284 4 Thermodynamics and Statistical Physics
=
e−nω/kT
e−ω/kT
1 −e−ω/kT
=
e−nω/kT
1
eω/kT− 1
=e−nω/kT
(
eω/kT− 1
)
Substituten= 10 ,
ω
k
=
8. 625 × 10 −^5
(1. 38 × 10 −^23 / 1. 6 × 10 −^19 )
= 1. 0
P(10,300)= 3. 2 × 10 −^3
In the limitT→0, the staten=0 alone is populated so thatn=10 state is
unpopulated.
In the limitT →∞, probability forn=10 again goes to zero, as higher
states which are numerous, are likely to be populated.
4.60 Consider a collection ofNmolecules of a large number of energy states,
E 1 ,E 2 ,E 3 etc such that there areN 1 molecules in stateE 1 ,N 2 inE 2 and
so on. The nature of energy is immaterial. The number of ways in whichN
molecules can be accommodated in various states is given by
W=
N!
N 1 !N 2 !...
(1)
The underlying idea is that the state of the system would be state ifWis a
maximum.
Taking logs on both sides and applying Stirling’s approximation lnW =
NlnN−N−ΣNilnNi+ΣNi
=NlnN−ΣNilnNi (2)
becauseΣNi=N (3)
ΣNiEi=E (4)
If the system is in a state of maximum thermodynamic probability, the varia-
tion ofWwith respect to change inNiis zero, that is
ΣδNi=0(5)
ΣEiδNi=0(6)
Σ(1+lnNi)δNi=0(7)
We now use the Lagrange method of undetermined multipliers. Multiplying
(5) byαand (6) byβand adding to (7), we get
Σ{(1+lnNi)+α+βEi}δNi=0(8)
Therefore
lnNi+ 1 +α+βEi=0(9)
orNi=Ce−βEi (10)