Ending global warming with Maxwell's demon

[MrFrety]

As a start for discussion I'd like to make the following bold claim:

Both the problems of enough usable energy and global warming can be resolved by implementing Maxwell's demon on a large scale.

To justify this bold statement somewhat, I'd like to point to a (relatively) recent successful (partial) experimental implementation of a Maxwell-demon:
https://phys.org/news/2016-02-physicists-photonic-maxwell-demon.html


Scenario for large scale implementation:

- Two giant neighboring enclosed containers of atmosphere

- Sorting gasparticles by velocity (velocities higher than average in, say, right capsule) (=Maxwell's Daemon)
(ideas to do so: Wien filter, Opening window of one container at highest temperature of the day and the other at lowest)

- Temperature difference can be used to do work, while cooling the gas
(Sorting doesn't have to be perfect, but the more we sort, the more we can violate the second "law" of thermodynamics, which is just a probability statement assuming we have no control whatsoever over the microstates of the gas.)

 

[Dale]

Here is a link to the original paper:
https://arxiv.org/abs/1510.02164
Note, you have dramatically misunderstood their paper. Their work does not in any way “disable the second law of thermodynamics”

A Maxwell’s demon is anything that uses some sort of measurement to achieve a reduction in entropy or to do work. In this case the amount of work done is related to the amount of information measured. The process of performing the measurements involves an increase in the entropy of the environment. The second law still holds with a Maxwell’s demon

Note: we do not discuss perpetual motion machines here. We can talk about the published Maxwell’s demon, which is not a PMM, but not the rest of your suggestion

 

[MrFrety]

A quote from the author on the linked phys.org page:
"Often we have more information available than thermodynamics supposes," Dahlsten said, explaining that things are normally not fully random and have a degree of predictability. "We can then use demon set-ups such as this one to extract work, making use of that information. Similarly, we can use extra information to reduce work costs of, for example, cooling systems. Personally I think that sort of technology will have a real impact on meeting the energy challenge facing the world."> In this case the amount of work done is related to the amount of information measured.
I agree.
But I disagree on the popular notion that there is a meaningful physical law linking the energy consumption of your sorting device to the maximum amount of energy that can be extracted from the system you're sorting. You can make your Maxwell demon in principle "arbitrarily" energy efficient imo...
For example, the energy required to open a window has nothing to do with the internal kinetic energy of the gas in the room.

> The second law still holds with a Maxwell’s demon.
On this I obviously disagree.
I think, the old counter argument was that a Maxwell demon can't be built...

 

[Dale]

I think, the old counter argument was that a Maxwell demon can't be built...

Certainly not. There have been several others built. The counter argument is about the entropy content in the information itself. I can look up other papers later, but tonight is too late.

This paper is not the first Maxwell’s demon constructed, it is the first one using light.

 

[MrFrety]

Certainly not. There have been several others built. The counter argument is about the entropy content in the information itself. I can look up other papers later, but tonight is too late.

I think what you refer to is what I'd refer to as the "new argument"...
But that's just a matter of naming conventions, of course...

 

[MrFrety]

Note: we do not discuss perpetual motion machines here. We can talk about the published Maxwell’s demon, which is not a PMM, but not the rest of your suggestion

Of course, you'd still never have a PMM... My claim would only mean that you can "cheat" the second law of thermodynamics to some degree but not perfectly... I'd always expect some entropy to be left, which increases again... The increasing of entropy, if your state is far from the equilibrium, is very drastic, too, because the equilibrium state is extremely much more likely... I'm thinking about this type of graph, which almost becomes a delta function for real world gas systems, where high Entropy means a high probability of the state...

 

[MrFrety]

> PPM

By the way, theoretically (!) - not in practice - a PPM would be completely trivial from the axioms of classical physics:

You simply look at a particle with constant momentum. It should move indefinetly.

But that's not what people were interested in back then, of course. They wanted usable systems like steam engines and there the second law of thermodynamics and friction spoil the fun...

 

[russ_watters]

> PPM

By the way, theoretically (!) - not in practice - a PPM would be completely trivial from the axioms of classical physics:

You simply look at a particle with constant momentum. It should move indefinetly.

But that's not what people were interested in back then, of course.

PMM is pretty much a term of art, and that isn't what it means. It explicitly refers to a device that violates one or more laws of thermodynamics:
https://en.m.wikipedia.org/wiki/Perpetual_motion

 

[Vanadium 50]

IBTL

My claim would only mean that you can "cheat" the second law of thermodynamics to some degree

No.
Just no.

 

[Dale]

Here is a different group developing a Maxwell’s demon. I particularly like this paper since it is so clear about the thermodynamics.

https://arxiv.org/abs/1507.00530

I disagree on the popular notion that there is a meaningful physical law linking the energy consumption of your sorting device to the maximum amount of energy that can be extracted from the system you're sorting

Pekola, the author of the above paper, showed exactly how information can be used as a physical fuel to drive cooling. In many ways information is a physical quantity, particularly in the context of thermodynamics. It has entropy and requires energy and work. Wishing it away doesn’t change the physics.

My claim would only mean that you can "cheat" the second law of thermodynamics to some degree but not perfectly

That claim is exactly a PMM of the second kind. We can not discuss it here, and real world Maxwell’s demons do not make that possible.