LECTURE 6

Properties of Ideal Gases

Ideal gases are a very simple system of noninteracting particles. The only energy
involved is the kinetic energy of the gas particles. There is no potential energy. Let’s
study this system as a way to illustrate some of the concepts that we have been discussing
such as internal energy, specific heat, etc.
First of all, the internal energy of an ideal gas is solely a function of its temperature
and is independent of its volume.
E = E(T ) independent of V (1)
Perhaps this is not surprising since the energy is solely kinetic and hence just depends on
the temperature. The energy does not depend on interactions between the particles, so
it doesn’t matter how close together the particles are, i.e., the density and volume don’t
matter.
One can prove that the energy is solely a function of the temperature in 2 different
ways. One way is microscopic and uses phase space (Reif section 2.5); the other way is
macroscopic and just uses the equation of state pV = νRT (Reif section 5.1). Let’s go
over the microscopic proof. Let ri denote the position of the ith particle and let pi be
it’s momentum. The number of states Ω(E) lying between the energies E and E + dE is
proportional to the volume of phase space contained between the surface in phase space
with energy E and the surface with energy E + dE:
Z E+dE
Ω(E) ∝ d3 r1 ... d3 rN d3 p1 ... d3 pN (2)
E

E + dE

Since there is no interaction between the particles (U = 0), the integrals over position
are trivial: Z
d 3 ri = V (3)
So
Ω(E) ∝ V N χ(E) (4)
where Z E+dE
χ(E) ∝ d3 p1 ... d3 pN (5)
E
is independent of V since the particles’ momenta does not depend on their position, so
that the integral does not depend on the volume. The total energy of the ideal gas is
simply the kinetic energy of the particles:
3
N X
p2iα
X
2mE = (6)
i=1 α=1

where the sum over α is the sum over the x, y, and z components of the momenta.
The sum contains 3N = f terms. For E= constant, eq. (6) describes a sphere in the
f −dimensional space of momentum components. The radius of the sphere is

R(E) = (2mE)1/2 (7)

(This is just like a spherical surface in 3D which is described by the equation x2 +y 2 +z 2 =

constant = R2 .) The volume of the sphere in f −dimensional space is proportional to

Rf = (2mE)f /2 (8)

The number of states Ω(E) lying in the spherical shell between energies E and E + δE
is then given by
Ω ∝ Rf −1 ∝ E (f /2)−1 ∝ E (3N/2)−1 (9)
Combining this with our factor of V N , we obtain

Ω(E) = BV N E 3N/2 (10)

where B is some constant independent of E and V .

It’s easy to calculate the inverse temperature β.
3N
ln Ω = ln B + N ln V + ln E (11)
2
So
∂ ln Ω 3N 1
β= = (12)
∂E 2 E
Solving for E yields
3N 3N
E= = kB T (13)
2β 2
Notice that the energy is just a function of the temperature, and not of the volume as
advertised. If Na is Avogadro’s number, then N = Na ν where ν is the number of moles.
Then
3 3
E = ν(Na kB )T = νRT (14)
2 2
where R = Na kB is the gas constant.
Specific Heat of an Ideal Gas

2
 We can now go on and calculate the specific heat at constant volume. It’s important to
specify what macroscopic parameters are being kept constant. As we mentioned earlier,
the specific heat we measure at constant volume is different from the specific heat we
measure at constant pressure. In either case we start with

dQ = dE + pdV (15)

At constant volume, dV = 0 and

dQ = dE (16)
So the molar specific heat at constant volume is
! !
1 dQ 1 dE
cV ≡ = (17)
ν dT V
ν dT V

Using our result E = 3νRT /2, we see that

3
cV = R (18)
2
Now let us compare the specific heat at constant pressure cp to that at constant
volume cV for an ideal gas. We start with

dQ = dE + pdV (19)

To obtain cV , we note that at constant volume dV = 0. So

dQ = dE (20)

Hence ! !
1 dQ 1 ∂E
cV ≡ = (21)
ν dT V
ν ∂T V
In general, for any system, not just an ideal gas, the energy is a function of both
temperature and volume:
E = E(T, V ) (22)
So ! !
∂E ∂E
dE = dT + dV (23)
∂T V
∂V T
For an ideal gas where the energy is just a function of T , this reduces to
!
∂E
dE = dT (24)
∂T V

or
dE = νcV dT (25)

3
for an ideal gas.
Now let us obtain an expression for the specific heat at constant pressure. We start
with

dQ = dE + pdV
= νcV dT + pdV (26)

At constant pressure the equation of state tells us

pdV = νRdT (27)

Substituting this into (26) yields the heat absorbed at constant pressure

dQ = νcV dT + νRdT (28)

By definition !
1 dQ
cp = (29)
ν dT p

By (28) this becomes

cp = cV + R (30)
Note that cp > cV in agreement with earlier statements. The ratio γ of the specific heats
is then given by
cp R
γ≡ =1+ (31)
cV cV
Using eq. (18) which states that cV = 3R/2 for a monatomic ideal gas, we have
3 5
cp = R + R = R (32)
2 2
and
cp 5
γ≡ = (33)
cV 3
Adiabatic expansion or compression for an ideal gas
For an ideal gas, we can make certain statements concerning p, V and T when the gas
expands under certain conditions. If the gas expands quasi-statically under isothermal
conditions (constant temperature), the equation of state tells us that

pV = constant (34)

Suppose, however, that the gas is thermally isolated so that no heat is absorbed
during the expansion, i.e., dQ = 0. This is called an adiabatic expansion. Adiabatic
conditions correspond to thermal isolation. When the gas expands, it will do work and
lose internal energy. As a result, its temperature will decrease.

4
 How will the pressure be related to the volume of the gas under adiabatic conditions
in a quasi-static process? We start with
dQ = dE + pdV (35)
Adiabatic means dQ = 0. We also have dE = νcV dT for an ideal gas. So
0 = νcV dT + pdV (36)
Now let’s eliminate T in favor of p and V . From the equation of state
d(pV ) = νRdT (37)
or
dp V + V dp = νRdT (38)
Use this to substitute into eq. (36) for dT .
cV cV cV
 
0= (pdV + V dp) + pdV = + 1 pdV + V dp (39)
R R R
or
(cV + R)pdV + cV V dp = 0 (40)
Since cp = cV + R, we can write
cp pdV + cV V dp = 0 (41)
Dividing both sides by cV pV yields
dV dp
γ + =0 (42)
V p
where
cp
γ= (43)
cV
For most gases it is a pretty good approximation to assume that γ is independent of
temperature over a limited temperature range. Then we can integrate eq. (42) to obtain
γ ln V + ln p = constant (44)
or
pV γ = constant (45)
Since γ > 1 (cP > cV ), p will vary more rapidly with increasing V than in the isothermal
case where pV = constant.
We can use the equation of state to eliminate p in favor of V and T . Using p = νRT /V ,
we get
V γ−1 T = constant (46)
Note that Eqns. (45) and (46) are true for adiabatic quasi-static processes.