IBM trying to move quickly to make solar cheaper than fossil fuels

Like it? Share it!

IBM trying to move quickly to make solar cheaper than fossil fuels

Even though solar power doesn’t have as much oomph as burning fossil fuels, it’s one of the most environmentally friendly forms of power generation. The problem right now is making solar generators that are cheap and highly efficient, which is exactly what IBM is trying to achieve— solar power that could replace fossil fuels in cost and efficiency.

An alternate approach to dropping the cost per Watt involves what’s called concentrating photovoltaics. There are some forms of photovoltaic chips that have a far higher efficiency than silicon, but are too expensive for use in standard, flat-panel arrays. To make these economical, you have to send them as much light as possible—the hardware that concentrates the sunlight on the chip is what gives the technology its name.

Bob Sandstrom, who provided a tour of some of the test equipment IBM has set up outside of Watson, said that it’s possible to buy triple-junction photovoltaic devices from a division of Boeing called Spectrolab that can exceed 36 percent efficiency. But this hardware is significantly more expensive than the same-sized silicon cell.

To get a better return out of these cells, it’s possible to use lenses to send these devices more photons. Doing that creates separate problems: the lenses can be expensive, and a good lens can send temperatures on the chip up to hundreds of degrees Celsius.

IBM first got interested in the chip cooling problem when it developed a gallium-indium alloy that was liquid across a broad range of temperatures, and found it could be used as an efficient thermal couple that links a chip to a heat exchanger. Although the first thought was to use it for cooling computer processors, the people in IBM’s Smarter Planet group decided it could also be used to draw the heat off one of Spectrolab’s chips. To develop the system, IBM partnered with the King Abdulaziz City for Science and Technology (KACST) in Saudi Arabia, which was interested in increasing its use of renewable power, including the possibility of using it to run a desalination plant.

Sandstrom showed off what he called the first generation design: a collection of large aluminum cones, with a fresnel lens across the wide end to focus the sunlight on the chip, concentrating it to 1,600 suns. At back was a bulky heatsink that looked like it had been pulled from a server made in the Pentium 4 days. A set of these cones were positioned on a large tracker that keeps them directed toward the Sun.

The whole thing was an impressive feat of engineering, but it was also costly.

Nearby, however, was a second-generation device. The cones were gone, replaced by a rectangular aluminum box with just a few simple fins on back to serve as a heat sink (the box itself also radiates heat). Instead of a single, large lens, the box had a sheet of material with rows of individual lenses imprinted on it. Each of these fed a small glass cone that focused the light even further. The liquid metal cooling system that got the whole project started was also gone (though Sandstrom wasn’t at liberty to tell me what replaced it).

Aside from the design changes, the intervening years had also made the inverters that convert the output to alternating current much more efficient. The net result of all of this? It’s approaching $2/Watt (but it’s not quite there yet).

Via