Revision of Boltzmann statistics for a finite number of particles
Author:
- 1 January 2009
- journal article
- Published by American Association of Physics Teachers (AAPT) in American Journal of Physics
- Vol. 77 (1) , 48-53
Abstract
The Stirling approximation, ln ( N ! ) ≈ N ln ( N ) − N , is used in the literature to derive the exponential Boltzmann distribution. We generalize the latter for a finite number of particles by applying the more exact Stirling formula and the exact function ln ( N ! ) . A more accurate and analytical formulation of Boltzmann statistics is found in terms of the Lambert W-function. The Lambert-Boltzmann distribution is shown to be a very good approximation to the exact result calculated by numerical inversion of the Digamma-function. For a finite number of particles N the exact distribution yields results that differ from the usual exponential Boltzmann distribution. As an example, the exact Digamma-Boltzmann distribution predicts that the constant-volume heat capacity of an Einstein solid decreases with decreasing N . The exact Digamma-Boltzmann distribution imposes a constraint on the maximum energy of the highest populated state, consistent with the finite total energy of the microcanonical ensemble.Keywords
This publication has 5 references indexed in Scilit:
- Analytical approximations for real values of the Lambert W-functionMathematics and Computers in Simulation, 2000
- The Validity of Stirling's Approximation: A Physical Chemistry ProjectJournal of Chemical Education, 1999
- A Simple Derivation of the Boltzmann DistributionJournal of Chemical Education, 1999
- The Boltzmann DistributionJournal of Chemical Education, 1996
- A New Derivation of Stirling's Approximation to n!The American Mathematical Monthly, 1990