Thermoelectrics Revisited — Again

The bytes were barely dry on the 14 Oct UFTO Note about thermoelectrics (“There is a tantalizing hope that someday someone will come up with a real breakthrough in direct heat-to-electricity conversion.”), and the following day, just such a possible breakthrough came to my attention in an item in EV Progress. (www.EVProgress.com)

The article talked about dramatic claims made at the September Global Powertrain Congress in Ann Arbor, Michigan for a “Power Chip” that would recover from 10-70 KW of the waste heat of a car’s engine as electric power. The technology is a new variant on direct thermal conversion.

Here is a portion the Power Chip’s own press release:
——————–
“Power Chips are discs comprising two electrodes separated by a gap of less than 20 nanometers, through which the hottest (most energetic) electrons tunnel to create an electrical current. Power Chips are silent, nonpolluting, solid-state devices that are scalable as arrays to meet any size power load. They can generate electricity from heat produced by any primary energy source……

They are projected to operate at 70% of the maximum theoretical [Carnot] efficiency for energy conversion, even when converting low-grade waste heat. The only other technology capable of converting such heat directly to electrical output is thermoelectric (Peltier) devices, but the efficiency of production thermoelectric systems is only 5-8%.

Power Chips™ are protected by an extensive patent portfolio covering general theory and specific techniques for quantum thermotunneling and thermal energy conversion. More details are available on the Power Chips plc Website http://www.powerchips.gi, including the full text of issued patents and photographs of prototype Power Chips.

Power Chips™ were invented and are being developed and licensed by Power Chips plc, a majority-owned subsidiary of Borealis Exploration Limited (BOREF). Both companies are incorporated in Gibraltar. Borealis’ business is reinventing the core technologies used by basic industries, including electrical power generation, cooling and thermal management, electric motors, and steel production.”
———————

Not mentioned was the obvious point that if you could do that, you wouldn’t bother with the IC engine in the first place. The company is in discussions with GM, who invited them to participate in the Powertrain conference.

I contacted the company, executed an NDA, and learned a great deal more about it through extensive conversations with management. Actually, the first product is going to be for cooling. (As with thermoelectrics, this process can be used either as a heat pump or a power generator.) It has attracted serious attention of major defense contractors for cooling of critical electronic components.
(See: http://www.boeing.com/news/releases/2001/q4/nr_011130a.html)

The parent company is Borealis, an unusual company with a colorful history dating back to an oil company founded in 1924.. There are over 100 employees scattered all over the world, and they draw on many additional institutions and people. The CoolChips subsidiary is already public (COLCF), and PowerChips and other subsidiaries are poised to go public as well. The long technology development has mostly been funded privately by private/family money of the principals, however they now recognize the need to broaden the base of support and involvement. A private offering memorandum is available from the company.

A great deal of technical and business information is available in various areas of the companies’ interlinked websites, powerchips.gi, coolchips.gi, and borealis.gi. The cooling technology was presented at the recent Long Beach 21st International Conference on Thermoelectrics, and another paper is being given today at the “Thermal Management” conference in Dallas. (Both events were cited in the 14 Oct UFTO Note).

Note in particular a new version of their technical overview dated Oct-28 (this is what is being presented in Dallas). Two nanotechnology milestones were reached recently: the fabrication of large conformal pairs of electrodes, and electrodes with excellent local smoothness. The document includes new detailed electron microscope data of the surfaces.
(http://www.coolchips.gi/technology/coolchipstech28Oct02v2.pdf)

The quantum tunneling theory is described in a paper by a group of Stanford materials researchers (I have the pdf if anyone would like to see it–it’s not easy reading unless you’re a quantum physicist–and even then it’s no walk in the park):

“Refrigeration by Combined Tunneling and Thermionic Emission in Vacuum: Use of Nanometer Scale Design”, Y. Hishinuma, T. H. Geballe, and B. Y. Moyzhes, Applied Physics Letters, Vol 78, No. 17, 23 April 2001.

According to their calculations, the basic tunneling process is ideally capable of delivering 95% of Carnot efficiency. The technical overview then goes through detailed analysis of losses, and comes up with a final figure of 70-80% of Carnot overall.

The physics theory is one thing; making a device is another. The company says it has developed reliable means to build such devices — with the unheard-of narrow gaps. Two small production lines are being debugged and ramped up currently.

First deliveries of the initial product are anticipated in a matter of months. It will be a several watt cooling chip, which will be offered for sale at a very high price. The device is said to be capable of delivering temperature differences of over 400 deg K, cooling down to 150 deg K with a hot side of 250 deg C.

If these claims bear out, even partially, it would truly be a game changer. If the devices can be made reliably and cheaply, then little would stand in the path, in every arena of refrigeration, power production and transportation, not to mention electronics. Time will tell.