Thanks to visit codestin.com
Credit goes to www.scribd.com

Scribd
Upload
41 views2 pages

University of Central Florida - Department of Physics

This document provides an overview of the key principles of statistical physics that were covered in Lecture 1 of the University of Central Florida's Physics 5524 - Statistical Physics course in Spring 2016. The 3 main points covered are: 1) Statistical distributions arise when a subsystem interacts with the rest of the system over a long period of time, causing it to occupy every possible state multiple times and approach a uniform probability distribution. 2) Statistical independence allows subsystems that weakly interact to be treated as independent, with their combined distribution being the product of the individual distributions. 3) Liouville's theorem states that the probability distribution function in phase space is conserved over time for an isolated system, demonstrating that the distribution is

Uploaded by

Firoz Khan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
41 views2 pages

University of Central Florida - Department of Physics

This document provides an overview of the key principles of statistical physics that were covered in Lecture 1 of the University of Central Florida's Physics 5524 - Statistical Physics course in Spring 2016. The 3 main points covered are: 1) Statistical distributions arise when a subsystem interacts with the rest of the system over a long period of time, causing it to occupy every possible state multiple times and approach a uniform probability distribution. 2) Statistical independence allows subsystems that weakly interact to be treated as independent, with their combined distribution being the product of the individual distributions. 3) Liouville's theorem states that the probability distribution function in phase space is conserved over time for an isolated system, demonstrating that the distribution is

Uploaded by

Firoz Khan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

University of Central Florida Department of Physics

Physics 5524 Statistical Physics


Spring 2016

Lecture 1, January 12
The fundamental principles of statistical physics
Statistical distribution.
Subsystem interacts with the rest of the system. During a sufficiently long time, the subsystem will
be at every state many times. When times T goes to infinity, the part of the time during which the
system was in a given part of the volume of the phase space is finite:

lim

One can also introduce the density of the probability , so


,, , ,,

,, , ,,
- probability to find system in the given volume of the phase space,
,, , ,,


,
.
The statistical distribution does not depend on the initial distribution neither of the whole system
nor of its parts, since the system goes through all states many times.

This averaging let us to avoid looking for time evolution of the function
average in time: f lim
.

.
p, q and taking its

For large systems the probabilistic nature of the statistics is not a problem, since physical quantities
of the macroscopic bodies practically do not change with time, after time long enough after the
initial time. In this sense the prediction of the statistical physics is deterministic.
Statistical independence.
Large subsystems for a not too long time can be regarded as quasi-independent, since the number
of bulk particles is much larger than the number of surface particles that interact with the rest of
the system. However, after sufficiently long-time this weak interaction will lead to statistical
equilibrium. Thus subsystems that weakly interact can be regarded as statistically independent
the state of one subsystem does not affect the statistical probabilities for the states of the
other sybsystems,
.Then, f f
f f . Since for the system with N small parts f
f , f f f f f (we used f f ~ ), one gets

Similar result holds for one homogeneous system, divided on N parts.

Liouvilles theorem.
Consider a large number of identical subsystems that are represented by a point in the phase
space that moves with time. At initial time they occupy points A , A , . At any time these points
will have the same distribution p, q as at t=0. The motion of the points can be regarded as a

last equality follows from the equations of motion), one gets

0, or

steady-flow gas,

div v

0, and since div v

0 (the

Liouvilles theorem:

The significance of energy.


Since the distribution function is conserved, it is an integral of motion. Since log
log
log , the log of the distribution function as an additive integral of motion. Since there are only 7
additive integrals of motion, one can write: log

E p, q
P p, q
M p, q . The
integrals of motion the energy, momentum and angular momentum completely define the
statistical distribution.
is normalization constant for , and
subsystems, can be obtained from seven integrals of motions.
p, q

constant

E p, q

E P p, q

P M p, q

, ,

are the same for all


the

microcanonical

distribution (conserved E, P, M). This distribution is valid only when at long times the trajectory of
the closed system passes arbitrarily close to the points defined by the arguments of the deltafunction (ergodic hypothesis). In general it is not true.
Momentum and angular momentum depend on the motion of the system as a whole. Therefore,
,

An integral of motion, the energy,


completely defines the statistical distribution.
,
,
the microcanonical distribution (conserved , ,

).

When the system is in partial equilibrium, the parameters , , of the subsystems relax to the
unique values for the system, when the equilibrium is reached. In the partial equilibria, the parts
of the system can be described by macroscopic quantities, which do not change for long time.

You might also like