Average number of particles/subsystems in a state

1 reply · 3K views · last reply
devd
Messages
47
Reaction score
1

Homework Statement


A system in thermal equilibrium at temperature T consists of a large number
mimetex.gif
of subsystems, each of which can exist only in two states of energy
mimetex.gif
and
mimetex.gif
, where
Mathematical expression: E_2 - E_1 = ε > 0, stating that the energy of state 2 exceeds that of state 1 by a positive amount ε.
. In the expressions that follow, k is the Boltzmann constant.

For a system at temperature T, the average number of subsystems in the state of energy
mimetex.gif
is given by

  1. 2$.gif
  2. Formula for the average number of subsystems in the lower-energy state: N_0 divided by (1 + e^(-epsilon/kT)).
  3. Handwritten expression N_0 e^(-epsilon/kT), a Boltzmann-type factor with N_0 multiplied by an exponential of negative epsilon over kT.
  4. Handwritten fraction: N_0 divided by (1 + e^(epsilon/kT)), the average number of subsystems in the lower-energy state for a two-state system.
  5. Handwritten expression N_0 e^{-ε/kT}, the average number of subsystems in the lower-energy state, with N_0 over e raised to minus epsilon over kT.

Homework Equations


Probability of a system to be in a system-microstate of total energy ##E_R##,
##P_R = \frac{e^{-\beta E_R}} {\sum_{R} e^{-\beta E_R}} ##

The Attempt at a Solution


We have the constraint ##\sum_r n_r= N_0 ## and ## \sum_r n_r \epsilon_r = E_R##
Where, r labels the single particle states.
Therefore, the average number of particles in the sth 1 particle state,
##\langle n_s\rangle = \frac{\sum_R n_s e^{-\beta (\sum_r n_r \epsilon_r)}}{\sum_{R} e^{-\beta (\sum_r \epsilon_r)}}##

To proceed one needs the nature of the particles.
For example,
## \langle n_s\rangle = \frac{1}{e^{\beta \epsilon_s} -1}## for Photons
## \langle n_s\rangle = \frac{1}{e^{(\beta \epsilon_s -\mu)}+1}## for FD statistics etc.

How do i proceed without further info? The question seems to conflate states of the total system with the subsystem states. I think the question is problematic and ambiguous. Anyhow, the supplied 'correct' answer is option (B).
 
The given options only makes sense if by 'system' they mean ensemble and by 'subsystem' they mean members of the ensemble.
Then, ##P(E_1)= \cfrac{e^{-\beta E_1}}{e^{-\beta E_1}+e^{-\beta E_2}}##
Again, ##P(E_1)= \cfrac{N_1}{N_0}. ## Therefore, ##N_1 = N_0\left(\cfrac{e^{-\beta E_1}}{e^{-\beta E_1}+e^{-\beta E_2}}\right)##
And hence, ##N_1 = N_0\left(\cfrac{1}{1+e^{-\beta \epsilon}}\right)## , where ## E_2-E_1= \epsilon## .

But, if they're talking about a single system, then the options don't make sense to me. But, this question appeared in the GRE, so they aren't likely to make such errors. So, what am i missing?