Lecture 1: Black-body radiation and Planck’s law

Learning objectives
• Derive Planck’s spectral distribution from counting EM modes in a cavity and using
the quantum occupancy.

• Convert the Planck distribution between frequency/energy and wavelength forms

carefully (Jacobian steps shown).

• Compute numerical values (worked example) and practice unit checks.

Prerequisites
Calculus (differentiation/integration), basic statistical idea of average occupancy, and
familiarity with constants h , c , k .

1. Physical picture and assumptions

We consider a cavity at temperature T in thermal equilibrium. The electromagnetic field in
the cavity can be treated as a set of standing wave modes (harmonic oscillators). Each
mode of frequency ν has energy levels n h ν (n=0 , 1, 2 , …). The average energy of a
harmonic oscillator at frequency ν in thermal equilibrium (Bose–Einstein statistics, zero
chemical potential) is
hν
⟨ E ⟩= h ν/ ( k T )
.
e −1
(We will later multiply this average energy by the density of modes per unit volume to
obtain the energy density.)

2. Counting modes in a cavity (detailed)

Consider a large cubic cavity of side L. Standing waves satisfy boundary conditions
producing discrete wavevectors k . The number of modes with wavevector magnitude
between k and k + d k (both polarizations) in the volume V =L3 is
V 2
3
⋅ 4 π k d k×2,
(2 π )
where:
V 2
• 3
4 π k d k is the number of states in k -space shell (standard result),
(2 π )

• the factor 2 accounts for two independent polarizations of the electromagnetic

wave.
Thus number of modes per unit volume between k and k + d k is
2
1 2 k
gk d k= 3
4 π k d k ×2= 2 d k .
(2 π ) π

Now use the dispersion relation for light in vacuum:

2π ν 2π
k= ⇒ d k= dν.
c c
Substitute:

( )( )
2 2
1 2πν 2π 8π ν
gν d ν= 2
d ν = 3 d ν.
π c c c
So the number of modes per unit volume in frequency interval [ ν , ν +d ν ) is
2
8π ν
3
dν.
c

3. Energy density u ( ν )
hν
Each mode has average energy ⟨ E ⟩= . Multiply by the number of modes per unit
h ν/ ( k T )
e −1
volume to get the energy density per unit frequency (energy per unit volume per unit
frequency):

( )
2 3
8π ν hν 8π h ν 1
u (ν ) d ν = 3
⋅ h ν /( kT ) d ν= 3 h ν /( k T )
dν.
c e −1 c e −1

Thus
3
8 π hν 1
u (ν )= 3 h ν / ( kT )
.
c e −1

4. From energy density to emitted flux (intensity)

The energy density u ( ν ) is the energy per unit volume. The energy flux (power per unit area
per unit frequency) leaving a black surface, which we denote I ν , is related to u ( ν ) by angular
averaging over the hemisphere above the surface. For isotropic radiation the average
projection factor is ⟨ cos θ ⟩ integrated over a hemisphere giving factor c /4 (standard
derivation: flux ¿ ∫ u c cos θ d Ω/4 π ). Therefore
c
I ν= u ( ν ) .
4
Substitute u ( ν ) :
 3 3
c 8π h ν 1 2π h ν 1
I ν ( ν )= ⋅ 3 h ν/ ( k T )
= 2 h ν /( k T )
.
4 c e −1 c e −1
So we have the Planck distribution in frequency form:
3
2π h ν 1
I ν ( ν )= 2 h ν /( k T )
.
c e −1

5. Energy variable E=h ν form

If we want the spectral flux per unit energy I E ( E ) so that I E d E equals the flux between
energy E and E+ d E , note that ν=E/h and d ν =d E /h. Using I ν d ν=I E d E,
dν 1
I E ( E )=I ν ( ν ) =I ν ⋅ .
dE h

Substitute I ν and replace ν 3=( E/h )3 :

3 3 3
1 2π hν 1 2πν 1 2π E 1
I E ( E )= ⋅ 2 E / (k T )
= 2 E /( k T )
= 3 2 E /(k T ) .
h c e −1 c e −1 h c e −1
So the energy form is
3
2πE 1
I E ( E )= 3 2 E /( k T )
.
h c e −1

6. Change of variable: wavelength form I λ ( λ )

We often want spectral flux per unit wavelength I λ ( λ ) with I λ d λ=I ν d ν . Use the relation
c c
ν= , d ν =− 2 d λ .
λ λ
Take absolute value for densities; thus

I λ ( λ )=I ν ( ν ) |dd νλ )=I ⋅ λc .

ν 2

3
2π hν 1
Substitute I ν = 2 h ν / ( kT ) and replace ν=c / λ :
c e −1

( )
3 2
2π hν 1 c 2 π hc 1
I λ ( λ )= 2 h ν/ ( k T )
⋅ 2= 5 h c/ ( λk T )
.
c e −1 λ λ e −1

Hence the Planck law in wavelength form:

 2
2π hc 1
I λ ( λ )= .
( )
5
λ hc
exp −1
λk T

Instructor note: watch the Jacobian; I ν and I λ are different functions and have different
peak locations because of the nonlinear change of variables.
Suggested insertion: Put Figure 2.1 here (black-body intensity vs wavelength) to show
the curve and peak shift. :contentReference[oaicite:3]index=3

7. Units / dimensional check (careful)

Take I λ expression and check units:

[ h )=J ⋅ s , [ c )=m ⋅s −1 , [ λ )=m , [ k T )=J .

Then 2 π h c 2 / λ 5 has units:
−1 2
J ⋅ s ⋅ ( m⋅ s ) J ⋅s ⋅ m2 ⋅ s− 2 J W
5
= 5
= 3 = 3.
m m m ⋅s m
But I λ is energy flux per unit area per unit time per unit wavelength so its SI units are
−2 −1 −3
W m m =W m , which matches. (If you prefer μm for wavelength, convert accordingly.)

8. Worked numerical example (showing arithmetic steps)

Compute I λ at λ=500 n m=5.00 ×10− 7 m for a black body at T =5800 K (solar surface
temperature approximation). Use:
− 34 8 −1 − 23 −1
h=6.62607015 ×10 J s , c=2.99792458 ×10 m s , k=1.380649 ×10 JK .
Step 1: compute exponent argument
hc
x= .
λk T
Numerically:

h c=( 6.62607015× 10−34 ) ( 2.99792458 × 108 )=1.98644586 × 10−25 J ⋅ m.

Now

λ k T =( 5.00 × 10−7 ) ( 1.380649 ×10− 23 ) (5800 )

First compute k T :
− 23 − 20
k T =1.380649 ×10 ⋅5800=8.0077642 ×10 J.
Then λ k T :
−7 −20 − 26
λ k T =5.00 ×10 ⋅8.0077642× 10 =4.0038821 ×10 J ⋅m.
Now the argument:
− 25
1.98644586 ×10
x= − 26
≈ 4.962 .
4.0038821 ×10
Step 2: compute exponential factor
x 4.962
e ≈e ≈ 142.78 .
Step 3: compute prefactor 2 π h c 2 / λ 5:
8 2
c =( 2.99792458 × 10 ) =8.98755179 × 10 m s .
2 16 2 −2

−7 5
λ =( 5.00 ×10 ) =3.125 ×10
5 −32 5
m .

2 π h c 2=2 π ⋅6.62607015 ×10− 34 ⋅8.98755179 ×10 16=1.1973669927 ×1016 ( units: J m 2 s− 1 ) .

Dividing by λ 5:
2 16
2 π h c 1.1973669927 ×10 47 −3
5
= −32
=3.8315743766 × 10 W m .
λ 3.125× 10

Step 4: divide by ( e x −1 ) :
47 47
3.8315743766 ×10 3.8315743766 ×10 45 −3
I λ= = ≈ 2.703× 10 W m .
142.78 −1 141.78
This is per metre of wavelength. Converting to the common unit W m−2 μ m− 1:
−1 −6 −1
1 m =10 μ m ,
so
45 −6 39 −2 −1
I λ ≈ 2.703 × 10 × 10 =2.703× 10 W m μ m .

Instructor note: these absolute numbers are large because this is flux emitted per unit area
of the surface of the black body (solar surface); when observed at Earth the geometric
2
dilution (factor ( R⊙ /d Earth ) 2 ≈ ( 6.96 ×10 8 /1.496 ×1011 ) ∼2.16 ×10 −5) reduces the flux
enormously.
(Numerical values above show all intermediate arithmetic step-by-step for
classroom transparency.)

9. Quick in-class activities

1. Units check: Give students the expression for I E and have them show units are
−2 −1
Wm J .

2. Plotting exercise: Ask students (pair work) to numerically plot I λ ( λ ) for T =5800K
and T =300K between λ=0.1 μm and 104 μ m and comment on peak positions.
10. Homework problems
1. Starting from the expression for I ν derive Eq. (1.3) again, showing every algebra
step (turning ν into λ ).

2. Using the Planck distribution, compute the total emitted power per unit area
(Stefan–Boltzmann law) by integrating I ν over all frequencies and show the integral
∞ 3 4
∫ e xx−1 d x= 15
π
is used.
0