How Can Maxwell's Relations Be Applied to Thermodynamic Equations?

Dewgale
Messages
98
Reaction score
9

Homework Statement


This is question 2.18 from Bowley and Sanchez, "Introductory Statistical Mechanics" .

Show with the help of Maxwell's Relations that
$$T dS = C_v dT + T (\frac{\partial P}{\partial T})_V dV$$
and
$$TdS = C_p dT - T( \frac{\partial V}{\partial T})_P dP.$$

Then, prove that
$$(\frac{\partial U}{\partial V})_T = T (\frac{\partial P}{\partial T})_V - P$$

Homework Equations


The four common Maxwell's Relations. They can be found here: https://en.wikipedia.org/wiki/Maxwell_relations.

The Attempt at a Solution


The issue for me is the first part. I have managed to prove that the second is true.
My attempt thus far:
Using the fact that ##dU = TdS - PdV##, we can rewrite the equation as $$TdS = dU + PdV.$$
We can then multiply and divide dU by dT to get $$TdS = (\frac{\partial U}{\partial T})_V dT + PdV.$$ However, ##\frac{\partial U}{\partial T}_V = C_v##, and so we have $$TdS = C_v dT + PdV.$$
This then leaves me with the issue of how to convert the term PdV ##\to T (\frac{\partial P}{\partial T})_V dV##. I was considering using the relation ##\frac{\partial P}{\partial T} = \frac{\partial S}{\partial V}##, but I'm not sure how to get that to work. I'll keep working at it, but any help is appreciated!
 
Physics news on Phys.org
Dewgale said:
We can then multiply and divide dU by dT to get $$TdS = (\frac{\partial U}{\partial T})_V dT + PdV.$$
Well, this is certainly not true! Recall that given a multivariable function ##f(x,y)##, we have
df(x,y) = \left(\frac{\partial f}{\partial x}\right)_{y} dx + \left(\frac{\partial f}{\partial y}\right)_{x} dy
 
Fightfish said:
Well, this is certainly not true! Recall that given a multivariable function ##f(x,y)##, we have
df(x,y) = \left(\frac{\partial f}{\partial x}\right)_{y} dx + \left(\frac{\partial f}{\partial y}\right)_{x} dy

I solved it!
Consider the fact that $$ dU = (\frac{\partial U}{\partial S}) dS + (\frac{\partial U}P\partial V}) dV.$$ Then since ##dU = TdS - PdV##, we can see that ##\frac{\partial U}{\partial S} = T##, or ##\frac{\partial S}{\partial U} = \frac{1}{T}##. Now, consider a function ##S=S(U,V)##.
$$dS = (\frac{\partial S}{\partial U}) dU_V + (\frac{\partial S}{\partial V}) dV$$
Using one of the Maxwell relations, we know that ##\frac{\partial S}{\partial V} = \frac{\partial P}{\partial T}##. We also know that ##dU_V = C_V dT##. Therefore, we have
$$dS = \frac{1}{T} C_V dT + (\frac{\partial P}{\partial T}) dV$$
and multiplying through by T gives
$$TdS = C_V dT + T(\frac{\partial P}{\partial T}) dV$$

A similar process using S(U,P) will give the other. Thank you!
 
Hi, I had an exam and I completely messed up a problem. Especially one part which was necessary for the rest of the problem. Basically, I have a wormhole metric: $$(ds)^2 = -(dt)^2 + (dr)^2 + (r^2 + b^2)( (d\theta)^2 + sin^2 \theta (d\phi)^2 )$$ Where ##b=1## with an orbit only in the equatorial plane. We also know from the question that the orbit must satisfy this relationship: $$\varepsilon = \frac{1}{2} (\frac{dr}{d\tau})^2 + V_{eff}(r)$$ Ultimately, I was tasked to find the initial...
The value of H equals ## 10^{3}## in natural units, According to : https://en.wikipedia.org/wiki/Natural_units, ## t \sim 10^{-21} sec = 10^{21} Hz ##, and since ## \text{GeV} \sim 10^{24} \text{Hz } ##, ## GeV \sim 10^{24} \times 10^{-21} = 10^3 ## in natural units. So is this conversion correct? Also in the above formula, can I convert H to that natural units , since it’s a constant, while keeping k in Hz ?
Back
Top