New Thin Film Thermoelectric Can Harvest More Waste Heat

Lawrence Livermore National Laboratory (LLNL) material scientists have created thermoelectric generators that can harvest waste heat from previously inaccessible sources, such as pipes with complex geometries.

Thirteen quadrillion (3600 terawatt-hours) BTUs of energy is lost annually through waste heat by U.S. industry.

The team cold-sprayed a bismuth-telluride powder on substrates ranging from stainless steel to aluminum silicate and quartz. The sprayed material had a randomly oriented microstructure largely free from pores and the cold-spray deposition was achieved without substantial compositional changes.

A ZT of 0.3 can probably only recover 2-4% of the heat as electricity, but it will be from places where we could recover nothing before

Advanced Processing and Additive Manufacturing of Functional Magnetic Materials – Cold Spray Deposition of Thermoelectric Materials

A new application of cold-spray deposition has been demonstrated that can fabricate bulk pieces of thermoelectric Bi2Te3 on a wide variety of substrates, without loss of structural integrity. By entraining particles in a supersonic gas flow, millimeter-thick blocks can be built up on flat or curved surfaces in a matter of seconds, providing a pathway to waste heat recovery from narrow pipes or other, more complicated, shapes. The sprayed material is composed of randomly oriented crystallites that match the chemical composition of the precursor billet and is > 99.5% dense. The Seebeck voltage and thermal conductivity are comparable to or better than bulk Bi2Te3, while the as-deposited resistivity is an order of magnitude higher. A simple annealing treatment at 400°C removes defects and decreases resistivity, increasing ZT to ~ 0.3 at operating temperatures of 80°C. Integrated TEG measurements of Bi2Te3 sprayed on copper flats or pipes suggest performance comparable to simple devices fabricated from blocks of the bulk material, demonstrating that cold spray is a viable alternative to traditional manufacturing approaches for thermoelectric materials.

SOURCE – LLNL, Advanced Processing and Additive Manufacturing of Functional Magnetic Materials
Written By Brian Wang,

38 thoughts on “New Thin Film Thermoelectric Can Harvest More Waste Heat”

  1. I don't see it mentioned but one of the biggest applications is cooling electronics, especially CPUs. Instead of costing more power it will generate a bit, win-win.

  2. You are quite right. This is why education is important; it let's you search more efficiently and understand the results faster/better. Those that claim that you do not need knowledge – everthing is searcheable – don't grasp this point.

  3. I wonder if it's a setup and maintenance issue? Cool steam down and it's nice, safe, low pressure, water.
    Cool supercritical CO2 down and it's still high pressure, asphyxiating, CO2.

  4. Way more than 15 sec in my case, as I wouldn't be satisfied with just a quick yes or no. And that's assuming I entered the correct search phrase on my first try.

  5. It would be the second part of the combined cycle. Current combined cycle means you have a turbine that burns the fuel, getting power like a jet engine, then a second system takes that heat created and boils some water generating steam which is then fed through another turbine. That hot air, I would think, could just as easily heat Supercritical CO2 instead of water and be fed through a turbine designed for that.
    I am not sure why this has not been done.

  6. Everything that extracts energy from a temperature differential is subject to Carnot. That's why direct ion beam to electricity of some nuclear fusion approaches are attractive. (not to downplay other issues with fusion)

  7. S-CO2 is great, hope it shows up some day. Good to remember that a combined cycle S-CO2 plant that pulls out 70% of electricity won't have much waste heat for the poor thermoelectrics. Then again its waste heat so an extra 1-2% for a passive, no maintenance, attachment is a win.

  8. Focus Fusion also does alpha ion beam to direct electricity.

    Since i'm a Focus Fusion investor here's what I remember from the August update for the reading pleasure of y'all:

    Upgrading their switches so that they will fire better with more current, and cut down on resonance. Resonance carries in to the plasma, makes plasmoid formation erratic.
    Having a new Beryllium anode manufactured.
    Impurities in test firings are low, this is good.
    No Beryllium ash present in the DPF even with test shots (a good sign, Beryllium is toxic so ash is a nuisance to work around in a test device).

  9. Supercritical CO2 has been going nowhere, slowly, forever. The promise is there, the potential is there, the product is nowhere. Very confusing why this hasn't been fast tracked. I assume because it can't improve on combined cycle.

  10. No, don't put it right after combustion. Thermoelectrics aren't as efficient as anything else which means that everything else should have prima nocta with the virgin heat.

    Sorry to put it that way, need more coffee.

  11. Coat the inner wall of an AP1000 containment unit. Lots of heat, nice heat sink (air). Temperature differential is large enough and the area being coated is uniform.

  12. I think Carnot's insight applies to all processes. Steam systems have used thermoelectric generation for years to take heat from 250 to 450 degree F steam pipes and make sufficient power for a phone to report conditions in man holes, and the electric yield is higher from the hotter pipes.

  13. They’re in the $20 per kg range only due to lack of demand. Spray this stuff on everything and prices would skyrocket. So not really abundant and situationally affordable. Any solution that requires millimeter thick bismuth telluride can’t scale well enough. Get it micron thin like thin film solar and that’s another story.

  14. The physics are never going to be all that good for thermoelectrics. Carnot efficiency on a solar cell is limited by the temperature of the light (like 5000 Kelvin) which is massively higher than 373 Kelvin of boiling water. And you can layer solar cells with sensitivity in difference light ranges.
    Not a candidate for the kind of research money solar should get.

  15. Frankly never cheap enough at 2-4%. You’ll never see this tech on a steam stage*. Only right after a combustion stage where the temperatures and thus carnot efficiency are high enough. Think catalytic converters on cars to replace an alternator and natural gas turbines where it replaces the steam part in combined cycle plants.

    *Maybe nuclear but then expense would be justified as a safety device

  16. A lot of those peaker plants are run only like 400 hours per year. Paying extra to make them combined cycle wouldn’t make any financial sense.

  17. Get it inexpensive enough and you can replace the diesel backup generators with thermoelectrics and turn a Gen II plant into something closer to “walk away safe”.

  18. Waste heat is a big market indeed and not very exploited.
    For anyone interested in state-of-the-art in waste heat recovery, take a look at Climeon.
    They are one of (if not THE) market leaders in commercial waste heat recovery systems. Their product is based on ORC and they manage to recover 10 percent of 70 – 130C waste heat. I think they started by going after passenger ships and they have several successful implementations.

  19. Energy properties include quality and quantity, so there are theoretical limits to the % of energy quantity that could be converted into electricity. The limit id the difference in degrees kelvin between the process inlet temperature and the heat sink temperature, divided by the degrees kelvin of the inlet temperature per Sadi Carnot. The average electric-only generating efficiency of 34% is not a flaw of steam turbines, but a failure to capture the thermal energy.
    Thermoelectric devices start with relatively low inlet temperatures, hence low efficiency. But the new process appears to pull electricity out of thermal temperatures to low for most thermal processes, making fuel-free electricity from any hot surface. Neat advance.
    Tom Casten

  20. Yep. Pipes of hot liquid are usually wrapped in an insulator. Also given the notably lower efficiency you would want to install the thermoelectrics downstream of the traditional turbine.

    I keep thinking of big nuclear plants where the waste heat goes to the atmosphere. the part of the heat exchanger that carries hot water and is sprayed with cold water. Put the thermoelectrics there.

  21. I'm hoping Tri-alpha energy and others like it end up becoming successful and proving direct conversion of electricity. It could end up improving nuclear in a similar way to how solar has improved the last couple decades.

  22. The reality is that thermoelectric stuff is very inefficient. You are not going to recover much of that wasted heat. However, every bit helps.

    It is however one of those technologies that we should spend a lot of money to try to reach higher efficiencies with. The payoff would be great if we could get to something like 10% efficiency. 20% or 30% would be World changing.

    The first commercial solar cells were only 2% efficient, but with a lot of research we are at 22.6%. I am hopping we can do the same with thermoelectric…but it well take a lot of research, if it is possible at all. AI, and supercomputer stuff may accelerate development, but there is no shortcut around figuring out how it all works at the nano level.

  23. We use combined-cycle too for natural gas, at least most of the new power plants. They can be upwards of 60% efficient. Utilities should, in fact, not be permitted to build simple-cycle natural gas plants, combined-cycle is just so much better. And the old simple-cycle natural gas plants should be replaced. My understanding is that most of these are converted oil burners, but some of these things are still being made just because they are cheaper. But they are a gamble even for the utilities because if natural gas prices go up, they are going to feel the burn. Also, long term they will make less money almost guaranteed. They don't save that much building them.
    They might be able to reach 70% with supercritical carbon dioxide, and perhaps a bit more with the thermoelectric stuff on top of that.

  24. Seems like a good thing to spray on the material under solar panels. They must get pretty toasty sitting under a black cell.

  25. Yep, "waste heat" is unavoidable (in terms of thermodynamics) as the physical infrastructure of the generating mechanism will absorb heat during the transfer between the boiler (or other heat source) and the turbine, as well as the exhaust gases. The question is, can this coating be applied economically, both in terms of initial application and in how long the coating will last in a harsh environment?

  26. There are limited places where this sort of thing makes sense, such as exhaust pipes on cars. Most of the time, if you have a hot pipe, it's because you're transporting something hot from one place to another, and want it to still be hot when it reaches its destination.

  27. The good news is that Bismuth and Tellurium are abundant and cheap. Shouldn't be too hard to tie into the grid next to boilers and industrial sites.

  28. If collectively, humanity is truly losing 3600 terawatt-hours of generation, perhaps it's time to give a second look to our boil water, create steam, rotate turbine schemes. That absurdly wasteful.

  29. OK, but what would the cost be per W of installed power on a "waste" heat source? In what temperature range is most of the 3600 TWh lost..? If Brian would add this information the article would an order of magnitude more informative…

Comments are closed.