# The Naked Scientists Forum

### Author Topic: How does entropy work?  (Read 1960 times)

#### eternity

• Jr. Member
• Posts: 27
##### How does entropy work?
« on: 03/04/2014 07:24:13 »
Can someone please help me to understand this concept becasue it seem counter intuitive to me (ie - entropy with regard to the future state of the universe).

A dictionary definition is this: lack of order or predictability; gradual decline into disorder.

The problem as I understand it (and this is where i think I need correcting) is that, if all objects and elements break down into their constituent parts (atoms etc), there will be chaos?
How is this possible?
Surely if everything is equal then eventually everything will be evenly spaced and have no effect on anything else?. This will be absolutely the opposite of chaos wont it?
You could argue that the current state of things are more chaotic couldn't you? I mean, isn't the basis for quantum science that it is unpredictable?

gradual decline into disorder - If everything is placed around an infinite space without having any effect on any other part, how is that disorder? There is order in the fact that atoms (or whatever we are talking about) are all predictably floating around because we know they cant/wont interact.

And since we dont know the future of the universe anyway (for example an expanding univers), how can we predict entropy as its final state?

« Last Edit: 07/04/2014 08:32:59 by chris »

#### eternity

• Jr. Member
• Posts: 27
##### Re: Entrophy
« Reply #1 on: 03/04/2014 07:25:15 »
thank you

#### evan_au

• Neilep Level Member
• Posts: 4130
• Thanked: 249 times
##### Re: Entropy
« Reply #2 on: 03/04/2014 09:58:51 »
Entropy can be measured in several different ways:
• An isolated region with hot and cold areas will, over time, converge to a more uniform, medium temperature. ie the temperature is initially non-random, but becomes more random.
• You can generate power from the differences between hot and cold areas. You can't generate power when everything is at a medium temperature.
• An isolated region with high and low pressures will, over time, tend to converge to a more uniform, medium pressure.
• An isolated region with mountains and valleys will, over time, converge to a more uniform, medium height.
• A magnet will, over time, lose its magnetism, ie the the magnetic field will become more random.
• A radio signal will, as it propagates, lose signal strength and become more random.
However, something which is perfectly "evenly spaced" is not really random - this arrangement is unlikely to occur by chance (unless there is an energy advantage, such as gravity which tends to pull a planet into a ball with uniform height).

Quantum theory states that nothing can be perfectly still, even as it approaches absolute zero. So there is a fundamental limit to how evenly spaced matter can become.

We see the universe today with large differences in temperature and composition. It is expected that these differences will tend to even-up over time, although in an expanding universe, these differences may take a long time to become undetectable.

See: http://en.wikipedia.org/wiki/Introduction_to_entropy#Origins_and_uses

#### chiralSPO

• Global Moderator
• Neilep Level Member
• Posts: 1879
• Thanked: 145 times
##### Re: Entropy
« Reply #3 on: 03/04/2014 15:51:04 »
Increasing entropy does not make everything equal, just indistinguishable.

Imagine a black and white tv. It has some large number of pixels, each of which can be either on (white) or off (black). Imagine that 50% of the pixels are on, and 50% are off. The most "ordered" way to arrange these black and white pixels is into two halves: one grouping of white and one grouping of black. There are many different ways that this could be, but there are so many more ways that it could be more jumbled.

For a tv with 1366x768 pixels, there are 21049088 possible images. Most of them would look like "static" or "white noise." If a tv were to cycle through these different possible images, never repeating one, it would essentially look like the image is not changing except for the extremely rare instances in which the pixels form some arrangement we see as "ordered." It would then, almost certainly, change to another image that is totally "disordered." This has nothing to do with anything other than statistical probabilities.

Randomly cycling through the different possibilities is not a perfect analogy to the real world, in which the past and present states play a role in determining the next state. But there are randomizing forces (not Newtonian forces), and as the state of the universe evolves (not Darwinian evolution), it approaches this state of homogeneous disorder.

You can also think about flipping a coin 1000 times. It is possible to get 500 heads and 500 tails (with a fair coin), but it is more likely that it would be 501 of one and 499 of the other. Even if it did come out exactly evenly, it is highly unlikely that one would flip 500 heads and then 500 tails, or that it would alternate HTHTHTHTHT.... or that there would be any sort of discernible pattern. It is no more likely to flip HHTHTTHTHTT... than it is to flip any other ordering, we just wouldn't recognize that outcome as "special" in any way, just like HTHTTTHTHHHTT... These last two cases are not identical, but we see them as having similar levels of disorder.

#### eternity

• Jr. Member
• Posts: 27
##### Re: Entropy
« Reply #4 on: 04/04/2014 09:50:17 »
Thank you very much both for replying.

I thin this term 'homogenous disorder' is what Im having a problem with. By its own definition, 'homogenous' would suggest that it isnt disordered, even though it is not in a specific state at any given time -it is more predictable than the current state which is defined (by default) as 'ordered'.

I give up, but thanks for trying  :)

#### The Naked Scientists Forum

##### Re: Entropy
« Reply #4 on: 04/04/2014 09:50:17 »