A Norwegian company is developing silent diesel generators based on a new kind of fuel cell. Nordic Power Systems, which is making the generators for that country’s military, has successfully tested a 250-watt solid-acid fuel cell developed by SAFCell, a spinoff from Caltech. The companies are now working on a 1.2-kilowatt system.

Solid-acid fuel cells are still at an early stage of development. But SAFCell says that they’re simpler than conventional fuel cells, and the key components (such as the electrolyte) can be made from relatively cheap materials. The researchers developing the technology think it could be cheap enough to replace the turbines used in high-efficiency power plants. (The high cost of existing fuel cells limits them to niche applications, such as backup power.)

The new generators work by producing hydrogen gas from diesel in a process called reforming (the fuel is heated, but not combusted, and mixed with air and steam). The hydrogen is then fed into the fuel cell to make electricity. Unlike the fuel cells that have been tested in cars, the new ones can tolerate impurities, such as carbon monoxide, that are present in hydrogen made from diesel. In large-scale production, the new fuel cells could also be significantly cheaper than high temperature solid-oxide fuel cells, such as those being sold by Bloom Energy, because they operate at lower temperatures, and so don’t require expensive heat-tolerant materials, says Calum Chisolm, SAFCell’s CEO.

Solid-acid fuel cells were first demonstrated in the lab 10 years ago. They’re based on solid acids that are good at conducting hydrogen ions, or protons, a class of chemicals that were discovered in the 1980s, but were thought to be impractical for fuel cells because they dissolve in water, which is produced when fuel cells combine hydrogen and oxygen. Sossina Haile, a professor of materials science and chemical engineering at Caltech, and her colleagues found a simple way around this problem: operate the fuel cells at temperatures high enough to turn the water into steam, which doesn’t dissolve the solid acids.

The resulting fuel cells combined the benefits of two main types of fuel cells: Polymer-electrolyte-membrane fuel cells and solid-oxide cells. Polymer-electrolyte-membrane fuel cells, the type GM and other car companies use in their prototype fuel-cell vehicles, are convenient because they run at low temperatures. But at these low temperatures, carbon monoxide can collect on catalysts and prevent them from doing their job. This requires them to use purified hydrogen fuel, which isn’t widely available. The new solid-acid fuel cells can run at higher temperatures (250 °C instead of 90 °C) at which carbon monoxide isn’t a problem, so they can run on hydrogen made on the spot from natural gas and even relatively dirty fuels such as diesel, which is far more readily available than hydrogen.

In their ability to use a range of fuels, the new fuel cells are like solid-oxide ones. But the latter typically operate at high temperatures–800 °C to 1,000 °C–and require expensive materials. The new fuel cells, once in commercial production, are expected to cost about as much as solid-oxide fuel cells being sold by Bloom Energy, Chisolm says, but costs could come down quickly to about a tenth of the cost of the Bloom technology as the company develops and implements a range of cost saving measures. At that point, the fuel cells would be cheap enough to be competitive with high-efficiency turbines used in power plants.

One key challenge is reducing the amount of platinum catalyst used, says Robert Savinell, a chemical engineering professor at Case Western Reserve University. Haile and the researchers at SAFCell have already identified a platinum-palladium catalyst and catalyst deposition methods that both reduce the amount of platinum required and increase power output, but the amount of platinum needs to be reduced more. They’re developing new catalysts that take advantage of the fact that the system works at relatively low temperatures.

Another option is recycling the platinum, a relatively simple process because of the chemical composition of the fuel cells, Chisolm says. That, combined with a good financing plan, could allow the fuel cells to hit the $1,000-per-kilowatt milestone widely regarded as the point at which fuel cells will see mass adoption, he says.