# Homework Help: Virial coefficients of the ideal gas equation.

1. Sep 2, 2007

### EricVT

1. The problem statement, all variables and given/known data

Even at low density, real gases don't quite obey the ideal gas law. A systematic way to account for deviations from ideal behavior is the virial expansion. Just working with the first order virial expansion we have:

PV=nRT(1 + B(T)/(V/n))

B(T) is a virial coefficient. A table is given with some measured values of B at different temperatures for nitrogen at atmospheric pressure (temp on left, B value on right separated by "/"):

T(K) / B(cm^3/mol)
-------------------
100 / -160
200 / -35
300 / -4.2
400 / 9.0
500 / 16.9
600 / 21.3

The problem asks to find the value of B(T)/(V/n) for each value of T given in the table.

2. Relevant equations

PV=nRT (ideal gas equation)

3. The attempt at a solution

Since B(T) is given in the table, it seems that only (V/n) needs to be found and then a simple division of terms should yield the answer for each T that you insert into the equation.

First I divided both sides by n and P, and set (V/n)=x, and set the quantity R/P = C

==> x = CT(1+B/x)
==> x/(1+b/x) = CT
==> x^2 = CTx +CTB
==> x^2 - CTx - CTB = 0

This is a quadratic for x, which I solved to get:

x = (1/2)(CT + sqrt[(CT)^2 -4(-CTB)] ==> x = (1/2)(CT + sqrt[(CT)^2 + 4(CTB)]

or x = (1/2)(CT - sqrt[(CT)^2 + 4(CTB)]

However, (CT)^2 is a very small number and for negative B terms such as -160 (the first value in the given table) you can't get real solutions to this quadratic. This doesn't seem correct since these virial terms represent corrections to real intermolecular interactions, so I feel like this technique must not be suitable for this problem.

Is there some misunderstanding on my part about what is being asked? The problem gives B as a function of T alone, but it seems like you can manipulate the ideal gas law to show B as a function of P,V,n, and T. This makes me think there is something I am not understanding about what is given in this problem.

1. The problem statement, all variables and given/known data

Another equation of state is given by the van der Waals equation:

(P + a*n^2/V^2)(V - n*b) = n*R*T

where a and b are constants that depend on the type of gas. Calculate the second virial coefficient B for a gas obeying the van der Waal's equation, in terms of a and b.

2. Relevant equations

Hint: The binomial expansion says that (1 + x)^p ~ 1 + p*x +(1/2)(p)(p-1)(x^2) for |px|<<1. Apply this approximation to the quantity [1-(n*b/V)]^(-1).

3. The attempt at a solution

First of all, is the virial expansion for the van der Waal's equation just

(P + a*n^2/V^2)(V - n*b) = n*R*T(1 + B/(V/n)) like for the ideal gas law virial expansion? I assumed it was and worked from there. I basically just divided out terms until I got

[1-n*b/V]^(-1) = [n*R*T*(1+B/(V/n))]/[P*V+ a*n^2/V)

Then applied the given hinted approximation and followed through all the boring algebra to finally get:

B = [P*V^2 + a*n^2]/[R*T*n^2 + ((R*T*b*n^3)/V)(1+n*b/V)] - V/n

Yikes! Doesn't seem too convincing (the units don't even make sense!), plus I'm not sure if my answer was only supposed to be in terms of "a" and "b" or if the other variables are fine as well.

Any help on this would be appreciated. This looks like a great community, so hopefully I can fit right in.

Last edited: Sep 2, 2007
2. Sep 2, 2007

### EricVT

One thing I've just thought of.

Can we approximate B/(V/n) as B/(RT/P), since this is a first order approximation? That could make the first question a bit easier and avoid the troubles of non-real solutions to the quadratic. But I have to wonder how accurate this approximation would be.