Cool Chips plc wants to start a revolution in refrigeration. The company claims that its devices — the size of a compact disk — could chill an entire house.
At this week’s Nanotech Planet Spring 2002 conference in San Jose, Calif., the company spoke for the first time about the science behind its cool technology. It also revealed prototypes of its cooling wafers, circular sandwiches of electrode surfaces separated by a gap of about 15 nanometers.
With such a infinitesimal gap, electrons follow the laws of quantum mechanics, “tunneling” across and moving heat from one side of the wafer to the other in the process. One side of the chip gets hot, the other cool. The small thin devices would have no moving parts or use environmentally unfriendly chemicals.
 |
If Cool Chips can leap some of its commercial hurdles with the help of corporate partners, within a year it could begin making energy efficient, low-cost devices that would displace large, heavy compressors in everything from refrigerators and air conditioners to cars and infrared sensors in satellites. The entire cooling industry is about a $100 billion business.
One of the company’s first applications could be cooling computers, electronic equipment and food galleys in commercial airliners — a billion dollar market opportunity, according to Cool Chips President Isaiah Cox. Boeing’s Phantom Works advanced R&D group evaluated the technology in November, 2001. The group found that the scientific principles underlying the technology were sound, but further development was needed.
Eric Gulliksen, a MEMS analyst with Venture Data Corp. in Natick, Mass., thinks that if Cool Chips can deliver on its promise, it has a significant market opportunity for cooling microprocessors alone.
“Heat dissipation is a huge problem in all kinds of computing devices,” he said. As chip circuitry continues to shrink and run ever faster, computer chips are generating more and more heat. “Chips in high-end video cards are even worse than CPUs.” He added that there are billions of embedded microprocessors in devices like microwaves and stereos that could become more powerful if effectively cooled.
The company believes that Cool Chips will be substantially more efficient than either conventional compressors in cooling systems or thermoelectric devices that employ the Peltier Effect, first discovered in the 19th century. Companies such as Tellurex Inc. and Coolworks Inc. make Peltier-based cooling modules that have about a 5-8 percent Carnot efficiency, a measurement of the effectiveness of a heat pump or cooling device.
Cool Chips projects that its devices could have a 70-80 percent Carnot efficiency. In comparison, high-performance compressors in air conditioners and refrigerators measure 45-50 percent.
However, those eye-popping Carnot numbers should be taken with a small helping of salt, according to an engineer familiar with Cool Chips. While electron transport of heat across a gap would indeed be highly efficient, some efficiency would be lost throughout the device, such as at the edges of the material and through imperfections in the electrode material.
Even at 60 percent Carnot efficiency, Cool Chips may have other commercial advantages. For example, if the devices work as promised, they would produce a lot of cool in a small area. The company says its can potentially deliver cooling power of about 5,000 watts per square centimeter. Put in perspective, today’s top computer chips need about 100 watts of cooling power per square centimeter.
Cox said that a square inch panel of Cool Chips could power a household refrigerator. He also noted that as much as 20 percent of electricity generated goes to running the cooling compressors.
Cox also believes his chilly chips could be manufactured for pennies per watt, and with high profit margins. He said that the chips can be made with relatively inexpensive materials in about 10 steps in a chip fab. “This is a beautiful technology for someone like Motorola,” he said. “The cost of material is infinitesimal compared with silicon.”
With the right development and production partners such as an IBM or Intel, Cox believes that the company could be in production in as little as 10 months. “We want to collaborate with the appropriate company to solve production issues such as surface roughness of the electrodes that will speed our path to mass production.”
The company has followed its own long path to this week’s announcement. Cool Chips is a majority-owned subsidiary of mining company Borealis Exploration Limited and publicly traded on the over the counter (OTC) securities market. Cool Chips has done all its own research and kept itself under the radar for years.
“Most scientists want the fame that goes with making important discoveries,” Cox said. “We wanted the fortune that we hope commercializing such a technology will bring Cool Chips.”
Still, for all the tantalizing promise Cool Chips presents, the road ahead is not certain. As the engineer familiar with the technology concludes, “The potential is great, but the devil is in assumptions, God is in the details. They’ve showed that the technology is at a point of becoming practical and commercially feasible. But the real work is ahead.”