AMBER Archive (2001)

Subject: Re: Maxwell-Boltzmann distribution of velocities

From: Andreas Svrcek-Seiler (svrci_at_tbi.univie.ac.at)
Date: Fri Oct 26 2001 - 05:46:36 CDT

> According to classical statistics, the Maxwell Boltzmann
> distribution of velocities is applicable to a classical system
> of non-interacting particles (an ideal gas), does anybody know why is it
> applicable
> in the case of atoms in a molecule?

...I *think*
1) because if you write down the hamiltonian (total energy),
you see that the partition function can usually (except if the potential
is velocity-dependent) factorized into a potential an a kinetic term,
the latter looking the same whether particles interact or not.
(that's for averages)
2) if there's an interacton between the particles (e.g. atoms),
velocities must of course be correlated. therefore, after 'throwing the
dice' to get the initial velocities, an equilibration period is needed
before the simulation deserves its name.

hope this helps,
correct me if I'm wrong.

regards
andreas 

-- 
             
            )))))
            (((((
           ( O O )
-------oOOO--(_)--OOOo-----------------------------------------------------
              o        Wolfgang Andreas Svrcek-Seiler  
              o        (godzilla) 
                       svrci_at_tbi.univie.ac.at
      .oooO            Tel.:01-4277-52733 
      (   )   Oooo.    
-------\ (----(   )--------------------------------------------------------
        \_)    ) /
              (_/