Environment

MIT's flow battery could store solar and wind power on the cheap

MIT's flow battery could store solar and wind power on the cheap
MIT's high-performance membranelss flow battery could be used to store green energy for the grid (Image: Felice Frankel/MIT)
MIT's high-performance membranelss flow battery could be used to store green energy for the grid (Image: Felice Frankel/MIT)
View 2 Images
A diagram of a vanadium flow battery
1/2
A diagram of a vanadium flow battery
MIT's high-performance membranelss flow battery could be used to store green energy for the grid (Image: Felice Frankel/MIT)
2/2
MIT's high-performance membranelss flow battery could be used to store green energy for the grid (Image: Felice Frankel/MIT)

Researchers at MIT have come up with a new design for a rechargeable flow battery that does away with the expensive and ineffective membrane of previous designs. The device could prove the ideal solution for effectively storing energy from intermittent power sources such as solar and wind power.

What is a flow battery?

A diagram of a vanadium flow battery
A diagram of a vanadium flow battery

A flow battery is a special type of rechargeable battery in which two liquids with opposite electric charge (electrolytes) exchange ions, converting chemical energy directly into electricity. The electrolytes are usually separated by a thin membrane that lets them exchange ions without mixing.

The electrolytes are stored separately from the cell itself, in two big tanks, and the electrolyte is pumped into the cell as needed. This means the system can be scaled easily, simply by changing the size of the tank. Doing so can produce systems of vastly different capabilities, from a few kWh up to several MWh.

Scalability aside, flow batteries come with many more perks: they can stay idle for long periods of time without losing their charge, they have a quick response time, and they can charge and discharge quickly just by replacing the electrolyte fluid. For these reasons, over the past few years some people have advanced them as a way to quickly refuel electric cars.

On the flipside, flow batteries are more complicated than standard batteries, each requiring their own system of pumps and sensors; moreover, energy densities are usually lower than those of your standard Li-ion battery.

A battery for the future

The main challenge in developing an effective flow battery has been to find a good balance between performance and costs. The electrolytes used are typically not very expensive, but they tend to eat away at the costly membrane, shortening the battery's lifetime. The MIT team's solution circumvents the issue in perhaps the most elegant of ways – by taking out the membrane altogether.

The small flow battery prototype built by the researchers uses a curious phenomenon in fluid dynamics called laminar flow: if both liquids are kept at low enough speeds and other conditions are satisfied, the two electrolytes won't mix, making the membrane superfluous.

Pumping liquid bromine over a graphite cathod and hydrobromic acid and hydrogen gas over a porous anode, the researchers created a reservoir of free electrons that can be discharged or released at will.

While other teams had attempted a membraneless design before, this is the first one in which the battery can be recharged as well as discharged. Their flow battery produced up to 0.795 watts per square centimeter: that's three times as much as other membraneless systems, and about 10 times higher than most lithium-ion batteries.

Future developments

The researchers also generalized the manufacturing parameters of the flow battery, by creating a mathematical model that they can use to optimize the system and eventually build larger-scale devices that are better suited to grid applications.

Previous membraneless systems have been largely unpractical, but the scaled-up version of the device could have a substantial real-word impact because it could be used to produce energy for a very competitive US$100 per kilowatt-hour. "Most systems are easily an order of magnitude higher, and no one’s ever built anything at that price," says William Braff, who was part of the research team.

One area in which this technology could be put to good use is the storage of renewable energy. Since sunlight and wind are highly unpredictable power sources in the short term, being able to store large quantities of clean energy to use as a buffer is essential if green energy is going to continue to satisfy a larger and larger portion of our energy needs.

The team's research appears in the journal Nature Communications.

Sources: MIT, Electropaedia

9 comments
9 comments
Max Kennedy
Please explain "produce energy for a very competitive US$100 per kilowatt-hour." If as stated that is bloody expensive energy. If it means $100/kWh storage capacity that is an entirely different beast!
Slowburn
While they can get it to work in the lab, the real world is a different beast all together.
Dale Whitworth
What about use in a vehicle? Mixing would appear to occur as a result of vibration and bouncing. And I distinctly remember seeing this same design in the 70's in PopSci or Mechanics Illustrated. It had the same problem then; worked OK stationary, failed when mobile. As for grid use, it seems like a great design.
notarichman
dale, they aren't predicting the use of the flow battery for vehicles...only for storage of electrical power such as the water behind a dam. Max, i believe the article is stating the cost of the equipment, not the cost of the electrical power.
this method could also be used to store power produced via tidal generators.
my questions are: 1. how long did they test the battery? 2. ease of construction, i.e. could a home owner build one to replace his storage batteries? 3. Is MIT licensing the method? and to which corporations? 4. How does this method compare to systems using fuel cells in cost? 5. same question with other storage batteries?
Dan Alger
Expecting energy storage for $100/kWh (or per kilowatt-hour) is the really big news. It should be your lead.
Except for big hydropower facilities, which depend upon very favorable geography, large amounts of energy "storage" needed for the grid are now provided most inexpensively with natural gas power plants sitting and waiting to be turned up, and these idling facilities cost about $1000/kWh.
These batteries even have an additional advantage in that really large amounts of power could be discharged more quickly than natural gas plants can be fully powered up, which is about 15 minutes. Many of the batteries now being put into place with wind, like those from GE, are to provide a fast response within this 15 minutes.
piperTom
The article states that the battery can supply "up to 0.795 watts per square centimeter", a measure of area. This raises the first question: area of what? From the context, it must be the area where the two electrolytes are in contact (but do not mix). But we are given no hint as to how much mechanism is required to produce and control the precise flows needed and thus, no idea how this relates to the overall size, weight, or cost of the battery. So the number is useless. A very precise THREE significant digits of useless!
Goran Pocina
Well, based on wishful thinking, rather than access to the original paper, $100 per 1 kW power capacity might be more relevant for a flow battery, whose capacity can be altered by altering tank size. Still $100 per kW/hr could be an indicator of how much the tank and electrolytes cost.
Roman Smetak
big oil will destroy any action on this in no time at all
Joel Detrow
This reminded me of the Liquid Metal Battery, from the company now known as Ambri. I wonder how it compares?