Statistical Thermodynamics Flashcards
(22 cards)
Number of ways of putting n different balls into n boxes
No. of microstates = n!
Number of ways of putting r different balls into n boxes (r<n)
No. of microstates = n!/(n-r)!
Number of ways of putting r identical balls into n boxes
No. of microstates = n!/r!(n-r)!
Ensemble average
- Take a single system and follow it over a long time; so that the system goes through all possible configurations
- Take a large number if identical but randomly configured systems and average over these at a single snapshot in time (this is the ensemble average)
Micro canonical ensemble
Isolated system with fixed U,V,N
Canonical ensemble
System is in contact with a thermal reservoir. V,N,T fixed for the system
Grand canonical ensemble
System can exchange heat and particles with the reservoir. V,T, mu fixed for the system
Principle of equal equilibrium probability
When a thermally isolated system comes into thermal equilibrium then the state probabilities of any set of mutually accessible states become equal. In a micro canonical ensemble there is equal probability of any state being occupied
Ergodic hypothesis
Given enough time the systems will explore all possible microstates and will spend an equal amount of time in each of them.
Implication: in equilibrium an isolated system will adopt the macro state with the most microstates
Boltzmann equation for entropy
S = kb * ln Omega
Boltzmann distribution function
P(Ei) proportional to exp(-Ei/kbT)
Properties of Boltzmann Distribution Function
Lower T- lower energy states more likely to be occupied
Infinite temperature- all states equally likely
In 3D the number of states between k and k+dk
g(k)dk = 4pik^2dk * V/(2pi)^3
The partition function for indistinguishable atoms: What conditions are required for the approximation of N! to hold?
As long as we can neglect any atoms which have the same quantum numbers. Okay as long as the number of possible states is much larger than the number of atoms
Sackur-Tetrode Equation
S = Nkb [ln(V/lambda^3 *N) +5/2]
Gibbs Paradox (Question)
What if the particles have some property X which we can’t measure (but someone else could). So they appear to be identical but they are not. Would the entropy change when we are able to measure the difference.
Gibbs Paradox (answer)
Yes, entropy depends on our knowledge of the system. The way to interpret the extra entropy is that after mixing A,B to return the system to its original state we would need to do work on the system equal to at least T*DeltaS
Fermions
Particles with half integer spin
Pauli Exclusion Principle
We can only have 1 fermion in any single quantum state
Bosons
Particles with integer spin
Classical limit
At low occupancy the fermi-Dirac, Bose-Einstein and Boltzmann distribution functions all become equal, because there is very little chance of multiple occupancy so the quantum nature of the particles is irrelevant