the Air Vent

Because the world needs another opinion

Thermoelectric Generation

Posted by Jeff Id on September 26, 2010

Catching heat is the purpose of every internal combustion engine ever developed.  Anywhere you see a hear gradient (difference in temp) between two objects, there is energy flow available and energy to be captured.  Think of it as a pre-charged battery.  Warm ocean surface water vs cold subsurface, hot engine vs cold air etc. The problem is that small differences in temperature are hard to work with.

I received a link today from a reader I won’t name without permission, he linked to an article on what seems to be a breakthrough in thermoelectric generation.

The claim is:

“We anticipate the thermoelectric voltage using our design to be about 100 times larger than what others have achieved in the lab,” Stafford added.

Now thermoelectric completely stinks as a generator these days but a 100 times improvement is pretty substantial to say the least.  I’m not enough of an expert on the cutting edge but if you want to even get 10 milliamps at 3 volts out of a $50 pieltier effect device, you had better have a hundred degrees of temp difference.  I’m sure many people who read that will have corrections and additions, but that has been my experience.

The news article at University of Arizona is here.

The paper is online here.

Again, I’m not enough of an expert to validate any claims or even judge their significance, but it does sound interesting for at least one application I can think of.

10 Responses to “Thermoelectric Generation”

  1. Jeff,
    I didn’t find it convincing. First, they haven’t actually made anything. It’s all calculation (not that there’s anything wrong with that :).

    Secondly, it seems unthermodynamic. They talk of molecule-thick layers. But as you said, the energy depends on the temperature difference between the electrodes, and you couldn’t get much over that distance. In fact, there’s a bind – thin for best electrical effect and low material cost, but thick to force the flux to cross a big temperature differential.

    This is a common bind in waste heat proposals. Waste heat is waste, and you generally want it to go away. But recovery obstructs its exit, because you want the max temp diff across that part of the path.

    I note the paper is unpublished (arxiv).

  2. Al S. said

    There’s certainly potential for heat engines and thermoelectic systems of some types. The old soviet orbital military radars were supposed to be powered by (nuke-heated) thermionic emission devices, like stacked vacuum diodes; I suspect they would convert some of the gamma to electric power also, potentially with efficiencies greater than 50%. We aren’t talking waste heat recovery here, though.
    And there are thermo-acoustic heat engines, for natural gas liquification and whatnot, that would save lotsa energy, if we ran ’em on a heat source like a molten-salt nuke.

  3. Steve Fitzpatrick said


    I’m with Nick on this one. They have made nothing physical (not even single molecules!). All they have is quantum chemistry calculations. I would not give this even a 10% chance of ever turning into something microscopic, never mind macroscopic. I am reminded of Maxwell’s demons and molecular one-way trap-doors….. which also ended up being nothing but imaginary. Don’t hold your breath waiting for a simple way to by-pass the well known limitations of thermoelectric devices.

  4. Alexander said

    In the late 1970s, GE produced 23% efficient BiTe thermoelectric devices. They were used in the Hawaii OTEC experiments.

  5. hempster said

    Few years ago there was a company called ENECO with similar claims:*/

    Behind this (scam?) was Peter Hagelstein,

  6. mrpkw said

    “Using computer simulations, Bergfield then “grew” a forest of molecules sandwiched between two electrodes and exposed the array to a simulated heat source.”

    We know how well those models work.
    Sounds like they may be fishing for more grant money.

  7. Retired Engineer said

    Nuclear thermal power is used in most deep space probes. Too far for PV solar. Apollo left some similar devices on the Moon. But it doesn’t generate all that much power. Sterling engines run on heat. And need a rather large difference to do much of anything. So count me on the skeptic side of this one.

    However, it is possible to make cold from hot, propane or LP gas powered refrigerators come to mind, in RV’s or remote cabins without electricity. So it may not be totally unreasonable to produce electric power this way. Thermal solar, perhaps using Sterling engines, does work.

    Pieltier devices are useful for generating local cooling of small things. No moving parts, and lousy efficiency. I’ve used them to cool detectors. Generate power? Easier to push a wet string up a steep hill.

  8. Don Keiller said

    it’s a model.
    Nuff said.

  9. BillyBob said

    All I know is that if Nick Stokes believes it is impossible, it probably is possible.

  10. John M said

    Whew, that press release once again reinforces what I think should be rule number one for academic researchers:

    Insist on review and veto power or you will end up sounding like an idiot!

    “Could” make CFCs obsolete? Helloooo!

    “You could just take a pair of metal electrodes and paint them with a single layer of these molecules,”

    Oh sure, like falling off a log!

    And the last line in the paper itself:

    “The efficiency of PPE-based devices increases with ring
    number and is only limited by the electronic coherence
    length, suggesting that highly efficient molecular-based
    thermoelectric devices may soon be realized.”

    Reminds me of the overhyped field of “molecular electronics”.

    Anyone been able to buy any devices yet?

    Note the date of the “mid-life crisis”.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: