Can heat be controlled as waves?


Georgia Tech assistant professor Martin Maldovan holds a tiny thermoelectric device that turns cold on one side when current is applied. Recent research has focused on the possibility of using interference effects in phonon waves to control heat transport in materials. CREDIT Couertesy of John Toon, Georgia Tech

A growing interest in thermoelectric materials — which convert waste heat to electricity — and pressure to improve heat transfer from increasingly powerful microelectronic devices have led to improved theoretical and experimental understanding of how heat is transported through nanometer-scale materials.

Recent research has focused on the possibility of using interference effects in phonon waves to control heat transport in materials. Wave interference is already used to control electronic, photonic and acoustic devices. If a similar approach can be used in thermal transport, that could facilitate development of more efficient thermoelectric and nanoelectronic devices, improved thermal barrier coatings, and new materials with ultralow thermal conductivity.

A progress article published June 23 in the journal Nature Materials describes recent developments and predicts future advances in phonon wave interference and thermal bandgap materials.

“If you can make heat behave as a wave and have interference while controlling how far it moves, you could basically control all the properties behind heat transport,” said Martin Maldovan, an assistant professor in the School of Chemical and Biomolecular Engineering and School of Physics at the Georgia Institute of Technology, and the paper’s author. “This would be a completely new way to understand and manipulate heat.”

In the classic definition, heat consists of vibrations in the atomic lattices of materials. The more vibrations in a material’s structure, the hotter the material. And in the same way that white light is actually composed of many different colors of light, these thermal phonons are made up of many different frequencies — each carrying varying amounts of heat.

Recent developments have shown that thermal phonons can interfere with their own reflections. The observation suggests that thermal phonons must exist as waves similar to electronic, photonic or acoustic waves. This interference could potentially be used to modify the velocity of phonons and the density of states, creating energy bandgaps that are forbidden for phonon waves. Utilization of similar bandgaps in optical and electronic materials has been key to developing a wide range of useful devices.

Until now, heat transport in nanostructured materials has largely been controlled by introduction of atomic-scale impurities, interfaces, surfaces and nanoparticles that reduce heat flow by scattering the phonons diffusely. Controlling wave effects could facilitate new approaches involving the specular reflection and transmission of thermal vibrations at interfaces.

“Considering the remarkable success achieved when using electronic, photonic and phononic wave interference to manipulate electrons, light and sound waves, it is certainly valuable to extend these theories to thermal vibrations, thereby creating a fundamentally new approach for manipulating heat flow,” Maldovan wrote in the paper.

Thermoelectric materials capture waste heat from sources such as automobile exhausts or industrial processes to produce electricity. Improving these materials will require further reducing thermal conductivity to improve their efficiency.

On the other hand, microelectronics designers want to increase thermal conductivity to transfer heat away from powerful and tiny devices. Developers of fuel cells and other conversion devices also need to improve the control of heat.

Maldovan wrote the article to clarify issues involved in thermal transport, and to interest others into the field. Ultimately, researchers will use this new information about heat transport to design better materials.

“These new wave phenomena can be used to create materials with low thermal conductivity,” said Maldovan. “We are trying to create a thermal bandgap, but that is not so easy to do.”

The search for thermal phononic wave materials will focus on semiconductors much like those used in microelectronics, Maldovan said. But while the silicon used in microelectronics had a natural bandgap, scientists had to create a band gap in photonics and acoustic materials, and the same will be true for thermal materials. Likely materials include silicon-germanium, gallium and aluminum arsenide and certain oxide superlattices.

Researchers have for many years focused on how far heat may be transported in materials. For the future, research will address the velocity of that transport, and how much heat is moved in the process, Maldovan predicted. He compares heat transport to a more familiar issue — human transportation.

“If you want to move a lot of people, you need a bus that will carry a lot of people,” he said. “You also want a vehicle that can move quickly because if you move faster, you can carry more people farther in less time.”

The next few years should bring about significant clarification concerning the role of interference and bandgaps in thermal materials, Maldovan predicted. That will allow continued progress in the materials needed for thermal control.

“It’s now a very cool thing to understand heat,” he said.


Citation: Martin Maldovan, “Phononic wave interference and thermal bandgap materials,” (Nature Materials, 2015).

59 thoughts on “Can heat be controlled as waves?

  1. Always good to see stuff like this happening. Sure they won’t be able to get all that waste heat, but this could lead to some very big energy savings in the future 🙂 .

    • The value of technology needs to be assessed throughout its life cycle from recovery to reclamation. The appearance of energy savings, ecological harmony, etc. may be deceptive. Still, development is an evolutionary process through trial, assessment, and modification.

  2. Light , of whatever colour, is directional. Photons and phonons are directed energy. Heat is random movement, not directional. Maybe this is badly explained but the two do not seem to be comparable.

      • Random, even chaotic, processes can be bounded by a sufficiently defined envelope. The unpredictable internal states of individual particles or energy may be less relevant than the cumulative state of the stream.

    • If I understand correctly, the random movement comes in part from vibrations bouncing off amorphous structures or atoms of varying weight. The regular structure of isotopically pure diamond can maintain wave propagation throughout its structure, so perhaps some of the envisioned work includes things like doping materials to change the equivalent of the index of refraction to make lenses.
      The mention of band gaps suggests they may be looking at something like a heat transistor. That would be a rather neat device with some really interesting applications.

      • Like you, I was thinking along the lines of possible thermal circuitry. Waste heat recovery just an obvious application. It will be interesting to hear the speculations. GK

  3. These people may be dead wrong, or they may be on to something wonderful. The thing is — they are doing real science where the proof will come when they make it work or they have to give up. God loves engineering; it is so honest.
    I wish them success and I note that anthropogenic CO2 has nothing to do with it. 🙂

    • Like the famous Richard Feynman quote:
      It doesn’t matter how beautiful your theory is, it doesn’t matter how smart you are. If it doesn’t agree with experiment, it’s wrong.

    • The article is so nebulous, it’s hard to tell is this is the Next Big Thing, or yet another hyped dud. Base on the history of ‘science’ press releases, I’d bet on the latter.

  4. yes Yet Another Peltier Element… but then different. I have to faint feeling this is also not going to pan out as expected efficiency cost/benefit wise.

  5. If we can control and perhaps harvest waste heat, then that would be good thing for all. This kind of research is really important and is really where these green Billions should be being spent – It really could be a game changer.

    • You wouldn’t be getting something for free, you’re essentially just be trying to increase the efficiency of the device, one would expect there to be a limit, beyond which entropy would happily still be king.
      We already do this on a grander scale, where we use secondary cycles to extract waste heat in power stations and where for instance some of the waste heat is sometimes used to heat homes etc. Energy is and will always still be lost, we’d just be able to grab a little more of it on it’s way out of the door….
      There would come a point where it’s so disordered that nothing would be re-claimable for useful work

      • I’d suggest looking up the Carnot Efficiency, the maximum work that can be done by a heat engine. It’s a basic property of entropy.

      • Hi Ben,
        If you were dealing with an engine that was already equivalent to an ideal carnot heat engine you might have a point, but you’re not
        If your statement was intrinsically correct then no heat engine could ever improve it’s efficiency – a kettle, would never have become a steam engine, a tweak to steam turbine blade design or to turbine expansion chambers would make no difference to the efficiency of steam turbine etc etc.
        You can make heat do more work for you, if get better at capturing it…. this would just be another method of capture, it would not capture everything, it would be lossy and it would be progressively more difficult to capture it and to do anything useful with it. It would not approach the efficiency of an ideal engine, but it would get a bit better… waste heat is only waste, once you’ve exceeded your abilities to do anything with it and then it is ultimately lost to us.
        You can reclaim it mechanically as per a turbine, you can reclaim it electronically (see anything on thermoelectric effects) or you can use it as a radiant heat source….
        There’s always a limit and entropy is always king…. In other words, you can approach an ideal, but you’ll never get there…..
        Don’t think it’s me that needs to read more, still reading and understanding were never actually the same thing….

  6. Wow! If I’m understanding this, it is saying they can make electricity directly from heat. That would be incredible. Scale it up. Industrial power production without the need for steam. Waste heat from industry turned into electricity. Where do we invest?

    • Darren
      This is already done but the summing of the many small soldered parts means it is expensive per Watt, so far. The efficiency is 1-2%. They Peltier devices melt easily because of the solder. They are therefore only run in safe ranges below their optimal efficiency which makes the cost problem worse.
      Whatever is learned about heat and sound can as well be applied to thermoacoustic devices which are 30-40 times more efficient and easier to make and don’t suffer from solder issues. Thermoacoustic devices used for refrigeration are above 40% efficient and have power above 100kW. A Large TEG, on the other hand, is 25 Watts and melts at 225 C.
      See Dutch work on travelling wave thermoacoustic generators (TACs) for state of the art hardware.

    • Darren Heat is electrons at work
      In 1975 the schottky diode and a rectenna allowed this to take place 1975 NASA JPL Goldstone Demo of Wireless Power Transmission
      “Piezoelectric materials exhibit both a direct and a reverse piezoelectric effect. The direct effect produces an electrical charge when a mechanical vibration or shock is applied to the material, while the reverse effect creates a mechanical vibration or shock when electricity is applied.”

  7. Could one use the wave properties of heat to remove energy more quickly from micro-electronic devices? My experience is that you can move a whole lot more heat by conduction than you can by radiation.
    Having said the above, I have seen huge energy densities at microwave frequencies and optical wavelengths. If you could find a way to convert random molecular vibrations to a coherent wave you should be able to move a lot of heat. It would be a real game changer for the electronics industry (where getting rid of heat is a ‘big deal’).

  8. But while the silicon used in microelectronics had a natural bandgap, scientists had to create a band gap in photonics and acoustic materials.

    All WUWT – Please.
    Did I miss something? A photonics band gap implies that a LASER can exist, and they do indeed. Is there such a thing as a bandgap in acoustics? Such a bandgap would imply the (at least) possibility of an acoustic equivalent of a laser or a maser.
    A bandgap for waste heat, or any generic heat, I call BS.
    I see now:
    phonon waves are just a mathematical construct. It is just energy propagating through mater. No bandgap, no stimulated emission effect, just thermal conductivity.

    • The thing is, though, that right or wrong – this is useful research. While we may not find a way to make heat waves as this research suggests, proving or disproving it will both likely lead to an increased understanding of thermal properties that can help us in other ways.
      All theories help, as long as they are pursued properly!

    • You can buy a heat powered ‘sonic laser’ kit from Prof S Garrett at Penn State U for about $20. I don’t think it works on a ‘band gap’. It is just a heat powered resonator that sends out a directional deafening sound.

    • In the 80’s, one of my ic design tool customers was the Naval Research Lab, one of the guys there showed me what they were working on, interdigitized conductors on a slab of quartz crystal, same as used in a clock, but he explained to me that depending on the spacing it worked just the same as an filter circuit, except instead of having to tune them by hand with lossy components, they could be mass produced, they were the very first surface acoustic wave filters, SAW devices.
      There was no technical reason they weren’t developed 30, 50 years sooner.
      There are lots of sources of waste heat maybe some clever stack of interleaved peltier coolers and maybe some heat pipes (another really clever device made for the military ).
      Good hunting!

  9. I echo some previous comments. Whatever they come up with will have to adhere to the second law of thermodynamics. Entropy, like death and taxes, cannot be avoided. This proposal sounds a bit like Maxwell’s Daemon. Good luck to them.

  10. “A phonon is a quantum mechanical description of an elementary vibrational motion in which a lattice of atoms or molecules uniformly oscillates at a single frequency.”
    Funny how my thermo and heat transfer professors never mentioned them…

  11. “Phonon” is a useful concept only in CRYSTALS. Metals are almost never a single crystal, but rather a collection of many small crystals randomly oriented with respect to each other, called “grains.” Somebody at Georgia Tech is misleading us.

      • Photons are the quantum mechanical description of light waves, in analogy phonons are the quanta of sound waves.The existence of sound waves has been demonstrated in crystals (metals and isolators) up to the cut off frequency. It has also been demonstrated that they exist in amorphous materials up to very high frequencies. Thermoelectric generators consisting of layered materials are also nothing new. The problems to be solved are efficiency, costs and wear.

    • Metals are not the only issue here. Semiconductors are also and they are almost always crystals perturbed by small amounts of impurities.
      Controlling the crystalline structure of metals is a big part of both historical and modern metallurgy. As with most things, we understand it better than we used to and that makes it more tunable to a specific application.
      The important part of the article is the discovery of what appears to be interference effects in heat transfer. If that is not an illusion, there will be interesting opportunities to take advantage of the wave nature of heat.

    • “Likely materials include silicon-germanium, gallium and aluminum arsenide and certain oxide superlattices.”
      These are all used in crystaline semiconductor manufacturering.

  12. Heat pipes can be constructed to mimic diodes, schmitt triggers, bandpass filters, etc. Could we see arrangements of heat pipes as oscillators, amplifiers, gates and so on? Heat-computers, even?

  13. This is interesting, but the idea depends on “specular reflection” and coherence. This might be the flaw in the argument. The more coherence and specular reflection they achieve the more reversible the process. So, a phonon of some frequency will travel unimpeded along a particular path, but also be able to travel equally well backward along the same path. Something like this has turned out to describe the flaw in Maxwell’s demon.

    • Yup. Even if it were possible to organize the thermal energy, and keep it coherent, you have that other annoying problem of keeping the unorganized thermal energy out from the outside.

  14. Cool.
    I’ve got my doubts about making something useful of phonon wave theory, but like fusion, thorium, zero point, and a host of other schemes, I’ll be happy to climb aboard when they get it retailed.

    • Thorium breeding/converting unlike fusion has already been demonstrated to provide net energy gain. It’s really just a question of economics and engineering. Fusion still requires invention and zero point requires theory and invention.

  15. I foresee a super-heat-conducting cylinder disguised as a rectal thermometer turning Al Gore into “Blue Boy”. It’s farfetched, I know, but a man’s gotta’ dream!

  16. The key sentence is “We are trying to create a thermal bandgap, but that is not so easy to do.”
    I think it’s great they’re trying, and they honestly admit their doubts.
    As somebody said earlier, they are doing real, not “climate” science.
    Our language always changes. I wonder if some future vernacular will include phrases like “stop climating”, “don’t tell climates”, or “you’re nothing but a dirty climater”.

  17. Diamond is an excellent thermal conductor, and I wonder if you can change its index of refraction as you mentioned Rick.
    I’ve done some analysis of microwave strip line circuits and how pin diodes just by changing the impedance makes for all sorts of interesting functionality.

Comments are closed.