Consider a system of non-interacting particles. Energy in the states of a single particle are denoted as {E1, E2,...}, then a system with corresponding {n1,n2,...} numbers of particles in the states has energy
E=i=1∑niEi thus the probability for the system to be in state with numbers {n1,n2,...}:
P∝exp(−kT∑i=1niEi)=exp(−kTn1E1)⋅exp(−kT∑i=2niEi) So, the probability to have N particle in the arbitrary chosen first state regardless the state of the rest of the system:
P(n1=N)∝exp(−kTNE1){ni}∑exp(−kT∑i=2niEi)here sum over {ni} means sum over all possible sequences of {n2,n3,...}. This sum is just a numerical factor and does not depend on N, so
P(n1=N)=exp(−0E1/kT)+exp(−1E1/kT)exp(−NE1/kT) where we use the information that for Fermi statistic N can be only 0 or 1. The average number of particles in this state is
⟨N⟩=N=0,1∑NP(n=N)=1+exp(−E/kT)0+exp(−E/kT) or
⟨N⟩=expkTE+11 If there is a chemical potential, the distribution turns to
⟨N⟩=expkTE−μ+11
Comments