning in 2026. Round-trip efficiency is 50 to 60%, lower than pumped hydro or gravity but higher than iron-air. LAES has no geographic constraints, uses no water in the storage process, and can be sited at existing industrial locations.

The LCOS for LAES at current scale is estimated at $0.12 to $0.18 per kWh at 10-hour duration, above the DOE's $0.05 target. Highview Power projects that costs will decline to $0.08 to $0.10 per kWh at scale as manufacturing volumes increase and plant sizes grow to 200+ MW (Highview Power, 2025).

What's Working

Iron-air batteries are generating the most commercial momentum. Form Energy has announced contracts or partnerships with utilities including Georgia Power, Great River Energy, Xcel Energy, and Central Maine Power, totaling more than 3 GW of pipeline capacity. The technology's extreme low cost on an energy-capacity basis makes it uniquely suited for multi-day storage applications that other technologies cannot economically address.

Pumped hydro continues to deliver proven reliability. The Bath County Pumped Storage Station in Virginia, the largest PHS facility in North America at 3,003 MW, has operated since 1985 with availability exceeding 95% and no measurable degradation in storage capacity. For regions with suitable geography, PHS remains the lowest-risk option.

Flow batteries are finding traction in industrial and microgrid applications where daily cycling at 4 to 12 hours of duration and 20+ year lifetimes justify the higher upfront cost. The DOE's Pacific Northwest National Laboratory reported in 2025 that non-vanadium flow chemistries using iron-chromium and zinc-bromine electrolytes are approaching $100 per kWh in energy capacity costs, potentially resolving the vanadium supply concern (PNNL, 2025).

What's Not Working

Advanced CAES has struggled to move beyond pilot scale. The only operational A-CAES demonstration, a 1.75 MW system by SustainX, was decommissioned in 2016 after failing to achieve commercial efficiency targets. Hydrostor's projects represent the most credible path forward, but none are yet operational at full scale.

Gravity storage systems have not demonstrated clear cost advantages over lithium-ion at durations below 8 hours, and face fundamental physics limitations at durations above 12 hours. Energy Vault's pivot to incorporating lithium-ion battery modules alongside its gravity system suggests the pure gravity approach may not stand alone commercially.

Hydrogen-based LDES, while theoretically capable of seasonal-scale storage, currently suffers from low round-trip efficiency (30 to 40% via electrolysis, storage, and fuel-cell reconversion) and high capital costs for both electrolyzers and fuel cells. This approach may ultimately serve a role in seasonal balancing, but it is not cost-competitive for 8 to 100 hour storage applications against the alternatives described above.

Key Players

Established: NextEra Energy (largest PHS operator in North America), Voith Hydro (PHS turbine manufacturer), Rongke Power (world's largest flow battery installation), Siemens Energy (CAES and thermal storage systems)

Startups: Form Energy (iron-air batteries, 100-hour duration), Hydrostor (advanced compressed air energy storage), Energy Vault (gravity-based storage), Highview Power (liquid air energy storage), Invinity Energy Systems (vanadium redox flow batteries), ESS Inc. (iron flow batteries)

Investors: Breakthrough Energy Ventures (Form Energy, Hydrostor), SoftBank Vision Fund (Energy Vault), Goldman Sachs (Hydrostor project finance), Ara Partners (industrial decarbonization storage projects)

Action Checklist

• Map grid reliability gaps by duration: identify whether the primary need is 8 to 12 hour daily cycling, 24 to 72 hour weather events, or seasonal balancing, as different LDES technologies optimize for different durations
• Issue technology-neutral LDES procurement solicitations that specify duration, response time, and LCOS targets rather than prescribing specific technologies
• Require round-trip efficiency reporting using standardized measurement protocols to enable valid cross-technology comparisons
• Evaluate siting constraints early: assess local geology for CAES, topography for PHS, and industrial land availability for battery and thermal systems
• Include supply chain risk in evaluation criteria: vanadium price volatility for VRFBs, concrete/steel supply for gravity, and iron ore sourcing for iron-air
• Monitor Form Energy's Georgia Power pilot results in 2026 as a critical proof point for iron-air commercial viability

FAQ

Q: Which LDES technology has the lowest projected cost per kilowatt-hour? A: Iron-air batteries from Form Energy target $0.02 to $0.03 per kWh at 100-hour duration, the lowest of any LDES technology. However, these figures are based on projected manufacturing costs, not demonstrated commercial pricing. Pumped hydro at $0.05 to $0.08 per kWh remains the lowest-cost proven technology for durations of 8 to 24 hours, though new projects face long development timelines. Flow batteries and CAES cluster in the $0.08 to $0.15 per kWh range at 10 to 12 hour durations.

Q: Does low round-trip efficiency disqualify a technology? A: Not necessarily. Round-trip efficiency matters most when the input electricity has significant cost. In grids with high renewable penetration, midday solar electricity is frequently curtailed or priced near zero. In those conditions, a technology with 40% round-trip efficiency but very low capital cost (like iron-air) can deliver a lower LCOS than a technology with 80% efficiency but higher capital cost (like PHS). The relevant metric is total cost of stored and delivered energy, not efficiency in isolation.

Q: How should policymakers structure LDES incentives? A: The most effective approach is duration-differentiated capacity payments that increase in value for longer discharge durations, reflecting the greater grid reliability contribution. California's CPUC decision in 2025 to create a separate procurement category for 8+ hour storage, distinct from the 4-hour lithium-ion category, is a useful model. Investment tax credits should apply to all LDES technologies without picking winners, and demonstration funding should prioritize technologies that are past laboratory stage but pre-commercial, where public risk-sharing has the highest leverage.

Q: When will LDES reach large-scale deployment in North America? A: The LDES Council projects that 30 to 40 GW of LDES capacity will be needed in North America by 2040. Near-term deployment (2026 to 2030) will likely be dominated by pumped hydro expansions and early iron-air and flow battery installations totaling 5 to 10 GW. Significant scale-up to 20+ GW will require manufacturing capacity buildout, supply chain maturation, and regulatory frameworks that properly compensate multi-day reliability services, most likely materializing in the 2030 to 2035 timeframe.

Sources

• BloombergNEF. (2025). Long-Duration Energy Storage: Market Outlook and Investment Forecast 2025-2040. London: BNEF.
• California Independent System Operator. (2025). Multi-Day Net Load Deficit Events: 2021-2025 Trend Analysis. Folsom: CAISO.
• Hydrostor Inc. (2025). Willow Rock Advanced Compressed Air Energy Storage Project: Technical and Economic Summary. Toronto: Hydrostor.
• Energy Vault Holdings. (2025). Rudong Commercial System: First-Year Operational Performance Report. Lugano: Energy Vault.
• Highview Power. (2025). CRYOBattery Technology: Cost Trajectory and Scale-Up Roadmap. London: Highview Power.
• International Hydropower Association. (2025). Hydropower Status Report 2025: Pumped Storage Chapter. London: IHA.
• Pacific Northwest National Laboratory. (2025). Flow Battery Cost and Performance Benchmarking: Vanadium and Non-Vanadium Chemistries. Richland: PNNL.
• LDES Council. (2025). Net-Zero Power: The Role of Long Duration Energy Storage in Decarbonized Grids. Zurich: LDES Council.