Company Claims New Chip Converts Heat To Electricity

Dster76 writes to tell us that the startup, Eneco, has invented a solid state energy conversion chip which they claim will be able to convert heat directly into electricity or reach temperatures of -200 C when given an electrical current. While such a device could revolutionize many aspects of computing I'll keep my skeptic hat on for the time being.

Comments

Energy conversion devices

[erroneus]

I don't know why the notion should be so foreign. If someone told you they created a solid state device that could convert light energy directly into electrical energy would you believe them? Yeah, probably, because you have seen these in action already. They are on just about every calculator out there now. But there was a time when they were just an idea and the topic of fiction.

The notion of using heat is so different? Surely the technology is quite different I'm sure, but I would not be quite so quick to be skeptical.

Re:Energy conversion devices

[Anonymous Coward]

Nobody doubts it can be done, see Peltier. They're not terribly efficient (I thought they were 15% efficiency capable, but I guess not..)

Re:Energy conversion devices

[Zork the Almighty]

There is nothing that this chip could do that could not be done by simply making current designs more efficient. In fact, the use of such a chip necessarily wastes energy.

Re:Energy conversion devices

[rolfwind]

People are quite right to be skeptical or they will be fleeced every time a con artist announces a promises a great sounding technology. (BTW, this isn't the first time I read about someone promising similiar solid state chips on /.)

Extraordinary claims require extraordinary proof.

That said, just because someone is a skeptic doesn't mean we are impossible to convince. Just show us the tech - put up or shut up, that simple. I think that is a fair test.

Afterall, it's good enough for skeptic James Randi with paranormal claims, it's good enough for me.

Re:Energy conversion devices

[kfg]

Something just sounds fishy about this; like a scheme to power your car with it's own exhaust.

No, you can't run your car that way, but you can use the exhaust to turn a fan to turn a compressor to force induction to increase power.

Well call this a "turbocharger."

KFG

Re:Energy conversion devices

[MillionthMonkey]

If someone told you they created a solid state device that could convert light energy directly into electrical energy would you believe them? Yeah, probably, because you have seen these in action already. They are on just about every calculator out there now. But there was a time when they were just an idea and the topic of fiction. The notion of using heat is so different? Surely the technology is quite different I'm sure, but I would not be quite so quick to be skeptical.

The Earth receives high energy, low entropy photons from the sun. It reradiates low energy, high entropy photons back into space. These reradiated photons are not very useful in a 300 K environment, which is in thermodynamic equilibrium with them. This is similar to how you'd find it much harder to extract work from sunlight if you were on the surface of the sun, an environment in thermodynamic equilibrium with that light. (Yes I know everything would melt you nitpickers but the point remains.)

The reason those calculators work is because they are exchanging energy with the sun's surface and they are not in thermodynamic equilibrium with it. On the earth's surface, if you try to make a solar cell to catch low infrared from objects on our own planet, you'll find that your cell radiates away the photons you are trying to capture, just by being at room temperature.

Re:Energy conversion devices

[kiatoa]

Sorry. Doesn't work that way. The application of this chip will push the temp of the device being cooled up even higher. If you care about the life of your semiconductor devices you will be trying to get them to run as cool as possible. Yes, even semiconductors wear out. Look up electromigration. Take any scenario and apply the thermoconversion device to it (heat direct to elec. is not new btw, don't all geeks make a copper/iron thermoelectric candle powered radio as a kid?) and compare with an equivalent cost best of breed heat sink and the theroconversion device will lose because the chip being cooled will be running hotter. It is the same thing as those guys who think they can put a steam or stirling engine into their cars and get better gas mileage. Sure you can do it but it doesn't make economical sense.

Re:Energy conversion devices

[Ungrounded Lightning]

Peltiers are just thermocouples/thermopiles made of semiconductors. They are inefficient mainly because the material they're made of is a good enough heat conductor that it conducts most of the heat they've pumped back across the temperature gradient. Absent that they should be able to reach carnot cycle efficiency. Meanwhile, if you are willing to feed 'em the extra power (or accept that they generate

You can get cooling down to cryogenic temperatures just by building a pyramid of peltier cells (with progressively fewer couples in each layer), all interconnected electrically. This was done 'way back when they were first invented.

This device is a more efficient vacuum-tube version, using nanostructure field-emission needles for the cathodes and built in a microscopic form-factor using integrated-circuit manufacturing techniques. It does the same thing, but using electrons in vacuum. (The heat kicks them off the emitter with a momentum high enough for them to pass through a field to a more-negative collector plate.) A vacuum is a GREAT insulator, so the efficiency is much better. (Or pump heat by applying a voltage to encourrage the electrons to jump off the needles at thermal vibration peaks, cooling them, and smack into the collectors, heating them.)

Also: Since it is apparently built of metals and ceramics rather than semiconductors you can run it very hot - like at the focus of a solar concentrator. That can beat photovoltaics by a bunch.

I've seen reports of this device before. I presume this one means either they need more funding or they've just solved a manufacturing problem, bringing them a step closer to commercial rollout.

Computing?

[PreacherTom]

Revolutionize computing? How about revolutionizing LIFE. If true, this would be larger than controlled fusion.

Long term plan ... what were they thinking?

[Josh Lindenmuth]

The device sounds legit (it certainly doesn't break any laws of physics), but Eneco's plan for its longterm usage is just loopy. They say they'll initially try to improve battery life by coupling it with processors to recoup energy lost as heat. Great startup plan, but then it goes downhill ... from the article:

Brown also sees the chips ultimately replacing batteries altogether. He argues that by linking the modules to a microburner - a catalytic burner that produces between 275 and 600 degrees centigrade you can heat the chips and generate enough power to run the device.

In theory this approach would be far cleaner as the burners that Eneco is planning to employ use Ethanol

So in other words, Eneco plans to replace our laptop batteries with small Ethanol burning stoves that run hotter than a car engine? How would this ever fly, given people are worried about their current laptops catching fire? Also, who wants to fill up their laptop with gas every couple days? Energy coming from the grid at least in theory can be from renewable sources (wind, solar, tides, etc.). Why push Ethanol, a fuel which cannot be used on a large scale (and arguably requires more energy to produce than it provides)? The only reason I can think of is that they are trying to ride the "Ethanol investing wave" that hit markets over the past couple years (and appears to be waning).

Hopefully investors will see through the zany longterm plan and focus on the merits of the product, it really does appear to be valuable across a wide range of industries.

Re:Long term plan ... what were they thinking?

[anonimouskiller]

actually ethanol from soy or sugar cane is a very clean energy source, and it doesn't require more energy to produce than it provides. Brazil has been using ethanol for more than 2 decades now, it's still cheaper then gasoline and polutes less. although the 'stoves in the lap' idea seems kinda dumb.

Where is the energy going?

[jandrese]

Ok, so it converts latent heat into electricity, presumably working like a heat engine with the cold side fixed at absolute zero somehow? If you add energy, it gets even colder and produces...more energy? Is it just me or does this thing sound a lot like a perpetual motion machine component? Either this thing is bogus or the article is misleading as to what it actually does.

Re:Where is the energy going?

[lhbtubajon]

No, you misunderstood. There are two apparent functions that are totally separate:

1) Extract heat and use heat differential to generate electricity.

2) Use electricity supply to cool down to -200.

Either one or the other, but not both at the same time.

Re:Where is the energy going?

[wsherman]

Use electricity supply to cool down to -200 C.

That's not exactly a fundamental science discovery but if it's true it's actually pretty neat.

Oxygen condenses at -183.0 C and nitrogen condenses at -195.8 C so if these things became widely available you could make your own liquid oxygen and liquid nitrogen.

Unfortunately, liquid hydrogen is down at -252.8 C so you wouldn't be able to condense the hydrogen gas you got from electrolysis of water to make your own liquid hydrogen and oxygen rocket engine.

Peltier?

[Deflatamouse!]

The description sounds like a peltier to me. Apply some current, and the device generates a temperature differential.

Can a temperature differential cause the device to operate in reverse?

Thermocouple

[gus goose]

Solid-state device that converts heat to electricity....

http://en.wikipedia.org/wiki/Thermocouple

Invented 1821 - Prior art?

gus

P.S. Yes, I know that TC's rely on a temperature differential, not just a temperature... ;-)

Dupe

[Ancil]

Dupe from at least 2002. Both the slashdot article and the technology.

Similar work been done before

[davidmcn]

A few years ago (6 I believe) a company called Cool Chips LLC (which was traded on PinkSheets.com back then) claimed to have done the same thing. Unfortunately outside of the first round of announcements (which may have even been on Slashdot), nothing more was mentioned. In the comments back then it was hypothesized that an energy conglomerate or oil company would buy Cool Chips out to keep the technology from ever coming to the market. Me wonders if that might have happened, or if some of the primaries from Cool Chips are now a part of this venture.

Woohoo I have two options

[McNihil]

Burn a fiery death of an exploding battery.

OR

Massive Freezer burn on my lap and thus gonads.

This is truly astonishing.

I do not believe a word of this.

Intel announces new chip

[Anonymous Coward]

Intel announces new chip to turn electricity into heat, I believe it's called Pentium or something like that. It's apparently very very VERY very good at it.

Peltier-Seebeck

[tigre]

See wikipedia for more. Seebeck is the reverse effect.

Very silly idea

[Ancient_Hacker]

"Thermionic energy" sounds really wizzy, until you think about it a bit. You are trying to get electrons to boil off a hot surface and plonk themselves onto a cooler collector plate. Which means you need a hot emitter, a cool colector, and in between something that will pass electrons, but not too much heat. Basically, a losing proposition, as anything that passes electrons is almost by definition an excellent conductor of heat. Try to think of somethign that conducts electricity but insulates heat. Hard to come up with isnt it?

There are thermionic devices already around, you're probably looking at one. Vacuum tubes and CRT's are thermionic devices. Not very powerful ones--a typical tube only boils off microamps of current at under a volt, while requiring several watts of electrical power to heat the emitter. Not very impressive.

Re:-200C ?

[SEMW]

What? No you wouldn't. Ever heard of... Well, a freezer? That's a device capable of turning electricity into a temperature differential, and as far as I know, doesn't break any laws of Thermodynamics. It's called a heat pump. The device in TFA can also act as a heat pump, probably using the Peltier effect.

thermodynamics

[bcrowell]

According to the laws of thermodynamics, the conversion of heat to other forms of energy requires access to thermal reservoirs at two different temperatures, and there's a limit on the possible efficiency of the process, which is 1-T(low)/T(high). Their press release doesn't seem to be claiming anything that violates this, so it's not obviously voodoo science or anything. However, any such heat engine is only going to be useful when (a) you have cheap access to hot and cold reservoirs, (b) the temperature difference is fairly high, and (c) the efficiency of the heat engine is superior to the other practical heat engines that you have to choose from, or there's some other practical reason why this particular heat engine is better for your application.

Human battery

[MECC]

"Brown also sees the chips ultimately replacing batteries altogether."

Especially if implanted in people. From birth. In vast crops...

ahh that's nothing

[lubricated]

I've had a chip in my computer that converted electricity into heat. It was called a p4.

not bogus, not necessarily "disruptive technology"

[jetpeach]

So the technology is definitely hyped up in the article, but this is not bogus like oh so many of these types of articles on slashdot are. I'm in an electrical engineering PhD program and the ideas presented in the article are sound (i.e. there isn't any breakage of the 1st law of thermodynamics and no magic magnets involved!). The obvious question is what is this material that replaces a vaccum, this "properly selected semiconductor thermoelectric that is thick enough to support a significant temperature differential between the emitter and the collector in order to achieve efficiencies of practical interest" as this is the key to the technology. If they indeed have found a material to do this this is a very interesting technology that probably will make it into our consumer products, and possibly "soon".

Re:Second Law of Thermodynamics

[djh101010]

The problem is that converting heat energy directly into electricity violates the Second Law of Thermodynamics, not unlike perpetual motion machines.

Can you explain how heat (infrared photons, right?) is different in this regard than visible light (as in a photovoltaic cell)? I'm not busting your chops here, I just don't understand why the wavelength of the light matters in this context.

Re:Amd vs Intel

[hAckz0r]

Lets just think about this for a minute. CPU's generate heat, which is not a 100% efficient process. The device is 20-30% efficient even then. To get by on using an inefficient power source the CPU would have to be as efficient as possible. Once it is efficient enough to run on reduced power it will generate less heat, and thus less electrical power. See where this is going? It can't be a perpetual device if it runs at a net loss. If it can't be the only power source for the CPU why build it in? At best it will only convert SOME of the waste heat to run some other part of the computer. Cooling comes to mind because the power to run the cooling is directly related to the heat generated, but turning a mechanical fan would take perhaps more energy than the device could generate at that given temperature. The device itself needs cooling to work. It does not sound to me to be practical even for running that. At best it will require an external source of BTU's and cooling to power the device for a small payback in terms of electrical power.

What would make a difference if such a device could work for all wavelengths of radiation converting all nearby sources of light, radio, static RF, and heat into usable power. Not just a "solar cell" but a radiation rectifier. Even at 20% efficiency there would be plenty of energy to harness if the spectrum was wide enough.

Re:eno2001 Claims Stomach...

[Warg! The Orcs!!]

then it's a Koobrewop

Re:Second Law of Thermodynamics

[LWATCDR]

No this device doesn't.
I read the link. It looks like an improved thermocouple. It uses a heat-sink and a heat source just like an RTG.
As one person said to discredit the story "it is like powering your car with it's exhaust". A gas turbine engine does exactly that.
This wouldn't be a perpetual motion machine since it would still require a power source. What this device does is simply recovers some of the wasted energy from the hot chip and feed it back into the battery.
The only "questionable" part is this mystery semiconductor that conducts electrons a lot better than it conducts heat.

Re:Second Law of Thermodynamics

[volpe]

Agreed on the temperature differentials part, but I don't think I agree with the characterization of temperature as energy content and heat as its flow. Heat is the thermal energy content. It need not flow. An object that isn't at absolute zero contains "heat". Temperature is the average kinetic energy of the constituent particles. A brick at 100 degrees C contains more heat than a grain of sand at 100 degrees C, even though they are the same temperature. And that statement about heat is a statement about a static condition, with no flow involved.

Re:Second Law of Thermodynamics

[hacksoncode]

Because infrared radiation is not heat, it's infrared radiation. It can be produced by hot things, and it can make other things hot, but it is not, itself, "heat".

Heat is the energy contained in random motion of particles. The key here is *random". If you extract energy from pure heat that's just sitting somewhere, you're reducing the entropy of the hot thing, practically by definition. In order for this to not be a violation of the Laws of Thermodynamics, you would have to create even more entropy somewhere else. The easiest way to do this would be to generate more heat than you removed, but then you're up against conservation of energy. There are other ways to create entropy, though, so it's not technically impossible.

The reason you can grab energy out of heat moving from a hot location to a cooler location is that that net motion is not random, so you can increase the entropy of the system by randomizing the non-random element.

Note: yes, all the above is a dramatic over-simplification.

Re:Second Law of Thermodynamics

[ozbird]

The only "questionable" part is this mystery semiconductor that conducts electrons a lot better than it conducts heat.

Their patent on the solid state energy converter mentions that they have been experimenting with indium antimonide (InSb).
They also hold a patent for a way to make N-type semiconducting diamond, which may hint to where they're heading with this (or not.)

HERE's the real story:

[Ungrounded Lightning]

HAH! Found it on their web site!

Re:Second Law of Thermodynamics

[Thomas Miconi]

Heat is the thermal energy content. It need not flow. An object that isn't at absolute zero contains "heat".

In everyday language, sure. But not in scientific language.

From the wiki article: "In physics, heat, symbolized by Q, is defined as energy in transit."

Heat is the amount of thermal energy that is flowing between two bodies at different temperatures. The "thermal energy content" (roughly) is temperature itself. GP was quite correct.