New Alloy Can Convert Heat Directly Into Electricity

[Azazal]

Very interesting. Curious to see how and if this ever plays out

Let there be science

A new alloy with unique properties can convert heat directly into electricity, according to researchers at the University of Minnesota. The alloy, a multiferroic composite of nickel, cobalt, manganese and tin, can be either non-magnetic and highly magnetic, depending on its temperature.

Multiferroic materials possess both magnetism and ferroelectricity, or a permanent electric polarization. Materials with both of these properties are very rare; check out this explainer from the National Institute of Standards and Technology if you’re interested in the electron orbital arrangements that cause these phenomena.

In this case, the new alloy — Ni45Co5Mn40Sn10 — undergoes a reversible phase transformation, in which one type of solid turns into another type of solid when the temperature changes, according to a news release from the University of Minnesota. Specifically, the alloy goes from being non-magnetic to highly magnetized. The temperature only needs to be raised a small amount for this to happen.
When the warmed alloy is placed near a permanent magnet, like a rare-earth magnet, the alloy’s magnetic force increases suddenly and dramatically. This produces a current in a surrounding coil, according to the researchers, led by aerospace engineering professor Richard James. Watch a piece of the alloy leap over to a permanent magnet in the video clip below.

A process called hysteresis causes some of the heat energy to be lost, but this new alloy has a low hysteresis, the researchers say. Because of this, it could be used to convert waste heat energy into large amounts of electricity.

One obvious use for this material would be in the exhaust pipes of vehicles. Several automakers are already working on heat transfer devices that can convert a car’s hot exhaust into usable electricity; General Motors is using alloys called skutterudites, which are cobalt-arsenide materials doped with rare earths.

Rare earth magnets are already a necessity in many hybrid car batteries, so heat-capture devices made of the new multiferroic compound could be placed near the magnets.
The material could also be used in power plants or even ocean thermal energy generators, the researchers said.

A paper on the alloy was published in the journal Advanced Energy Materials.

[erik_t]

Oh hey, Dick James! Kind of suspected as much once I saw U of M and metallic hysteresis. I'll have to walk downstairs next week and talk with him about this.

[Purple]

This begs the obvious question of costs. Is there any word on how expensive this material is to make?

[erik_t]

Manganese isn't super cheap, but it's an alloy rather than a carefully-molecularly-constructed compound. It shouldn't be vastly more expensive than the constituent elements, I wouldn't think.

[Surlethe]

A couple of basic questions.

(a) Does B go as temperature?
(b) Will the material have to be subjected to an alternating field? Because otherwise it seems like you'd get a burst of electricity as the magnetic field switches on, and then the magnetic field would stabilize.

PS- You're a showoff, Erik.

[erik_t]

Completely shameless. Dr. James rollerblades to work, if anyone was curious.

I'll ask him if he's got a few free minutes (he will). Not next week, though, I'm away for a conference.

[Singular Intellect]

This technology would also have very obvious applications in solar power Eegeneration.

[U-95]

Fascinating at least. What applications could have it in domestic use?

[Imperial528]

This technology would also have very obvious applications in solar power Eegeneration.

Or really any kind of power generation. I'm looking to see if it could be useful in nuclear (fission, or preferably fusion) power generation applications, since it could potentially drastically cut down on reactor size, especially if it is more efficient than water-based systems employed now. Of course, since it probably won't cool the reactor as fast as a water based system would then fuel load may have to be made smaller at the same time, but if at the end we get a cheaper reactor that's for the better, I think. And the smaller fuel load would mean that a nuclear accident would be easier to prevent, especially if the reactor is designed to run without active cooling.

[Purple]

This technology would also have very obvious applications in solar power Eegeneration.

Or really any kind of power generation. I'm looking to see if it could be useful in nuclear (fission, or preferably fusion) power generation applications, since it could potentially drastically cut down on reactor size, especially if it is more efficient than water-based systems employed now. Of course, since it probably won't cool the reactor as fast as a water based system would then fuel load may have to be made smaller at the same time, but if at the end we get a cheaper reactor that's for the better, I think. And the smaller fuel load would mean that a nuclear accident would be easier to prevent, especially if the reactor is designed to run without active cooling.

Don't use it for cooling the reactor. Rather, use it to cool the water after it leaves the steam turbines. That way you get both the power generated from the turbines and extra while doing away with the massive cooling towers.

[Hamstray]

Or really any kind of power generation.

So what is the conversion efficiency of this process versus the carnot, brayton or even stirling cycle?

[Hamstray]

Don't use it for cooling the reactor. Rather, use it to cool the water after it leaves the steam turbines. That way you get both the power generated from the turbines and extra while doing away with the massive cooling towers.

What 'extra' energy do you propose is to be gained out of this? and why would you stick with your thermal neutron economy water cooled reactors if you can potentially beat the efficiency of their carnot cycle?

[Purple]

Mostly becouse water cooling has the advantage of being fail safe. From what I understand it's easier to pump more water over the reactor than it is to go in and replace an alloy plate if something happens. It all depends thou as you said in how the system compares in efficiency to he standard operating cycle.

Now, my knowledge of how reactors work might not be perfect. But from what I understand the water that condenses from vapor from the steam turbines is stored in massive water towers until it is cold enough to be safely returned to the environment. If these water towers were replaced or augmented by some heat to electricity system that could salvage extra energy and increase the overall efficiency of the plant without requiring engineers to develop a massively novel approach to nuclear reactors.

[Imperial528]

This technology would also have very obvious applications in solar power Eegeneration.

Or really any kind of power generation. I'm looking to see if it could be useful in nuclear (fission, or preferably fusion) power generation applications, since it could potentially drastically cut down on reactor size, especially if it is more efficient than water-based systems employed now. Of course, since it probably won't cool the reactor as fast as a water based system would then fuel load may have to be made smaller at the same time, but if at the end we get a cheaper reactor that's for the better, I think. And the smaller fuel load would mean that a nuclear accident would be easier to prevent, especially if the reactor is designed to run without active cooling.

Don't use it for cooling the reactor. Rather, use it to cool the water after it leaves the steam turbines. That way you get both the power generated from the turbines and extra while doing away with the massive cooling towers.

I was thinking of using it to more directly draw energy from the reactor by covering the reactor in heat sinks made of this material, or with a component inside the sink made of it to create electricity, rather than using it to cool it, hence why it would work best on smaller, cooler reactors. For large scale reactors your idea is probably better, though, since it would make cooling less of a headache.

Or really any kind of power generation.

So what is the conversion efficiency of this process versus the carnot, brayton or even stirling cycle?

Since there are less steps involved I would assume it has greater efficiency then current methods, but since I don't have any numbers on the material I really couldn't say for sure, and I'm not exactly the person to figure it out anyhow.

[Skgoa]

Drawing power from the reactor IS cooling it. The reaction makes heat and that heat is taken away to be turend into electricity.

[Chardok]

ONE STEP CLOSER TO MY FURTURE IDEA OF MINING THE HEAT THAT THE GROUND ABSORBS BY RUNNING TINY METAL GRIDS THROUGH ALL ROAD SURFACES IN THE US! I AM VINDICATED! You could put this shit on ANYthing! you know how heavily urbanized areas are heat islands because of all the concrete? Well, just coat the rebar in new construction with this shit and BOOM-O! (depending on the output, I guess) you got yourself a bunch of sweet, sweet free electricity! the future is upon us! Print it, ship it, like it on Facebook, tweet it, retweet it, and serve it on toast, humanity is saved!


Seriously, though - how far out if this stuff from mass implementation and why isn't this plastered on the front page of the internet? This sounds like it COULD be the game-changer we've been looking for.

[Norade]

Do we have enough of the rarer bits of metal to cover any significant area with the stuff? Also, the US can't even keep bridges from collapsing, let alone even dream of resurfacing even a fraction of the total road system even in just the southern most 10% of the nation.

[Chardok]

Do we have enough of the rarer bits of metal to cover any significant area with the stuff? Also, the US can't even keep bridges from collapsing, let alone even dream of resurfacing even a fraction of the total road system even in just the southern most 10% of the nation.

Look, you take grids of this shit, and just pound it the hell into the pavement/sidewalks, then a quick splash of asphault and bob's your uncle. you don't have to do it all at once, and you don't have to do it for all the roads, just the ones in the urban areas for now. The earth sucks up enough heat every day from the sun that it seems foolish to *not* mine it in some form. You doul even just run freaking pipes under the ground filled with ammonia or some other liquid with an absurdly low boiling point, circulate it through the pipes, back to a radiator made of that shit to harvest the heat/make electricity, then you just let the condensed liquid flow back to continue the loop.

[Norade]

Do we have enough of the rarer bits of metal to cover any significant area with the stuff? Also, the US can't even keep bridges from collapsing, let alone even dream of resurfacing even a fraction of the total road system even in just the southern most 10% of the nation.

Look, you take grids of this shit, and just pound it the hell into the pavement/sidewalks, then a quick splash of asphault and bob's your uncle. you don't have to do it all at once, and you don't have to do it for all the roads, just the ones in the urban areas for now. The earth sucks up enough heat every day from the sun that it seems foolish to *not* mine it in some form. You doul even just run freaking pipes under the ground filled with ammonia or some other liquid with an absurdly low boiling point, circulate it through the pipes, back to a radiator made of that shit to harvest the heat/make electricity, then you just let the condensed liquid flow back to continue the loop.

Once again, if they could do this then they would have roads in decent shape and not falling to shit everywhere.

[Chardok]

Come on, you're being too pessimistic - by your logic we shouldn't have built the space shuttle because our roads are poo. I think you're just jealous because the sun hates Canada.

[Magis]

I think someone needs to inject some reality into this discussion.

Firstly, this thing has zero application in a nuclear reactor. For starters, this alloy "works" by oscillating its temperature about some point, so that you get that magnetic switching effect. In other words, the alloy itself has to be cooled, since once you heat it up past the magnetic threshold, you need to cool it down past that threshold in order to heat it up again (unless I'm totally missing something here). Secondly, you can't put a nickel alloy into a nuclear reactor - end of story. Nickel, compared to other metals, is extremely prone to neutron radiation damage. Statistically, a nickel material in a thermal fission reactor would experience at least one dislocation of every single nickel atom in the material every six months, due to neutron interactions (a dislocation is when the atom is ejected from its position in the crystal lattice of the metal). This leads to a bunch of material degradation phenomena, like creep (along all axis), large loss in ductility, large decrease in yield strength, large increase in the temperature dependency of yield strength, accelerated fatigue crack growth, etc. Nickel just isn't used in these cores.

Other types of thermoelectric effects are already well-known (like the Seebeck effect), and haven't found any widespread power applications, though do find some use in niche application, like in space probes. Their efficiencies are also completely abysmal, and don't come close to even approaching the efficiencies attainable via thermodynamic cycles (like boiling water + steam turbines). My best guess is that this alloy will maybe find some niche uses, but won't be adopted for widespread use by anyone.

[Imperial528]

I think someone needs to inject some reality into this discussion.

Firstly, this thing has zero application in a nuclear reactor. For starters, this alloy "works" by oscillating its temperature about some point, so that you get that magnetic switching effect. In other words, the alloy itself has to be cooled, since once you heat it up past the magnetic threshold, you need to cool it down past that threshold in order to heat it up again (unless I'm totally missing something here).

Upon rereading the article, again, it seems like you've got it spot on. Which makes it appear to me that essentially for all intents and purposes aside from very specific applications this is just a new way to make a thermocouple*, at least so far as electrical generation goes.

*Obviously it is quite different from a thermocouple in practice, but unless it is much more efficient I can't imagine a power generation situation where a thermocouple couldn't do the same job.

Secondly, you can't put a nickel alloy into a nuclear reactor - end of story. Nickel, compared to other metals, is extremely prone to neutron radiation damage. Statistically, a nickel material in a thermal fission reactor would experience at least one dislocation of every single nickel atom in the material every six months, due to neutron interactions (a dislocation is when the atom is ejected from its position in the crystal lattice of the metal). This leads to a bunch of material degradation phenomena, like creep (along all axis), large loss in ductility, large decrease in yield strength, large increase in the temperature dependency of yield strength, accelerated fatigue crack growth, etc. Nickel just isn't used in these cores.

Good to know, henceforth all my points about using this alloy in a nuclear reactor are conceded.

[someone_else]

So, this thing must work in pulses. Heated to above its "trick temp" and generates a jolt of power, then cooled to below its trick temp and returns inert. Rinse and repeat.

Anyone has any idea of a way to do so that does not involve buttloads of moving parts and fluids?

[Solauren]

So, this thing must work in pulses. Heated to above its "trick temp" and generates a jolt of power, then cooled to below its trick temp and returns inert. Rinse and repeat.

Install them in smelting plants near the molten metal.

[someone_else]

it needs a cooling system, otherwise it won't work (more than once).

I was thinking of a rotating frame, where above there is the hot part and below the cold part, the rotor contains the whole gimmick, or a battery of gimmicks.

It should have small radius, but there is no problem about lenght. The rotor is part of the cooling system and has to be moved somehow.