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

Does physics forbid such a device; a heat destroyer

  1. Jul 2, 2012 #1
    Hello physics forum crew, I would like some help coming up with one of those "is this possible" type questions.

    Basically, I'm writing a sci-fi story that is has the strict limitation that all technology must be feasible within the laws of physics as we currently know them. However, there is some leeway as you can assume that man has advanced our technology for many thousands of years and might find a way to use old physics any new ways.

    Anyways, the question is this: can a "heat destroyer" be made?

    As I define it, this device takes simply converts heat into some other form of energy, either EM or perhaps electricity. Of course, this can already be done in many ways today, but what we're talking about is a matter of degree. The amount of power it generates is not important, nor is the efficiency, but the important part is it can do so "infinitely" i.e. you turn the device on and it brings itself to near absolute zero, I suppose a somewhat higher minimum cap is ok. Another limitation is, other than the heat, it can't be fed any other energy, except maybe for some control or other higher level stuff, but the key here is it's not like you have to feed this thing a huge amount of energy for it to work, it just "eats" the heat.

    My only lead is the carnot's work, perhaps the formula making clear that close delta T's make for very little work.


    But it it specifically talks about work and efficiency. I can't find a way to use the limit to definity rule out such a device.

    Any input is much appreciated.

  2. jcsd
  3. Jul 2, 2012 #2
    I suggest looking at http://en.wikipedia.org/wiki/Coefficient_of_performance to get an idea about the theoretical limits on a heat engine being used to cool a reservoir. The bottom line is that it takes an infinite amount of energy to get all the way to absolute zero, but you can get arbitrarily close with merely finite amounts of energy.

    Either way, though, you need energy to do this. You're moving heat from a cold reservoir to a hot one. This is like trying to push a ball up a hill. Energy is required to do this. It certainly can't suck all the heat out of a region and produce energy in the process.

    Rather, perhaps the solution to your problem is to use the concept of negative temperature. A system with a negative temperature is actually hotter than any system with a positive temperature, and heat will naturally want to flow from it to a positive temperature reservoir. Thus, the usual concepts of a heat engine producing energy apply--it makes the negative temperature reservoir colder and yet it extracts heat from the system and can do work.

    The downside is that most systems don't experience negative temperatures. You couldn't apply this idea to any arbitrary material or reservoir of gas, only very specific objects that could experience such a phenomenon.
  4. Jul 3, 2012 #3
    I can see where you are coming from, but I don't think that answers the fundamentals for this problem.

    It isn't a heat exchanger problem. The primary thought here is that the is machine converting heat to another form of energy, and can do this "perfectly", so any kinetic energy in the form of heat that hits the business end of this machine is converted to electric current. Converting energy from one form to another doesn't exactly "take" energy, but there are losses; so the question is does physics disallow the converting of increasingly up to (nearly)infinitely smaller quanta of kinetic energy.

    Here's an example: let's say you you have some sort of exotic "gas", lets not worry too much about what the gas is, but it's some sort of monoatomic or subatomic ion or subatomic particle, the gas is contained within some sort of nano-mechanic system that is basically is series of nano-magnetic structures in a conductive matrix. Every time a ion passes through the field it would general some(very small) amount of current, and loss some of it's kinetic energy, as it successive passes through more nano-magnetic structures, it would lose all of it's energy. I can't find a specific law that would forbid this time of system.

    Or am I missing something?
  5. Jul 3, 2012 #4
    Within the right conditions you might be able to get a blend of gasses to absorb a select
    spectrum of heat, and then other gasses in the blend radiate it in a different spectrum.
    This may not sound like much, but a microwave oven does this in reverse. (A narrow spectrum heats the water molecules, which then transfer the heat to other portions of the heated object.
    Also look at the breaks on a modern locomotive. The drive motor is reversed and the current generated is converted to heat via a resistor. The large resistor is air cooled.
    So no wearable parts!
  6. Jul 3, 2012 #5
    You can't convert heat into any other form of energy freely without putting something (energy not in the form of heat) in. It is impossible for any machine to just take heat from a reservoir at a single temperature and convert it into work, that's one way of stating the 2nd law of thermodynamics.

    Imagine you could do what you're suggesting, then if you wanted you could put your machine in a cold reservoir say and turn loads of its heat into some form of work, say making a wheel rotate. You can then use this rotating wheel to drive a conveyor belt in a hot reservoir against friction, producing heat in the hot reservoir and raising its temperature. You have now built a machine which takes heat from a cold reservoir to a hot reservoir at no cost, and that is definitely forbidden.

    I know you don't talk about two reservoirs in your example, but once you've turned heat into work, there's nothing stopping you from turning it back into heat again and dumping it wherever you want to, and nature can't give you the option to do that or you'd violate the 2nd law in the form it's more usually stated in.

    I think that's right anyway, if I've understood what you're saying right and thermodynamics right.
  7. Jul 3, 2012 #6

    I'm not sure I agree. Conversion does not "take" energy, but there are always losses. A photovoltaic cell for instance, converts light into electrical energy. It doesn't take any energy other than the photon striking the cell in order to generate the electricity, although some of the energy is converted into heat, rather than electrical energy. The loss can be reduced considerably, for instance if the wavelength of the light carried exactly the energy required to knock the electron into the conduction band and if the circuit was super conducting; in this case the vast majority of the energy of the photon would be converted into electrical energy, without any outside force.

    I am also not sure I understand your second example, but if I understand it correctely, in your example the transfer was not "for free" as the temperature of the cold reservoir decreased, energy was taken from it and not all of it was transferred to the hot reservoir as there would certainly be losses due to friction, entropy, etc, so it is doubly not for free.
  8. Jul 3, 2012 #7


    Staff: Mentor

    I would suggest a much simpler resolution. Just have your device somehow be in thermal contact with deep space. Then your Tc is just 2.7 K. It isn't "perfect", but it is pretty close, and it is the kind of thing which doesn't violate any known laws of thermo. You can sweep the details about how the device is in thermal contact with deep space under the "future tech" rug.
  9. Jul 3, 2012 #8
    No that part of my post I am certain of. There are only two ways of taking heat out of a reservoir, or 'destroying' the heat to use your terminology. The first is if you have an even colder reservoir handy, in which case just put the two in contact and it will happen. If you don't have that then the only possibility you have left is to use a heat pump where you take the heat from your reservoir and dump it still as heat in a hotter reservoir, you also have to then put yet more extra energy in to do this which is why freezers need to be plugged in. The greater the difference in temperatures, the more extra energy you will have to put in to power your heat pump. Those are the only two ways of getting rid of heat. It is impossible to do it without a second reservoir, you can't simply convert heat to some form of energy that isn't heat at no cost, for the reasons I outlined in my first post.

    Well if your machine worked there would be nothing in principle stopping my example from being 'for free'. All those little losses like friction all end up as heat, and so even the losses are heating up the hot reservoir while cooling the cold reservoir, which shouldn't be allowed. Even if it did only transfer 1% of the energy to the hot reservoir and the rest somehow found its way back to the cold one, that's still a result that is forbidden by the 2nd law, because a process has occurred which has the sole net result of taking some heat from a cold reservoir and transferring it to a hotter one.

    I really like DaleSpam's suggestion though, that seems like a really neat way of building exactly the machine you want to build.
  10. Jul 3, 2012 #9
    This is actually won't work. You can't be "in contact" with deep space, there is nothing to contact with. Although the average temperature of deep space probably what you state, it is not very effective at thermal dissipation, the (pretty much)only way an object cools off in space is via black body radiation. Thus a spaceship with a very energetic power source, let's say a fusion generator, would actually have quite a problem, the heat would need to be dissipated and space doesn't help much at all, it has zero thermal conductivity, you would have to construct some rather massive fins, and gain as much surface area on the ship as possible to maximize black body bleed off.

    You are may be correct about the specifics of thermo dynamics to to heat reservoir but this case does not simplify to the scenario that the thermo that you cite describes. You state that there are only two ways to remove heat from a reservoir, first of all, this problem as stated above, doesn't simplify to a simple heat reservoir; there are many more ways that I can think of off to remove heat from the theoretical machine that I can think of off the top of my mind, I feel this is probably a pretty strong indication the reservoir model doesn't fit well here.

    For example, all objects above absolute zero emit black-body radiation. A object that represents the "cold reservoir" in your example would be emitting photons, it is entirely conceivable for a cold reservoir to emit a photon which would then strike the object that represents the "hold reservoir" which is then absorbed and converted into thermal energy. I think this is not the best example, but in the case an assertion is made (that this situation is accurately described by a simple hot/cold reservoir problem) and can be contradicted, the assertion may be invalid.
  11. Jul 3, 2012 #10


    Staff: Mentor

    I didn't say "in contact", I said "in thermal contact". And yes, you can be in thermal contact with deep space. Heat transfer in deep space is via radiation to a 2.7 K thermal bath called the cosmic microwave background radiation.

    Yes, that is just an engineering problem, not a fundamental violation of known physics. So you can just sweep it under the "future tech" rug as I said earlier: they develop some elegant way to increase the effective surface area for thermal contact. The point is that it doesn't violate known physics like a "heat destroyer" would.

    Look, it is your story. You can accept or reject your "feasible within the laws of physics as we currently know them" rule at your own whim. But what you described simply does not fit that rule. Either the device or the rule has to go, they are incompatible.
  12. Jul 3, 2012 #11
    I haven't seen any evidence that the "heat destroyer" violates physics, nor has anybody presented me with conclusive evidence of such. There was a reason the "heat destroyer" was in quotes, if you read the first post, you would see that it is somewhat of a misnomer because it doesn't set out to destroy heat, we all know that is impossible, but to convert it to other forms of energy.

    Going back to your other statement, whether you are "in contact or in thermal contact" is the same thing in this case, there is nothing to be in contact with or in "thermal contact" with. Heat transfer(which is not an accurate term at all for this) is much more accurately said to be as I said it was: energy lost to black-body radiation, I have no idea what you are getting at. You are obviously going pointy-side-up, which frankly is a very bad sign for intelligent discourse, so I very much doubt any productive discourse is to be had between us.
    Last edited: Jul 3, 2012
  13. Jul 3, 2012 #12


    Staff: Mentor

    It violates the second law of thermo.

    A device which converts heat to other forms of energy is called a heat engine. They obey the second law of thermo and it's consequences. Including the equation that you cited in the OP.

    Last edited: Jul 3, 2012
  14. Jul 3, 2012 #13

    Wrong, a heat engine is, and I quote, "heat engine is a system that performs the conversion of heat or thermal energy to mechanical work"
  15. Jul 3, 2012 #14


    User Avatar
    Staff Emeritus
    Science Advisor

    Which converts the thermal energy to another form, such as kinetic energy for example.
  16. Jul 3, 2012 #15
    I'll give my thoughts on this, but it's up to the OP and other physicists to decide if I'm right (or wrong for that matter).

    I'm not a physicist, and I apologise for the potential blunders I might make in the following statements.

    A point I'd like to address is that the conversion of energy does indeed involve, or as I might say, "take" energy. Otherwise, there would be a spontaneous interconversion of energy. For example, I possess chemical energy (please pardon the poor wording;physically this might be incorrect, but you get the idea), and I have a system that readily converts some of this chemical energy to mechanical energy (a bicycle). If it didn't "take" energy for the interconversion, why isn't there a sporadic conversion of chemical energy to mechanical? Without me even attempting to pedal?

    Let's take your example: you have plenty of photons (light energy) and also a system capable of converting light energy to electrical energy (a solar cell). Then why doesn't the conversion take place instantaneously? It is imperative that the photons MUST be directed and made to strike on the solar cell, and I believe this involves energy of some sort.

    Just my thoughts on the problem.
  17. Jul 3, 2012 #16
    From the wiki entry on perpetual motion:

    "There is a scientific consensus that perpetual motion in an isolated system violates either the first law of thermodynamics, the second law of thermodynamics, or both. The first law of thermodynamics is essentially a statement of conservation of energy. The second law can be phrased in several different ways, the most intuitive of which is that heat flows spontaneously from hotter to colder places; the most well known statement is that entropy tends to increase (see entropy production), or at the least stay the same; another statement is that no heat engine (an engine which produces work while moving heat from a high temperature to a low temperature) can be more efficient than a Carnot heat engine.

    In other words:

    1. In any isolated system, one cannot create new energy (first law of thermodynamics)

    2. The output power of heat engines is always smaller than the input heating power. The rest of the energy is removed as heat at ambient temperature. The efficiency (this is the produced power divided by the input heating power) has a maximum, given by the Carnot efficiency. It is always lower than one

    3. The efficiency of real heat engines is even lower than the Carnot efficiency due to irreversible processes.

    The statements 2 and 3 only apply to heat engines. Other types of engines, which convert e.g. mechanical into electromagnetic energy, can, in principle, operate with 100% efficiency."
    Last edited: Jul 3, 2012
  18. Jul 3, 2012 #17


    Staff: Mentor

    I would just drop the pretense that you want to stick to the laws of physics as we currently know them. It is clearly not something that you really want to do.
  19. Jul 3, 2012 #18


    User Avatar
    Staff Emeritus
    Science Advisor

    Deeviant I have no idea what you are trying to get at. A heat engine converts thermal energy into work, as you said. However this also necessitates the conversion of that energy into other types of energy as well, otherwise no work could have been done.
  20. Jul 3, 2012 #19
    I repeat:

    In other words(2nd law of thermo says...):

    1. In any isolated system, one cannot create new energy (first law of thermodynamics)

    2. The output power of heat engines is always smaller than the input heating power. The rest of the energy is removed as heat at ambient temperature. The efficiency (this is the produced power divided by the input heating power) has a maximum, given by the Carnot efficiency. It is always lower than one

    3. The efficiency of real heat engines is even lower than the Carnot efficiency due to irreversible processes.

    The statements 2 and 3 only apply to heat engines. Other types of engines, which convert e.g. mechanical into electromagnetic energy, can, in principle, operate with 100% efficiency."

    In the ionic gas and nano-magnetic structure example, heat can simply be looked as mechanic energy(the ions are moving, that's all heat is, the more heat, the higher the average kinetic energy of the ions). The ions are moving, a moving charge creates a magnetic field(a moving magnetic field), somehow* the moving charges are channeled or in some way pass by the conductive surface of the chamber,the interior of the chamber could consist of some sort of porous material that exposes a huge amount of surface area to the gas, thus converting some of the energy into electric current reducing the kinetic energy of the ion, thus the average kinetic energy of the ion gas, thus the heat of the system.

    In this case, other that circuit/voltage/etc losses the conversion is 100% efficient, no energy is required to reduce the heat of the system as others have said in this thread before. No laws of thermo have been broken as still others have asserted.

    The question is this: the ion goes through the channel, loses some of it's energy which it is converted into electrical current and continues to bounce around until it is runs past another conductor, has some more of it's kinetic energy converted into electrical energy and so on, until it reaches some limit, of which I do not know what is the limiting factor, what part of such a mechanism is disallowed by physics?

    *one assumption made is that any engineering issues can be assumed to be solved, as long as they do not violate fundamental laws.
    Last edited: Jul 3, 2012
  21. Jul 3, 2012 #20


    User Avatar
    Staff Emeritus
    Science Advisor

    The general idea is feasible, however I don't see how this applies to your device in your original post. This requires charged particles, so unless you are going to place it next to a star you're going to have to use energy to ionize whatever you want to use for power.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook