Legendre Transformations: arbitrary dimensionality
∂Y
Y = Y ( X 0 , X 1 ,...) Pk =
∂X k
ψ = Y − ∑ Pk X k
Legendre transformation: k
Inverse Legendre transformation:
Y = ψ + ∑ X k Pk
k
One may also have partial Legendre transformations, when the function Y
is transformed only with respect to some of its coordinates.
Legendre transform contain all information in the original fundamental
relation.
Notes
Thermodynamic potentials
By applying Legendre transformations to the fundamental relation in the energy
representation, we can obtain various thermodynamic potentials
~ ~
U = U (S , V , N ...) U = U (T , P, µ ...)
Transformation with respect to S only: Helmholtz free energy F
∂U
U = U (S , V , N ...) T=
∂S
Solve these two equations to eliminate S and U F (T , V , N ,... ) = U − TS =
(express S and U as functions of the rest of the U (S (T , V , N ), V , N ... ) − TS (T , V , N ... )
variables). Then substitute these S and U into
the Legendre transform:
Notes
Thermodynamic potentials
Transformation with respect to V only: Enthalpy H
∂U
U = U (S , V , N ...) −P=
∂V
Solve these two equations to eliminate
V and U (express V and U as function of H ( S , P, N ,...) = U + PV =
the rest of the variables). Then
substitute these V and U into the
U (S , V (P, S , N ...), N ...) + PV (P, S , N ...)
Legendre transform:
Notes
Thermodynamic potentials
Transformation with respect to both S and V: Gibbs free energy G
∂U ∂U
U = U (S , V , N ...) −P= T=
∂V ∂S
Solve these two equations to eliminate S, V and U (express S, V
and U as function of the rest of the variables). Then substitute
these S, V and U into the Legendre transform:
G (P, T , N ,...) = U − TS + PV =
U (S (P, T , N ,...), V (P, T , N ,...), N ...) − TS (P, T , N ,...) + PV (P, T , N ,...)
Notes
How to calculate the Helmholtz free energy
A recipe of calculating the free energy from the fundamental relation in
the energy representation U=U(S,V,N,..) (Legendre transformation):
∂U
1. Calculate T as a function of S, V, N: T (S , V , N ) =
∂S
2. Solve the two equations: U = U (S , V , N ...) and T = T (S , V , N )
in order to express U and S as functions of T, V and N:
U = U (T , V , N ...) S = S (T , V , N ...)
3. Substitute these functions into the Legendre transform for the
Helmholtz free energy:
F (T ,V , N ,...) = U (T , V , N ,...) − TS (T , V , N ,...)
Notes
Example: Helmholtz free energy for van der Waals fluid
The fundamental relation in the entropy representation:
s = R ln(v − b ) + cR ln(u + a / v ) + s0
We can rewrite it in the energy representation:
s − s0 1 a
(v − b ) c −
−
u = exp
cR v
1
−
S − S 0 V c a 2
Eliminating molar quantities: U = N exp − b − N
cRN N V
Notes
Example: Helmholtz free energy for van der Waals fluid
1
−
T (S , V , N ) =
∂U 1 S − S 0 V c
1. T= exp − b
∂S cR cRN N
1
−
2. S − S 0 V c a 2
U = N exp − b − N
cRN N V Solve for U and S
1
−
1 S − S 0 V c
T= exp − b
cR cRN N
a 2 V
U = cNRT − N S = NR ln − b + cNR ln(cRT ) + S 0
V N
Notes
Example: Helmholtz free energy for van der Waals fluid
3. F = U − TS
a 2 V
F = cNRT − N − T NR ln − b + cNR ln (cRT ) + S 0
V N
V a 2
F (T , V , N ) = cNRT (1 − ln (cRT )) − NRT ln − b − N + TS 0
N V
Notes
The minimum principle for Helmholtz free energy
We know that U(S,V,N,…) of a composite system is at minimum with respect to
unconstrained extensive parameters in equilibrium. What about F(T,V,N,..) ?
We will now show that F(T,V,N,…) of the system of interest is also at minimum
with respect to it unconstrained extensive parameters if the system is in thermal
equilibrium with a thermal reservoir at a temperature T.
Therefore, F(T,V,N,…) is ideal for finding equilibrium states of systems of interest
that are in contact with thermal reservoirs. It is simpler than finding the minimum
of U(S,V,N,…) since T = const and thus we have effectively reduced the number
of thermodynamic coordinates by one, also we do not have to worry about the
thermal reservoir. We have traded generality (consider only systems in contact
with a thermal reservoir) for simplicity (reduced the number of coordinates and
eliminated explicit consideration of thermal reservoir ).
Notes
Proof of the minimum principle for Helmholtz free energy
Consider a composite system in thermal contact with a thermal reservoir.
U – internal energy of the system of interest
Ur – internal energy of the reservoir gas
gas
S – entropy of the system of interest
reservoir
Sr – entropy of the reservoir
System of interest
Notes
The internal energy minimum principle
If we know the entropy S of the system + reservoir,
then there is only one point on the line S = const that
S describes an equilibrium state of the composite
system.
min U
U This point corresponds to the maximum of entropy with
respect to the unconstrained variable.
From geometrical considerations and the fact that
dS/dU > 0, the same point is the minimum of internal
energy at S = const
Xi(1)
Therefore, the state of equilibrium of system + reservoir
for a given total entropy is achieved at the minimum of
internal energy U of the system + reservoir.
Notes
Proof of the minimum principle for Helmholtz free energy
We know that the internal energy of the composite system + reservoir is at
minimum in equilibrium:
Stotal
Utotal
d (U + U r ) = 0 d (U + U r ) > 0
2
On the surface of constant entropy: d (S + S r ) = 0 Xi(1)
We can also show that the internal energy of the system of interest itself is at minimum:
d 2 (U + U r ) > 0 d 2U > 0 because d 2U >> d 2U r
Notes
Proof of the minimum principle for Helmholtz free energy
To prove: d 2U >> d 2U r
2 r
∂ U ∂ 2U r Nr 1
d 2U r = ∑ r r
dX r
j dX r
k
and
r r
~ r r ~ r
j ,k ∂X j ∂X k ∂X j ∂X k N N N
∂ 2U r ∂ 2U
Thus r r
<<
∂X j ∂X k ∂X j ∂X k
This condition is true because the reservoir is much larger than the
composite system of interest.
Notes
Proof of the minimum principle for Helmholtz free energy
d (S + S r ) = 0 dS r = − dS
d (U + U r ) = ... + TdS + T r dS r = ... + TdS − T r dS = 0
For arbitrary small dS:
d (U + U r ) = ... + (T − T )dS = 0
r T =Tr
Notes
Proof of the minimum principle for Helmholtz free energy
Since no work is done on the reservoir
(walls with thermal reservoir are rigid): d (U + U r ) = dU + T r dS r = 0
since d (S + S r ) = 0 dU − T r dS = 0
Because T r is constant: (
d U −T rS = 0)
Therefore, the quantity U −T rS has an extremum in equilibrium.
Notes
Proof of the minimum principle for Helmholtz free energy
(
d U −T rS = 0)
One can also show that ( )
d 2 U −T rS > 0 so the extremum is a minimum
We know that T =Tr in our case of the system in contact with
the thermal reservoir. Therefore we can replace Tr with T.
d (U − TS ) = dF = 0
Notes
Utility of the Helmholtz free energy: Example
For example, let us consider a system that
consists of two chambers filled with N1 and gas
gas
N2 moles of ideal gas and separated by a
movable impermeable piston. We need to
find the equilibrium state of the system. reservoir Tr
The only unconstrained extensive variable of the free energy F(T,V,N) is volume
∂F1 ∂F2
dF = 0 = d ( F1 + F2 ) − = 0 = −(P1 − P2 )
∂V1 T ∂V2 T
RT RT r RT r V1 V2
P= = v1 = v2 =
v v1 v2 N1 N 2
Notes
Utility of the Helmholtz free energy: Example
V1 V2
=
N1 N 2
In addition V1 + V2 = Vt where Vt is the total volume of the composite system.
N1 N2
Therefore: V1 = Vt V2 = Vt
N1 + N 2 N1 + N 2
Here we solved two equations with two unknowns (V1 and V2). If we were to minimize
the internal energy, we would have to solve five equations with five unknowns (V1, V2,
S1, S2 and Sr).
Notes
Why is the Helmholtz potential called “free energy”?
The Helmholtz potential F(T,V,N,…) characterizes the maximum work one can extract
from a system at constant temperature (in contact with a thermal reservoir). It is the
energy that is “free” to be converted into work at constant T.
To find maximum work, we need to consider a
reversible process: dS = - dS r
System A→B T=Tr
dU = UB – UA= – δWRWS – δQRHS
δWRWS = – dU – δQRHS
δQRHS δWRWS
δWRWS = – dU – T r dS r
δWRWS = – dU + T r dS Thermal Work sink
δWRWS = – (dU – T dS) reservoir
δWRWS = – dF
Work is equal to the decrease of free energy of the system!
Notes
