CML1001: Afternotes pt.

3
Particle in a box

Narayanan Kurur
IIT Delhi

January 2026
Contents
1 Schrödinger equation: Some exactly solvable cases 3
1.1 Particle in a box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.1 Electronic spectroscopy of conjugated alkenes . . . . . . . . . . . 10
1.1.2 Particle in multi-dimensional boxes . . . . . . . . . . . . . . . . . 11
.

2
1 Schrödinger equation: Some
exactly solvable cases
1.1 Particle in a box
We will now look at the solutions of the Schrödinger equation for exactly solvable
cases. First, we consider a particle of mass m confined to move along the x-axis
between 0 to L. This is achieved by making the potential V = 0 between x = 0 and
x = L and V = ∞ for x < 0 and x > L (see Figure 1.1). In quantum chemistry this
model is referred to as particle in a box (PIB) of length L.
In case you are wondering that this model is just that, a model, that is not the
case. This is a reasonable approximation for the situation experienced by the π
electrons in a conjugated polyene, the simplest example of which is 1,3-butadiene,
CH2 CH CH CH2. The four π-electrons in butadiene, one from each of the four
sp2 hybridized carbon atoms, are essentially delocalized over the σ framework made
up by the four carbon atoms. So, each of the four electrons can be thought of as mov-
ing independently on a line of length given by the length of the molecule, which can
be approximated as L = 2 ∗ rdb + rsb , where rdb and rsb are, respectively, the lengths
of a C C and C C bond.
Before we set up and solve the Schrödinger equation let us apply de Broglie’s
approach to this problem. de Broglie associates a wave with every material particle
traveling with momentum p. The wavelength of the wave is
h
λdB = . (1.1)
p
The classical analog of a particle in a box is a string that is fixed at both ends. When
such a string is plucked, we know that the amplitude of the oscillations at the fixed
ends is zero. In other words, an integer number of half-wavelengths must fit in the
length of the box. Applying this idea to the present case, we find that
λdB h
n =n =L n = 1, 2, · · · (1.2)
2 2p
Solving for p we get
nh
p= (1.3)
2L
The particle feels no potential energy so all its energy is in the form of kinetic energy.
As a result
p2
E= . (1.4)
2m
Substituting for p, we get
n2 h2
En = n = 1, 2, · · · (1.5)
8mL2

3
 1 Schrödinger equation: Some exactly solvable cases

V=∞ V=0 V=∞

0 L

Figure 1.1: The one-dimensional infinite potential well of length L. It is divided into
three regions, with regions I and III having V = ∞ and region II having
V=0

4
 1 Schrödinger equation: Some exactly solvable cases

Let us now proceed to analyze this problem quantum mechanically. The Schrödinger
equation is
Ĥψ( x ) = Eψ( x ), (1.6)
where V ( x ) = ∞ in regions I and III and V ( x ) = 0 in region II. The results are in
no way affected if in region II the potential is V ̸= 0 because it only has the effect
of altering the zero of energy and so without loss of generality we will assume that
V = 0 within the box.
In regions I and III the wave function is identically zero since the potential is infinite
so we will only have to consider the solution within the box. This is because a particle
cannot have infinite potential energy. As a result, it cannot penetrate into a region of
space where V = ∞. We focus our attention now on the Schrödinger equation written
in the form given in equation 1.6. Both E and ψ on the RHS are finite while the LHS
has one term, V, in regions I and III that is infinite. The equality in equation 1.6 can
only hold when ψ is identically zero, which implies that the wave function goes to
zero when the potential becomes infinite. This result is true in any region where the
potential V is infinite. We also note that since V exhibits a second order discontinuity
(infinite jump) across the border, d2 ψ/dx2 will also exhibit this discontinuity, so that
dψ/dx will be discontinuous there.
We are looking for the solutions to the equation

h̄2 d2
− ψ( x ) = Eψ( x ). (1.7)
2m dx2
We rewrite this differential equation in the form

d2 ψ
= kψ. (1.8)
dx2
q
2mE
where k = . The general solution to this differential equation is
h̄2

ψ( x ) = A cos(kx ) + B sin(kx ) (1.9)

where A and B are constants.

The boundary conditions that the wave function should satisfy are

ψ (0) = 0 (1.10)

and
ψ( L) = 0. (1.11)
The first of these conditions implies that A = 0. The second condition yields B sin(kL) =
0 which implies that
kL = nπ, n = 1, 2, 3, · · · . (1.12)
In other words k = nπ/L. The wave function has the form
 nπx 
ψn = B sin . (1.13)
L
The first five wave functions are plotted in figure 1.2.
The number of nodes, that is, the number of zero crossings, increases as the quan-
tum number increases. The ground state, n = 1, has no node, the first excited state,

5
 1 Schrödinger equation: Some exactly solvable cases

0.5

ψ( x )
0

−0.5

0 L/2 L

Figure 1.2: The first five wave functions, n = 1 to n = 5, of a particle confined to a

box. The number of internal nodes is n − 1. The wave functions show
even-odd symmetry about the mid-point of the box.

n = 2, has one node, and so on. As noted earlier, the curvature of the wave function
increases as the quantum number increases because the kinetic energy increases. The
increase in the number of nodes with increase in energy is a characteristic of quantum
mechanical systems.
The wave functions also show interesting symmetry properties. About the mid-
point of the box, the ground state wave function is symmetric (even), while the first
excited state is antisymmetric (odd), and so on. In general, all the states with n = odd
are symmetric, while the states with n = even are anti-symmetric - there is alternating
even-odd symmetry. In general, wave functions of symmetric potentials show even-
odd symmetry.
The constant B in the wave function is yet to be determined. The normalization
condition on the wave function yields B. Normalization refers to the requirement
that Z L Z L  nπx 
ψ2 dx = B2 sin2 dx = 1 (1.14)
0 0 L
from which we determine that r
2
B= . (1.15)
L
You would have noticed that the application of the boundary conditions naturally
leads to quantization. In addition, observe that we do not consider n = 0 as a
possibility because that would imply that the wave function is identically zero for
all x or in other words there is no particle! q
2mE nπ
Because E is related to k by the relation k = and k = we hence obtain that
h̄2 L

h̄2  nπ 2 n2 h2
E= = , n = 1, 2, 3, · · · (1.16)
2m L 8mL2
The energies of the first few states are shown in the figure below.
Interestingly, the energies obtained by solving the Schrödinger equation are identi-
cal to that obtained from a qualitative analysis using the de Broglie principle. Observe

6
 1 Schrödinger equation: Some exactly solvable cases

E5 25h2
8mL2

E4 16h2
8mL2

E3 9h2
8mL2

E2 4h2
8mL 2
E1 h2
8mL2
0 L

Figure 1.3: The energies of the first five states of a particle in a box. The energies are
directly proportional to the square of the quantum number, and inversely
proportional to the second power of the length, and first power of the
mass.

h 2
that the lowest state has an energy of 8mL 2 , which is different from zero. This is called
the zero point energy. It implies that even when the system is in the ground state it
is undergoing ceaseless motion.
Interestingly, we expect a zero point energy based on Heisenberg’s principle. What
is the zero point energy predicted by the indeterminacy principle? The particle is
confined to a box of length L, which means that the uncertainty in the position is
L - the particle has to be somewhere in the box. The minimum uncertainty in the
momentum from the Heisenberg relation is

h̄
∆p = (1.17)
2L
Heisenberg’s principle then imposes a limit to ∆p ≥ h̄/2L. We will consider the
equality because we are interested in an order of magnitude estimate. Because the
average momentum is zero, the expectation value of p2 would equal the minimum
uncertainty in the momentum (why?), that is p2 = ∆p2 . So, the minimum energy such
p2 2
a particle would have is E = 2m = 8Lh̄2 m In other words, according to the Heisenberg
principle, we expect that the particle would never be at rest - it would have the
minimum energy given above.
Example 1.1.1. Find the energy level separation and the distance between adjacent
probability maxima in the case of a particle of mass 1 × 10−10 kg (e.g., a small grain
of salt) confined to a box of half-width 10−4 m at a temperature of 1 K and comment
on their physical significance.
Solution The energy levels of this system are

En = n2 (6.625 × 10−34 J s)2 /(8 × 10−10 kg × (10−4 m)2 ) = 1.4 × 10−50 Jn2

so the separation between adjacent levels, (dEn /dn), is 2.8 × 10−50 J n. The separa-
tion between adjacent probability maxima is 2 × 10−4 m/n because we have to fit n

7
 1 Schrödinger equation: Some exactly solvable cases

0.6

0.4

ψ ( x )2 0.2

0
0 L/2 L

Figure 1.4: The probability density of the first five states of a particle in a box. Con-
trast this with the classical case where the particle is equally likely to be
found across the whole length of the box. At high quantum numbers, the
classical probability is reproduced by the quantum particle, as expected
from the Bohr correspondence principle.

times half the wavelength within the box. At a temperature

q of 1 K the energy is about
1.4×10−50 J
1.4 × 10−23 J leading to a value for n of around 3 × 1013 ( 1.4×10−23 J
). The separation
between adjacent energy levels would then be about 10 − 36 J and the separation be-
tween adjacent peaks in the probability distribution would be about 10−17 m. These
quantities are so small that an experiment of the accuracy required to detect any
effects due to energy quantization would be completely impossible in practice. An
experiment to define the position of the particle to this accuracy or better would be
similarly impossible. Thus to all intents and purposes, quantum and classical me-
chanics predict the same results—all positions within the well are equally likely and
any value of the energy is allowed—and the correspondence principle is verified. ◀
Having determined the wave function for this problem let us put it to some use.
We start by looking at the probability density, which is given by ψ2 . The probability
density of the first five states, whose wave functions were plotted earlier, is plotted
below.
In the ground state, the probability density, given by ψ12 , is maximum at the mid-
point of the box implying that the particle is most likely to be found at the mid-point
of the box. In the first excited state, at the mid-point of the box, the probability
density is zero and the particle has a negligible probability of being found there.
Often we might be interested in the probability of finding the particle in a particular
region of the box. Imagine that we have a particle in the ground state of the box and
we wish to determine the probability of finding it in the middle-third of the box, that
is between L/3 and 2L/3.
One of the things that we are often interested in is the position of the particle if it
is present in a particular state. Notice that the states that we solved for are energy
eigen states. Moreover, they are not eigen states position. That is
xψn ̸= constant × ψn (1.18)
In cases where the wave function is not an eigenfunction of the operator correspond-

8
 1 Schrödinger equation: Some exactly solvable cases

0.6

0.4

ψ ( x )2 0.2

0
0 L/3 2L/3 L

Figure 1.5: The probability of finding the particle in the middle-third of the box when
in the ground state is given by the shaded region.

ing to the observable, we will be concerned with the statistical mean of the measured
values, averaged over a large number of measurements, the expectation value. The
expectation value is defined of an operator is defined as
ˆ
R ∗
ψ Opψdτ
⟨Op⟩ = R ∗ . (1.19)
ψ ψdτ
The expectation value of the position for a particle in the n-th state of a particle in a
box would hence be
2 L 1 L
 nπx   nπx  Z   
2nπx
Z
⟨x⟩ = sin x sin dx = 1 − cos x dx (1.20)
L 0 L L L 0 L
Because the wave functions are normalized the denominator is unity. You can show,
either by symmetry arguments or by explicit integration, that the result is L/2, which
agrees with the classical result.
This calculation may be performed for any other observable. For example, the
average momentum or the expectation value of the momentum of a particle in the n-
th state of the box. The only difference in the procedure to determine the expectation
value of position is to replace the position operator with the momentum  operator.

d
That is, multiplication by x is replaced with the momentum operator, −ih̄ dx . The
average momentum is then

2 L 2 L  nπx  
 nπx   nπx  
d  nπx 
Z Z
⟨ p⟩ = sin pˆx sin dx = sin −ih̄ sin dx.
L 0 L L L 0 L dx L
(1.21)
By symmetry arguments, we find that the average momentum is zero - the particle
has equal probability of traveling in either the + x or − x direction and its average
momentum is zero.
Example 1.1.2. Verify the uncertainty principle for the particle in a box.
∆x∆p ∆x
p
Solution We need to verify that x ≥ h̄/2, where = ⟨ x2 ⟩ − ⟨ x ⟩2 and
similarly ∆p x = ⟨ p2x ⟩ − ⟨ p x ⟩2 . Three expectation values are easily obtained: ⟨ x ⟩ =
p

9
 1 Schrödinger equation: Some exactly solvable cases

L/2 (because of symmetry of the integral about L/2), ⟨ p x ⟩ = 0 (the particle has equal
2 2 2
probability of going to the right or left), ⟨ p2x ⟩ = 2mEn = n πL2 h̄ (because energy is
L 2 4π 2 n2
RL
 
solely kinetic). That leaves ⟨ x2 ⟩ = L2 0 x2 sin2 πx
L dx = 2πn 3 − 2 ◀

Exercise 1.1.1. Solve the Schrödinger equation for the case where the box extends from − L/2
to L/2.

1.1.1 Electronic spectroscopy of conjugated alkenes

One of the applications of the particle in a box is to explain the electronic spectra of
conjugated alkenes. For example, the strong red-orange color of carrots is because of
a pigment called β-carotene whose structure is given below:

Figure 1.6: The color of carrots is attributed to the pigment, β-carotene, whose struc-
ture is given. The alternating single and double bonds in this pigment is
the reason for the color.

The structural element in the pigment β-carotene which leads to the color are the
alternating single and double bonds. Cyanine dyes are other examples with this
structural element;they are generally represented as
 ⊕ ⊕
 
R2 N CH CH CH CH CH NR2′ R2 N CH CH CH CH CH NR2′
n n
As we know from organic chemistry, the π- electrons in such systems are delo-
calized across the entire σ-framework because of resonance. If the π-electrons are
considered to not interact with each other, or feel the potential of the carbon nuclei,
then each electron can freely move across the entire length of the molecule. In other
words, each π-electron is in a one-dimensional box, the length of which is deter-
mined by the σ-bonded framework. This method is called the free electron molecular
orbital model. A similar model was proposed by Sommerfeld for metals. The overall
length of the box can be determined by summing up the carbon-carbon bond lengths;
a correction is made for any extra length at the ends of the box. For carbon, the bond
lengths in conjugated systems are in the range between 148 pm and 133 pm. As an
average, we use a bond length of 140 pm. Accounting for the extension of the electron
wave beyond the last carbon nucleus in the chain can be obtained by comparison to a
more exact calculation (see Section 8.4), and a value of 140 pm beyond the outermost
C atoms is a good choice. Therefore, we determine the box length L for π-electrons
by the total length of the C C bonds involved in the π-bonding and one bond length
beyond each of the outermost C atoms.
The simplest hydrocarbon which could be studied by this model is 1, 3-butadiene.
For butadiene, the length, L, of the box is 5 × 140 pm = 700 pm. Butadiene has four
π-electrons. The four carbon atoms in butadiene can be thought of as being at L/5,
2L/5, 3L/5, and 4L/5. The edges of the box are unoccupied because we have added
a correction factor to the molecule length. Each state in the PIB model is treated like

10
 1 Schrödinger equation: Some exactly solvable cases

an orbital and is occupied by two electrons. In other words, each PIB state resembles
an orbital. The four electrons in butadiene would then occupy the first two states;
two in the ground state, and a further two in the first excited state. The energy
difference between the two states, (E2 − E1 ) for an electron in a box of length 700 pm,
is 3.79 × 10−19 J or 2.3 eV. The experimental value is 2.5 eV; the agreement is good
considering the simplicity of the model.
Example 1.1.3. Use the FEMO model to discuss the π-bonding for the amidinium
ion, by resonance between the two structures
⊕ ⊖ ⊖ ⊕
H2 N CH NH2 H2 N CH NH2
Solution The box length in this case is L = 4 × 140 pm = 560 pm (two bonds and
two ends), and the system will have four π-electrons occupying the two lowest energy
orbitals, n = 1 and n = 2. According to the expression for the energy of a particle in
a box, we have
h2
En = n2 2
= n2 1.92 × 10−19 J
8mL
The four electrons would occupy the two lowest states with energies 1.92 × 10−19 J
and 7.68 × 10−19 J ◀
Any electron in any atom or molecule can undergo a transition from one state to an-
other. If the frequency of the electromagnetic radiation falling on an atom or molecule
matches some energy level separation in it, the radiation is absorbed by the molecule.
Electronic spectroscopy is mainly concerned with the ultraviolet (200 nm to 400 nm)
and visible (400 nm to 700 nm) portion of the electromagnetic spectrum. Among or-
ganic compounds, aromatic and other conjugated molecules and their substituted
derivatives absorb in this part of the spectrum. Some heteroatomic compounds, even
with only one double bond, such as ketones and aldehydes, also have bands in this
area.
Figure 1.7 shows how the spectrum arises with the help of the energy level diagram
according to the free electron model. The lowest energy absorption arises due to a
transition from the highest occupied level, E2 , to the lowest unoccupied level, E3 . In
butadiene this occurs in the UV region of the electromagnetic spectrum. Butadiene is
colorless as a result.
The case of β-carotene is similar although the number of electrons involved is much
larger, and so is the length of the box. Once again, an electronic transition occurs
from the highest occupied state to the lowest unoccupied state. Because there are 18
π-electrons this is between E9 and E10 . The energy difference in the case of β-carotene
corresponds to visible light; the color which we see is complementary to the color of
light which is absorbed.
Early experimental spectra of some dienes and β-carotene are given in the figure
1.8

1.1.2 Particle in multi-dimensional boxes

The Schrödinger equation of a particle of mass m in a rectangular box is given by

h̄2 ∂2 ∂2
 
− + ψ( x, y) = Eψ( x, y), (1.22)
2m ∂x2 ∂y2

11
 1 Schrödinger equation: Some exactly solvable cases

9h2 E3
8mL2

hν = E3 − E2

4h2 E2
8mL2

h2 E1
8mL2

Figure 1.7: The energy level diagram of butadiene in the free electron model. An elec-
tronic transition is caused by light corresponding to the energy difference
between E3 and E2 .

(a) Electronic spectra of some dienes. (b) Electronic spectrum of β-carotene.

Figure 1.8: Representative experimental spectra of conjugated alkenes: dienes and β-

carotene. The units are wavenumbers.

12
 1 Schrödinger equation: Some exactly solvable cases

where x extends from 0 to L x and y from 0 to Ly .

Separation of variables
Separation of variables is a useful technique for solving a number of partial differ-
ential equations (PDEs) that we come across in quantum mechanics. It is one of
the most fundamental techniques for solving partial differential equations and is ap-
plied to linear PDEs. Here we will highlight the separation of variables to solve the
Schrödiner equation for a particle in a rectangular box (2D) given in equation 1.22.
The separation of variables approach proceeds as follows. We assume the solution
ψ( x, y) can be written as a product ψ( x, y) = X ( x )Y (y). Substituting this in the
Schrödinger equation yields
 2
∂2

∂
+ X ( x )Y ( y ) = E ′ X ( x )Y ( y ) , (1.23)
∂x2 ∂y2

where E′ = − 2m2 E. The left hand side of the equation becomes

h̄

∂2 X ∂2 Y
Y (y) + X ( x ) . (1.24)
∂x2 ∂y2
Dividing through by X ( x )Y (y) gives

1 ∂2 X 1 ∂2 Y
+ = E′ . (1.25)
X ∂x2 Y ∂y2
On regrouping the terms we obtain the result

1 ∂2 X 1 ∂2 Y
= − + E′ . (1.26)
X ∂x2 Y ∂y2
This equation must hold for all values of x and y. This is only possible if both sides
of the equation are equal to a constant. We call this constant Ex′ . So this equation then
separates into two equations
1 d2 X
= Ex′ (1.27)
X dx2
and
1 d2 Y
− 2
+ E′ = Ex′ (1.28)
Y dy
with Ex′ being called the separation constant. Note that now partial derivatives do
not appear in the equation since the former equation is only a function of x, while
the latter is dependent only on y. The latter of the two equations can be rewritten as

1 d2 Y
= E′ − Ex′ = Ey′ . (1.29)
Y dy2
In each dimension the differential equation that needs to be solved is identical to that
solved for a particle in a one-dimensional box so that the solutions and energies can
be written by analogy as

Xnx = A x sin(n x πx/L x ) with Enx = n2x h2 /(8mL2x ) (1.30)

13
 1 Schrödinger equation: Some exactly solvable cases

and similarly for the y-dimension. We note that the eigenfunction of the two-dimensional
problem is a product of the eigenfunctions in the x and y dimensions

2
ψ( x, y) = sin(n x πx/L x ) sin(ny πy/Ly ) (1.31)
L x Ly

while the energies are additive

h2  
E = Ex + Ey = (n x /L x )2 + (ny /Ly )2 . (1.32)
8m
Example 1.1.4. Extend this argument to the case of a particle in a 3D box.
Solution Note that you will have to do this variable separation in two steps. Let us
say that you separate out the x equation first and the separation constant is Ex′ . At
this stage an equation containing y and z would be equal to E′ − Ex′ . Now you could
go ahead and separate the y equation by an analogous procedure with a separation
constant, say Ey′ . Thus finally you will end up with the following three equations

1 d2 X
= Ex′ ,
X dx2
1 d2 Y
= Ey′ ,
Y dy2
and
1 d2 Z
2
= E′ − Ex′ − Ey′ = Ez′ .
Z dz
◀
Exercise 1.1.2. Show that cos ax cos by cos cz is an eigenfunction of the Laplacian operator

∂2 ∂2 ∂2
 
2
∇ = + + .
∂x2 ∂y2 ∂z2

14