In recent years, as the global energy transition deepens and the surging demand for computing power driven by AI model training accelerates, the quest for new energy storage technologies that are safer, more durable, and more cost-effective has intensified. According to relevant data, global new energy storage installations surpassed 113.3 GW in 2025, with cumulative capacity now accounting for 56.2% of the overall power storage market. Projections indicate that total global installed storage capacity will reach 1,545 GW by 2034. The industry is evolving from a predominantly liquid lithium‑ion battery landscape into a diverse, multi‑technology arena, with applications expanding rapidly across various scenarios.
From a Single Lithium‑Ion Path to a Multi‑Technology Matrix
Today, new energy storage technologies are advancing at an unprecedented pace. In China, for example, the 200 MW/400 MWh advanced electrochemical energy storage station in Shanwei, Guangdong, has pushed semi‑solid battery technology toward large‑scale grid‑level deployment and was recognized by the National Energy Administration as one of the “Top Ten Scientific and Technological Innovations in the Energy Sector for 2025.” In the United States, Google has partnered with Form Energy to integrate ultra‑long‑duration, high‑safety iron‑air batteries into data center power systems, offering a fresh solution for the energy backbone of the AI era. In Japan, South Korea, and the European Union, solid‑state batteries are viewed as a strategic lever to reshape the future energy landscape, with multiple nations channeling policy support, funding, and industrial‑chain coordination to secure a lead in this technological frontier.
Energy storage is a critical pillar of the new power system, fundamentally altering the traditional paradigm where electricity is generated, transmitted, and consumed simultaneously. Conventional storage is dominated by pumped hydro; new energy storage refers to all storage technologies other than pumped hydro that primarily output electricity. The key differentiators are shorter construction periods, greater siting flexibility, faster response times, stronger regulation capabilities, better alignment with renewable integration, and the ability to be deployed in distributed fashion at generation, grid, and consumer sides—embedding deeply into every segment of the power system.
From a technical perspective, new energy storage falls into two main categories: electrochemical and physical. Electrochemical storage relies on chemical reactions within batteries to store and release energy, offering rapid response and flexible deployment; physical storage uses media such as air, water, or thermal mass, making it more suitable for large‑scale, long‑duration applications.
Over the past decade, liquid lithium‑ion batteries—with their high energy density and mature supply chains—have dominated the electrochemical storage market, accounting for over 96% of new energy storage capacity. But as real‑world demands have multiplied, the research and industrial communities have come to expect far more from storage than simply “holding electricity.” The system must act as a versatile all‑rounder, simultaneously meeting the grid’s complex needs across safety, economics, resource availability, lifespan, and discharge duration. No single technology can address all these dimensions; instead, a diverse matrix covering different time scales (from millisecond‑level responses to seasonal storage) and application requirements is essential.
Electrochemical and Physical Storage: Each Shining in Its Own Right
In the electrochemical arena, new pathways—solid‑state batteries, sodium‑ion batteries, flow batteries, and iron‑air batteries—are rapidly emerging, each offering distinct advantages.
Solid‑state batteries are widely regarded as the “ideal technical form” for storage and a key direction for next‑generation high‑performance batteries. Unlike conventional lithium batteries that use liquid electrolytes, solid‑state batteries employ solid electrolytes for ion conduction, fundamentally reducing the risk of fires and thermal runaway. Moreover, they can potentially incorporate lithium‑metal anodes to further boost energy density. Semi‑solid (mixed solid‑liquid) storage lithium batteries developed by the Institute of Physics of the Chinese Academy of Sciences and WeLion New Energy have already achieved large‑scale grid‑level applications, while all‑solid‑state batteries are expected to begin commercialization around 2028. Toyota, Samsung, Solid Power, and others are also accelerating R&D and scaling up manufacturing. Breakthroughs in material costs and fabrication processes for solid‑state batteries will profoundly influence the future of both grid‑scale storage and electric vehicles.
As a “complementary sibling” to lithium batteries, sodium‑ion batteries have drawn considerable industry attention. Sodium and lithium share similar chemical properties, yet sodium is far more abundant and widely distributed—often dubbed the “budget‑friendly resource.” They also perform well across a broad temperature range from ‑30°C to 50°C, showing great potential for high‑altitude storage, grid peaking, and telecom base stations. The Institute of Physics and HiNa Battery have jointly built mass‑production lines and demonstrated hundred‑MWh‑scale sodium‑ion storage systems. France’s Tiamat is focusing on fast‑charging sodium‑ion cells for transportation and stationary storage.
Flow battery commercialization has also accelerated notably in recent years. Unlike conventional batteries that store energy within the electrodes, flow batteries store energy in external electrolytes, offering long cycle life and high safety—particularly well‑suited for large‑scale, long‑duration storage. China has commissioned and operates the world’s largest vanadium flow battery storage project—the Dalian Flow Battery Energy Storage Peak‑Shaving Power Station, whose first phase is 100 MW/400 MWh. The U.S. Department of Energy continues to support flow battery R&D and demonstration projects to advance long‑duration storage, while Australia and other countries are also incorporating flow batteries into renewable‑hub storage plans.
Iron‑air batteries operate on the reversible reaction of rusting and de‑rusting. During discharge, the battery takes in oxygen from the air and oxidizes iron metal; during charging, the rust is converted back to iron and oxygen is released. While conventional lithium batteries typically store energy for 2 to 10 hours, iron‑air batteries can discharge slowly for up to 100 hours. The core material—iron—is cheap and safe, with future breakthroughs hinging on electrode innovation and electrolyte optimization. The U.S. Department of Energy has designated it as a priority area, aiming to provide multi‑day power reserves for the grid during extreme weather events.
In physical storage, a variety of technologies also demonstrate unique strengths. Compressed‑air energy storage uses surplus electricity to compress air and store it in underground caverns, then releases it to drive turbines when needed; flywheel storage stores kinetic energy in a high‑speed rotating rotor, offering high power burst capability; molten‑salt thermal storage and cold storage enable energy regulation through thermal media. China’s domestically developed compressed‑air systems continue to set new records for single‑unit power output, while the U.S. and Europe are actively exploring adiabatic compressed‑air schemes that co‑operate with wind and solar power.
Overall, different storage technologies are progressively forming a differentiated division of labor: short‑duration frequency regulation (within half an hour) relies mainly on flywheels, supercapacitors, and high‑power lithium batteries; medium‑duration storage (up to 10 hours) is currently dominated by lithium batteries, evolving from liquid to safer semi‑solid and all‑solid‑state versions, while sodium‑ion batteries without resource constraints are also advancing rapidly; long‑duration storage on the generation and grid sides is supported by flow batteries, compressed air, iron‑air batteries, molten‑salt thermal storage, and cold storage. This complementary synergy across multiple technology routes provides comprehensive support for the new power system.
Accelerating into Large‑Scale Application
As technologies mature, new energy storage is moving from demonstration projects to large‑scale deployment, gradually integrating into power systems, industrial production, and the digital economy. Renewable generation is currently the most important application scenario—storage captures surplus power and releases it during peak demand, smoothing the load curve and providing essential support for high‑penetration renewable grids.
AI is emerging as a new driver of storage demand: the rapid growth in computing for large‑model training and AI data centers has prompted tech giants like Google, Microsoft, Meta, and Amazon to invest heavily in long‑duration storage projects. New energy storage is becoming a foundational capability for the AI era. At the same time, storage is being rapidly adopted in industrial parks, zero‑carbon zones, green buildings, and remote microgrids, helping enterprises optimize electricity use, increase self‑sufficiency, and reduce fossil‑fuel dependence.
AI itself is now deeply involved in the entire lifecycle of storage—from R&D and manufacturing to operation. In R&D, AI accelerates the screening and prediction of novel materials, shortening development cycles; in manufacturing, digital twins and machine vision are driving intelligent, unmanned production; in operation, AI continuously monitors equipment status, forecasts lifespan decay and safety risks, and dynamically optimizes charge/discharge strategies based on weather, load, and electricity prices.
From laboratory breakthroughs to critical infrastructure supporting renewable grids, AI data centers, and zero‑carbon industrial parks, new energy storage is rapidly scaling up. Yet several major challenges remain: on the technical front, the core goals of safety, long life, and low cost have not been fully achieved; on the industrial side, supply chains for key materials, equipment manufacturing, recycling, market standards, and technical specifications still need further strengthening.
We are confident that as diverse technology pathways continue to mature, a more flexible, efficient, and low‑carbon energy system will gradually take shape, opening a new chapter in the energy revolution for humanity.
By Li Hong
People’s Daily, June 29, 2026, Page 14