Rippling Graphene Harvests Thermal Energy?

[bob012345]

TL;DR Summary

It is claimed 2D Graphene sheets can be used to harvest thermal energy while being at the same temperature as the surroundings.

It is claimed 2D Graphene sheets can be used to harvest thermal energy while being at the same temperature as the surroundings in seeming contradiction to Feynman's *argument (in the popular accounts). However, the academic papers referenced below make the claim everything is well within known physical laws. Yet they claim power is still possible from random motions of Graphene sheets. I include two formal papers from major journals and three news accounts of the teams work for reference.

What I am trying to understand about this work is where is the energy coming from if there is no temperature change and if this really works, how is it not violating at least the 2nd law of thermodynamics if not the first as they seem to claim they can extract energy from the ambient through the graphene sheets (if I understand correctly).

Since this work is published in Phys. Rev. E. and phys. Rev. Letters, I hope I am not violating PF policy bringing it up for discussion since it seems a sensitive topic to say the least. I selected the 'advanced' prefix mainly because the papers would be read by professionals.

Popular;
https://physicsworld.com/a/rippling-graphene-harvests-thermal-energy/
https://scitechdaily.com/physicists...f-graphene-to-generate-clean-limitless-power/
https://newatlas.com/graphene-motion-limitless-energy/52319/

Academic;
https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.117.126801
https://journals.aps.org/pre/abstract/10.1103/PhysRevE.102.042101*Feynman's Ratchet and Pawl Argument;
https://www.feynmanlectures.caltech.edu/I_46.html

 

[hutchphd]

As I recall Feynman's Ratchet and Pawl does not have a battery attached to it...

 

[bob012345]

As I recall Feynman's Ratchet and Pawl does not have a battery attached to it...

When I first saw this a couple of months ago I saw the battery and I said the same thing to myself. They claim to separate out the effects of the battery which provides a bias voltage and the motion of the Graphene sheet. I wonder if the bias voltage is just for the STM and any potential application would not have it?

I discovered a video here where professor Thibado explains the circuit including the bias voltage.

 

[hutchphd]

I think this is perhaps a slightly interesting transient phenomenon. If it didn't have graphene in the title no one would care IMHO. If you find some substance to it please post.

 

[bob012345]

Now I understand the bias voltage. It is merely to charge the Graphene sheet but it is not the source of the energy. The source of the energy is the Graphene sheet's random motion forcing charge to move around the circuit. The group is researching energy harvesting circuit configurations in conjunction with the Graphene being a source of one dimensional motion. The paper below summarizes this;

Low power energy harvesting circuits, which utilize variable capacitance as a source of power, have been investigated. Five different circuit topologies were simulated, and the most efficient one was experimentally tested. The most efficient circuit utilizes two transistors for rectification and two storage capacitors. The storage capacitors can be charged to any voltage set by a DC power supply. The source of power comes from the driving force behind the varying capacitance and not from the DC power supply. https://cpb-us-e1.wpmucdn.com/wordpressua.uark.edu/dist/3/316/files/2017/05/aip_ad_eh.pdf

I find this interesting.

 

[hutchphd]

My guess is that they are creating an electronic form of Maxwell demon...but if they have the numbers then we shall see.

 

[Lord Jestocost]

My guess is that they are creating an electronic form of Maxwell demon..

I don't think so. In case one can uphold a slight temperature difference between the "fluctuating" graphene sheet and the load resistor (which produces itself thermal electrical noise thus feeding back to the capacitor), I don't see any problems. Or one biases the system by means of a battery which finally does the work. As John D. Norton remarks in his paper "All Shook Up: Fluctuations, Maxwell’s Demon and the Thermodynamics of Computation":

"The molecular-scale world is quite unlike that of our macroscopic experience. Each molecular-scale component has its own thermal energy that leads it to bounce around in all its degrees of freedom. Those molecular motions overturn our macroscopic intuitions about how delicate, molecular-scale machinery operates and defeat apparently natural designs for demons...

... For each process that could accumulate violations of the second law, there would be another, also derived from fluctuations, that would undo it."

 

[hutchphd]

With apologies, I was using a shorthand. I did not mean to imply that the Demon is ultimately successful in this circumstance. The demon somehow obfuscates that this is simply a heat engine. Frankly unless they start spontaneously harvesting real power, the details are not interesting to me.

 

[bob012345]

A couple of quotes from the scitechdaily.com article referenced above.

Thibado’s team found that at room temperature the thermal motion of graphene does in fact induce an alternating current (AC) in a circuit, an achievement thought to be impossible.

According to Kumar, the graphene and circuit share a symbiotic relationship. Though the thermal environment is performing work on the load resistor, the graphene and circuit are at the same temperature and heat does not flow between the two. That’s an important distinction, said Thibado, because a temperature difference between the graphene and circuit, in a circuit producing power, would contradict the second law of thermodynamics. “This means that the second law of thermodynamics is not violated, nor is there any need to argue that ‘Maxwell’s Demon’ is separating hot and cold electrons,” Thibado said.

 

[bob012345]

I don't think so. In case one can uphold a slight temperature difference between the "fluctuating" graphene sheet and the load resistor (which produces itself thermal electrical noise thus feeding back to the capacitor), I don't see any problems. Or one biases the system by means of a battery which finally does the work. As John D. Norton remarks in his paper "All Shook Up: Fluctuations, Maxwell’s Demon and the Thermodynamics of Computation":

"The molecular-scale world is quite unlike that of our macroscopic experience. Each molecular-scale component has its own thermal energy that leads it to bounce around in all its degrees of freedom. Those molecular motions overturn our macroscopic intuitions about how delicate, molecular-scale machinery operates and defeat apparently natural designs for demons...

... For each process that could accumulate violations of the second law, there would be another, also derived from fluctuations, that would undo it."

If I understand correctly, there is no temperature difference and it's not the battery that does the work.

 

[hutchphd]

an achievement thought to be impossible.

I would love to see a reference for this (I am not quibbling here I really would like to know what this refers to...Josephson junctions happily do this in the cold )

 

[Lord Jestocost]

...and it's not the battery that does the work.

Why is then at all a battery in the circuit? A battery is an active electric component.

Every passive electric circuit element produces current noise when at finite temperatures. Connecting two such elements, for example a resistor and a capacitor in parallel operation, nothing happens on the average when both are at the same temperature (an electric version of Feynman's Ratchet and Pawl machine). Due to thermal electric fluctuations, sometimes the capacitor releases energy to the resistor, sometimes the resistor releases energy to the capacitor. That’s all.

 

[TeethWhitener]

Full disclosure: There was discussion of this paper among the mentors/SA's a while back, in anticipation that it might attract a lot of crackpottery.

I've been over this paper several times now and I think the important takeaway is that the pop-sci press has blown these results way out of proportion. The paper itself claims no violation of any of the laws of thermodynamics. In fact, as far as I can tell, the specific scientific advance made in this paper is simply to theoretically describe and then experimentally observe a more generalized version of Nyquist noise. Two very important foundational papers to read (if you have access) are Nyquist's original 1928 paper:
https://journals.aps.org/pr/abstract/10.1103/PhysRev.32.110
and Brillouin's classic 1950 paper on the electric version of Feynman's ratchet and pawl:
https://journals.aps.org/pr/abstract/10.1103/PhysRev.78.627.2
In both cases, power generated from thermal fluctuations in one part of the circuit is shown to be dissipated in another part of the circuit, so that the laws of thermodynamics are not violated.

Thibado's paper shows that Nyquist's result is a first-order approximation to a more general result (this part is not new), and they observe the "super-Nyquist" behavior in the graphene-coupled circuit (this is the new part--I think). But again, at no point is thermodynamics violated.

To their credit, Thibado's paper specifically states "Our model provides a rigorous demonstration that continuous thermal power can be supplied by a Brownian particle at a single temperature while in thermodynamic equilibrium, provided the same amount of power is continuously dissipated in a resistor. Here coupling to the circuit allows electrical work to be carried out on the load resistor without violating the second law of thermodynamics." But I do think they're playing really fast and loose with the concepts in some of their interviews (and really even in the paper itself).

 

[bob012345]

I would love to see a reference for this (I am not quibbling here I really would like to know what this refers to...Josephson junctions happily do this in the cold )

I think the statement comes from quotes like this or similar;

“It was predicted in the ’50s that if you had this type of what we call Brownian motion, this kind of thermal motion of the graphene, that for one it couldn’t be used to power a circuit,” Thibado said. “And it turned out that that theory that was done in the ’50s was completely wrong.”

https://fulbrightreview.uark.edu/th...ates-listeners-of-short-talks-about-his-work/

What do JJ's happily do in the cold?

 

[Twigg]

Frankly unless they start spontaneously harvesting real power, the details are not interesting to me.

Talk about high standards for novel results (just messing with ya)

 

[hutchphd]

“It was predicted in the ’50s that if you had this type of what we call Brownian motion, this kind of thermal motion of the graphene, that for one it couldn’t be used to power a circuit,” Thibado said. “And it turned out that that theory that was done in the ’50s was completely wrong.”

Is there any actual scientific reference? A talk show is somewhat less than any known standard. This all begins to sound like a duck.

 

[bob012345]

Full disclosure: There was discussion of this paper among the mentors/SA's a while back, in anticipation that it might attract a lot of crackpottery.

I've been over this paper several times now and I think the important takeaway is that the pop-sci press has blown these results way out of proportion. The paper itself claims no violation of any of the laws of thermodynamics. In fact, as far as I can tell, the specific scientific advance made in this paper is simply to theoretically describe and then experimentally observe a more generalized version of Nyquist noise. Two very important foundational papers to read (if you have access) are Nyquist's original 1928 paper:
https://journals.aps.org/pr/abstract/10.1103/PhysRev.32.110
and Brillouin's classic 1950 paper on the electric version of Feynman's ratchet and pawl:
https://journals.aps.org/pr/abstract/10.1103/PhysRev.78.627.2
In both cases, power generated from thermal fluctuations in one part of the circuit is shown to be dissipated in another part of the circuit, so that the laws of thermodynamics are not violated.

Thibado's paper shows that Nyquist's result is a first-order approximation to a more general result (this part is not new), and they observe the "super-Nyquist" behavior in the graphene-coupled circuit (this is the new part--I think). But again, at no point is thermodynamics violated.

To their credit, Thibado's paper specifically states "Our model provides a rigorous demonstration that continuous thermal power can be supplied by a Brownian particle at a single temperature while in thermodynamic equilibrium, provided the same amount of power is continuously dissipated in a resistor. Here coupling to the circuit allows electrical work to be carried out on the load resistor without violating the second law of thermodynamics." But I do think they're playing really fast and loose with the concepts in some of their interviews (and really even in the paper itself).

I really hope you are not suggesting that merely being interested in this paper makes someone under suspicion of being prone to crack pottery!

If I understand it you are making the point that the work dissipated in the load resistor can never ever be used to do anything useful without violating the laws of thermodynamics as we know them. Correct? It seems clear to me that Thibado believes the work referenced from the 1950's is flawed and such power could be harnessed. I would have a hard time understanding why he would make such clear statements suggesting the work could be harnessed if he did not truly believe it? Here is a recent interview.

https://fulbrightreview.uark.edu/th...ates-listeners-of-short-talks-about-his-work/

So, of course he may be entirely mistaken.

 

[bob012345]

Is there any actual scientific reference? A talk show is somewhat less than any known standard. This all begins to sound like a duck.

I quoted two scientific references by the author in the first post. Did you look at them at all?

 

[bob012345]

Why is then at all a battery in the circuit? A battery is an active electric component.

Every passive electric circuit element produces current noise when at finite temperatures. Connecting two such elements, for example a resistor and a capacitor in parallel operation, nothing happens on the average when both are at the same temperature (an electric version of Feynman's Ratchet and Pawl machine). Due to thermal electric fluctuations, sometimes the capacitor releases energy to the resistor, sometimes the resistor releases energy to the capacitor. That’s all.

To charge the graphene sheet as I understand it but my understanding is not the metric you should go by. Read the articles and papers and if interested email prof. Thibado and ask.

 

[bob012345]

My original questions were simply these;

What I am trying to understand about this work is where is the energy coming from if there is no temperature change and if this really works, how is it not violating at least the 2nd law of thermodynamics if not the first as they seem to claim they can extract energy from the ambient through the graphene sheets (if I understand correctly).

The answers I think are suggested are that there is no useful energy going anywhere, it just sloshes around the system and thus there are no violations of any laws (not that anyone was suggesting there were).

So my refined question becomes this;

At the most fundamental level random thermal motion is not harvestable. But what of larger scales? Wind is in some sense organized motion of air molecules on top of random motions of course. Even Brownian motion as originally observed is not individual molecular collisions but the statistical aggregation of many thousands of hits on a large particle such as a pollen grain which Robert Brown observed. Could not this graphene sheet buckling behavior be thought of as large scale aggregate motions (10um x 10um sheets contain a lot of atoms!).

So I wonder, is this phenomenon really more about engineering on a small scale but physics on a much larger scale than fundamental?

 

[TeethWhitener]

I really hope you are not suggesting that merely being interested in this paper makes someone under suspicion of being prone to crack pottery!

Of course not, but it seems to be a case of a researcher overstating his findings coupled with the imprimatur of peer review, which is a perfect storm for crackpottery.

I try to give people in interview situations like this the benefit of the doubt, because these questions are asked in real time and everyone's trying to generate interest, but claims like this guy is making in the last part of that interview (comparing ambient thermal energy with wind and solar power, for instance) are really slippery and poorly defined. If I'm understanding the actual paper correctly, the researchers assert that there should be usable work done on a load resistor; however, in the experiment, there was no load resistor aside from the diodes wired in parallel. This goes to @hutchphd 's point about needing to show that power is actually harnessed in the device.

It seems clear to me that Thibado believes the work referenced from the 1950's is flawed and such power could be harnessed.

Yeah this is that slipperiness I was talking about. Depending on how you choose to define "ideal diode," you'll get different predictions. The more I read about this from interviews, the more sales-y and detached from the actual findings it starts to sound.

@Lord Jestocost also makes a very good point. The paper clearly states that the overall equilibrium power from the circuit is zero, but then follows it up by saying that this is because the power harvested from the graphene is dissipated in the resistors. The problem with this is that it disguises what makes the ratchet and pawl fail: namely, that at equilibrium, sometimes power is harvested from the resistors and dissipated in the graphene. So it's really unclear how you're going to get any useful work from this setup (aside from temporary fluctuations that get averaged out to zero over time).

The battery in the circuit makes all the talk of equilibrium properties even more frustrating (and also is probably honestly the ultimate source of the power). If you take the battery away, the apparatus doesn't work. Why not? Because, as you said, the battery is biasing the graphene relative to the STM tip. This doesn't happen at equilibrium. Which means the system is not in equilibrium. Just like every other system that we use to draw power from (including wind and solar).

Like I said, I think the main result comes in the extension to the Nyquist noise term, and the authors seem to assert as much in the PRE paper (NB--there's a non-peer-reviewed arxiv version of the paper floating around that's significantly different from the published, peer-reviewed PRE paper). I don't quite know what to make of all the rest of the hoopla.

 

[TeethWhitener]

Wind is in some sense organized motion of air molecules on top of random motions of course.

Wind is caused by the sun heating the air. The sun and Earth aren't in thermal equilibrium, and this imbalance can be harnessed to perform useful work.

Even Brownian motion as originally observed is not individual molecular collisions but the statistical aggregation of many thousands of hits on a large particle such as a pollen grain which Robert Brown observed

Yes and on average the pollen grain doesn't go anywhere. It bounces in random directions and its average position basically stays the same over time. It's a system in equilibrium and therefore can't be harnessed to perform useful work.

The takeaway message is that equilibrium is boring.

 

[bob012345]

Wind is caused by the sun heating the air. The sun and Earth aren't in thermal equilibrium, and this imbalance can be harnessed to perform useful work.

Yes and on average the pollen grain doesn't go anywhere. It bounces in random directions and its average position basically stays the same over time. It's a system in equilibrium and therefore can't be harnessed to perform useful work.

The takeaway message is that equilibrium is boring.

But equilibrium is not boring uniformity at all. It is a distribution. Also, one of the points of the papers is that sometimes there is a huge jump in position. The buckling is a large collective motion. What causes that? Could it be a quantum effect and not a thermal effect? Since the work was originally done in vacuum, it's not air molecules. What is the meaning of temperature on a 2D surface?

Also on a large scale, truly random motions can be utilized. I'm sure we all have seen examples.

Anyway, the Thibado team seems to believe they can commercialize this. They are making chips. If they don't work there will certainly be egg of their collective faces. Supposedly the first chip will provide a constant 10 milliwatts from a 12mm square device.

https://www.ntsinnovations.com/graphene-energy-harvesting

 

[TeethWhitener]

But equilibrium is not boring uniformity at all. It is a distribution.

This is more playing fast and loose with language. Thermodynamic equilibrium is defined as no net flow of matter or energy.

Also on a large scale, truly random motions can be utilized. I'm sure we all have seen examples.

Try telling your boss you were late for work because you were waiting for your car to tunnel from your house to the parking lot. Truly random motions can't be used in any sort of reliable sense. Playing semantic games won't make it any less true.

 

[hutchphd]

Yes and on average the pollen grain doesn't go anywhere. It bounces in random directions and its average position basically stays the same over time.

Just a small correction: in dimensions higher than 1D the random motions cause any particular atom to move on average (this is diffusion!)

Also on a large scale, truly random motions can be utilized. I'm sure we all have seen examples.

I'm sure you can't name a useful one. Large random energy excursions happen excruciatingly infrequently. I also don't carry emergency oxygen in case of atmospheric fluctuation.. If there is evidence please provide it.

 

[bob012345]

Just a small correction: in dimensions higher than 1D the random motions cause any particular atom to move on average (this is diffusion!)

I'm sure you can't name a useful one. Large random energy excursions happen excruciatingly infrequently. I also don't carry emergency oxygen in case of atmospheric fluctuation.. If there is evidence please provide it.

I was referring to the quote from @TeethWhitener Truly random motions can't be used in any sort of reliable sense. I was referring to the concept of using random motion in general. Sorry for any confusion. I believe some ocean Buoys use random wave motion to generate power. If not I think I could design one that could use accelerations to oscillate internal masses that generate and store power even if the device was free floating in completely random motion.

 

[hutchphd]

I believe some ocean Buoys use random wave motion to generate power. If not I think I could design one that could use accelerations to oscillate internal masses that generate and store power even if the device was free floating in completely random motion.

The title of the thread is

Rippling Graphene Harvests Thermal Energy?​

Thermal.

 

[bob012345]

The title of the thread is

Rippling Graphene Harvests Thermal Energy?​

Thermal.

It's a matter of scale. The Graphene sheet is not at the fundamental scale of atoms, it's a collective motion of millions of atoms. Experimentally according to the Thibado team, it happens and since motion is confined to a 2D sheet it seems nature is giving us something new to consider and the motion might be useful.

If you want another thermal example I can give you a thought experiment to consider. Take my Buoy concept and scale it to pollen grain size ~100 microns. In principle one might design a device on that scale to generate tiny amounts of energy by collisions or accelerations in a fluid or air and store it internally for later use.

 

[TeethWhitener]

Just a small correction: in dimensions higher than 1D the random motions cause any particular atom to move on average (this is diffusion!)

Yeah, what I meant was that $\langle \vec{x} \rangle = 0$ for a Brownian particle.

 

[TeethWhitener]

The Graphene sheet is not at the fundamental scale of atoms, it's a collective motion of millions of atoms.

You would do well to take your own advice and read the PRE paper. The convex/concave flip is modeled as a particle in a double potential well (with a few other things thrown in). At any rate, the STM tip which provides the second “plate” of the variable parallel plate capacitor is generally nearly atomically sharp and will thus only be probing a handful of atoms at a time.

 

[berkeman]

Thread is closed for Moderation...

UPDATE -- After a discussion in the Science Advisor forum, this thread will remain closed. Thank you everyone for your contributions.