2nd law - in a closed system, entropy increases.
I
concur.
Corollary - in an open system, the 2nd law does not
apply.
Like some of the previous posters I am going to have to
admit to it being a while since I studied thermodynamics. So take what follows
with whatever amount of salt you wish, but I don't believe the second law
ceases to function simply because a system is open. It just no longer prohibits
a descrease in entropy for the system under consideration (and hence is not a
valid argument against evolution).
I seem to recall that entropy inreasing
in a closed system is but one of at least two equivalent formulations of the 2nd
law. I also seem to remember that one of the consequences of the 2nd law has to
do with it requiring an expenditure of energy to move heat against a thermal
gradient. I.e. you can't get air conditioning for free. (I came up with a
"brilliant" plan to do just that when I was in Jr. High. Would have worked as
well as all other perpetual motion machines!)
My initial reaction, like
somebody's above, was that this would violate the 2nd law. Perhaps more
analysis is needed about whether the expenditure of energy is actually enough to
satisfy the 2nd law. I do recall that finding the flaw in perpetual motion
machines (which is what violation of either the 1st or 2nd law leads to) can be
subtle.
A couple of other observations from my increasingly vague
recollection:
-- The 2nd law as stated above is from classical
thermodynamics and uses a defintion of entropy based on absolute temperature and
energy transfer for a reversible process.
-- Statistical mechanics (aka
statistical thermodynamics) uses a different definition of entropy, which can be
viewed as a measure of randomness. Unlike classical thermodynamics, statistical
mechanics does not forbid a decrease in entropy in a closed system. But
for a system of any significant size (in terms of the number of particles) it is
very unlikely. (As its name would suggest, statistical mechanics deals in
probabilities as opposed the absolutes of classical thermodynamics.) [ Reply to This | Parent | # ]
| 
