A former Sony TV factory near Pittsburgh is coming to life again after lying idle for four years. Whirring robotic arms have started to assemble a new kind of battery that could make the grid more efficient and let villages run on solar power around the clock.

Aquion, the startup that developed the battery, has finished installing its first commercial-scale production line at the factory, and is sending out batteries for customers to evaluate. It recently raised $55 million of venture capital funding from investors including Bill Gates. The money will help it ramp up to full-speed production by this spring.

Jay Whitacre, the Carnegie Mellon professor of materials science and engineering who invented the new battery, says it will cost about as much as a lead-acid battery—one of the cheapest types of battery available—but will last more than twice as long. And while lead is toxic and the sulfuric-acid electrolyte in lead-acid batteries is potentially dangerous, the new battery is made of materials so safe you can eat them (although Whitacre says they taste terrible). Nontoxic materials are also a good fit for remote areas, where maintenance is difficult.

Most importantly, by providing an affordable way to store solar power for use at night or during cloudy weather, the technology could allow isolated populations to get electricity from renewable energy, rather than from polluting diesel generators. Combining solar power and inexpensive batteries would also be cheaper than running diesel generators in places where delivering fuel is expensive (see “How Solar-Based Microgrids Could Bring Power to Millions”).

The batteries could allow the grid to accommodate greater amounts of intermittent renewable energy. As Aquion scales up production and brings down costs, the batteries could also be used instead of a type of natural gas power plant—called a peaker plant—often used to balance supply and demand on the grid. When recharged using renewables, the batteries don’t need fuel, so they’re cleaner than the natural gas power plants.

In some places, concerns over pollution make new natural gas plants hard to build, which could create an opening for Aquion’s technology, even if it’s somewhat more expensive.

Much of the sprawling factory where Aquion is setting up shop is derelict, with potholes in the floor and piles of abandoned ductwork and manufacturing equipment left over from making old-fashioned cathode ray tube TVs. Aquion has tidied up a section of the factory and installed equipment that’s ordinarily used for making aspirin tablets or wrapping chocolates in foil and arranging them in boxes. Now the equipment stamps out and precisely assembles battery electrodes along with foil current collectors to make batteries the size of briefcases.

By using cheap equipment originally developed for large, existing industries, Aquion is keeping down costs, Whitacre says.

The battery is made of inexpensive materials including manganese oxide and water. In concept, it operates much like a lithium-ion battery, in which lithium ions shuttle between electrodes to create electrical current. But the new battery uses sodium ions instead of lithium ones, which makes it possible to use a salt water electrolyte instead of the more expensive—and flammable—electrolytes used in lithium-ion batteries.

The trade-off is that the batteries store less energy by weight and volume than lithium-ion batteries do, so they’re not practical for cars or portable electronics. But space isn’t as much of an issue for stationary applications, where batteries can be stacked in warehouses or shipping containers. For storing large amounts of power from the grid, success is “all about cost,” Whitacre says.

Aquion will need to compete with companies such as GE and Fluidic Energy, which are also manufacturing novel batteries for the grid (see “GE’s Novel Battery to Bolster the Grid” and “Years in the Making, Promising Rechargeable Metal-Air Batteries Head to Market”).