Show that for a van der Waals gas, ((delC_V)/(delV))_T = 0, where C_V = ((delU)/(delT))_V?

enter image source here

1 Answer
Jan 22, 2017

By definition, the constant-volume heat capacity was:

C_V = ((delU)/(delT))_V," "" "bb((1))

where U is the internal energy, and T and V are temperature and volume, respectively, as defined in the ideal gas law and other gas laws.

To show that for a van der Waals gas, the constant-volume heat capacity does not change due to a change in volume at a constant temperature, i.e. ((delC_V)/(delV))_T = 0, we first write out the van der Waals (vdW) equation of state:

bb(P = (RT)/(barV - b) - a/(barV^2) = (nRT)/(V - nb) - (an^2)/(V^2)), " "" "bb((2))

where barV = V/n is the molar volume, R is the universal gas constant, a and b are the vdW constants for intermolecular forces of attraction and excluded volume (respectively), and P is the pressure.

This means we'll need to have an expression for dU that includes P somehow, or even a partial derivative.

We can use the Maxwell relation for the internal energy in a closed system undergoing a reversible process:

dU = TdS - PdV,

If we divide by (delV)_T for the entire equation, we get:

((delU)/(delV))_T = T((delS)/(delV))_T - Pcancel(((delV)/(delV))_T)^(1) " "" "bb((3))

This will be the main expression we'll work with. Note that T and V are the natural variables of the Helmholtz free energy, A, whose Maxwell relation is:

dcolor(red)(A) = -Scolor(red)(dT) - Pcolor(red)(dV)

Using the fact that the Helmholtz free energy is a state function (just like G, H, S, etc), there is a cyclic relationship we can use:

-((delS)/(delV))_T = -((delP)/(delT))_V

Plugging into bb((3)) gives:

((delU)/(delV))_T = T((delP)/(delT))_V - P " "" "bb((4))

which is something we can relate back to the vdW equation of state, using ((delP)/(delT))_V. We're almost there.

Given that U is a state function (and this is very important!), the order of partial differentiation does not matter:

(del)/(delV)[((delU)/(delT))_V]_T = (del)/(delT)[((delU)/(delV))_T]_V

or

((del^2U)/(delVdelT))_(V,T) = ((del^2U)/(delTdelV))_(T,V)

This will become relevant if we take the partial derivative of C_V with respect to V at constant T and plug in the definition of C_V from bb((1)):

((delC_V)/(delV))_T = (del)/(delV)[C_V]_T

= (del)/(delV)[((delU)/(delT))_V]_T

= (del)/(delT)[((delU)/(delV))_T]_V" "" "bb((5))

Now consider the right side of bb((5)) and plug in bb((4)):

((delC_V)/(delV))_T

= (del)/(delT)[T((delP)/(delT))_V - P]_V

= (del)/(delT)[T((delP)/(delT))_V]_V - ((delP)/(delT))_V " "" "bb((6))

Now, we should figure out what ((delP)/(delT))_V is so we can use it! Plugging in the vdW equation of state from bb((2)):

((delP)/(delT))_V = (del)/(delT)[(nRT)/(V - nb) - (an^2)/(V^2)]_V

= (nR)/(V - nb)

Plugging this back into bb((6)), we get:

color(blue)(((delC_V)/(delV))_T) = (del)/(delT)[(nRT)/(V - nb)]_V - (nR)/(V - nb)

= (nR)/(V - nb) - (nR)/(V - nb)

= color(blue)(0)