# Entropy Difference of an Unknown Gas (not an ideal gas)

Tags:
1. Dec 17, 2015

### albertov123

1. The problem statement, all variables and given/known data
Temperature, pressure and volume measurements performed on 1 kg of a simple compressible substance in three stable equilibrium states yield the following results.

State 1 (T1=400 C , V1= 0,10 m3, P1=3 MPa)
State 2 (T1=400 C , V1= 0,08 m3, P1=3,5 MPa)
State 3 (T1=500 C , V1= 0,10 m3, P1=3,5 MPa)

Estimate the difference in entropy S2-S1

2. Relevant equations
We don't know the gas. So I can't assume this is an ideal gas and I can't go to thermodynamics tables. I don't know the relevant equation.

3. The attempt at a solution
First, I didn't get why the question identifies state 3. I think we can completely ignore state 3 because the question is entropy difference between state 2 and state 1.

This is clearly a compression process, (volume decreases, pressure increases) but temperature stays still. But when a gas is compressed, its pressure and temperature rises. So there must be heat transfer going on.

Entropy change = Sgen + Q/T
But we don't know the entropy generation so we can't go from here.

I assume I need a relation related with volume and pressure that yields entropy but in constant temperature. Is there a relation like that?

2. Dec 17, 2015

### Staff: Mentor

Yes. There is a relation for the partial derivative of entropy with respect to pressure at constant temperature in terms of the P-V-T properties of a gas.
Have you studied that yet? Are you learning about the Maxwell equations in your course yet? The derivation starts off with dG=-SdT+VdP.

3. Dec 17, 2015

### albertov123

Yes we are learning maxwell equations in fact. That derivation must be dG=-SdT+VdP+∑μidNi from my notebook.

The relation you have mentioned is probably -(∂S/∂P)T=(∂V/∂T)P and with that equation my solution would include state 3. But how could i go with this equation I'm not clear. Although I'm very doubtful, can I think the above equation as -(ΔS/ΔP)between state 1-2=(ΔV/ΔT)between state 2-3? Otherwise I don't know what to do.

4. Dec 17, 2015

### Staff: Mentor

Yes, that's what you do. They said "estimate" the entropy change, so using finite differences is OK.