I have been given a problem to derive a relation for change in entropy from change in Gibbs free energy:

Starting with the isothermal equation

$$G(p_2) = G(p_1) + nRT\ln\frac{p_2}{p_1}\label{eqn:1}\tag{1}$$


$$S(V_2) = S(V_1) + nR\ln\frac{V_2}{V_1}\label{eqn:2}\tag{2}$$

using the relation

$$\mathrm dG = V\,\mathrm dp - S\,\mathrm dT.\label{eqn:3}\tag{3}$$

The derivation wants you to assume that pressure is constant and thus $\mathrm dp = 0,$ and \eqref{eqn:3} can be arranged to

$$\left(\frac{\partial G}{\partial T}\right)_p = -S.\label{eqn:4}\tag{4}$$

Now I understand that then differentiating \eqref{eqn:1} w.r.t $T$ and using the fact $p_1 V_1 = p_2 V_2$ (for constant $T)$ can produce the required equation \eqref{eqn:2}.

However, I don't understand quite how we can justify the derivation of an isobaric equation from an isothermal one as this method appears to. Equation \eqref{eqn:1} describes isothermal processes and is a function of pressure, $p_2.$ How is it valid then to suddenly say now pressure is constant and therefore use \eqref{eqn:4} when in equation \eqref{eqn:1} the pressure is clearly not constant?

Might it have something to do with Gibbs free energy being a state function meaning, once there is a relationship for it, constraints (for different state variables) can be changed to derive new relationships using a different constraint?

  • $\begingroup$ Who says you are assuming that the pressure is constant? $\endgroup$ Feb 24, 2021 at 0:16
  • $\begingroup$ @ChetMiller The answer to the question says it. Is it wrong? $\endgroup$
    – user246795
    Feb 24, 2021 at 9:16

2 Answers 2


For an ideal gas at constant temperature $\Delta H=0$, so $\Delta G=-\Delta (TS)$, and, at constant temperature, $$\Delta G=-T\Delta S$$

  • $\begingroup$ Is the derivation set out in the question flawed then? I know there are many ways to derive the equation, I'm just unsure about the method behind the original post derivation. $\endgroup$
    – user246795
    Feb 24, 2021 at 9:21
  • $\begingroup$ I don't follow the derivation set out in the question. $\endgroup$ Feb 24, 2021 at 12:27

Maybe, what they expect you to do is to write: $$-\left(\frac{\partial S}{\partial P}\right)_T=\frac{\partial ^2 G}{\partial P\partial T}$$From Eqn. 1, this gives, for an ideal gas $$-\left(\frac{\partial S}{\partial P}\right)_T=\frac{nR}{P}$$


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.