Researchers at the University of Arkansas claim to have found a way to harvest energy from heat (not heat gradients) using graphene and a nonlinear circuit. See here:
https://journals.aps.org/pre/abstract/10.1103/PhysRevE.108.024130 / https://arxiv.org/abs/2308.09083
https://arkansasresearch.uark.edu/the-power-of-motion-harvesting-energy-from-freestanding-graphene/
I was pretty sure this would violate the laws of thermodynamics. They argue that it doesn't using reasoning I don't understand. I will say that it "works" and resolve YES if external trustworthy organizations determine that their technology is in fact able to extract useful work from ambient heat (i.e. not heat gradients, ambient radiation, etc, and not by releasing energy added in the production process in some way) by close. Otherwise this resolves NO. As this is somewhat subjective I won't trade in this market.
1,000
3.00What is the functional difference of harvesting "work" from ambient heat or ambient radiation? I would want this clarified before any bet. @osmarks
@JoaoPedroSantos I don't know that much thermodynamics, but I think the relevant thing is whether it can get energy out of a system which is otherwise in thermal equilibrium. As I understand it, if you have an object at the same temperature as your equipment then you can't get any useful energy out of radiation from the object.
@osmarks Ok fair enough, is just that for me ambient heat or ambient radiation are the same thing and you description was saying that only one of them would be counted xD, and yes I think getting work from any of them should be akin to a Maxwell demon and be almost or completely impossible.
New Question from me and Answer from the first author on the paper:
Q: What is the physical low entropy source that this pulls energy from in a thermal bath without violating the second law? We know that entropy is increasing (with enough time) so that this doesn't violate the 2nd law, but what is the low entropy source? I'm trying to figure out a physical intuition for what it might be. The only thing I can come up with is the energy you're harvesting comes from the physical disconnect you must make for the circuit? But if that is true, then you'll always be limited to that tiny bit of energy.
A: The “useful” work is the charged capacitor. You can remove it from the circuit and go power something. Remember we started with the shorted capacitor. If you measure the charge on the shorted capacitor, you get zero every time. So, not only is the average zero but the variance is zero (variance is trivially related to entropy, and here it is the sum of all the squared values we measure, which were zero). So, the low entropy source is the shorted capacitor. Once it is rewired to a resistive source the capacitor finds itself in a non-equilibrium situation (very far from equilibrium). This is because the variance (entropy) is too low (Boltzmann proved it needs to be kB T C, for an RC circuit and not zero). Increasing the entropy is the driving force that charges the capacitor. The capacitor is trying to get to equilibrium. After a long enough time, it will reach equilibrium (we proved that), and this is why our system satisfies the second law. I hope that helps.
@AdamTreat So it sounds like the work is coming from connecting a capacitor a resistor? Doesn't sound promising in terms of practical applications.
Sounds like a kid trying to game rules.
Throwing techo-babble around to confuse the audience.
Hopefully the world expert in graphene will be at my department’s pickup basketball game Saturday morning so I can pick his brain about this, haha.
My thoughts are that this will work but will produce an inconsequential amount of power. It probably doesn’t violate the laws of thermodynamics as the energy spent in the production of pure, monolayer graphene is being converted very slowly (ie, the device works for a finite time and the graphene must be replaced/re-stretched/etc).
@BenjaminShindel I'm still unclear where the low entropy source is physically embodied in this thermal bath. What precisely is it pulling energy from without violating the second law in a thermal bath.
@JonathanRay "However, we just discovered that our graphene-based devices harvest solar energy as well." -> Inspired by this I assume?
Ok, I just received an email reply from the first author on the new paper. Here are my questions and his answers:
Q: Do you consider your work groundbreaking or having near term commercial impact?
A: Our work is groundbreaking. We are continuing to make device structures, package them, and test them. Our current objective is to boost the output power (it is too low right now for applications). However, we just discovered that our graphene-based devices harvest solar energy as well. So, we are also working on a hybrid energy harvesting device.
Q: What do you say to the layman who looks at what you've done and thinks that you're claiming breaking of laws of thermo?
A: Entropy is similar to the variance of a set of numbers. Imagine a shorted capacitor. If you measure the charge on the shorted capacitor in time you always get zero. So, the numbers give you zero as the average and zero as the variance. Now remove the short and replace it with a linear resistor. The resistor-capacitor circuit must now have a charge variance of kB T C (to match the Johnson-Nyquist result). The average is still zero, but the variance must change from zero to kB T C. This will take some time and the time is a few factors of R C. Finally replace the resistor with a diode. The diode has a non-zero resistance, so the variance will still need to grow from zero to kB T C. However, the diode allows charge to move more easily (less resistance) in one direction than the other (higher resistance). So, as the charge variance grows, it grows asymmetrically. As a result, the capacitor develops an average charge different from zero. Eventually the average zero reaches zero, but in the transient phase it is non-zero. By eventually reaching zero we satisfy the second law.
Anyone have any further or more pointed questions? Leave comments below...
@AdamTreat Interesting, I wonder what the way to think about the entropy here is then. I'm also confused about how presumably the capacitor starts at zero charge, ends up at non-zero charge, then returns to zero charge. What is different about the final state which prevents it from going non-zero again?
@dayoshi they disconnect the circuit? i'm wondering if the energy harvested is going to be equal to the energy needed to disconnect the circuit...
@AdamTreat Yeah it feels like it has to be that, or the energy needed to set it up etc. Obv I'm not a physicist so not sure, but it makes me suspicious that they don't have a clean story about what low entropy source is being used to produce the work.
@AdamTreat probably obvious, but it doesn't have to be equal each cycle, sometimes you get excess, and sometimes the capacitor ends up charged below average, like the "disconnector" has no means to time the right moment.
@33cb I wonder if this won't work for the same reason as Maxwell's demon? The explanation I've heard of that one is that the "demon" needs to have information about the state of system being manipulated, and that implies that the joint demon/box system is actually extremely low entropy.
@AdamTreat I might only have a BEng but I'm quite convinced that's just not how any of that works? As the voltage on the capacitor rises the size of the noise spike necessary for the diode to be polarized in conduction will rise, while the amount of electrons able to leak backwards through it will do the same, and it will reach net zero at a really low voltage.
A previous paper by the same authors that they cite claims that, for a circuit involving their graphite sheet and a resistor:
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.
This parses to me as an energyless heat-pump: the graphite sheet absorbs heat and supplies power which is then dissipated by the resistor, which heats up. This is a straightforward violation of the second law of thermodynamics and you would probably be able to generate infinite energy by attaching this to a thermocouple.
But I might be misunderstanding something.
maybe we can ask for alexey's input https://twitter.com/alexeyguzey/status/1701441107736101200