04.08.2013 Views

Statistical Mechanics - Physics at Oregon State University

Statistical Mechanics - Physics at Oregon State University

Statistical Mechanics - Physics at Oregon State University

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

128 CHAPTER 6. DENSITY MATRIX FORMALISM.<br />

At this point we encounter an important difference between functions and<br />

oper<strong>at</strong>ors. We cannot interchange the order of oper<strong>at</strong>ors if they do not commute,<br />

and hence we have to write something of the form A 2 ′ = A ′ A+AA ′ ! In general,<br />

an oper<strong>at</strong>or and its deriv<strong>at</strong>ive do not commute. Therefore:<br />

and<br />

∂<br />

∂x τ n (x) =<br />

n<br />

τ m−1 <br />

∂τ<br />

(x) τ<br />

∂x<br />

n−m (x) (6.46)<br />

m=1<br />

∂<br />

∂x eτ(x) = 1<br />

n!<br />

n<br />

n<br />

τ m−1 <br />

∂τ<br />

(x) τ<br />

∂x<br />

n−m (x) (6.47)<br />

m=1<br />

and we cannot say much more. But, in our example we need to take the trace<br />

of this equ<strong>at</strong>ion, and we can use Tr (ABC) = Tr (CAB). Therefore:<br />

Tr ( ∂<br />

∂x eτ(x) ) = 1<br />

n!<br />

which finally leads to<br />

or in other words<br />

n<br />

n<br />

Tr (τ m−1 <br />

∂τ<br />

(x) τ<br />

∂x<br />

n−m (x)) (6.48)<br />

m=1<br />

Tr ( ∂<br />

∂x eτ(x) ) = 1<br />

n!<br />

n<br />

Tr ( ∂<br />

∂x eτ(x) ) = <br />

Tr ( ∂<br />

∂x eτ(x) ) = Tr ( <br />

n<br />

n<br />

n<br />

Tr (τ n−1 <br />

∂τ<br />

(x) ) (6.49)<br />

∂x<br />

m=1<br />

1<br />

n! nTr (τ n−1 (x)<br />

1<br />

(n − 1)! τ n−1 (x)<br />

Tr ( ∂<br />

∂x eτ(x) ) = Tr (e τ(x)<br />

Tr ( ∂ρ<br />

) = Tr (ρ<br />

∂x<br />

and using x = Rnm:<br />

<br />

∂ log(ρ)<br />

Tr (ρ<br />

) = Tr ( ∂ρ<br />

∂Rnm<br />

∂ log(ρ)<br />

∂Rnm<br />

∂x<br />

<br />

∂τ<br />

) (6.50)<br />

∂x<br />

<br />

∂τ<br />

) (6.51)<br />

∂x<br />

<br />

∂τ<br />

) (6.52)<br />

∂x<br />

<br />

) (6.53)<br />

∂Tr ρ<br />

) = ( ) = δnm<br />

∂Rnm<br />

Our final result for the partial deriv<strong>at</strong>ive is therefore<br />

<br />

∂X<br />

= −kB 〈m| log(ρ) |n〉 + (λ − 1)kBδnm<br />

∂Rnm<br />

and setting this equal to zero gives<br />

(6.54)<br />

(6.55)

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!