Scribd
Upload
31 views3 pages

Ensemble Theory

1. The document discusses different statistical ensembles used to model thermodynamic systems, including the microcanonical ensemble, canonical ensemble, and grand canonical ensemble. 2. The microcanonical ensemble models a system with a fixed number of particles N, volume V, and energy E. The canonical ensemble models a system in contact with a heat reservoir at a fixed temperature T, allowing the system's energy E to fluctuate. 3. The grand canonical ensemble further extends this by allowing both the system's energy E and number of particles N to fluctuate, with their average values corresponding to the thermodynamic quantities. Probability distributions depend on temperature T and chemical potential μ.

Uploaded by

mehdiramezani200
Copyright
© Attribution Non-Commercial (BY-NC)
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
31 views3 pages

Ensemble Theory

1. The document discusses different statistical ensembles used to model thermodynamic systems, including the microcanonical ensemble, canonical ensemble, and grand canonical ensemble. 2. The microcanonical ensemble models a system with a fixed number of particles N, volume V, and energy E. The canonical ensemble models a system in contact with a heat reservoir at a fixed temperature T, allowing the system's energy E to fluctuate. 3. The grand canonical ensemble further extends this by allowing both the system's energy E and number of particles N to fluctuate, with their average values corresponding to the thermodynamic quantities. Probability distributions depend on temperature T and chemical potential μ.

Uploaded by

mehdiramezani200
Copyright
© Attribution Non-Commercial (BY-NC)
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/ 3

Ensemble theory

prepared by Huidong Xu July 12, 2008


A statistical system with given macrostate (N, V, E), at any time t, is equally likely to be in any one of an extremely large number of distinct microstates. As time passes, the system continually, switches from one mircrostate to another. Each microstate is a phase point (qi , pi ) as a representative point in the phase plane. Now, we consider an ensemble of systems. As time passes, every member of the ensemble undergoes a continual change of microstates, i.e., the representative points move along their repective trajectories. We dene a density function (q, p, t) such that, at any time t, the number of representative points in the volume element (d3N qd3N p) around the point (q,p) of the phase space is given by the product (q, p, t)d3N qd3N p. Clearly, the density function sympolizes the manner in which the members of the ensemble are distributed over all possible microstates at dierent instants of time. At equilibrium, an ensemble is said to be stationary if does not depend explicitly on time, i.e., =0 t According to Liouvilles theorem, d = + [, H] = 0. dt t (2) (1)

This means the local density of the representative points, as viewed by an observer moving with a representative point stays constant in time. So for a system at equilibrium, we have
3N

[, H] =
i=1

qi + pi qi pi

=0

(3)

Microcanonical Ensamble

One way to satisfy Eq. 3 is to assum that is independent of the coordinates (q, p), i.e., (q, p) = const. (4) This is the microcanonical ensemble. In this ensemble the macrostate of the the systems is dened through a xed number of particles N , a xed volume V and xed energy E. The basic problem then consists in determining the number (N, V, E) of distinct microstates accessible to the system. From this number, complete thermodynamics of the system could be derived. 1

Canonical ensemble

However, the total energy E is hardly measured and hard to control. A far better alternative is a xed temperature T of the system which is controllable by keeping the system in contact with a heat reservoir. This is the canonical ensemble in which the macrostate of the system is dened through the parameters N, V and T . In this ensemble, the energy E of a system is a variable; in principle, it can take values anywhere between zero and innity. What is the probability that, at any time t, a system in the ensemble is found to be in one of the states characterized by the energy value Er ? We denote this probability by the symbol Pr . There are two ways to nd the dependence of Pr on Er . First, we consider a system A with energy E in a very large heat reservoir A with energy E and E + E = E (0) const. Since the heat reservoir is very large, Er is almost constant and have number of microstates (Er ). The probability of the reservoir assuming that particular energy value Er is proportional to that number of microstates
Pr (Er ) (E (0) Er ).

(5)

The expansion of its logarithm gives:


ln (Er ) =

ln (E (0) ) + const Er . ln E

ln E

(Er E (0) ) + E =E (0)

(6) (7)

= where

;
N,V

(8)

Note that, in equilibrium, = = a/kT . So Pr exp(Er ) Or, Pr = exp(Er ) r exp(Er ) (10) (9)

The second way is to regard the system as a member of a canonical ensemble (N, V, T ), in which an energy E is being shared by N identical systems constituting the ensemble, and studying the statistics of this sharing process. If nr denotes the number of systems which, at any time t, have the energy value Er , then n exp(Er ) (11) Pr = r = N r exp(Er ) Now, the partition function is dened as QN (V, T ) =
r

exp(Er /kT )

(12)

Alternative expression for partition function can be:

QN (V, T ) =
0

eE g(E)dE, 2

(13)

where g(E) is the density of states. For a classical system with Hamiltonian H(q, p) in phase space (q, p), the ensemble average is < f >= f (q, p)(q, p)d3N qd3N p (q, p)d3N qd3N p (14)

where (q, p) denotes the density of the representative points in the phase space, and (q, p) exp (H(q, p)). (15) So we can write the partition functino as QN (V, T ) = = 1 N !h3N VN N !h3N eH(q,p) d3N qd3N p

(16)
N

ep
0

/2mkT

4p2 dp

(17)

Grand canonical ensemble

It comes from the realization that not only the energy of a system bu t the number of particles as well is hardly ever measured in a direct manner. Therefore, we may regard both N and E as variables and identify their expectation values, < N > and < E >, with the correponding thermodynamic quantities. Now we get Pr,s exp {(Nr Es )} (18)

You might also like