Forgot your password?
typodupeerror
Power Transportation Science

'Thermoelectrics' Could One Day Power Cars 174

Posted by Soulskill
from the hot-wheels dept.
sciencehabit writes: "Fossil fuels power modern society by generating heat, but much of that heat is wasted. Researchers have tried to reclaim some of it with semiconductor devices called thermoelectrics, which convert the heat into power. But they remain too inefficient and expensive to be useful beyond a handful of niche applications. Now, scientists in Illinois report that they have used a cheap, well-known material to create the most heat-hungry thermoelectric so far (abstract). In the process, the researchers say, they learned valuable lessons that could push the materials to the efficiencies needed for widespread applications. If that happens, thermoelectrics could one day power cars and scavenge energy from myriad engines, boilers, and electrical plants."
This discussion has been archived. No new comments can be posted.

'Thermoelectrics' Could One Day Power Cars

Comments Filter:
  • by noh8rz10 (2716597) on Wednesday April 16, 2014 @04:14PM (#46772833)

    technically, you would still need an energy source (gasoline, natural gas, batteries) to power the cars. thermo electrics could make it more efficient by recycling waste heat. but the thermoelectrics themselves would not power the cars.

    • Fuel provides energy. Power comes from the conversion rate of the engine. You "power" things with an engine, but the net energy comes from fuel. Some engines can handle multiple fuel types, e.g.

    • thermo electrics could make it more efficient by recycling waste heat. but the thermoelectrics themselves would not power the cars.

      If they are sufficiently efficient, they could power a car directly. An internal combustion engine is typically only about 15-20% efficient, so the bar is not too high. Using thermoelectrics directly could have several advantages: being solid state, they would be reliable and require little or no maintenance; and since the fuel is just used to create heat, it could use cheaper grades of fuel.

      • by noh8rz10 (2716597)

        ok but then the car is still powered by fuel. in your scenario you would still need an external combustion engine to produce heat.

        • Or a RTG [wikipedia.org]
          • by Guspaz (556486)

            It's not as silly as you might think. I believe you get roughly 500W of heat per kilo of plutonium-238. A Tesla Model S driving at normal speeds consumes something like 15KW. If you could get 50% efficiency for your thermoelectrics, you could build an RTG-powered model S with 60 kilos of plutonium. You'd need capacitors for surge demand, obviously.

            Of course, this would be completely insane, but I don't see why it's not theoretically possible, since the battery pack on the car that you'd be replacing already

            • There are several problems with your scenario:
              1. Current TEs are no where close to 50% efficient. More like about 5%.
              2. Pu238 is available in very limited quantities from reactor fuel reprocessing
              3. You can't "turn-off" an RTG. They have to run continuously.

              • by Guspaz (556486) on Wednesday April 16, 2014 @08:01PM (#46775235)

                1. I realize that they're currently at 5%, the whole point of my scenario was examining what sort of changes a large increase in efficiency would produce... that's the whole point of the article, after all. Efficiency would need to be somewhere around 50% to justify replacing ICEs with thermoelectric engines. Is that possible? I've got no idea, TFA gives zero layman-friendly information about what sort of efficiency improvements are foreseen.

                2. Supply isn't as big a problem as the incredible safety issues. I acknowledge in my post that the idea is totally insane, which is why I doubt that, even with a big improvement in efficiency, you'd probably never see RTGs used outside of military applications.

                3. That's not necessarily a problem. They conveniently provide power that can be used for active cooling. Cooling them in a vacuum is an issue (hence the giant heat dissipation fins), cooling them in an atmosphere isn't as much of an issue.

                I suspect that sufficiently efficient thermoelectrics might find their way into military UAVs, which could remain airborn for extended periods of time, for example. Or as an alternative to shipping diesel to remote outposts (although they're currently looking into robotic trucks to solve that problem).

                • by tragedy (27079)

                  2. Supply isn't as big a problem as the incredible safety issues. I acknowledge in my post that the idea is totally insane, which is why I doubt that, even with a big improvement in efficiency, you'd probably never see RTGs used outside of military applications.

                  The safety issues aren't really that bad. You could put 60 kilos inside a casing that would easily block the radiation down to negligible levels and would be effectively indestructable in the worst conceivable accident. Worries about "dirty bombs" are ridiculous considering the large array of easily available substances that would be much more dangerous (not very) in such a bomb. As for a nuclear weapon, I'm not exactly sure how fissionable it is, but I do know that you would need a massively powerful nucle

              • by tragedy (27079)

                1. Current TEs are no where close to 50% efficient. More like about 5%.

                The article was about new, higher-efficiency materials. Still not high enough, and not near 50% efficiency, but certainly getting up there. Good enough so that, if you could get hold of the material, you could at least use it to charge your electric cars batteries currently, even if you couldn't power the car directly. Of course, at present, you'd be better off with a Stirling engine.

                3. You can't "turn-off" an RTG. They have to run continuously.

                Presumably, you could plug them into the power grid in most places you park them

            • In your 60600 analysis you're forgetting the shielding and cooling requirements.
            • you might see the military using RTGs.

              Yeah, I can't wait to have explosive ordinance being flung at vehicles powered by a giant box of ionizing radiation. Great idea.

              • by Guspaz (556486)

                This is different to throwing explosive ordinance at current vehicles powered by nuclear reactors how?

        • by gman003 (1693318)

          It is completely valid to say "a car is powered by an engine". The engine is powered by fuel, but the car's power comes from the engine. Replacing the reciprocating engine with a thermoelectric engine allows for the headline to, in fact, be an accurate statement.

          • by noh8rz10 (2716597)

            if you have a "thermoelectric engine" then what is producing the heat to power the thermoelectrics?

            • if you have a "thermoelectric engine" then what is producing the heat to power the thermoelectrics?

              A flame.

              • by noh8rz10 (2716597)

                what are you burning to produce a flame?

            • Re: (Score:3, Interesting)

              by Em Adespoton (792954)

              The power plant -- just like in Diesel Electric trains; you have the electric engines that power the train and the power plant that powers the engines. Diesel fuel powers the power plant, and it in turn was powered by solar energy. The sun is powered by hydrogen fusion reactions; the hydrogen fuel was provided by gravitational attraction, which was powered by time and space.

              I'll leave it up to the reader to determine who/what powered time and space.

            • When you plug in a microwave oven, do you say that it's powered by the wall socket, or that it's powered by coal / methane / uranium / solar / hydroelectric / wind / geothermal?

              Because if you say the latter, you sound like an idiot. And, by the way, that's the exact semantical argument you're making here.

        • by Newander (255463)

          If the car is electric it could be powered by waste heat from industrial processes and primary power generation.

        • by sgt scrub (869860)

          Is an exothermic reactor considered internal combustion?

          • Probably not. Internal combustion only means that the combustion occurs inside the engine. Contrast this to a classic example of external combustion: a steam engine, where the fuel is burnt well away from where the work is produced. Jet engines are internal combustion; most stirlings are external.

            One might have a good case for arguing that thermoelectric engines are neither IC nor EC, since it needn't be combustion that provides the heat.

      • So say we can double that. That makes the fuel 40% efficient as we use some of the heat towards efficiency. That will double the gas mileage. However if you need a smaller engine, then it will be producing less heat. That is good if it is 1 for 1. However if their needs to be a particular heat starting limit then it may cause an issue. Unless you go with a bigger car.

        The idea as the engine gets more efficient people buy bigger cars, is economically sound and proven. A large truck today can do about the

    • by fuzzyfuzzyfungus (1223518) on Wednesday April 16, 2014 @04:36PM (#46773179) Journal
      My (admittedly pretty hazy at this point) memory of heat engines is that their theoretical peak efficiency depends on the thermal delta they manage to achieve. Exactly the same resource that thermoelectric materials scavenge (albeit at miserable efficiency) into electricity.

      Anybody who actually has some grasp of the matter want to chime in on where and why you would use thermoelectrics (and how efficient they would have to be) rather than simple insulation or one of the various waste-heat-recovery systems that transfer some amount of the heat remaing in outgoing exhaust gases into incoming working fluids?

      Is the thermoelectric advantage purely that, assuming material reliability is OK, they are a 100% solid state, trivial to scale from 'handle with tweezers and magnification' to 'pretty large', and their output is easy to transfer and useful for all kinds of things after just a little DC-DC cleanup, or are there actually situations where they might be absolutely more efficient than insulation and heat recovery, rather than just easier to tack in almost anywhere in a design that you have a few extra cubic centimeters and expect a temperature difference?
      • by Dare nMc (468959)

        > insulation or one of the various waste-heat-recovery systems that transfer some amount of the heat remaing in outgoing exhaust gases into incoming working fluids?

        Several reasons this isn't done in the engine intake. The main power conversion in a ICE is through the thermal expansion of the gasses trapped in the cylinder, so heating it before the intake valve is closed only reduces the density of the air taken into the cylinder (PV=nRT so at the same Pressure and volume, the higher the temperature, the

      • Anybody who actually has some grasp of the matter want to chime in on where and why you would use thermoelectrics (and how efficient they would have to be) rather than simple insulation or one of the various waste-heat-recovery systems that transfer some amount of the heat remaing in outgoing exhaust gases into incoming working fluids?

        In IC engines you want the intake air to be cold so the density is higher (more oxygen); you may have noticed that cars with turbo charges often have something called an intercooler for this reason (compressing the intake air heats it). If you're going to scavenge anything from the exhaust gases it will be low-grade heat (delta-T is smaller, moreso with an efficient engine). Using heat from exhaust to pre-heat the intake is more commonly done with gas or steam turbines, where you want to reduce the amount

    • Maybe Solar Heat?
    • If the thermoelectrics are significantly more efficient that than internal combustion engine, removing it completely would save a lot of weight and may result in a more efficient system.

  • by neonv (803374) on Wednesday April 16, 2014 @04:15PM (#46772837)

    We better speed up this global warming thing so we can power our thermo cars!

    • Re:Hotter Earth (Score:5, Informative)

      by ShanghaiBill (739463) on Wednesday April 16, 2014 @04:19PM (#46772923)

      We better speed up this global warming thing so we can power our thermo cars!

      That doesn't work. TEs aren't powered by heat, but by heat gradients. So if everything is uniformly heated by the same amount, there is no benefit.

      • So... sink a steel pipe half a mile into the ground, it isn't that hard to create a heat gradient. That's of course, if it's possible to hit anywhere near decent efficiencies with standard materials, which is something I'll have to see in production before I fully believe.

        • So... sink a steel pipe half a mile into the ground, it isn't that hard to create a heat gradient.

          That would give you enough of a gradient to generate a micro-watt from a ton of TEs. In a perfect ideal TE, the efficiency is (1 - Th/Tc) where Th= Hot side in Kelvins, Tc = Cold side in Kelvins. Existing TEs are no where close to ideal, and the earth's heat gradient is about 0.025K/meter. A negligible amount of heat would flow through the TE, and far less than 1% of that would be converted to electricity.

        • by ceoyoyo (59147)

          Why yes, I have been to Iceland.

        • The problem with that is that it will cool down the inside of the Earth.
          You can have a look at the movie "The Core" to see what will happen then...

      • Well, you could build a thermoelectric shell around the entire earth... Of course, if you can do that, 'stop global warming' is probably something you did in high school as a busy-work lab project at your school for particularly gifted hypersentient disembodied intelligences.
        • you guys just really don't get it, do you?

          Global warming is a non-issue. Elon Musk is going to put up a orbital sunshade and hold the world ransom to turn the lights back on.
          And there's NOTHING you can do to stop him.
      • It's not that there's no benefit : it's a net loss. Any heat engine works better between 300K and 400K than between 400K and 500K, even though the temperature difference between two states is the same.
        Pedantic remarks : There's no such thing as "heat gradient". You probably meant "temperature gradient". And thermoelectrics generators really are powered by heat.

    • by Immerman (2627577)

      Good luck with that. Human energy production is directly a miniscule factor in global warming - it's the CO2 byproducts that's the problem. In the 90+ years it takes for a unit of CO2 to be removed from the atmosphere it will capture roughly 1,000,000x as much solar thermal energy as was produced by the burning of fuel that created it. If we generated 1000x as much heat directly, but without producing the CO2, then global warming would be a non-issue.

  • by Anonymous Coward on Wednesday April 16, 2014 @04:17PM (#46772901)

    Several vehicle manufacturers have been experimenting with supplemental power generation systems [wikipedia.org] in their cars. BMW for instance has a steam turbine. [wikipedia.org] Honda's doing thermal recovery more efficient than regenerative braking.

  • Ah, the clickbait (Score:2, Informative)

    by Anonymous Coward

    Now, scientists in Illinois report that they have used a cheap, well-known material to create the most heat-hungry thermoelectric so far

    Because it's soooooo hard to actually state what the well-known material is

    Ultralow thermal conductivity and high thermoelectric figure of merit in SnSe crystals

    Oh, I guess it's not hard at all. A salt made of Selenium and Tin.

  • ...from my cold, dead hands...
  • You expect slashdotters to generate heat... on a mattress?
    • by PIBM (588930)

      I guess they could use your BBQ, in some ways ?

  • by bobbied (2522392) on Wednesday April 16, 2014 @04:44PM (#46773295)

    I debunked this LAST time it was posted..

    Look, these things are NOT going to get you thermodynamic efficiency gains on anything of value. Any system which is designed to be efficient now, will not benefit from this kind of heat to electricity device. Thermodynamic rules demand a maximum efficiency that is as good as you can do. Most industrial scale energy production is pretty darned good compared to the maximum possible. So you are NOT going to be able to just hook up these things and get electrical energy for *free* (even without the device costs). Any energy you manage to get, will be lost someplace else because you put these devices in the heat flow. Don't even bother trying this, it simply won't work. Don't let them fool you with all this "waste heat" garbage, at least until you understand the Thermodynamic laws that govern all this and can explain what a heat engine is.

    As I concluded before, in situations where you have less than ideal conditions, like in cars with internal combustion engines, you MIGHT get a little bit of energy, but I ask you is it going to be worth it? Are you sure you are going get enough gain to make it worth the weight, cost and complexity? Where I'm not so sure that answer is a good one, I'm willing to entertain that it *might* be possible for internal combustion engines. Go ahead and work on that idea, but I'm fairly sure it's not going to work very well.

    I'd also suggest that there are more efficient heat engines you might consider. These heat flow direct to electricity devices are horribly inefficient compared to the ideal.

    • by sconeu (64226)

      Lisa, in this house, we obey the LAWS OF THERMODYNAMICS!!!!

    • by rasmusbr (2186518)

      The one application that I've heard about that sounds semi-plausible is sandwiching something like this between a solar cell and a liquid cooler. The difference in temperature between the PV cell and the cooler might be enough to yield meaningful amounts power and the waste heat that the cooling system captures could be used for heating.

      • by bobbied (2522392)

        I'm not sure how the cooling works for PV cells. But, if they are actively cooling the liquid there is little to be gained from this arrangement and it is likely going to lead to hotter cell temperatures and cooler liquid temperatures.

        So it *might* work, but only if the temperature differential they can stand is high enough and they are not expending energy to cool the liquid through some heat engine.... (I.e. if they use something like a swamp cooler or something.)

    • Any energy you manage to get, will be lost someplace else because you put these devices in the heat flow.

      You sir, are ignorant as fuck. It's a sad comment on the state of affairs that a clueless bullshit comment like your could be moderated informative.

      We've been extracting energy from waste heat, without incurring extra losses, for over a century now - it's been a standard practice in steam [wikipedia.org] engineering [wikipedia.org] since the 1800's. In the same way, if you put these devices in an IC engine's exhaust you can recove

      • by bobbied (2522392)

        Any energy you manage to get, will be lost someplace else because you put these devices in the heat flow.

        You sir, are ignorant . It's a sad comment on the state of affairs that a clueless bullshit comment like your could be moderated informative.

        We've been extracting energy from waste heat, without incurring extra losses, for over a century now - i

        Calm down and think about what I said. Your average power plant is pretty darn good efficiency wise (which is what I said if you don't mind reading), which is exactly what you are saying too. Yea, we've come a long way from just dumping waste steam, we have optimized things very well actually. I'm saying that there is very little room for improvement left at this point and there is NO FREE LUNCH here. These devices that convert heat flow directly to electric power are NOT going to increase the efficienc

        • We're talking about cars here, right? At least, TFA is.

          Currently, cars dump hot exhaust gases out the back end without doing anything with the heat, because it's a byproduct of the mechanical force created through ignition of fuel. IC engines don't use the heat very much - they use the expansion of gases to create mechanical work, and the "waste heat" blows out the tailpipe as, you know, waste.

          If you could harness that heat in some way to create electrical or mechanical power, you're more efficient than t

          • by bobbied (2522392)

            The article mentions power plants near the end.

            Where I'm dubious about the usefulness of this technology in cars, I would agree that there is at least a possibility it could help. Thermodynamics would allow it. However, in a power plant, Thermodynamics tell me they are not useful.

            I also wonder if the original author and the researchers involved actually understand the issues, given they try and present this as a way to improve a power plant. On that point they are totally and obviously mistaken. If th

    • by cusco (717999)

      There's a lot of energy available from an IC engine. If you doubt me let your car run for anything more than two minutes and then touch the exhaust manifold. Bumped one with my arm when I was in high school and it took twelve years for the scar to fade.

      • by bobbied (2522392)

        There's a lot of energy available from an IC engine. If you doubt me let your car run for anything more than two minutes and then touch the exhaust manifold. Bumped one with my arm when I was in high school and it took twelve years for the scar to fade.

        Which is why I stipulate that these heat to electricity devices MIGHT be of value for internal combustion engines. The heat dumped by them is significant and the temperature differential quite high. There is at least opportunity to get something that would normally just get dumped. I just openly wonder if for a car or truck it will be worth the cost, weight and complexity it will add. I strongly suspect that it's not worth it, but I'm not totally sure.

        • by cusco (717999)

          Then I misunderstood your previous post, which I took to mean that there wasn't enough to bother with. It would be interesting to me to see an estimate of how much it would cost to make an exhaust manifold of thermocouple material, and what the estimated output would be. With hybrid vehicles like the Prius, which just uses the IC engine to charge the batteries that actually propel it, it might well be worth it.

          • Back a few years there was a quest for an all ceramic car engine. Fuel could burn a lot hotter so your could get a lot more power per litre. It seemed obvious. A few partial successes happened and then Mercedes actually built one. On a test bed the performance was wonderful. In a car the extra weight required to keep it cool made the performance of a 1936 Chevy look better. In the end some ceramic parts are in use, mostly in trucks, but making it all ceramic to get those higher temperatures is impract
    • by Immerman (2627577)

      The maximum limit on an an arbitrary heat-conversion system is that doesn't break accepted theory is the Carnot-cycle heat engine, where eff 1 - T_cold / T_hot (as measured from absolute 0). But it's a rare real-world engine that gets anywhere near the Carnot efficiency limit - a car engine might run at 1100K for an ideal efficiency of around 73%, but the reality in most cars is closer to 25%. Being solid-state a thermoelectric device could potentially operate at very near the ideal (no mechanical losses

      • by bobbied (2522392)

        The maximum limit on an an arbitrary heat-conversion system is that doesn't break accepted theory is the Carnot-cycle heat engine, where eff 1 - T_cold / T_hot (as measured from absolute 0). But it's a rare real-world engine that gets anywhere near the Carnot efficiency limit - a car engine might run at 1100K for an ideal efficiency of around 73%, but the reality in most cars is closer to 25%. Being solid-state a thermoelectric device could potentially operate at very near the ideal (no mechanical losses), roughly tripling the efficiency. Assuming 90% efficient electric wheel motors the total system efficiency could be nearly as high.

        Don't be fooled that "hey they are solid state and convert directly to electricity". Deep down, it's still a physical process that produces electricity, even if the moving parts are not something you can see. In actual practice, what happens with these things produces horrible efficiency.

        These electronic devices are semiconductor junctions that you get heat to flow through in hopes the electrons will bounce their way across the junction into the cooler side and get stuck... They are not efficient from a

    • by Vellmont (569020)

      Umm.. so the article was focused on the abstract idea of increasing efficiency of thermoelectric generators. The practical idea (and even the article title) was about how it might be able to power a car more efficiently. But yet you focus right in on how it's never going to work. (Why yes, I DO understand the carnot limit of heat engines).

      The article never talked about massive gains in heat efficiency for power plants, just scavenging waste heat. Right now we have massive cooling towers at power plants

      • by bobbied (2522392)

        The article never talked about massive gains in heat efficiency for power plants, just scavenging waste heat.

        You are falling for the waste heat argument. Power is generated by the transfer of heat from a high temperature source to a low temperature sink. In a power plant, we've managed to engineer them pretty well and we get pretty close to the ideal. Any new device will have to exceed the existing efficiency or it simply will not help, but hurt. Remember it is the TRANSFER of heat that is used to produce power, not the heat energy itself. So the "dumping of waste heat" is not a power loss in the system where

  • the first application of such devices would be more like a solar cell, power plant, backup generator etc. Putting a new device in car can take decades, but putting in these can be done much more quickly as the number of approvals needed is far few. Whenever, someone uses "car" where it is not justified, I know the innovation is most likely worthless showoff or it is decades away from practical use. Yes, one day all cars will run on fusion power. Thanks.

  • E = (T2-T1) / T1 (Score:4, Informative)

    by Animats (122034) on Wednesday April 16, 2014 @05:33PM (#46773857) Homepage

    E = (T2-T1) / T1

    Everyone with an engineering degree knows this. Trying to extract much energy from low-grade heat at the output end of an engine is inefficient. This was figured out a long time ago. Here it is in The Manual of the Steam Engine [google.com]. It's possible to increase steam engine efficiency by compounding, where the exhaust from each cylinder feeds a larger, lower pressure cylinder. This is cost-effective up to about 3 cylinders ("triple expansion"). Engines up to quintuple-expansion have been built, but the additional power from the last two cylinders in the chain isn't worth the trouble.

  • Replace current refrigeration/cooling technology (essentially heat pumping or heat moving) with this - I get paid to provide cooling or refrigeration, not in cash but in Watts. Put in calories (which I wanted to get rid of anyhow), pull out watts.

    And - no, I don't want to hear about perpetual motion. The only perpetual motion machine in the Universe is the Universe, and the jury's still out on that one.

  • they have used a cheap, well-known material to create the most heat-hungry thermoelectric so far

    Did they do it with one weird trick discovered by a mom?

  • Waste heat is often a lot and usually needs to be moved away fast.
    If you let the heatflow of a car pass through these tiles you's need a lot more surface area to get sufficient heat out of the engine.
    If you try to cool a CPU through these things the CPU will overheat quite soon.
    These things have "Ultralow thermal conductivity" according to the article. That means they act as thermal insulators. Not what you want when using waste heat.
    This is useful for other places.
    1. Places where you can replace insulation

The opposite of a correct statement is a false statement. But the opposite of a profound truth may well be another profound truth. -- Niels Bohr

Working...