A decrease in entropy .. and magnets and solvation

Here is a cool thought experiment I haven't given any thought yet-


Take a beaker and fill it with an appropriate solvent. The solvent is meant to solvate a ferromagnetic specie (any will do, as long as they do not cluster together).

Now drop a magnet into this beaker.


Do the magnetic particles cluster around the magnet? Doesn't this lead to a decrease in entropy of the beaker?

Clearly, if entropy decreases the reaction can't be spontaneous, so the particles will not actually cluster around the magnet. This is Law.

 

Log in or Sign up to hide all adverts.

Hang on, if the ferromagnetic species is solvated, how strong a magnet do you need? What's the Ksp of the solvent/solution?

And even if the magnet is strong enough to attract solvated molecules, why would this decrease entropy in the solvent or the magnet, or anywhere else? You know that magnetic fields do work, right?

 

Log in or Sign up to hide all adverts.

I suppose the magnet could be any strength as long as it attracted at least some of the ferromagnetic particles (perhaps it would be possible to have a magnet strong enough to attract all of the solute). This would lead to a decrease in the density of solute in the solvent, which would certainly lead to lower entropy. Lower entropy because the solute is confined to a region close to the magnet- without the ability to diffuse everywhere in the solvent.

As the solute initially dissolves, it will release/withdraw energy of solvation. This energy can be positive or negative, and the deciding factor is the change in gibbs free energy, I think. There has to be a net change in gibbs free energy for the solute to dissolve.

...

Magnetic fields do work, of course.. a magnet has the potential to bind only a certain amount of solute. Binding its maximal weight of solute, the magnet will have expended all potential work? After this, work must be done detaching the bound solute from the magnet, and only this will allow the magnet to do work again. hmmmmm this is something I will have to explore

So a magnet entering the solution will have the potential of lowering the entropy of that solution by expending its potential for work - by raising the gibbs free energy of the solution. Or something like that.

 

Log in or Sign up to hide all adverts.

Here is another issue- there is no correlation between how much entropy the magnet can decrease and the strength of the magnet.

Follow this trail of consciousness and see what you learn about solvation.

Take a beaker, partitioned so that all the ferromagnetic particles are only on one side. The entropy of this system is smaller than if the partition was removed and the ferromagnetic specie was allowed to diffuse across the entire beaker.

If you drop a magnet of sufficient strength into this solution, all of the ferromagnetic particles will bind to it (as far as I know). This should be true for both of the beakers since the weight of solute is the same. However the entropy reduction will be different in relationship to the volume of the beaker, meaning the same magnet can alter entropy in a way that is completely independent of its strength, size, and properties.



Is it possible that the magnet will bind different amounts of solute depending on the volume of the beaker / density of the solute ?

This doesn't seem impossible since there may be a Ksp between the magnet and the solute, ie the particles bind to the magnet non-permanently and there is certain probability that a bound particle will detach from the magnet. This way density will dictate at what rate particles bind, with the solvent providing a constant value for how quickly magnetic particles detach, with the net weight of bound solute being at an equilibrium with the solute in solution. This doesn't sound right to me though, the idea that a particle bound to a magnet will spontaneously detach.

Please Register or Log in to view the hidden image!

 

Bind in what way? You mean the molecules which are in solution will "stick" to the magnet?

Again, how does a magnetic field applied to the solution alter the entropy, why does it "decrease"?
All I can imagine a magnetic field doing is making the molecules dance around a bit differently.

What if instead of a solution of ferromagnetic "particles" you had a metallic gas? Are these particles suspended in solution, or are they dissolved - there's a major difference. You can extract metallic "particles" from a solution with magnets, extracting dissolved species means lowering their Ksp.

 

A reaction CAN be spontaneous and decrease entropy if it releases energy (google "gibbs free energy"). In this case, you are releasing magnetic potential energy that exists between the ferromagnetic particles and the magnet when they are far apart from each other.

It's very similar to the entropy decrease you get if you drop a high-mass object into a cloud of gas, causing the gas to fall toward the massive object (in that case the energy released would be gravitational potential energy).

 

Yes, binds as in sticks to the magnet. However, if the particles are strongly ligated by the layer of solvation, then this layer will remain intact.

Entropy decreases because the ferromagnetic particles are not free to move around randomly through the solvent. At any time you have molecules concentrating in one place entropy will be decreased. This can happen spontaneously, ie all the air molecules in a box have a certain probability of all being on one side of the box at once - a very low probability. The magnet would be concentrating all magnetic particles on one side of the beaker, decreasing entropy, and I suppose releasing heat with respect to nasor's comment.

What is the difference between dissolved and solvated ? EDIT: okay, this is the difference between a solution and a suspension. I will study the hell out of it tonight.

I see what you mean by ksp. If you have an equilibrium between solvated magnetic particles (ie saturated solution), and then you place a magnet in that solution the magnetic particles will stick to it and the solution will no longer be saturated. This leads to more solute dissolving for the solution to be saturated and equilibrium to be reached.

Yeah, it seems gibbs free energy is an important concept when it comes to this stuff. So if the solution gets warmer, the entropy is allowed to decrease? This makes sense since extra heat will indirectly contribute to entropy.

However, your example is a bit unsettling. We can decrease entropy as well as get energy? Thats like saying we can remove salt from the ocean and actually obtain energy by doing so (as long as the parallel between the gas and solution holds).

 

Start here:

1) dissolve or suspend the particles in solution; solvation energy will be greater than suspension energy.
2) apply magnetic field; the work done separating particles from solution lowers the free energy of solvation or suspension.
3) the entropy of the total system increases, unless "negative work" can be done to stop heat being generated.

....all yours

 

thank you

1) what is suspension energy?
2) the ferromagnetic particles won't necessarily seperate from the solution - they may still contain a layer of solvation. a ferrofluid remains suspended even after it is affected by a magnetic field
3) the heat could be siphoned away to do work - creating entropy outside of the solution

 

Suspension is what happens to particles small enough to diffuse through the solution, suspension energy is the energy the solvent supplies to keep particles suspended, it's way lower than solvation.

 

Thats interesting - if the particles are always suspended then they will always be affected by gravity. Wouldn't it require a continuous flow of energy to keep them suspended?

Okay, how about this- if there is a release of energy when entropy decreases, one could extract solutes from the oceans, decrease entropy and release heat that would drive some kind of engine. Once all the solutes are gone.. you just dump the ones you have extracted back in, they dissolve, and you can pull them out again. Where does the energy come from?

 

I think you've overlooked something. Brownian motion is the random jiggling of liquid molecules, this is what suspends particles which do not go into solution - the particles are in the solid phase but small enough that the random motion energizes them, which is free energy in the solution. Suspended particles are independent of gravity otherwise they wouldn't be suspended....

'sigh' - when energy is released entropy increases. It's physically impossible to "lower" entropy - entropy decreases only in a cycle, like a sort of temporary borrow, which has to be paid back. The cycles can't be loss-free either because the system always uses some. Lowering energy in a system lowers the system's entropy but not total entropy. In other words, it costs energy to lower a system's energy.

Entropy is simply a measurement of energy content, Energy is conserved but a system's entropy isn't, if it was nothing would have variable energy, the universe would be motionless.

 

I think you are right, but I wouldn't say they are independent of gravity since the gravitational field still exists in the beaker - so its still there.

I agree, entropy goes in cycles - if you decrease it you gain the potential of increasing it again. But it is possible to never use this potential energy and thus permanently decrease entropy.

Okay, so we agree that the solute could aggregate on the magnet, thereby increasing heat energy but localizing the solute. Now lets say this released heat energy is utilized by a heat engine to charge a battery. When you charge a battery, doesn't its entropy decrease?

 

Consider liquid water freezing into ice; the entropy of the water decreases as it freezes, and the freezing is an exothermic process, so the water will release energy and warn everything else up as it freezes.

No, because extracting salt from the ocean would be endothermic. If salt coming out of solution was exothermic, then yeah, you could obtain energy from sea water by taking salt of out it.

 

Well, you could get energy from an endothermic process as well - what you really need is a temperature difference. Concentrating salt would be an endothermic process which would provide a heat sink. This cold area could be coupled with a warm area to get work, which can be used to charge a battery which decreases entropy.

 

When you say "charge a battery which decreases entropy", you mean "in the battery", right?

Since a charged battery is in a state of less disorder, it's at a state of lower entropy than a discharged "disordered" one. But you can't get something for nothing - as you note charging a battery means doing work and doing work increases entropy.

As stated already, you can only lower entropy temporarily, and only relative to the environment, which is "everything except the battery".
Entropy doesn't "lower itself" spontaneously; it takes work to lower entropy in a closed system.

 

Yes, in the battery.

The work is supplied by the magnet that is decreasing mechanical entropy of the solution. This entropy decreases, heat is released, heat is converted to work, entropy of the battery decreases. The two reactions are 'coupled together'. If you had an ultra-efficient heat converter then the amount of heat released into solution could be significantly lowered, to the point where it would no longer offset the decrease in mechanical disorder of the ferromagnetic particles.

 

There's a not so "simple" answer too. Lorentz invariance and entropy-volume changes.