> The polycrystalline tin selenide the team makes is spiked with sodium atoms, creating what is known as a “p-type” material that conducts positive charges. To make working devices, researchers also need an “n-type” version to conduct negative charges.
This is literally inventing half a solution. Shockley didn't go to the press when he'd built half a diode.
(OK, this is snarky, but I think this is another case where people need to understand the difference between "a small but valid advancement in the art" and "actual products". As a materials science paper this is valid.)
> Zhao’s team recently reported making an n-type single-crystal tin selenide by spiking it with bromine atoms. And Kanatzidis says his team is now working on making an n-type polycrystalline version. Once n-type and p-type tin selenide devices are paired, researchers should have a clear path to making a new generation of ultra-efficient thermoelectric generators.
You don't understand the difference btw a research paper and a product launch. A paper is to provide inspiration. In this case, a high ZT p-type crystal may indicate a new mechanism that can be applied to searching / designing new thermoelectric materials. It's not for selling you a working device that can be put in your shopping cart.
(I did research in nano thermoelectrics and published a few papers in this field)
In defense of parent though it does not say “can” or “will” convert. News and views should have reflected that so it’s really not the fault of the original paper authors.
I get the snark but this is also Science and their easy to digest news release. It is literally a science journal, not a product release or even engineering journal. I wouldn't say it really fits into the "advancement in the art," or "actual products." It is scientific progress in the material science world (which I know is your last parenthetical aspect of your comment).
Kinda answered my own question with some searching - looks like p type has lots of electrons, n type has lots of holes, and graphene oxide (GO) makes an excellent and simple to use n type material.
Graphene can be modified by changing its topology to be p or n, but GO is n by default. Do electron and hole features have to match in these constructions?
> Those could be installed everywhere from .. water heaters .. to scavenge some of the 65% of fossil fuel energy that winds up as waste heat.
Modern condensing gas boilers for domestic use are better than 90% efficient, the exhaust gas is very cool so I don't think that thermoelectric devices have much to offer there.
Fossil fuel systems 'cheat' a little in efficiently metrics by using the 'Lower Heating Value' of burning the fuel. That's the amount of energy the fuel contains if you assume that the resulting water produced in combustion is released as water vapor.
However in real boilers, there is no need for this - by condensing that water vapor back into liquid water you can get more energy out. That's the 'Higher heating value'.
When measuring efficiency, we always use the lower heating value, whereas we ought to use the higher heating value. Therefore a "98% efficient furnace" is actually more like 88% efficient when considering the higher heating value.
Also, that's not necessarily more efficient for heating when compared to heat pumps. With a SCOP of 4, converting gas to electricity at 35% efficiency is better than using it for heat at 100%.
If you had CHEAP device, which uses exhaust at 70C to still generate some energy, this could be useful. This way, your exhaust could be lower temperature than your heated water. Currently this is not feasible, because such devices are not cheap enough.
Yeah, I think the more conventional way to integrate heat-to-electricty into heating applications is the reverse of that: feed the high-grade heat from fuel combustion into the converter, and use the rejected waste heat from it for heating. This gives a big enough thermal gradient to get useful amounts of electricity out, and the waste heat can still be plenty hot enough to be used for heating.
If the exhaust is too cold the water vapor condenses out left with a puddle of acid (the acid is from the trace sulfur in the Nat gas)
Also, cold exhaust doesn't exhaust (since it cant rise through a density change). So, if your exhaust is too cold, it will collect in your basement (CO2 should be a bit denser than air, so absent any drafts it'll collect)
The previous owners of my house installed my high efficiency stove improperly using the old brick flue. The fumes eventually ate away at the mortar and a few bricks fell I'm clogging it.
Unless you have a fan blowing smoke into flue. Without fan, my furnace would not work, exhaust is only slightly above water temperature and furnace is designed so that condensate pools in special container, which you have to empty or connect to sewage.
Most new and recently installed domestic gas boilers in the UK use balanced flues not chimneys and are condensing boilers. The exhaust is fan assisted and the temperature is very low.
If I understand the article correctly, you're talking about two different things. Something can be efficiently heated but if it cools down all of that energy is effectively lost. This device would allow you to recapture some of that energy that is lost to then reheat the water. Think of it more as active insulation than a better heating element.
I don't think that's quite how it works. This device (or any thermoelectric generator) converts a fraction of the heat energy flowing through it into electrical energy. That means energy has to flow through it for it to work - essentially the opposite of insulating. Using this device, by definition, will result in the original hot thing being cooler, and the "cold end" (normally the outside world) being ever so slightly warmer.
The place this would be useful is in capturing otherwise unusable energy from low grade heat waste. For example, the output water of a steam turbine might still be "hot", but there's very little you can do with it - it's in the wrong place for heating houses, it's too cold for steam turbines. Generally that energy is lost to the air.
Current peltier chips are barely a few percent efficient, break easily and require high temperature differentials. They're used in stove-top fans and RTGs. The latter should be Sterling engines to be ignorant mind. I don't doubt there are some interesting things to be accomplished by shaping the flow of heat conduction but I sincerely doubt it'll be energy generation directly. Thermodynamics isn't something you can just ignore!
The history of moving parts in space missions (RTGs) is why Kepler's mission is degraded: keeping reaction wheels and other spinning things lubricated and moving when you can't get in there and service them is hard.
It says a ZT of 3.1, isn't that approaching 25% efficiency on par with more standard heat engines? Also more in line with say a solar panel for light to electricity.
That kind of efficiency would be a big breakthrough if it where affordable, imagine the applications.
I suspect you would still need a fan to cool off the hot side, but very silent, very long lasting, and the only moving part could be generic easy to replace fans.
Most current fridges have a fan on the cold side to circulate the coldness all around, and rely just on convection to pull away the heat (excepting RV propane fridges, which don't have a cold-side fan and have very uneven temperatures, but do have a hot-side fan because otherwise they don't work on hot days)
Having had the evaporator fan go out in an old fridge once, I think you'll find fridge fans not as generic nor replacements as easy to find as you might hope.
> and rely just on convection to pull away the heat
Interesting, I've definitely seen a normal sized house fridge with a fan on the hot side, but I don't know what is typical.
> I think you'll find fridge fans not as generic nor replacements as easy to find as you might hope.
Current generation probably not, but there's no reason these can't be your typical $20 computer case fan, especially on the hot side where there's no worries about condensation... for peltier elements you could even stick a off the shelf CPU heat sink on top (though presumably at scale you would build the heat sink custom).
Technically if you are willing to go overkill you could potentially use huge passive radiators but that would probably be a bit of a niche application.
I wonder if you could get it small enough to use with CPUs. Could extend the battery of a laptop for example. I’m also thinking of AC units, where it could lower the power consumption at the same time as lowering the heat we diffuse into nature, especially in big cities where the most rooftops feel like saunas.
> I’m also thinking of AC units, where it could lower the power consumption
Nope, if you use this on hot side, AC will have to work more to get rid of heat, but this device will still provide less power than ac must now use. Design criterium for AC's is that hot side should be as cool as possible (so it has a big fan to cool it down), but thermoelectrics require hot side to be as hot as possible.
AFAIK compressor units are still way more efficient than this device, so no reason to bog them down with something with less efficiency.
The solid state methods can sustain a temperature differential with a given power/heat flow. If the hot side is cooler, the cold side can be cooler, or less power is required.
If you insert something between hot side and air to extract energy, there will be more heat resistance and lower heat transfer, but this resistance will be bigger than what you get from your energy extracting device.
If you use that new device as a cooling enchancer, you will have to provide some power. If your first cooler user more efficient method, you should just make bigger unit using more efficient method, because otherwise whole apparatus will be less efficient.
Thermoelectric elements add resistance. You are literally adding a giant resister between the hot and cold side, which decreases the thermal Flux for a given temperature difference.
When it was unusually hot over here, I was thinking about this. We can convert nearly every kind of energy into electricity and back. But for heat, for some bizarre reason, this only works one way. A lot of processes generate it, but you literally can't convert it into any other form of energy. You have to transfer it somewhere where it won't be as much of a problem.
It would be very nice to have an AC where you don't need to deal with that. I thought about many ways to get rid of heat without having to dump it somewhere, but thermodynamics is such a terrible thing.
> I thought about many ways to get rid of heat without having to dump it somewhere, but thermodynamics is such a terrible thing.
This is likely a stupid question but temperature is a statistical property describing the energy of molecules right? Is there no way to tap the energy of the molecules for useful work without having a temperature difference?
You might want to read about Maxwell's daemon [1]. I came across this a few weeks back, two links away from the HN homepage.
I found it intreaguing and read more about it but I can't quite umderstand the proposed solutions, likely due to my lack of background in thermodynamics.
Clearly, Maxwell was much better (and funnier) in articulating what I was asking. Looks Maxwell's demon can work some of the time at a molecular level, but on average, you need to do work to extract useful heat from matter at a macroscopic level. The hypothesis being that you need to do work just to identify hot/energetic molecules from the cold/low-energy molecules.
I once read in some article that CPUs have the same power consumption as equivalent output of a nuclear reactor, based on size, of course. So the 1 inch of a CPU takes as much energy as 1 inch of a nuclear fusion output from a nuclear reactor. Again, I don't know if this is true, but if it is, that's a lot of heat energy, given the number of computers in the world.
But maybe I remember wrong, this was probably 15 or 20 years ago that I read this.
In summer, the sun (rudely) gives me excess unwanted heat in my house. I need to expend even more energy to get that heat from inside to outside my house. (We call this process "Air Conditioning".)
What would be ideal is if the heat energy could be converted into some other form of energy that I could dispose of. Maybe if I could turn it into electricity I'd put a lightbulb in my window and shine the converted heat energy away.
I don't really care if that process is inefficient because that heat was completely free and unasked-for. I don't need to do anything useful with the converted heat energy, I just want it gone. If the process of converting heat to a disposable form of energy can be done for less energy than air conditioning needs, then its worth doing.
> What would be ideal is if the heat energy could be converted into some other form of energy
That would be nice, wouldn't it. If you ever build a universe, you should definitely implement that. In this universe the laws of thermodynamics specifically forbid destroying heat, or converting it into something else.
Heat will naturally spread out evenly over time, but if we would prefer to do the opposite, moving heat where we want it (or away from where we don't want it), we need to use a heat pump which expends energy to do that, and that energy produces further waste heat which, yes, we also need to move. This is a core principle of how our universe works. Nobody is going to invent a new material that violates the laws of thermodynamics, for the same reason nobody is going to find an even integer that was hiding between 2 and 4 somehow without us noticing it.
> In this universe the laws of thermodynamics specifically forbid destroying heat, or converting it into something else.
But wait, what does a Stirling engine run on? Or a steam engine? You clearly can get some energy out of heat.
Also, objects that are hotter than absolute zero radiate infrared light — and you, in principle, could convert light, regardless of its spectrum, into electricity.
They run on temperature differences, that's how you get usable energy out of it. Google Carnot engine, there's even a super simple way to find the maximum efficiency just from the temperatures!
Temperature difference is caused by heat. A heat engine converts heat into mechanical work. A Carnot heat engine is an idealized form of heat engine vs say a Diesel heat engine.
You can most certainly convert heat into something else. It's called the law of conservation of energy, not heat, which is just one form of energy.
But the limits to these transformations are governed by the 2nd law of thermodynamics, not the first. You can't arbitrarily convert heat into other forms of energy.
E.g., if your machine relies on all energy in my room to spontaneously gather in my coffee mug to heat it up, while simultaneously cooling my warm room, this would be perfectly fine with conservation of energy and super useful. Unfortunately, it would also violate the 2nd law and it's thus something that doesn't happen in our universe.
It sure is tempting to think of heat as just energy. I mean it is. Of course. But the kind of device you're talking about is like Maxwell's Daemon. The article is talking about a step towards making an inexpensive thermocouple. These require temperature differentials, not just heat. And you don't want to be drawing power from the temperature differential in your house ;)
Heat is call "Low Quality" energy in thermodynamics. Heat is basically disorganized particle movement. It is very hard to convert disorganized energy to organized energy like electrical current.
I don't see why people are downvoting you, as you are basically wanting to avoid wasting good energy with air conditioning.
Maybe people thought your tone was blasfemous?
Because what GP is proposing is thermodynamically impossible. The second law of thermodynamics states that heat naturally flows from hotter to colder regions. You can't just magically convert a static pool of heat to light and beam it somewhere else with no energy input. It's like suggesting that we generate electricity by rolling stones uphill; it just doesn't make sense.
I don't know about blasphemy. My intended tone was "telling a joke".
I know what I proposed would require our understanding of thermodynamics to be wrong, but this is in response to an article about a material that converts heat into electricity. I would love to hook up this material to a lightbulb in my window and save money!
I did get your snark but also half-thought that you were alluding to covering the house in solar panels. Which would, in an indirect way, remove the heat from your house and convert it to electricity :)
People are downvoting because billpg essentially wants a perpetuum mobile, something that creates useable energy just from the presence of a high temperature reservoir. This violates the 2nd law of thermodynamics. Heat does not equal usable energy, what you need to extract something useful is a temperature difference.
TE devices are SPECTACULARLY inefficient. Carnot's Law applies and the thermal tolerance of all TE material are limited. You are lucky to get 5% on a good day. You still still a heat source and cold sink just like ALL OTHER heat engines and the efficiency is limited by the difference in temperature. There is no cheating. There is no magic.
Selenium is moderately available but it is toxic waste when you need to replace it (all things have finite lifespans) and then need to dispose of the discard.
Depends entirely on how the system was laid out I guess. If you're in a cold region, say Iceland, and water-cool your miners entirely by running the hot water against a thermoelectric system with the one side out in the snow, maybe 60%?
I was trying to find article about some random guy who used mining rig to suprcharge his heat pump [0] but found commercial solutions [1]. Just search "miner heater" on google.
I once joked here about crypto showers. Electric showers are ubiquitous where I live, it's just a dumb resistance heating up water. Could be replaced with a ton of processors mining cryptocurrency and heating up the water as a side effect.
Are you telling me this sort of thing already exists today?
I wonder if these researchers would help productize their findings into a kick starter prototype kit, would this advance faster due to the hive effect?Seems like every innovation stops "3 feet from gold"
Unfortunately we'll always need the pipe to blow air outside. The heat has to go somewhere. Even with this invention, if they're converting heat to electricity, they will have to release at least as much heat as they converted to electricity.
Indeed, this can be demonstrated by running current through a plain old Peltier cooler without a big chunky heat sink attached. At first the cooler will have one side heat up and the other side get quite cold, but within a few minutes the entire block becomes evenly warm.
Eh, doesn't all modern air conditioner just do heat exchanging? no need to actually blow the air out - but yes you still need a pipe for the heat exchanging medium
For example you could convert the heat into photons (infrared radiation) and beam that out of the room without needing a pipe. You would probably need an infrared transparent patch on the window to aim at.
You can’t. The second law of thermodynamics prohibits it.
“It is impossible to devise a cyclically operating device, the sole effect of which is to absorb energy in the form of heat from a single thermal reservoir and to deliver an equivalent amount of work.”
Nitpick on nitpick -- you absolutely can convert heat energy into electric energy. A dynamo on a heat engine does exactly that. What suffices is temperature differential.
It is a terrible analogy, a modern physicist would prefer being caught dead rather than using it, but one can think of heat as a liquid (traditionally called caloric ) stored in a vessel. Temperature is the level of the liquid. For flow you need another vessel where the level is low, that would allow flow of heat from a vessel with higher level to a lower one and one can then convert that flow into power of a form one desires. Its possible to go in the opposite direction also, that's called a heat pump. Charge and voltage would be another (incorrect) way of thinking about heat and temperature.
This is literally inventing half a solution. Shockley didn't go to the press when he'd built half a diode.
(OK, this is snarky, but I think this is another case where people need to understand the difference between "a small but valid advancement in the art" and "actual products". As a materials science paper this is valid.)