next up previous
Next: Entropy, Energy and Free Up: Ensembles and Thermodynamic Potentials Previous: Ensembles and Thermodynamic Potentials

Aside: Quiz Solutions

1.
For some system the dependence of the probability of micro-state $\Prob(E)$ on energy was obtained for different temperatures:
\begin{figure*}
 \InputIfFileExists{prob.pstex_t}{}{}
 \end{figure*}
Check all statements that apply:
(a)
T1>T2>T3
(b)
T3>T2>T1
(c)
T2>T1
(d)
T3>0
(e)
T3<0

Solution:
Since $\Prob(E)\sim\exp(-E/kT)$, the log of probability depends on E linearly:

\begin{displaymath}
\ln\Prob(E) = \mathit{const}- \frac{E}{kT}
 \end{displaymath}

  • If T>0 this is a decreasing function, so T1>0, T2>0, T3<0
  • The greater is T, the less steep is the line (what happens if $T=\infty$?), so T1>T2>T3

2.
Prove that $\sqrt{\left\langle (x- \overline{x})^2\right\rangle} /\overline{x} \sim
 1/\sqrt{V}$. What happens with fluctuations in the thermodynamic limit?
Solution:
A is an extensive value , X is an intensive value. Therefore at $V\to\infty$ the scaling is $A\sim
 V$, $\partial A/\partial X\sim V$, $\partial^2 A/\partial X^2\sim
 V$,...

We obtain:

\begin{displaymath}
\left\langle (x- \overline{x})^2\right\rangle = \left\langle...
 ...ght\rangle^2 \sim V, \quad
 \left\langle x\right\rangle\sim V
 \end{displaymath}

and

\begin{displaymath}
\frac{\sqrt{\left\langle (x-
 \overline{x})\right\rangle}}{\...
 ...le x\right\rangle}\sim
 \frac{\sqrt V}{V} = \frac{1}{\sqrt V}
 \end{displaymath}



© 1997 Boris Veytsman and Michael Kotelyanskii
Tue Sep 9 22:39:08 EDT 1997