PDA

View Full Version : Prove that Entropy is Extensive...


Tsar_183
Oct8-11, 06:08 PM
1. The problem statement, all variables and given/known data

Show explicitly that Entropy as defined by the Gibbs Entropy Formula is extensive. That is, for two independent (noninteracting) systems A and B,

S(A,B) = S(A) + S(B)

where S(A,B) is the entropy of A and B considered as part of a larger system.

2. Relevant equations

S = -k \sum pi ln(pi)

3. The attempt at a solution

I honestly have no idea where to start! I tried letting pi = 1/Ω, to obtain,

S = k \sum (1/Ω)ln(Ω), and then tried summing S(A) and S(B) together to obtain S(A,B), but it didn't work out. I also tried just summing up S(A) and S(B) without writing in terms of Ω...didn't work either. I then tried,

S = -k \sum pi ln(pi) ==>
S = k \sum (1/Ω) ln(Ω) ==>
S = k (1/Ω) ln(Ω)\sum 1, \sum 1 = Ω
S = k (1/Ω) ln(Ω)Ω
S = k ln(Ω)
and then I summed up S(A) and S(B) which WORKED,
S(A,B) = k ln(Ω(A))+k ln(Ω(B)) = k ln(Ω(A)Ω(B)) = k ln (Ω(A,B)), but I don't think this argument works. Plus the prof derived the Gibbs Entropy Formula from k ln Ω... so I don't think i'm even on the right track! Any ideas or suggestions? Thanks!!