I am trying to derive an equation for chemical potential from Boltzmann entropy and this is what I have come up with so far:

$$S = k\ln{\left(\frac{N!}{N_\mathrm{up}!\,N_\mathrm{down}!}\right)} \label{eq:1} \tag{1}$$

and the derivation of chemical potential from the 1st law of thermodynamics,

$$\mu = -T\left(\frac{\partial S}{\partial N}\right)_{UV} \tag{2}$$

I have used Stirling's approximation and properties of logs and have come up with an answer, but after searching for my answer it is not online anywhere, so it feels like I might be going in the wrong direction.

From my computation of entropy which I found to be,

$$S = k_b\left( N\cdot\ln{N} - \sum(N_i\cdot\ln{N_i}) \right) \tag{3}$$

I got chemical potential to be

$$\mu = -Tk_b\left( \ln{N} +1 - \sum{(\ln{N_i} + 1)} \right) \tag{4}$$

Is this a derivation for entropy that anyone has ever seen before? Is it even possible to take the equation $\eqref{eq:1}$ and derive an equation for chemical potential?

Any answer or push in the right direction would be greatly appreciated.


Start with $$ S(N_1,N_2) = k \log\frac{(N_1+N_2)!}{N_1! N_2!} $$ Write the chemical potential as \begin{align} \mu &= - kT\left(\frac{\partial S}{\partial N_1}\right)_{N_2}\\ &= -kT \frac{S(N_1+1,N_2)-S(N_1,N_2)}{(N_1+1)-N_1} \\ & = -kT \log \frac{(N_1+N_2+1)!}{(N_1+N_2)!} \frac{N_1!}{(N_1+1)!} \frac{N_2!}{N_2!}\\ & = -kT \frac{N_1+N_2+1}{N_1+1} \end{align} Make the approximation $$ \frac{N_1+N_2+1}{N_1+1} \approx \frac{N_1+N_2}{N_1} = \frac{1}{x_1} $$ and write the final result as $$ \boxed{ \vphantom\int \mu = kT \log x_i } $$ with $$ x_i = \frac{N_i}{N} = \frac{N_i}{\sum_i N_i} $$


So the main conceptual problem that you are facing I think is that if say I am looking at a thermodynamic system of water molecules and salt molecules, then I expect there to be a chemical potential for water molecules and also a separate one for salt molecules. It doesn't make much sense for me to define a chemical potential for molecules overall.

In your case, you are writing this in the first part to suggest to me that you've got two different things, $N_{\text{up,down}}$ that presumably should each have a chemical potential. But then you seem to be calculating the overall chemical potential instead, which is not at all obvious to me.

Given these labels “up, down,” I guess maybe you have a system of spins? Or are they just “above” some divider vs below it? Basically I guess I am asking what this system is going to be in contact with, just as you start to ask about the temperature of a system when you are considering it sharing energy with its surroundings. A chemical potential is like sharing energy, except you're sharing particles of some sort or another.

If you do have particles, then there's a difference between being in contact with a reservoir of exclusively spin-up particles that you might want to add while holding $N_\text{down}$ constant, versus say a more “spintronics” application where $N$ remains constant but you are going to be in contact with a source of angular momentum such that $N_\text{up}-N_\text{down}$ can change, you will need a different notion of chemical potential for the two cases.

So it's worth understanding first how you think of these populations before you evaluate their respective chemical potentials.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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