1549380323-Statistical Mechanics Theory and Molecular Simulation

(jair2018) #1
Energy fluctuations 143

which measures the width of the energy distribution, i.e. the root-mean-square devia-
tion ofH(x) from its average value. The quantity under the square root can also be
expressed as


〈(H(x)−〈H(x)〉)^2 〉=〈

(


H^2 (x)− 2 H(x)〈H(x)〉+〈H(x)〉^2

)



=〈H^2 (x)〉− 2 〈H(x)〉〈H(x)〉+〈H(x)〉^2

=〈H^2 (x)〉−〈H(x)〉^2. (4.4.2)

The first term in the last line of eqn. (4.4.2) is, by definition, given by


〈H^2 (x)〉=

CN



dxH^2 (x)e−βH(x)
CN


dx e−βH(x)

=


1


Q(N,V,β)

∂^2


∂β^2

Q(N,V,β). (4.4.3)

Now, consider the quantity


∂^2
∂β^2

lnQ(N,V,β) =


∂β

[


1


Q(N,V,β)

∂Q(N,V,β)
∂β

]


=−


1


Q^2 (N,V,β)

[


∂Q(N,V,β)
∂β

] 2


+


1


Q(N,V,β)

∂^2 Q(N,V,β)
∂β^2

. (4.4.4)


The first term in this expression is just the square of eqn. (4.3.19) or〈H(x)〉^2 , while
the second term is the average〈H^2 (x)〉. Thus, we see that


∂^2
∂β^2

lnQ(N,V,β) =−〈H(x)〉^2 +〈H^2 (x)〉= (∆E)^2. (4.4.5)

However, from eqn. (4.3.27),


∂^2
∂β^2

lnQ(N,V,β) =kT^2 CV= (∆E)^2. (4.4.6)

Thus, the variance in the energy is directly related to the heat capacity at constant
volume. If we now consider the energy fluctuationsrelativeto the total energy, ∆E/E,
we find
∆E
E


=



kT^2 CV
E

. (4.4.7)


SinceCVis an extensive quantity,CV∼N. The same is true for the energy,E∼N, as
it is also extensive. Therefore, according to eqn. (4.4.7), the relative energy fluctuations
should behave as
∆E
E




N


N



1



N


. (4.4.8)


In the thermodynamic limit, whenN→ ∞, the relative energy fluctuations tend to
zero. For very large systems, the magnitude of ∆Erelative to the total average energy

Free download pdf