Thermodynamic potential of entropy, analogous to the free energy
A thermodynamic free entropy is an entropic thermodynamic potential analogous to the free energy . Also known as a Massieu, Planck, or Massieu–Planck potentials (or functions), or (rarely) free information. In statistical mechanics , free entropies frequently appear as the logarithm of a partition function . The Onsager reciprocal relations in particular, are developed in terms of entropic potentials. In mathematics , free entropy means something quite different: it is a generalization of entropy defined in the subject of free probability .
A free entropy is generated by a Legendre transformation of the entropy. The different potentials correspond to different constraints to which the system may be subjected.
The most common examples are:
Name
Function
Alt. function
Natural variables
Entropy
S
=
1
T
U
+
P
T
V
−
∑
i
=
1
s
μ
i
T
N
i
{\displaystyle S={\frac {1}{T}}U+{\frac {P}{T}}V-\sum _{i=1}^{s}{\frac {\mu _{i}}{T}}N_{i}\,}
U
,
V
,
{
N
i
}
{\displaystyle ~~~~~U,V,\{N_{i}\}\,}
Massieu potential \ Helmholtz free entropy
Φ
=
S
−
1
T
U
{\displaystyle \Phi =S-{\frac {1}{T}}U}
=
−
A
T
{\displaystyle =-{\frac {A}{T}}}
1
T
,
V
,
{
N
i
}
{\displaystyle ~~~~~{\frac {1}{T}},V,\{N_{i}\}\,}
Planck potential \ Gibbs free entropy
Ξ
=
Φ
−
P
T
V
{\displaystyle \Xi =\Phi -{\frac {P}{T}}V}
=
−
G
T
{\displaystyle =-{\frac {G}{T}}}
1
T
,
P
T
,
{
N
i
}
{\displaystyle ~~~~~{\frac {1}{T}},{\frac {P}{T}},\{N_{i}\}\,}
where
Note that the use of the terms "Massieu" and "Planck" for explicit Massieu-Planck potentials are somewhat obscure and ambiguous. In particular "Planck potential" has alternative meanings. The most standard notation for an entropic potential is
ψ
{\displaystyle \psi }
, used by both Planck and Schrödinger . (Note that Gibbs used
ψ
{\displaystyle \psi }
to denote the free energy.) Free entropies where invented by French engineer François Massieu in 1869, and actually predate Gibbs's free energy (1875).
Dependence of the potentials on the natural variables [ edit ]
S
=
S
(
U
,
V
,
{
N
i
}
)
{\displaystyle S=S(U,V,\{N_{i}\})}
By the definition of a total differential,
d
S
=
∂
S
∂
U
d
U
+
∂
S
∂
V
d
V
+
∑
i
=
1
s
∂
S
∂
N
i
d
N
i
.
{\displaystyle dS={\frac {\partial S}{\partial U}}dU+{\frac {\partial S}{\partial V}}dV+\sum _{i=1}^{s}{\frac {\partial S}{\partial N_{i}}}dN_{i}.}
From the equations of state ,
d
S
=
1
T
d
U
+
P
T
d
V
+
∑
i
=
1
s
(
−
μ
i
T
)
d
N
i
.
{\displaystyle dS={\frac {1}{T}}dU+{\frac {P}{T}}dV+\sum _{i=1}^{s}\left(-{\frac {\mu _{i}}{T}}\right)dN_{i}.}
The differentials in the above equation are all of extensive variables , so they may be integrated to yield
S
=
U
T
+
P
V
T
+
∑
i
=
1
s
(
−
μ
i
N
T
)
+
constant
.
{\displaystyle S={\frac {U}{T}}+{\frac {PV}{T}}+\sum _{i=1}^{s}\left(-{\frac {\mu _{i}N}{T}}\right)+{\textrm {constant}}.}
Massieu potential / Helmholtz free entropy[ edit ]
Φ
=
S
−
U
T
{\displaystyle \Phi =S-{\frac {U}{T}}}
Φ
=
U
T
+
P
V
T
+
∑
i
=
1
s
(
−
μ
i
N
T
)
−
U
T
{\displaystyle \Phi ={\frac {U}{T}}+{\frac {PV}{T}}+\sum _{i=1}^{s}\left(-{\frac {\mu _{i}N}{T}}\right)-{\frac {U}{T}}}
Φ
=
P
V
T
+
∑
i
=
1
s
(
−
μ
i
N
T
)
{\displaystyle \Phi ={\frac {PV}{T}}+\sum _{i=1}^{s}\left(-{\frac {\mu _{i}N}{T}}\right)}
Starting over at the definition of
Φ
{\displaystyle \Phi }
and taking the total differential, we have via a Legendre transform (and the chain rule )
d
Φ
=
d
S
−
1
T
d
U
−
U
d
1
T
,
{\displaystyle d\Phi =dS-{\frac {1}{T}}dU-Ud{\frac {1}{T}},}
d
Φ
=
1
T
d
U
+
P
T
d
V
+
∑
i
=
1
s
(
−
μ
i
T
)
d
N
i
−
1
T
d
U
−
U
d
1
T
,
{\displaystyle d\Phi ={\frac {1}{T}}dU+{\frac {P}{T}}dV+\sum _{i=1}^{s}\left(-{\frac {\mu _{i}}{T}}\right)dN_{i}-{\frac {1}{T}}dU-Ud{\frac {1}{T}},}
d
Φ
=
−
U
d
1
T
+
P
T
d
V
+
∑
i
=
1
s
(
−
μ
i
T
)
d
N
i
.
{\displaystyle d\Phi =-Ud{\frac {1}{T}}+{\frac {P}{T}}dV+\sum _{i=1}^{s}\left(-{\frac {\mu _{i}}{T}}\right)dN_{i}.}
The above differentials are not all of extensive variables, so the equation may not be directly integrated. From
d
Φ
{\displaystyle d\Phi }
we see that
Φ
=
Φ
(
1
T
,
V
,
{
N
i
}
)
.
{\displaystyle \Phi =\Phi ({\frac {1}{T}},V,\{N_{i}\}).}
If reciprocal variables are not desired,[ 3] : 222
d
Φ
=
d
S
−
T
d
U
−
U
d
T
T
2
,
{\displaystyle d\Phi =dS-{\frac {TdU-UdT}{T^{2}}},}
d
Φ
=
d
S
−
1
T
d
U
+
U
T
2
d
T
,
{\displaystyle d\Phi =dS-{\frac {1}{T}}dU+{\frac {U}{T^{2}}}dT,}
d
Φ
=
1
T
d
U
+
P
T
d
V
+
∑
i
=
1
s
(
−
μ
i
T
)
d
N
i
−
1
T
d
U
+
U
T
2
d
T
,
{\displaystyle d\Phi ={\frac {1}{T}}dU+{\frac {P}{T}}dV+\sum _{i=1}^{s}\left(-{\frac {\mu _{i}}{T}}\right)dN_{i}-{\frac {1}{T}}dU+{\frac {U}{T^{2}}}dT,}
d
Φ
=
U
T
2
d
T
+
P
T
d
V
+
∑
i
=
1
s
(
−
μ
i
T
)
d
N
i
,
{\displaystyle d\Phi ={\frac {U}{T^{2}}}dT+{\frac {P}{T}}dV+\sum _{i=1}^{s}\left(-{\frac {\mu _{i}}{T}}\right)dN_{i},}
Φ
=
Φ
(
T
,
V
,
{
N
i
}
)
.
{\displaystyle \Phi =\Phi (T,V,\{N_{i}\}).}
Planck potential / Gibbs free entropy[ edit ]
Ξ
=
Φ
−
P
V
T
{\displaystyle \Xi =\Phi -{\frac {PV}{T}}}
Ξ
=
P
V
T
+
∑
i
=
1
s
(
−
μ
i
N
T
)
−
P
V
T
{\displaystyle \Xi ={\frac {PV}{T}}+\sum _{i=1}^{s}\left(-{\frac {\mu _{i}N}{T}}\right)-{\frac {PV}{T}}}
Ξ
=
∑
i
=
1
s
(
−
μ
i
N
T
)
{\displaystyle \Xi =\sum _{i=1}^{s}\left(-{\frac {\mu _{i}N}{T}}\right)}
Starting over at the definition of
Ξ
{\displaystyle \Xi }
and taking the total differential, we have via a Legendre transform (and the chain rule )
d
Ξ
=
d
Φ
−
P
T
d
V
−
V
d
P
T
{\displaystyle d\Xi =d\Phi -{\frac {P}{T}}dV-Vd{\frac {P}{T}}}
d
Ξ
=
−
U
d
2
T
+
P
T
d
V
+
∑
i
=
1
s
(
−
μ
i
T
)
d
N
i
−
P
T
d
V
−
V
d
P
T
{\displaystyle d\Xi =-Ud{\frac {2}{T}}+{\frac {P}{T}}dV+\sum _{i=1}^{s}\left(-{\frac {\mu _{i}}{T}}\right)dN_{i}-{\frac {P}{T}}dV-Vd{\frac {P}{T}}}
d
Ξ
=
−
U
d
1
T
−
V
d
P
T
+
∑
i
=
1
s
(
−
μ
i
T
)
d
N
i
.
{\displaystyle d\Xi =-Ud{\frac {1}{T}}-Vd{\frac {P}{T}}+\sum _{i=1}^{s}\left(-{\frac {\mu _{i}}{T}}\right)dN_{i}.}
The above differentials are not all of extensive variables, so the equation may not be directly integrated. From
d
Ξ
{\displaystyle d\Xi }
we see that
Ξ
=
Ξ
(
1
T
,
P
T
,
{
N
i
}
)
.
{\displaystyle \Xi =\Xi \left({\frac {1}{T}},{\frac {P}{T}},\{N_{i}\}\right).}
If reciprocal variables are not desired,[ 3] : 222
d
Ξ
=
d
Φ
−
T
(
P
d
V
+
V
d
P
)
−
P
V
d
T
T
2
,
{\displaystyle d\Xi =d\Phi -{\frac {T(PdV+VdP)-PVdT}{T^{2}}},}
d
Ξ
=
d
Φ
−
P
T
d
V
−
V
T
d
P
+
P
V
T
2
d
T
,
{\displaystyle d\Xi =d\Phi -{\frac {P}{T}}dV-{\frac {V}{T}}dP+{\frac {PV}{T^{2}}}dT,}
d
Ξ
=
U
T
2
d
T
+
P
T
d
V
+
∑
i
=
1
s
(
−
μ
i
T
)
d
N
i
−
P
T
d
V
−
V
T
d
P
+
P
V
T
2
d
T
,
{\displaystyle d\Xi ={\frac {U}{T^{2}}}dT+{\frac {P}{T}}dV+\sum _{i=1}^{s}\left(-{\frac {\mu _{i}}{T}}\right)dN_{i}-{\frac {P}{T}}dV-{\frac {V}{T}}dP+{\frac {PV}{T^{2}}}dT,}
d
Ξ
=
U
+
P
V
T
2
d
T
−
V
T
d
P
+
∑
i
=
1
s
(
−
μ
i
T
)
d
N
i
,
{\displaystyle d\Xi ={\frac {U+PV}{T^{2}}}dT-{\frac {V}{T}}dP+\sum _{i=1}^{s}\left(-{\frac {\mu _{i}}{T}}\right)dN_{i},}
Ξ
=
Ξ
(
T
,
P
,
{
N
i
}
)
.
{\displaystyle \Xi =\Xi (T,P,\{N_{i}\}).}
Massieu, M.F. (1869). "Compt. Rend". 69 (858): 1057.