How may entropy be reversed?

[shynung]

Most physicists would say that entropy is the only law of physics that cannot be broken/reversed/bent.

It's been mentioned earlier in the thread that refrigerators reduce their local systemic entropy by raising the entropy of their surroundings.

[Brotoro]

Decrease the entropy in the Sun and rebuild to its youthful figure? Sure...given amazing technology. But you won't be able to do it without increasing entropy elsewhere.

[Starwhip]

As far as known physics goes. No.

Indeed.

[PB666]

It's been mentioned earlier in the thread that refrigerators reduce their local systemic entropy by raising the entropy of their surroundings.

Better than that, plants change entropy by intercepting light a 1 billionth of a second before it hits the ground, and that drives the increase in organization and system complexity as the plant grows and as plants and animals evolve. And yet to the photon the time that it traveled between the sun and the plant, or sun and ground is no different, no time passed at all. If the plant had not absorbed the photon, the ground would have and certainly converted that energy into lower frequency radiant energy across a broad spectrum back into space. The net flow of energy is from the center of the universe (invisible to us) to the outer bounds of an expanding universe (also invisible to us) and plants just get in the way of the entropic process, just like black holes. Plants are more or less a like the insert slot of an energetic Pachinko machine.

I wouldn't count on plants or refrigerators rescuing our sun.

[BlueCosmology]

The net flow of energy is from the center of the universe (invisible to us) to the outer bounds of an expanding universe (also invisible to us)

There is neither a center of the universe or an outer bound of the universe. The universe is expanding from all points. Just like how there is no center or outer bound of the surface of an expanding balloon.

[AC656]

Isn't there a fantastically small chance that entropy could actually run in reverse for some indefinite time? After all, isn't entropy at least mostly, or entirely, simply that "disordered" states are more common than "ordered" states, and once you have a huge number of particles, the chances of them progressing from order to disorder becomes, statistically, effectively certain? So the question is, if you waited an extremely long time, through random chance, could entropy reverse itself in a closed system? Like maybe you have to wait 10^100^100^100 years, but eventually, you could observe the entropy in a macroscopic system reverse itself- maybe you might see heat flow backwards (from cold to hot) for a few seconds?

correct but the universe will probably reach maximum entropy killing all life on it before any significant fluctuation takes place. But yes.

Awesome ted talk on this topic I recommend to everyone:

[bonyetty]

I have thought that life could be defined as reverse entropy. Also an interesting something I read once. At the current rate of human reproduction all matter in the universe will be people in 9,000,000,000 years. Also I think that stars fit into our current definition of life. It's a relatively wacky universe with relatively wacky ideas indeed. Great fun to think about though.

[Ralathon]

I have thought that life could be defined as reverse entropy.

Pretty much the exact opposite of reality. Life can exist only by producing more entropy. Entropy would increase slower if there was no life on this planet. So unless some lifeform figures out how to reverse entropy, life is only accelerating heat death.

Also an interesting something I read once. At the current rate of human reproduction all matter in the universe will be people in 9,000,000,000 years.

And 1 E-Coli bacterium will reproduce every 30 minutes or so. Run the maths and you come to the conclusion that the entire earth will be converted into E-Coli after 64.5 hours and after 163.5 hours the entire visible universe will be filled with e-coli.

Needless to say, looking at current birthrates and extrapolating is not a very accurate model.

Also I think that stars fit into our current definition of life.

wat.jpg

[magnemoe]

I have thought that life could be defined as reverse entropy. Also an interesting something I read once. At the current rate of human reproduction all matter in the universe will be people in 9,000,000,000 years. Also I think that stars fit into our current definition of life. It's a relatively wacky universe with relatively wacky ideas indeed. Great fun to think about though.

Plenty of processes who reverse entropy locally, life is the most common, but also heat pumps and probably solar cells do it.

However its locally, an planet is local in this setting.

Doing it on an universe scale pretty much require you to be God or have root access to the matrix as you have to change how the universe works.

Now changing the universe to stop entropy without killing everyone in it would be harder, as in harder than getting root access.

[bonyetty]

Pretty much the exact opposite of reality. Life can exist only by producing more entropy. Entropy would increase slower if there was no life on this planet. So unless some lifeform figures out how to reverse entropy, life is only accelerating heat death.

My understanding is that life is organized chemistry therefore more life more order.

And 1 E-Coli bacterium will reproduce every 30 minutes or so. Run the maths and you come to the conclusion that the entire earth will be converted into E-Coli after 64.5 hours and after 163.5 hours the entire visible universe will be filled with e-coli.

Yep how cool is maths and life. Big numbers are BIG one million seconds is about 12 days an American billion seconds is about 31 years.

Needless to say, looking at current birthrates and extrapolating is not a very accurate model.

I know. I don't beleve every thing I read. It can be fun and interesting discuss things I have read.

wat.jpg

Can't see the picture. I don't know of a widely agreed definition of life. Are viruses alive, are they life. What about Prions or software viruses?

I haven't read how to answer quotes properly in forums.

Edited March 3, 2015 by bonyetty

[K^2]

Isn't there a fantastically small chance that entropy could actually run in reverse for some indefinite time? After all, isn't entropy at least mostly, or entirely, simply that "disordered" states are more common than "ordered" states, and once you have a huge number of particles, the chances of them progressing from order to disorder becomes, statistically, effectively certain? So the question is, if you waited an extremely long time, through random chance, could entropy reverse itself in a closed system? Like maybe you have to wait 10^100^100^100 years, but eventually, you could observe the entropy in a macroscopic system reverse itself- maybe you might see heat flow backwards (from cold to hot) for a few seconds?

Nope. Just not how entropy works. Even in Classical Mechanics things are way more complicated. When you add QM into the mix, it gets really tricky. It's a trivial theorem in Quantum Mechanics that entropy is conserved. (<S> = Tr(ln(ÃÂ)), d<S>/dt = [H,S] = 0.) So we're back to the distinction between fine grain entropy and coarse entropy. And coarse entropy doesn't allow for fluctuations like these.

Edited March 3, 2015 by K^2

[PB666]

Is it possible? Well, first let me ask you what reference frame? Multiverse? Yes. Universe? No.

You see, entropy is always increasing for the whole mechanism, unless acted upon by another mechanism. If each mechanism is a universe, then you can move around entropy as you see fit (assuming you're a Type V civilization). But then you would have a multiversal entropy that's constantly rising. Everything is doomed to die. Kind of sad, really.

I knew someone would bring this up eventually. Its kind of a red herring argument, there is a belief out there that if multiverse actually exist, they could be caused by 'true vacuum' bubbles forming in empty space as isolated singularities, as the singularities break down their own laws of physics form with different flavors of particles, etc. (The multiverse physicist are rather creative in their thinking).

I kind of like the idea of a unique singularity, since it means that we need to focus on our own business before we start trying to reinvent the wheel as a square, triangle, sphere, and Fourier transform of the wheel.

- - - Updated - - -

There is neither a center of the universe or an outer bound of the universe. The universe is expanding from all points. Just like how there is no center or outer bound of the surface of an expanding balloon.

There is no visible landmark that describes the center (that we know of) to an expanding balloon and there is a leading edge to a spherical expansion. r_t = r₀ + f(t) where t is the time standing at the center of the universe, rofl.

There are various estimates that suggest the universe is between 40 billion light years across to 160 billion light years. The comoving estimate of the Universe is 45.7 billion light years across, whereas 78 billion light years is the 2003 estimate (Cornish, Spergel, Starkman, and Komatsu (2003). Phys. Rev. Lett. 92) making r = 39 billion light years across (plus or minus a few angstroms). Since the center is not moving and everything else is then the center of the universe is the place were 39 billion light years has elapsed since all hell broke loose. lol.

Btw there is a restaurant at the end of the universe, you need an infinite improbability drive to get there. I love astrophysics, its so full of BS.

- - - Updated - - -

Nope. Just not how entropy works. Even in Classical Mechanics things are way more complicated. When you add QM into the mix, it gets really tricky. It's a trivial theorem in Quantum Mechanics that entropy is conserved. (<S> = Tr(ln(ÃÂ)), d<S>/dt = [H,S] = 0.) So we're back to the distinction between fine grain entropy and coarse entropy. And coarse entropy doesn't allow for fluctuations like these.

If this is so how could the universe form to begin with. At some point there has to be a huge dS. And if fine grain entropy allows fluctuations then why cannot momentum be transferred through vacuum space to some unsuspecting target in the next mass layer? That conserves course entropy. hah-hah.

[Ralathon]

There is no visible landmark that describes the center (that we know of) to an expanding balloon and there is a leading edge to a spherical expansion. r_t = r₀ + f(t) where t is the time standing at the center of the universe, rofl.

There are various estimates that suggest the universe is between 40 billion light years across to 160 billion light years. The comoving estimate of the Universe is 45.7 billion light years across, whereas 78 billion light years is the 2003 estimate (Cornish, Spergel, Starkman, and Komatsu (2003). Phys. Rev. Lett. 92) making r = 39 billion light years across (plus or minus a few angstroms). Since the center is not moving and everything else is then the center of the universe is the place were 39 billion light years has elapsed since all hell broke loose. lol.

You're fundamentally misunderstanding how the universe works.

The scientists in the article that you referenced are trying to constrain the topology of the universe.They're trying to see if the universe is curved positive (like a sphere), flat (like a plane) or negative (Like a saddle point). Their data constrained the positive curvature to a radius of more than 24 Gpc. This does not mean that there's some kind of border between universe and not!universe once you go further than that.

Look at the expanding balloon example again. Say the balloon has a radius of 1 meter. This does not mean that an ant living on the surface encounters a barrier or an end to balloon if he walks more than 1 meter. If we inflate or deflate the universe the radius increases or decreases. But the ant won't be able to point at any point on the surface and say "It is inflating/deflating from this specific point!", the true center of inflation is in the 3rd dimension after all, something inaccessible to our 2 dimensional ant.

in much the same way our 3d universe does not have a point from which it expands. All we know is that if the universe is curved, it has a radius of at least 24Gpc. There is no special point from which the big bang happened, it happened everywhere at once, it's just that the radius of the universe was 0 at that point.

[K^2]

Fine grain conserves entropy absolutely. Coarse can increase, but it's averaged out, so there aren't any random fluctuations there, either. And S(t = 0) can be anything without violating conservation laws, so long as dS/dt = 0. Same as it is with energy.

Entropy doesn't really measure disorder. It is a measure of chaos, but in the Chaos Theory sort of sense. Picture a classical system in a particular state. I can encode coordinates and velocities of all particles as a single vector. So a state of the system at some time t is just a point. Now, I take a neighborhood around that point, and I watch how all of the states in the neighborhood evolve. The distribution of states at some t2 > t is a measure of entropy. Roughly speaking, the change in entropy corresponds to change in volume of that neighborhood. Of course, if we map all of the points one-to-one, volume is unchanged. (Liouville's Theorem.) On the other hand, if we take sort of a new bounding volume to see how much spread there has been, there can be a volume increase. This is fine grain vs coarse entropy. In mathematically precise statement, for fine grain, we only consider the exact mapping of initial to final states. For coarse entropy, we also consider an infinitesimal neighborhood of each state as part of the new volume.

It's this averaging out in coarse entropy that prevents fluctuations from being a factor. Or, again, if you want to be mathematically precise, because we consider a neighborhood of each point, you cannot end up with coarse entropy being smaller than fine grain entropy. Which means that coarse entropy can increase, but it can never decrease. At best, random fluctuations can keep it from increasing.

[Sean Mirrsen]

Can you boost entropy instead of reversing it? Or is it unalterable in all regards?

[K^2]

Sure. Pretty much anything that "consumes" energy does just that. There are limitations, but all of them are basically just the result of a requirement that total entropy increases, not just local entropy. For example, if you have a hot object and you want to make use of that heat energy, you have to make sure the heat flows from hot object to cold one.

[PB666]

You're fundamentally misunderstanding how the universe works.

The scientists in the article that you referenced are trying to constrain the topology of the universe.They're trying to see if the universe is curved positive (like a sphere), flat (like a plane) or negative (Like a saddle point). Their data constrained the positive curvature to a radius of more than 24 Gpc. This does not mean that there's some kind of border between universe and not!universe once you go further than that.

Look at the expanding balloon example again. Say the balloon has a radius of 1 meter. This does not mean that an ant living on the surface encounters a barrier or an end to balloon if he walks more than 1 meter. If we inflate or deflate the universe the radius increases or decreases. But the ant won't be able to point at any point on the surface and say "It is inflating/deflating from this specific point!", the true center of inflation is in the 3rd dimension after all, something inaccessible to our 2 dimensional ant.

in much the same way our 3d universe does not have a point from which it expands. All we know is that if the universe is curved, it has a radius of at least 24Gpc. There is no special point from which the big bang happened, it happened everywhere at once, it's just that the radius of the universe was 0 at that point.

I take it that my sarcasm went over your head (the plus or minus one Angstrom should have been a clue).

But to be more serious, since the latest measurements of CMBR anisotropy do not indicate the period of inflation and since space-time dimensionality is distorted at its beginning and quantum mechanics and its intrinsic uncertainty ruled its inflation and early expansion, and given that the 2003 paper is an estimate, but then there is still a limit to the universe in that the light travel time since the end of inflation cannot be exceeded then it is pointless to talk about universe that is larger than the light travel time at the margin of inflation at the end of inflation (should we ever know how big and long that was). The only thing you need is a central reference point and you know how large it is, but that estimate (2003) is probably wrong since the size or age of the inflationary bubble is not known and not knowable, so the answer is more or less academic. Imagine you are the first electron anti-proton pair, hit by an high-energy gamma and thrown into a spiral following a beam of light that occurred at the moment inflation ended. All are traveling radially outward (except the matter engaged in a timeless dance). That light beam never ages for its inflation ended Plank's time ago and those dancing pair, the big bang happened a moment in the past and there are no galaxies in its Universe. I can't even contemplate what its C_BR might look like.

If there is a multiverse, and if someday other instances of that multiverse are visible, then your statement has meaning, otherwise the constrained Universe (that is everything) to matter and energy. Now if someday it is determined that the Higgs field or any of its variant pre-existed the known universe and/or propogates faster than the speed of light then that would be a new limit. But for most of the literature Universe is the Visible Universe and that is the most limited constraint. This is the reason I was poking fun at the astrophysicist.

And BTW, our known Universe is Euclidean flat in every direction, and it is expanding about the same rate in every direction, and while certain physicist proclaim that in a post inflationary universe symmetry breaking is not allowed, we live in a material universe and not an anti-material universe. The constraints of your universe, paradoxically lead to speculation about a form of matter and physics that lie outside our visible universe, which makes for interesting humor.

[PB666]

Fine grain conserves entropy absolutely. Coarse can increase, but it's averaged out, so there aren't any random fluctuations there, either. And S(t = 0) can be anything without violating conservation laws, so long as dS/dt = 0. Same as it is with energy.

Hah, what a dodge, the universe is not an experiment. It exists, lol. So if space-time merge then basically entropy can be violated. We assume that there is not another external perspective on this and in this perspective entropy may not be conserved.

Entropy doesn't really measure disorder. It is a measure of chaos, but in the Chaos Theory sort of sense. Picture a classical system in a particular state. I can encode coordinates and velocities of all particles as a single vector. So a state of the system at some time t is just a point.

That is the point, but the point had no momentum in any recognizable form (either collectively or discrete), thus it could be anywhere. When inflation ends and the forces unravel then momentum appears, but can you directly connect that momentum back to a reaction at the center of inflation? no, its connected to something that is meaningless in our space-time.

momentum = Plancks constant/ wavelength

so what is the wavelength of a point when space and time are merged. There is no change of distance/change of time and no period.

and what are the composite wavelengths in and inflationary bubble where neither matter or light exist.

quantum mechanics solves this problem by inserting gradient and imaginary units. It can only be a definable gradient when expansion begins, prior to such how can any gradient be defined.

Now, I take a neighborhood around that point, and I watch how all of the states in the neighborhood evolve. The distribution of states at some t2 > t is a measure of entropy. Roughly speaking, the change in entropy corresponds to change in volume of that neighborhood. Of course, if we map all of the points one-to-one, volume is unchanged. (Liouville's Theorem.) On the other hand, if we take sort of a new bounding volume to see how much spread there has been, there can be a volume increase. This is fine grain vs coarse entropy. In mathematically precise statement, for fine grain, we only consider the exact mapping of initial to final states. For coarse entropy, we also consider an infinitesimal neighborhood of each state as part of the new volume.

It's this averaging out in coarse entropy that prevents fluctuations from being a factor. Or, again, if you want to be mathematically precise, because we consider a neighborhood of each point, you cannot end up with coarse entropy being smaller than fine grain entropy. Which means that coarse entropy can increase, but it can never decrease. At best, random fluctuations can keep it from increasing.

The problem is that energy permeates the universe, both in vacuum space and in occupied space, and it is not clear how much energy there is (ergo dark energy and dark matter) so that at all levels entropy are not possible to detect. That was my complaint in the previous thread, I reserve the right to decide whether or not there are 'pokes' in the quantum physics until we have a better understanding of all these forces. We know that the lowest energy instance of the Higgs permeates space, some suspect there may be 3 other variants, and no-one is quite certain if Higgs expanded with expansion or pre-expansion.

There is a much deeper problem that this, if varied states is a character of expansion (which I completely agree), then the problem arises that the progression statistically of the universe makes alot of sense now in our comoving universe all the way back to hydrogen, helium, smallest traces of lithium and energetic hv, but it makes almost no sense at the transition from inflation to end of the opaque phase.

[Sean Mirrsen]

Sure. Pretty much anything that "consumes" energy does just that. There are limitations, but all of them are basically just the result of a requirement that total entropy increases, not just local entropy. For example, if you have a hot object and you want to make use of that heat energy, you have to make sure the heat flows from hot object to cold one.

Can you cool off the hot object without transfering the heat to a cool object? That example is pretty much what I had in mind when I asked - an entropic cooling system, i.e. for spacecraft with plentiful energy but no capacity to radiate heat (i.e. in low solar orbit). Nevermind practical, would something like it be even possible?

[Mr Shifty]

Can you cool off the hot object without transfering the heat to a cool object? That example is pretty much what I had in mind when I asked - an entropic cooling system, i.e. for spacecraft with plentiful energy but no capacity to radiate heat (i.e. in low solar orbit). Nevermind practical, would something like it be even possible?

Prosaically, if it's a fluid you can lower the object's pressure (i.e. make it do work); its average temperature will decrease without transferring heat anywhere. Or you could melt ice, which uses up energy without raising temperature. More esoterically, you could maybe cause large nuclei to fuse, using up heat energy for atomic binding.

Edited March 4, 2015 by Mr Shifty
removed apostrophe from possessive its

[BlueCosmology]

I take it that my sarcasm went over your head (the plus or minus one Angstrom should have been a clue).

But to be more serious, since the latest measurements of CMBR anisotropy do not indicate the period of inflation and since space-time dimensionality is distorted at its beginning and quantum mechanics and its intrinsic uncertainty ruled its inflation and early expansion, and given that the 2003 paper is an estimate, but then there is still a limit to the universe in that the light travel time since the end of inflation cannot be exceeded then it is pointless to talk about universe that is larger than the light travel time at the margin of inflation at the end of inflation (should we ever know how big and long that was). The only thing you need is a central reference point and you know how large it is, but that estimate (2003) is probably wrong since the size or age of the inflationary bubble is not known and not knowable

Again, the universe does not have a center. The 3D spatial part of the universe we live on sits on the surface of 4D spacetime. Just like the 2D part of a balloon sits on the surface of the 3D volume. The estimates you're talking about are estimates of the observable universe.

http://spaceplace.nasa.gov/review/podcasts/transcripts/070523_universe.html

The universe does NOT have a center. Inflation was not just some expansion outwards of things in space.

[Technical Ben]

Would it not be more accurate to say "to all recent measurements the universe does not have a centre".

Or perhaps more specifically, as in the earth has no "centre" to stand on it's surface (taking into account 2d space along the surface of a 3d spheroid), the universe has no "centre" to sit in. However, while the earth has a centre at it's core (taking into account 3d space), the universe does not have a centre (taking into account 4d space?) as far as we can see so far from measurements.

[Bill Phil]

Can you boost entropy instead of reversing it? Or is it unalterable in all regards?

Yes.....

Life.

[BlueCosmology]

Would it not be more accurate to say "to all recent measurements the universe does not have a centre".

Or perhaps more specifically, as in the earth has no "centre" to stand on it's surface (taking into account 2d space along the surface of a 3d spheroid), the universe has no "centre" to sit in. However, while the earth has a centre at it's core (taking into account 3d space), the universe does not have a centre (taking into account 4d space?) as far as we can see so far from measurements.

It would definitely be accurate to say that, however I'd definitely feel that is not more accurate, but considerably less accurate. Not only do all current measurements indicate the universe does not have a center, but our most well tested theoretical models for space also predict it to be, e.g. general relativity / Friedmann metric. For an analogy, if you measure the position of some objects on a desk it's certainly accurate to say that "to all recent measurements objects left on their own stay still". However, that isn't more accurate then the models that govern it.

[PB666]

Again, the universe does not have a center. The 3D spatial part of the universe we live on sits on the surface of 4D spacetime. Just like the 2D part of a balloon sits on the surface of the 3D volume. The estimates you're talking about are estimates of the observable universe.

http://spaceplace.nasa.gov/review/podcasts/transcripts/070523_universe.html

The universe does NOT have a center. Inflation was not just some expansion outwards of things in space.

This point of view and its based on a incomplete view of the universe. The perspective we see is based on a conic section of 3.55' off a radial from the center to our position in space that ends before the edge of the material universe along that radial. Give the estimated size and our age of the universe. The isoquant we see is 0.25% or less of all possible isoquants and thus we are traveling away from most points in our universe very close to the speed of light. It is not possible to see the center or edge from this perspective, does not imply that neither exist. The CMB you see, as isotropic as it appears does not represent the afterglow of the entire universe as some would imply, just the comoving space close to the where our galaxies matter finally materialized. There is an assumption that any point in space would have the same perspective as our perspective, but that is just an assumption. I love astrophysics, it is so full of BS.