Equality of mixed partials, again
Every Maxwell relation is the same trivial fact — a smooth function's mixed second derivatives commute — applied to a thermodynamic potential. Take the Helmholtz energy F(T,V) with dF=-S\,dT-P\,dV. Reading off, S=-(\partial F/\partial T)_V and P=-(\partial F/\partial V)_T. Now demand \partial^2F/\partial V\,\partial T=\partial^2F/\partial T\,\partial V.
The Maxwell relation from F — an entropy derivative equated to a purely mechanical one.
That is the whole trick. Each of the four potentials $U,H,F,G$ has two natural variables, and equality of mixed partials gives one such identity apiece.
The four relations and the square
One Maxwell relation per potential. Note that entropy derivatives always sit on the left; P–V–T derivatives on the right.
The payoff is now visible. The left sides involve entropy, which no instrument reads directly. The right sides involve only P, V, T and their slopes — the mechanical equation of state you can measure with a gauge and a thermometer. The Maxwell relations trade the unmeasurable for the measurable.
Worked example: does an ideal gas's energy depend on volume?
Question: squeeze a gas isothermally into half its volume. Does its internal energy change? Kinetic theory whispers 'no' for an ideal gas, but let us prove it from thermodynamics alone — and see exactly where the argument breaks for a real gas.
- We want (\partial U/\partial V)_T. Start from the fundamental relation dU=T\,dS-P\,dV and divide through by dV at fixed T: (\partial U/\partial V)_T = T(\partial S/\partial V)_T - P.
- The nuisance term (\partial S/\partial V)_T is unmeasurable — but the Maxwell relation from F says (\partial S/\partial V)_T=(\partial P/\partial T)_V.
- Substitute to get a relation true for any substance: (\partial U/\partial V)_T = T(\partial P/\partial T)_V - P. This 'internal pressure' contains only the equation of state.
- For an ideal gas P=NkT/V, so (\partial P/\partial T)_V = Nk/V = P/T. Then T(\partial P/\partial T)_V - P = P - P = 0.
- Conclusion: (\partial U/\partial V)_T=0. An ideal gas's energy depends on temperature alone (Joule's law) — proven with no reference to atoms.
The 'internal pressure' or thermodynamic equation of state — exact for any substance, zero for an ideal gas.
The same machinery: heat capacities
The identical Maxwell-relation move yields the general gap between the two heat capacities — again in terms of measurable P–V–T derivatives only.
A general, exact result; for an ideal gas it collapses to C_P-C_V=Nk (equivalently nR).