BETTER BATTERIES are the key to electric cars that can drive for hundreds of miles between rechargings, but progress on existing technology is annoyingly incremental, and breakthroughs are a distant prospect. A new way of organizing the guts of modern batteries, however, has the potential to double the amount of energy such batteries can store.

The idea came to Massachusetts Institute of Technology professor Yet-Ming Chiang while he was on sabbatical at A123 Systems, the battery company he co-founded in 2001. What if there was a way to combine the best characteristics of so-called flow batteries, which push fluid electrolytes through the cell, with the energy density of today's best lithium-ion batteries, the kind already in our consumer electronics?

Flow batteries, which store power in tanks of liquid electrolyte, have poor energy density, which is a measure of how much energy they can store. Their one advantage is that scaling them up is simple: you just build a bigger tank of energy-storing material.

Chiang and his colleagues constructed a working prototype of a battery that is as energy dense as a traditional lithium-ion battery but whose storage medium is essentially fluid, like a flow battery. Chiang calls it “Cambridge crude”—a black slurry of nanoscale particles and grains of energy-storing metals.

If you could visualize Cambridge crude under an electron microscope, you would see dust-size particles made of the same materials that make up the negative and positive electrodes in many lithium-ion batteries, such as lithium cobalt oxide (for the positive electrode) and graphite (for the negative one).

In between those relatively large particles, suspended in a liquid, would be the nanoscale particles made of carbon that are the secret sauce of this innovation. Clumping together into a spongelike network, they form “liquid wires” that connect the larger grains of the battery, where ions and electrons are stored. The result is a liquid that flows, even as its nanoscale components constantly maintain pathways for electrons to travel between its grains of energy-storage medium.

“It's really a unique electrical composite,” Chiang says. “I don't know of anything else that is like it.”

The fact that the working material of the battery can flow has raised some interesting possibilities, including the idea that cars equipped with these batteries could drive into a service station and fill up on Cambridge crude to replace their charge. Chiang's collaborator on the project, W. Craig Carter of M.I.T., proposes that users might be able to switch out something resembling a propane tank filled with electrolyte rather than recharging at an outlet.

Transferring charged electrolyte into and out of his batteries is not the first commercial application that Chiang is pursuing, however. Along with Carter and entrepreneur Throop Wilder, he has already founded a new company, called 24M Technologies, to bring the team's work to market. Carter and Chiang are guarded about what the company will release first, but they emphasize the suitability of these batteries for grid-storage applications. Even a relatively small amount of storage can have a significant impact on the performance of intermittent energy sources such as wind and solar, Chiang says. Utility-scale batteries based on his design would have at least 10 times the energy density of conventional flow batteries, making them more compact and potentially cheaper.

Cambridge crude has a long way to go before it can be commercially viable, however. “A skeptic may say that this new design offers significantly more challenging problems to solve than benefits a potential solution may offer,” says the head of a major research university's energy-storage program, who spoke on condition of anonymity so as not to offend a colleague. All the extra machinery required to pump the fluid through the battery's cells adds unwanted mass to the system. “The weight and volume of the pumps, storage cylinders, tubes, and the extra needed weight and volume of the electrolyte and carbon additives could make [the technology heavier than] the state of the art.” These batteries may also not be as stable, across time and many cycles of charging and discharging, as conventional lithium-ion batteries.

A more fundamental issue is that charge times for these new batteries would be slower—two to four times slower, Carter says, than conventional ones. This creates a problem for cars, which require rapid transfers of power. One work-around could be pairing it with a conventional battery or an ultracapacitor, which can discharge its energy in a matter of seconds, to buffer transfers during braking and acceleration.

The new design has promise, however. A system that stores energy in “particulate fluids” should be compatible with almost any battery chemistry, says Yury Gogotsi, a materials engineer at Drexel University, making it a multiplier on future innovations in this area. “It opens up a new way of designing batteries,” Gogotsi says.