Skip to content Skip to navigation

Stanford scientists advance new way to store wind and solar electricity on a large scale, affordably and at room temperature

Stanford PhD candidates (from left) Geoff McConohy, Antonio Baclig and Andrey Poletayev have developed a new type of flow battery that could lead to affordable storage of renewable power. (Image credit: Mark Golden)
JULY 19, 2018

A new type of flow battery that involves a liquid metal more than doubled the maximum voltage of conventional flow batteries and could lead to affordable storage of renewable power.


A new combination of materials may realize the potential for a special type of rechargeable battery to store large amounts of renewable power to be delivered when needed to the electric grid quickly, cost effectively and at normal ambient temperatures.

When mixed, sodium and potassium become a liquid metal at any temperature above freezing
and normal pressure, as seen in this video. The alloy can contain 40% to 90% potassium by
weight. (Video credit: Antonio Baclig)

For years, researchers have pursued practical application of the technology, known as “flow batteries,” for the grid. These batteries store an electron donating fluid and an electron absorbing fluid in separate, large tanks and can flow the fluids together for a chemical reaction that produces electrical current when needed. Researchers have mostly experimented with electrically active molecules dissolved in water, but the water limits the chemicals that can be used to ones with low voltages. Other investigated fluids have required extremely high temperatures or used very toxic or expensive chemicals.


Stanford professor of materials science and engineering, William Chueh, his PhD student Antonio Baclig, and Jason Rugolo, now a technology prospector at Alphabet’s research subsidiary X Development, decided to try sodium and potassium (Na-K) as the fluid for the electron donor—or negative—side of the battery. When mixed, these elements form a liquid metal at room temperature. This liquid has at least 10 times the available energy per gram as other candidates for the negative-side fluid of a flow battery.

For the separator between negative and positive sides of their first cells, the team used a ceramic made from sodium and aluminum oxide. This membrane must conduct ions between the fluids as an essential part of producing current, without letting other chemicals pass. The Stanford scientists expected this membrane to conduct positive sodium ions to the positive side of the cell as the electrons exited the negative side to the circuit.

The group's research is featured on the cover
of  the July 18, 2018 issue of Joule.

In the first experiments, however, potassium from the Na-K fluid entered and quickly fractured the ceramic membrane, rendering the devices useless. The team suspected that changing the mobile positive ion from sodium to potassium could result in a stable, high-voltage device. All they had to do was change the membrane from one made of aluminum oxide and sodium to a less common one made of aluminum oxide and potassium.

It worked. While water-based flow batteries are limited to about 1.5 volts, the new device had open-circuit voltages of 3.1–3.4 V. Even non-aqueous, molten salt-based flow batteries, which operate only at extremely high temperatures, have been limited to a maximum of 2.3 V. And the new membrane remained stable with Na-K for thousands of hours of operation. The research is the cover story of the July 18 issue of Joule, a peer-reviewed journal on sustainable energy technology.

“We still have a lot of work to do,” said Baclig, “but this is a new type of flow battery that could affordably enable much higher use of solar and wind power using Earth-abundant materials.”

Higher voltage leads to lower cost


The open-circuit voltage of the battery measures how much energy is transferred by each electron. By doubling the voltage, for example, each electron that passes through the battery carries twice as much energy. This significantly increases the amount of energy stored relative to the battery’s size, as well as the speed with which the battery can supply power, known as power density. The new devices demonstrated maximum power densities of 65 milliwatts per cm2 at 22 degrees Celsius (72 F).

“This power density is promising because it shows that the membrane is really the only thing limiting the power, and we can improve that,” explained Chueh. “Previously, people looking at this battery would have thought that interfacial effects on both the Na-K side and the water-based positive side would have caused a much lower power density.”

Sodium-potassium alloy is a room-temperature liquid metal that
could unlock a high-voltage flow battery. The purple dots represent
potassium atoms and the blue dots are sodium. The ceramic
membrane conducts positive potassium ions to the positive side
of the battery during discharge, and back to the negative side
during recharging. (Image credit: Antonio Baclig)

Most significantly for a stationary battery, higher voltage results in lower manufacturing costs.

The team of Stanford PhD students involved, which in addition to Baclig includes Geoff McConohy and Andrey Poletayev, also found that the ceramic membrane very selectively conducts potassium ions, allowing virtually no sodium to migrate to the positive side of the cell. This is very encouraging for a long lifetime of Na-K batteries utilizing the new membrane.

However, while Na-K contains far more available energy per liter of fluid than most chemicals used in previous flow batteries, the new device’s membrane slowed its power output.

“Typically, these types of membranes are used at temperatures higher than 200 degrees Celsius, (392 F), which allows the potassium ions to move even more quickly through them. We wanted a room-temperature battery, which means the membrane resistance can be limiting,” said Baclig. “We experimented with a thinner membrane, which boosted the device’s power and showed that refining the membrane’s design is a promising path.”

Also, the various fluids the Stanford team tried on the positive side of the battery limited its ability to store accessible electricity. At a normal room temperature, the storage capacity shown was comparable with other types of flow batteries.

The work ahead

The study experimented with four different fluids on the positive side of the battery. Two used water-based solvents, and two used carbonate-based solvents. Previous work showed that water degrades this class of membranes. As expected, the ceramic membrane interacted with the water-based fluids, leading to an unusable device after just a few minutes. Chemical additives and coatings reduced the degradation considerably, but water presents another problem: It is highly reactive with Na-K.

The team is now investigating positive-side fluids, focusing on non-water-based ones.

“A new battery technology has so many different performance metrics to meet: cost, efficiency, size, lifetime, safety, etc.,” said Baclig. “We think this sort of technology has the possibility, with more work, to meet them all, which is why we are excited about it.”

This project was funded by Stanford’s TomKat Center for Sustainable Energy, the Anthropocene Institute, the State Grid Corporation of China through Stanford’s Energy 3.0 corporate affiliate program, the National Research Foundation of Korea, the U.S. National Science Foundation and Stanford Graduate Fellowships.


More about this project>>

Read more spotlights>>