The Infinite Square Well
Take a single electron and trap it. Not loosely — box it into a tiny region a few nanometres wide,
with walls so steep the electron simply cannot get out. Physicists do exactly this inside a
quantum dot, a speck of semiconductor so small that the electrons rattling around
inside it behave like a textbook problem come to life. And the textbook problem is the one on this
page: the infinite square well, the very first fully solvable model in quantum
mechanics, and the one every physicist meets on day one of confinement.
The setup is deliberately austere. A particle of mass m is confined to a
one-dimensional stretch of length L. Inside, it feels nothing — the
potential is flat and zero. At the two edges the potential leaps to +\infty,
an unclimbable wall. That is the whole apparatus. Yet from this bare box tumble out three of the most
important ideas in all of quantum theory: the particle's energy can only take certain
discrete values, its lowest possible energy is not zero, and the states it
lives in are standing waves. One box, taught thoroughly, and you understand why the
quantum world comes in rungs.
V(x) = \begin{cases} 0, & 0 < x < L \\ +\infty, & x \le 0 \ \text{or}\ x \ge L. \end{cases}
Boundary conditions force standing waves
Inside the well the potential is zero, so the time-independent
Schrödinger equation
is just that of a free particle:
-\frac{\hbar^2}{2m}\frac{d^2\psi}{dx^2} = E\,\psi, \qquad 0 < x < L.
The general solution is a mix of sines and cosines, \psi(x) = A\sin(kx) + B\cos(kx),
with k = \sqrt{2mE}/\hbar. On its own that allows any energy at all. What
pins everything down are the walls. The wavefunction must be continuous, and outside
the well — where the potential is infinite — the particle simply cannot be, so \psi = 0
there. Continuity then demands that \psi vanish exactly at both edges:
\psi(0) = 0 \qquad\text{and}\qquad \psi(L) = 0.
The first condition kills the cosine (\cos(0) = 1 \ne 0), leaving
\psi(x) = A\sin(kx). The second condition, \sin(kL) = 0,
is the crucial squeeze: a sine is zero only at whole multiples of \pi, so
kL is not free to be anything — it must be
n\pi for a whole number n:
k_n = \frac{n\pi}{L}, \qquad n = 1, 2, 3, \dots
This is exactly the same maths as a guitar string clamped at both ends: only those wavelengths that
fit a whole number of half-humps between the walls are allowed. The particle can only exist as one of
a discrete ladder of standing waves. Normalising so the total probability is
1 (setting \int_0^L |\psi|^2\,dx = 1) fixes
A = \sqrt{2/L}, and we have the stationary states:
-
The allowed wavefunctions are standing waves
\displaystyle \psi_n(x) = \sqrt{\tfrac{2}{L}}\,\sin\!\left(\frac{n\pi x}{L}\right),
for n = 1, 2, 3, \dots
-
The allowed energies are
\displaystyle E_n = \frac{n^2\pi^2\hbar^2}{2mL^2} = n^2 E_1, so they
grow as the square of n.
-
The standing-wave wavelength is
\lambda_n = \dfrac{2L}{n} — a whole number of half-wavelengths spans the
box.
Seeing the standing waves
The formulae are clean, but the picture is cleaner still. Below is the wavefunction
\psi_n(x) plotted across the box. Drag the slider to climb the ladder of
states and watch what happens: each step up in n squeezes in one more
half-wave, and the curve wiggles faster and faster while always returning obediently to zero at both
walls.
Two counts are worth burning into memory. A node is an interior point where
\psi_n = 0 (not counting the walls); state n has
exactly n - 1 of them. An antinode is a peak of the wave;
state n has n of them. The ground state
n = 1 is the gentlest possible wave that still fits: one hump, zero interior
nodes.
What the particle actually does is governed not by \psi_n but by its square,
the probability density |\psi_n(x)|^2. Switch to that
picture below.
This is where quantum intuition parts ways with classical. A marble rattling in a box would be found
with equal probability anywhere inside it. The quantum particle is nothing of the sort: in the ground
state it strongly prefers the middle and avoids the walls, and in excited states it clumps into a
landscape of n hills and n - 1 forbidden valleys
where it is never found at all.
The energy ladder: En ∝ n²
Because every allowed energy is E_n = n^2 E_1, the whole spectrum is
fixed once you know the single number
E_1 = \frac{\pi^2\hbar^2}{2mL^2}.
Every other level is just a whole-number-squared multiple of it:
E_2 = 4E_1, E_3 = 9E_1,
E_4 = 16E_1, and so on. Because of that n^2,
the rungs do not sit evenly spaced; they spread apart as you climb. The gap between
neighbouring levels is
E_{n+1} - E_n = \big[(n+1)^2 - n^2\big]E_1 = (2n+1)\,E_1,
which grows without limit. The picture below draws the ladder to scale — notice the widening gaps,
the opposite of the crowding levels you saw in the hydrogen atom.
The dependence on L is just as striking. Because E_1 \propto 1/L^2,
shrinking the box shoots every energy up. Halve the width and every level quadruples.
This is not a mathematical accident but a real, measurable effect: it is precisely why smaller
quantum dots glow with bluer (higher-energy) light than larger ones. Squeezing a quantum particle costs
energy — a kind of "quantum pressure" that resists confinement, and the ultimate
reason matter doesn't simply collapse.
Worked examples
Example 1 — energy ratios. An electron in a box sits in the ground state with energy
E_1. It absorbs a photon and jumps to the n = 3
state. How much larger is its energy now? Since E_n = n^2 E_1,
\frac{E_3}{E_1} = \frac{3^2 E_1}{E_1} = 9.
Its energy is nine times the ground-state value. And the energy of the absorbed photon
is the gap, E_3 - E_1 = 9E_1 - E_1 = 8E_1.
Example 2 — the standing wavelength. A box is L = 1.0\ \text{nm}
wide and the particle is in the n = 4 state. What is the wavelength of the
standing wave? Using \lambda_n = 2L/n,
\lambda_4 = \frac{2 \times 1.0\ \text{nm}}{4} = 0.5\ \text{nm}.
Exactly four half-wavelengths (4 \times 0.25\ \text{nm} = 1.0\ \text{nm})
fit inside the box — the standing-wave condition made visible.
Example 3 — counting nodes. How many interior nodes does the
n = 5 state have, and how many antinodes? Nodes: n - 1 = 4.
Antinodes (peaks): n = 5. The wavefunction crosses zero four times between the
walls and bulges into five humps.
Example 4 — shrinking the box. A particle's box is squeezed from width
L to width L/2. By what factor does the ground-state
energy change? Since E_1 \propto 1/L^2, halving L
multiplies E_1 by (1/(L/2)^2)/(1/L^2) = 4. Every
level jumps to four times its old value — the quantum pressure of tighter confinement.
Because a particle sitting perfectly still would violate the
uncertainty
principle. Confining it to a region of width L pins down its
position to within \Delta x \sim L, which forces its momentum to be
uncertain by at least \Delta p \gtrsim \hbar/L. A spread in momentum means
the particle must carry some kinetic energy — roughly
(\Delta p)^2/2m \sim \hbar^2/2mL^2, which is exactly the scale of
E_1. This irreducible leftover is the zero-point energy.
It is why liquid helium never freezes solid under its own vapour pressure however cold you make it, and
why matter has a size at all: the electrons in an atom cannot collapse to the nucleus, because
squeezing them into a smaller box costs more energy than they can lose. Absolute stillness is simply
not on the quantum menu.
Two classic traps. The first: it is tempting to start counting states at
n = 0, the way you might for many things in physics. But
n = 0 gives \psi_0(x) = \sqrt{2/L}\,\sin(0) = 0
everywhere — a wavefunction that is zero across the whole box describes no particle at
all (the total probability would be zero, not one). So n starts at
1, the lowest energy is E_1 \ne 0, and there is no
state of zero energy. (Negative n just flips the sign of the same wave and
adds nothing new.)
The second trap: assuming the particle is equally likely to be found anywhere in the box, the way a
classical marble would be. Not so. The probability density
|\psi_n|^2 = (2/L)\sin^2(n\pi x/L) is bumpy: peaks at the
antinodes, and exact zeros at the nodes where the particle is never found. The flat classical
distribution only emerges as an average in the limit of very large n — a
glimpse of the correspondence principle, quantum weirdness fading into classical smoothness only when
the quantum numbers get huge.