Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Heat to electricity question

  1. Mar 31, 2010 #1
    Someone told me that it is possible to directly translate heat into electric current.

    This seems like a fantasy to me, especially when I consider the claim that ambient room or outdoor heat could be made into electricity, and as a result the conductor would become cooler.

    That would mean that you could cool your house by running other appliances!

    Still, because electricity can be generated from light I started wondering if there might be some truth.

    The best I could come up with is that some sort of capacitor (or whatever it is that stores charge and then releases it) could be dissipated and would then recharge from a heat-conductor. If this effect could somehow be made regenerative, would the "electricity vacuum" of the capacitor draw charge from the heat conductor, causing it to cool whereby it would draw more heat in from the surrounding air, hot water, etc.?

    A bizarre and naive question I'm sure, but maybe someone has some insight. Thanks.
  2. jcsd
  3. Mar 31, 2010 #2


    User Avatar

    Staff: Mentor

    Yes, you can. It's called the seebeck or thermoelectric effect: http://en.wikipedia.org/wiki/Thermoelectric_effect
    No, that's not what it means. It doesn't take ambient heat and convert it to electricity (a violation of the 2nd law of thermodynamics), it takes a temperature difference (like any mechanical heat engine) and uses the heat transfer across that temperature difference to generate energy.

    One example is the RTG generators that satellites use. They take the heat from a radioactive decay source on one side of the thermoelectric device and radiate that heat into space on the other side, absorbing some of that heat transfer to generate electricity.
  4. Mar 31, 2010 #3
    A http://en.wikipedia.org/wiki/Thermocouple" [Broken] is a temperature sensor that utilizes the fact that any junction of dissimilar metals will produce an electric potential related to the temperature difference of the metals.
    Last edited by a moderator: May 4, 2017
  5. Mar 31, 2010 #4
    Thanks for the link. As for the temperature-differential/entropy issue, I had thought about that. I didn't know about the satellites.

    Still, I'm wondering if an electric current has some sort of momentum where, once moving, it could create some sort of "vacuum" that would draw energy into the current.

    If this were the case, couldn't an existing electric current be used to draw more electricity/energy out of something else, such as a hot conductor?

    I know this sounds like grasping, but I'm just exploring all the possibilities.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook