What Is a Flow Battery?

A flow battery stores its energy in liquids held in tanks, and it makes or absorbs power in a stack the liquids are pumped through. Sandia’s chapter in the DOE electricity storage handbook describes the arrangement: one tank holds the positive electrolyte, the other holds the negative electrolyte, and the liquids pass through a cell where they are oxidized and reduced. A membrane, often an ion-selective membrane such as the type the chapter names, lets the ions that balance charge through and is there to stop the two liquids mixing and to stop the electrodes shorting electronically. The stack sets the power. The tanks set the energy. That separation is the reason a flow battery is discussed for long-duration stationary storage, and the reason it is a poor fit in a hallway.

How a flow battery is different from a lithium pack

A lithium iron phosphate pack stores energy in solid electrodes inside a sealed box. Power and energy are tied together by how that box was built. You do not add three hours by pouring a larger tank, because there is no tank. A flow battery does, for the traditional designs the Department of Energy’s Storage Innovations assessment describes: power and energy are independent, and that independence does not automatically extend to hybrid flow designs that still have a solid electrode. If the quote says “flow” and also says the energy is in a solid plate, ask which architecture it is. The words are not the machine.

Vanadium is the chemistry Sandia’s chapter calls the most widely commercialized. Both sides are vanadium in acid, which matters because if some of one liquid crosses into the other, the capacity is not permanently destroyed the way it is when two different elements contaminate each other. The chapter gives the discharge reactions as V5+ gaining an electron on the positive side and V2+ becoming V3+ on the negative side, a nominal cell around 1.6 volts, vanadium concentrations on the order of 1 to 3 molar, and an energy density around 20 watt-hours per liter. That density is why the tanks are large. A 2012 Department of Energy fact sheet makes the same architectural point and adds the limits of that generation of vanadium batteries: sulfuric-acid electrolyte, a narrow temperature window, expensive membranes, low energy density, and a capital cost that, in that fact sheet’s judgment, blocked broad use. A 2012 judgment is not a 2026 price. It is a warning not to describe flow batteries as a cheap default.

The Storage Innovations assessment, a later DOE document, still treats aqueous vanadium as a technical success for the symmetric chemistry, because crossover is manageable and rebalancing deals with water movement over time. It also says vanadium’s cost and supply are what hold the market back. Iron-chromium, which Sandia notes was invented at NASA in the late 1970s, is the contrasting case: different elements on the two sides, so crossover is a permanent capacity loss, and the chromium reaction wants a catalyst. A cell voltage around 1.2 volts is the figure in that chapter for iron-chromium. None of these cell voltages is a battery you put on a 48-volt solar bus without a power-conversion stage sized for the stack.

Where a flow battery belongs in a solar plant

It belongs where the hours of storage are long enough that adding tanks is the point, and where there is a room for pumps, membranes, and maintenance. Solar.com’s marketplace writing, as cited on the chemistry comparison page, puts flow batteries with businesses and utilities rather than with house walls. This article does not need a second opinion on that placement. A house that needs to shift solar into the evening is the lithium iron phosphate conversation, in why LFP batteries are used for solar storage. A plant that needs many hours, and can tolerate pumps, is the flow conversation. Using one quote to answer the other job is the mistake.

The solar side is still an electrical interface. The stack does not connect to a module string because both are DC. Voltage, current, and the power electronics have to be designed. A grid-tied plant still has to cease energizing an island within 2 seconds under IEEE 1547-2018 when the utility is gone, unless a separate islanding design was built. A flow battery does not waive that. It also does not run without its pumps. A stopped pump is a stopped battery, even if the tanks are full. That auxiliary load is part of the round-trip, and it should be in the efficiency the vendor claims. If the efficiency number has no pump in it, ask again.

Safety is the electrolyte you can spill, not a slogan that liquids are harmless. The vanadium system in the handbook is an acidic solution. Secondary containment, eyewash, and a maintenance plan for seals are part of owning it. They are not part of owning a sealed LFP pack. Neither chemistry is “set and forget.” Warranties on the stack, the membrane, and the electrolyte can expire on different clocks. Ask which clock you bought.

What a flow battery specification should state

Cycle life claims for flow batteries are easy to overread. The symmetric chemistry can be rebalanced, which is the reliability argument in the Storage Innovations text. A membrane and a pump still wear. Ask what is included at year ten, and do not convert a tank-life sentence into a promise that nothing is serviced.

Frequently asked questions

Is a flow battery a lithium battery with a pump?

No. The energy is in dissolved chemicals in tanks, not in a solid lithium electrode. Hybrid designs exist and break the simple split. Ask whether both sides are liquid.

Why are flow batteries not on house walls?

Energy density around 20 watt-hours per liter, in the vanadium description Sandia gives, means large tanks, plus pumps and an acidic electrolyte. The house job is a sealed pack cycled daily. That job is the LFP page, not this one.

Does a flow battery last forever because the liquids can be rebalanced?

Rebalancing is why vanadium crossover is not the same failure as mixing two different elements. Pumps, seals, membranes, and the power electronics still have service lives. The Storage Innovations text treats supply cost, not immortality, as the market limit.

Sources and references

Tanks, stack, membrane, vanadium reactions, about 1.6 volts, 1 to 3 molar, about 20 Wh/L, and the iron-chromium contrast: Sandia, DOE electricity storage handbook, redox flow chapter. 2012 limits on density, temperature, membranes, and cost: US Department of Energy, vanadium redox flow battery fact sheet, October 2012. Symmetric chemistry, rebalancing, and vanadium supply: DOE Storage Innovations 2030, flow batteries. Islanding still applies: NREL and Sandia IEEE 1547-2018 primer.