The second law of thermodynamics sucks

#1

The second law of thermodynamics has been warped from a observation to an axiom. Physicists want to prove it at all costs. But they can’t trick me.

Here is a short intro to the 3 definitions and justifications for it.

The statistical argument assumes ergodicity (all microstates are equally probably over time). This is wild! Why should it be true?

The informational argument is… well I don’t understand it fully. The general idea, that you can never gain more information about a system, only lose it, does seem pretty convincing. Here is an interesting paper on it.

However! This type of entropy is not well defined in an universe with infinite energy, or one in which energy is not conserved. Our best guess (and it’s all pretty tenuous) is that the universe is expanding, and perhaps increasingly so. This expansion adds gravitational potential energy to the universe by moving them apart and removes energy from the universe by lowering the frequency of traveling light waves. But these (and other effects) need not cancel each other out. You could conceivably generate work from this constantly increasing grav. potential. (Here is an only slightly related interesting page on energy conservation in GR energy is not conserved in a general relativity universe)

So why do physicists stand by the second law if it isn’t even well defined? I think because it makes sense and is intuitive. That’s hardly a reason! It shouldn’t be a part of the physics bible.

Arguments?

#2

In our Thermo class everything was super hand wavy and approximations were everywhere. Which led me to believe that that’s how statistical/thermal physicists operate. Which who knows if that is true cause it was Liz McGrath. But it seems like they feel really really good about ergodicity.

Since we don’t have a unified theory of physics different types of physics are often operating under much different assumptions from others. And unless you’re like a cosmological thermal physicist or something if those exist you’re probably not worrying about general relativity.

This type of thing is what annoys me about physics. A lot of assuming things to make things work.

“It makes sense and is intuitive” + no hard experimental refutation is like gold for physicists. They love that shit.

#3

Yeah I don’t see why we can assume (approximate) ergodicity. I think the idea is that sufficiently complex and chaotic dynamical systems should be (are often? is this an argument about cardinality?) ergodic. Brownian motion is chaotic so the application to thermodynamics makes sense. But on larger scales the laws of the universe are very structured, not chaotic. And there are chaotic systems which are even periodic.

I might need to understand ergodic theory to see why it should or shouldn’t apply to the universe at large. There are theorems about when it applies.

Since we don’t have a unified theory of physics different types of physics are often operating under much different assumptions from others. And unless you’re like a cosmological thermal physicist or something if those exist you’re probably not worrying about general relativity.

Yeah, but then the second law shouldn’t be applied to such universal scales. Probably I am railing against popular physics rather than physicists themselves… but then there is this quote

" The law that entropy always increases holds, I think, the supreme position among the laws of Nature. If someone points out to you that your pet theory of the universe is in disagreement with Maxwell’s equations – then so much the worse for Maxwell’s equations. If it is found to be contradicted by observation – well, these experimentalists do bungle things sometimes. But if your theory is found to be against the second law of thermodynamics I can give you no hope; there is nothing for it but to collapse in deepest humiliation."

— Sir Arthur Stanley Eddington, The Nature of the Physical World (1927)

“It makes sense and is intuitive” + no hard experimental refutation is like gold for physicists. They love that shit.

I guess what else can you do?

#4

Yeah physicists like to be romantic and all-important about the stuff they believe in because many of them believe things like the 2nd law have been decreed by god or the universe or whatever. I don’t really know much of anything about ergodicity/ergodic theory.

#5

I’m glad you agree :D. Next up is Thomas!

Mathias and Will think I’m an idiot because even though they don’t understand the definition (nor do I…), the 2nd law is law. They may be beyond hope.

#6

sheep. they’ve consumed the thermodynamic kool-aid. or would that kill them? I’m not totally up on my cult details.

#7

You need to revisit the meaning of “law” in science. I’m not sure what you mean when you say that it has changed from an observation to an axiom. It is and has always been a pattern that we noticed. It cannot be proven in the same way (and for the same reason) that the rule of big numbers cannot be proven. It is obviously false, and yet almost always true.

Whatever it is that is causing the expansion of the universe has energy (dark energy?) When that influence increases the gravitational energy of galaxies, etc, its own energy decreases. This may seem arbitrary, but this is actually how we define all non-kinetic forms of energy. Total energy goes down when something goes up hill? Not anymore! now it stays constant because we introduce a term called GPE that is equal to whatever it needs to be in order for the total energy to remain constant.

Thought: I might be talking out of my ass because along with potential energies being described as above, they are also related to integrals of forces over distances (work). The expansion of the universe is not being caused by a force and thus no work is being done on galaxies as they move apart from each other. (Alan? Thoughts?)

And finally, I will make a Stephen Hawking argument/explanation of the second law:
Time as we know it is defined by the second law. We experience time as our bodies organize atoms and molecules and proteins in our brains to construct thoughts. This process would be meaningless in a world where things become spontaneously ordered. Thus, the order in which we experience things must be the order in which other things are becoming naturally disordered, because in that direction, the brain is becoming more ordered, and thus can remember the past. In a world where things become naturally ordered, life is unique in its ability to become naturally disordered, though the brain for this reason can only remember the future, and not the past. We, of course, live in both of these worlds. Time neither flows forwards or backwards, but we think we are moving along it because we can only remember in one direction.

To say that it should be removed from the physics bible seems insane from a teacher’s standpoint because it is such a salient and dependable (if not fundamental, from your point of view) part of our universe, which it is the physicist’s job to study.

1 Like
#8

The bit about the 2nd law determining the direction we experience time flowing was probably my favorite part of A Brief History of Time.

In regard to the expansion of the universe, I just want to throw in an alternative to dark energy, which is the possibility that the graviton has a very small (but non-zero) mass, which would result in the strength of gravity weakening over very large scales, and could explain what we observe as the acceleration of the expansion of the universe. This may be appealing in that it could eliminate the need for introducing a whole new weird type of energy into our picture of the universe, and I think it helps in the energy conservation picture Thomas was talking about above, but not totally 100% on that.

#9

I’m referring to quotes such as:

" The law that entropy always increases holds, I think, the supreme position among the laws of Nature. If someone points out to you that your pet theory of the universe is in disagreement with Maxwell’s equations – then so much the worse for Maxwell’s equations. If it is found to be contradicted by observation – well, these experimentalists do bungle things sometimes. But if your theory is found to be against the second law of thermodynamics I can give you no hope; there is nothing for it but to collapse in deepest humiliation."
— Sir Arthur Stanley Eddington, The Nature of the Physical World (1927)

The law of large numbers can be proven. It fact, it’s a mathematical theorem. But you’re right, as with other statistical properties like entropy, we can’t use it to definitely predict an outcome. We might prove the law of entropy in the same way, and I think that’s the goal of the information theoretic argument.

There is no reason to suspect this pool of dark energy will ever deplete. In which case, the energy in the universe is infinite, and so, I think, entropy is not well-defined.

Yeah entropy as the arrow of time is interesting. It is the irreversible process. I’m not sure how that fits in here. And I don’t really understand the connection to human cognition. It was unclear when I read that bit from Stephen Hawking as well.

Agreed, let’s not remove it from curriculum, as we don’t Newton’s laws of motion, even though we know they don’t actually hold. If fact, I don’t have any particular prescriptions based on my wild raving… Who knows what to do with this all. Except maybe open our minds :).

Yeah that’s very interesting. I should look into it. It would certainly complicated my perpetual motion machine that exploits what is essentially a constant force applied to distance objects… But perhaps you could construct another. Does the mass of objects shrink as they produce gravitrons? I suppose not. So we need to harvest the energy generated by the emission of these massive waves…

#10

Did I tell you that Dr. V doesn’t think your perpetual motion machine would work? Basically because two atoms bound together cannot be affected by the expansion of the universe because they are vibrating within a certain distance of each other, and the slow (read: SLOW) expansion between them would not cause them to ever actually move further apart.

I’m not sure about the details, but he felt very strongly about it, and his PHD was cosmology-focused. I’ve been trying to get him and some other Excel people to get on here, but I don’t think he has.

@alan, I thought gravity working differently on long scales was ruled out? but I thought that mostly based on this, so I don’t have strong convictions about it.

1 Like
#11

True, but two atoms a 1bn light years away will get pulled apart fast(er). If you had a big rope 2bn light years away, the rope would be under tension. I can’t imagine the exact mechanism, but this constant (increasing, even) tension can produce work.

I think there is actually the opposite problem, you would need a rope with strong enough atomic bonds to actually resist this tension. But the biggest concern is that you would need to grow the rope indefinitely (using matter and thus energy).

Edit:

On further thought, I’m not sure this would actually be a problem. If the tension is strong enough to rip apart atoms, it is certainly enough to generate work. You could just make your string shorter.

#12

Two points, one on you and the other at the edge of the observable universe, will rapidly begin moving away from each-other and will soon be moving away from each-other at greater than the speed of light, then exiting the observable universe from the perspective of one-another. If you place a third point between the two of them, then the new point will see both the “neighbors” moving away at great speed, but neither leaving what is technically observable. If you place a fourth point in between two of those, everything else will be moving away from it, so on and so forth – but this quickly becomes absurd when you think about it as a solid rope.

This is ultimately because objects bound by a local gravitational influence do not expand. This is a pretty core premise of a lot of cosmology, and why any cosmology (where the expansion of the universe is relevant) usually starts by measuring the smallest distances at 500 megaparsecs, wherein you can treat galaxies like point sources of light. At that distance you qualify all of space as homogeneous (because it is beyond the point where space is spotty, and then beyond the point where space is fractal) and you can treat the expansion of the universe as happening everywhere. In this case your rope is made of galaxies separated from each-other by 500 megaparsecs each.

If this rope worked, you could also deploy a rope in space by first sending out a few billion meters of rope and if it’s creating space-energy then I’d imagine it would find higher energy states by pulling out more. Haven’t thought that one through very much but it’s a fun thought that space would suck the rope into the abyss like a big spaghetti.

Wouldn’t very large objects just inherently heat up for no reason?

But that’s not the topic of the conversation. 2nd law of thermodynamics is definitely weird, but if you think it’s wrong we can take bets on different physical bodies. I’ll bet on entropy increasing, you can bet against it. I’ll put down my life savings.

Ergodicity more or less tells us that the likelihood of finding a particular type of state is more or less baked into its multiplicity. There are many more configurations of gas particles in a room where they’re just sort of spread apart evenly than there are where they’re all packed into a corner as tight as you can get them. And I don’t think anybody is telling you “if you measure a system at T=t, then at T=t+dt where dt is almost no time at all, then the state of the system will be entirely random and every atom might have fallen to the floor”. It’s more like, if you measure the state of a room at two periods of time that are greater than the amount of time it might take for the atoms to shuffle around ( a few seconds for a room at 70f, maybe) then you’ve really got no idea what exact system it evolved into. A lot of physics is super hand-wavy like that because the math gets super unsolvable obnoxiously fast.

Interested in disagreements. My source for the thermo stuff is only a couple classes, but I took a lot of independent studies in cosmology with a professor who spent a great deal of his professional life working as or with cosmologists. The universe at very large scales doesn’t behave like it does locally, and building a rope 500 megaparsecs long is just all kinds of broken.

#13

It may be absurd in that it is wild, but not that it is wrong. I don’t see the conclusion you are trying to make with this. Maybe it’s just a note.

This is my proposal.

I’m not sure about this. Possibly so. The reason would the expansion of the universe. As you noted, it would need to be very large. It’s easy to make a long string than a spherical body with that diameter.

I don’t believe in a cosmological ergodic hypothesis at all. I think it is the least convincing of the popular arguments for the 2nd law. It is a reductionist attempt to apply traditional thermodynamics to the macro scale. Not all macro behavior can be reduced to Brownian motion.

#14

You wouldn’t need a large object, then. Every object that is under a gravitational influence (or a physical bond?) would be under this universal tension, and would then be generating energy as heat. So then the universe would, as a function of its matter, be eternally generating energy as a non-stop universe-wide perpetual motion machine. What is special about a billion light-year long rope that couldn’t be emulated by the sum of all the matter in the universe being everywhere?

You also low-key ignored the part where cosmological theory applies at scales of hundreds of megaparsecs.

#15

It’s not clear to me that tension necessarily generates heat. Heat is motion, and balanced forces don’t cause motion. I’d like to note that even if this effect increases produces heat, it doesn’t mean a given objects net temperature is increasing. Anyway, I wouldn’t have a problem with this. I don’t believe in the 2nd law after all!

It’s difficult to harness work from all matter in the universe everywhere. It’s not suitable for my perpetual motion machine. The OP talks about the more general consequence of the universe expanding: indefinitely increasing gravitational potential.

I said my rope would be long! (2bn light years is ~600 megaparsecs). I don’t see your point here.

#16

That last part is fair, I may not have explained it properly – a solid object at 500 megaparsecs doesn’t exist in any cosmological theory. At that distance space is homogeneous, and all theory assumes that. Introducing a non-homogeneous object at that scale invalidates the theory we’re familiar that explains anything there and leaves us with unexplored ground. It’s kind of like taking the small angle approximation and trying to apply it to large angles – no one ever said it applied to both cases.

The generation of heat from tension is more related to the idea that if a rope is spontaneously under tension, then either the force of the tensions is enough to overcome the bonds of the rope or the energy will have to diffuse in some way. Energy that exits a system can be summed up as “waste heat”. It could show up in many ways, maybe future astronomers measuring the sky will have to account for the cosmic background radiation and the giant-fuckin-rope background radiation.

#17

It seems to invalidate the 2nd law!

Force need not produce energy. Think of a stationary object in a gravitational field.

Also, just a note, energy doesn’t need to diffuse from a system. Energy as matter is an obvious example. Another is an object with potential energy. How will it lose this energy? You might say black body radiation is inevitable, but this only converts heat energy to the emission of light. Even it an object cools to 0K it will still have potential energy. Also, black body radiation is caused by the movement of charged particles (protons and electrons). Neutrons alone do not have black body radiation, so if you had a ball of them there would be no black body radiation. You wouldn’t even lose thermodynamic energy! (There are some practical problems with building such a ball: neutrons decay quickly. I’m leaving out particle decay in general because I don’t understand it. It probably does change this picture.)