E’ possibile rappresentare lo stato del sistema, passando dall’ Entropia termodinamica, funzione delle variabili estensive passando ad una o più di quelle coniugate, sostanzialmente sfruttando la trasformata di Legendre.

Helmhotz free energy

We want to descrive the system using the variables . If our system needs to have constant temperature, it must be allowed to exchange energy with the environment, and thus is not isolated anymore. A usefull idea is that we can think of our system beeing in contant with an infinite thermal reservoir at fixed temperature, with which it can exchange energy, but not particles or volume.

Let and be the respective entropies. Since the system plus the reservoiur are isolated, the total energy is a constant . To find the equilibrium energies we need to maximize the total entropy:

Since with our hypotesis , we can Taylor expand the reservoir entropy:

thus the equilibrium entropy is

which is the same as minimizing

The themordynamic potential defined as

is called Helmhotz free energy. (The factor is due from conventions).

Since it’s basically a Legendre transform, which is involutive, it contains all the information contained in the entropy.

The infimum is obtained at , so that we can invert and write

which is the familiar

Properties It shares similar properties of the entropy:

  • is convex in the extensive variables ()
  • and concave in Proof linear minus concave is a convex function inf of convex function is convex, concavity follows from the inf definition.

Thermodynamic stability Like the entropy the derivatives of are physical quantities of interest.

from this follows a conforting fact:

since is convex w.r.t. . If this quanty where , we are in trouble…

The Grand potential

We can go further, and use the set of variables . Now we allow our system to also exchange particles with the reservoir. Again with the hypotesis we can approximate

to find the equilibrium number of particles we minize the free energy:

but using the definition of the free energy this is the same as

again we define the Grand potential as

as before

one cool fact: since is linear in the only extensive variable left, , it has the simple form of

an easy consequence of the Euler relation .