Hi everyone, this is my first post here but I have been "lurking" in the shadows for a while so I almost feel like I know the regulars
Anyway, to answer the OP I think the limitations of the laws (if limitiation is the right word) and the biggest misconception about the laws is that they are not absolute. Obviously nothing in science is absolute but rather an approximate model, but with Thermodynamics it's more than that. Even though it came about at the same time as classical (non-quantum) physics, it is itself grounded in quantum physics. Because of this it is statistical in nature (ie an average). The laws don't apply on a small scale level because it requires a massive number of particles for this behaviour to become as predictable as it is.
The example my Statistical Mechanics lecturer used to illustrate this was rolling dice. If you roll 2 dice (normal 6 sided) and add the scores together, the most likely score to get would be 7. If you plotted a probability distribution of the average score on each die it would be a smooth curve with a peak at 3.5, but with big probabilities that you could get other scores too. On the other hand, if you rolled a billion dice the probabilitiy distribution curve would look almost like a line at 3.5, dropping off to almost 0 either side of it. The more dice you roll, the more certain you can be that the average will be 3.5.
It's exactly the same with Thermodynamics. The reason why it is so deterministic is because it uses a huge number of particles (much more than a billion). The average behaviour then becomes very very predictable. The less particles that are involved, the more likely it is that you will observe something contrary to the laws of thermodynamics (just like the less dice you roll, the more likely it is you'll get an average other than 3.5).
This message has been edited by happy_atheist, 08-21-2004 11:41 AM
™ Version 4.2