Normalization and Probability Densities in the 1D Box

The normalized particle-in-a-box density is not a museum piece — it is the working model behind real spectroscopic and materials predictions. Cyanine dyes, in which a chain of conjugated carbons connects two nitrogen-containing end groups, are treated as one-dimensional boxes for their $\pi$ electrons; the normalized $|\psi_n|^2$ densities determine transition dipole moments that set absorption wavelengths used in photographic sensitizers, laser dyes, and biological fluorescent stains. Quantum dots — nanoscale semiconductor crystals — behave as three-dimensional analogs of the particle in a box, and the same normalization logic (adapted to three Cartesian directions) determines the electron and hole probability densities that control quantum-dot emission color, now commercially exploited in QLED displays and biomedical imaging labels. Carbon nanotubes and graphene nanoribbons, when short enough to show quantum confinement along their length, are modeled with box eigenfunctions to predict where along the ribbon the electron density concentrates, informing molecular electronics designs. Even in introductory computational chemistry courses, students verify Hartree-Fock or density-functional-theory orbital shapes for simple confined systems by comparing them qualitatively to the nodal patterns of $|\psi_n|^2$ from the exactly solvable box, since the box gets the topology (node count, density humps) right even where the quantitative energies differ from a real molecular potential.

Ask the tutor: How would the normalization constant change if the box had "soft" (finite-height) walls instead of infinite ones?

Start from what is already known: solving $-\dfrac{\hbar^2}{2m}\dfrac{d^2\psi}{dx^2} = E\psi$ inside an infinite square well of width $L$, with $\psi(0) = \psi(L) = 0$, gives the family of standing waves

Nothing in this differential equation constrains $A$. If $\psi_n$ solves the Schrödinger equation, so does $2\psi_n$, $-7\psi_n$, or $i\psi_n$ — the equation is linear and homogeneous, so any scalar multiple of a solution is equally valid as a solution to the differential equation. But Max Born's probabilistic interpretation of the wavefunction demands something the differential equation does not: that $|\psi_n(x)|^2\,dx$ be the probability of finding the particle in the infinitesimal interval $[x, x+dx]$. Probabilities over the entire accessible region must sum (integrate) to exactly 1, since the particle is somewhere inside the box with certainty. That single physical requirement,

is the normalization condition, and it is exactly one equation for exactly one unknown, $A$. It fixes $|A|$ completely; the phase of $A$ (an overall sign or complex phase) remains physically irrelevant because it cancels in $|\psi_n|^2$ and in every observable built from bilinear combinations of $\psi$. Once $A = \sqrt{2/L}$ is found (worked in full below), the object of real chemical interest is the probability density

This function is never negative — it is a squared quantity — and it integrates to 1 over the box by construction. It has $n$ humps (regions of high probability) separated by $n-1$ interior nodes where the probability density vanishes exactly, inherited directly from the nodes of $\psi_n$ itself. For $n=1$, the density is a single smooth hump peaking at the box center, already telling you the ground-state particle is most likely found in the middle and never at the walls. For $n=2$, the density splits into two humps with a node exactly at the center — a purely quantum feature with no classical analog, since a classical bouncing ball has some chance of being caught anywhere, including the midpoint. The chemical payoff is immediate: $\pi$ electrons in a conjugated chain, modeled as one-dimensional box particles, occupy specific $\psi_n$ orbitals, and the shape of $|\psi_n|^2$ tells you where electron density piles up along the chain — directly relevant to reactivity, spectroscopic transition moments, and the qualitative shape of frontier molecular orbitals in extended $\pi$ systems.

The three relationships that carry all the chemistry are the normalization integral, the normalized eigenfunction, and the resulting probability density.

where: $\psi_n(x)$ is the (as-yet unnormalized) eigenfunction, $L$ is the box length, and the integral runs over the entire physically accessible region since $\psi_n = 0$ outside the box. Reading: this is not an equation to solve for $x$ — it is a constraint equation for the leftover constant multiplying $\sin(n\pi x/L)$.

where: $n$ is the quantum number indexing energy levels and node count, $L$ is the box width in the same length units used everywhere else (commonly picometers or ångströms for molecular applications), and $\sqrt{2/L}$ is the universal normalization constant — the same for every $n$, which is a special convenience of the particle-in-a-box (not true for every quantum system). Reading: this is now a complete, physically meaningful wavefunction, ready to be squared, integrated, differentiated, or fed into any expectation-value formula.

where: $x_1, x_2$ bound the sub-interval of interest inside $[0,L]$, and the integrand is manifestly non-negative everywhere. Reading: this is the master formula for "what fraction of the time is the particle found in this region" — it reduces to 1 when $x_1=0, x_2=L$ and can be evaluated in closed form using the half-angle identity.

Example 1 — Foundational: normalizing $\psi_2$ for an electron in a 2.50 Å box and evaluating it at a specific point. An electron is confined to a one-dimensional box of length $L = 2.50\text{ Å} = 2.50\times 10^{-10}\text{ m}$. Write the normalized $n=2$ wavefunction and evaluate $|\psi_2(x)|^2$ at $x = 0.625\ \text{Å}$ (one quarter of the way along the box — deliberately not the midpoint, to keep the arithmetic honest). From the derivation, the normalization constant is universal:

So

At $x = 0.625$ Å $= 0.625\times 10^{-10}$ m, the argument of the sine is

So $\sin(\pi/2) = 1$ exactly, giving

and the probability density there is

Sanity check via the direct formula: $|\psi_2|^2 = (2/L)\sin^2(\pi/2) = (2/2.50\times10^{-10})(1) = 8.00\times10^9\ \text{m}^{-1}$, matching exactly. Note this quarter-point is in fact the location of the first interior maximum of the $n=2$ density (since $\sin^2$ peaks when its argument is $\pi/2$), which is a coincidence of choosing $x = L/4$ for $n=2$ — for other $n$ or other fractions, the value would differ, illustrating that reading off $|\psi_n|^2$ requires the actual quantum number and position, not a generic guess.

Example 2 — Applied: probability an electron sits in the "middle third" of a conjugated-chain box. Model a linear conjugated chain (e.g., a fragment of a polyene) as a box of length $L = 8.40$ Å, chemically realistic for a chain of about seven or eight conjugated carbons. Find the probability that a $\pi$ electron in the $n=1$ HOMO-type level is found in the middle third of the box, i.e., between $x_1 = L/3 = 2.80$ Å and $x_2 = 2L/3 = 5.60$ Å. Using the closed-form result from the second derivation with $n=1$:

The linear term:

The sine arguments: for $x_2 = 5.60$ Å,

so $\sin(240.0^\circ) = -0.8660$. For $x_1 = 2.80$ Å,

so $\sin(120.0^\circ) = +0.8660$. Substituting:

So there is about a 60.9% chance of finding the ground-state $\pi$ electron in the central third of the chain — substantially more than the naive classical estimate of $1/3 \approx 33.3\%$, precisely because the $n=1$ density is humped toward the center. This is the quantum-mechanical basis for saying that ground-state $\pi$ electrons in short conjugated systems are preferentially delocalized toward the middle of the chain, a qualitative fact used when rationalizing reactivity patterns (e.g., electrophilic attack tendencies) in simple Hückel-adjacent box-model reasoning, and foreshadows why average positions in the Box Model for Conjugated Molecules are computed with exactly this density.

Example 3 — Exam/competition level: probability in a narrow off-center slice at large $n$, and the classical limit. Consider $n = 12$ in a box of length $L = 5.00$ Å. Find the probability of locating the particle in the narrow window $x_1 = 3.00$ Å to $x_2 = 3.20$ Å (a slice of width $0.20$ Å, about 4% of the box, deliberately off-center and not aligned with any obvious symmetry point). Compare the exact quantum answer to the classical (uniform-density) estimate. The classical estimate simply uses width divided by box length:

The exact quantum probability uses the closed form with $n=12$:

Compute the sine arguments. For $x_2 = 3.20$ Å:

Reduce modulo $2\pi$: $15.36\pi = 7(2\pi) + 1.36\pi$, so the effective angle is $1.36\pi\ \text{rad} = 244.8^\circ$, giving $\sin(244.8^\circ) = -0.9048$. For $x_1 = 3.00$ Å:

Reduce modulo $2\pi$: $14.40\pi = 7(2\pi) + 0.40\pi$, so the effective angle is $0.40\pi = 72.0^\circ$, giving $\sin(72.0^\circ) = +0.9511$. Now assemble:

So the exact quantum probability is 6.46%, versus the naive classical 4.00% — a substantial 62% relative enhancement, because this particular slice happens to straddle a probability maximum of the rapidly oscillating $n=12$ density. If the slice had instead straddled a node, the quantum probability would fall below 4.00%, potentially close to zero. This is the key exam-level subtlety: at large but finite $n$, the oscillation term $\dfrac{1}{2n\pi}[\ldots]$ does not vanish for an arbitrary narrow window — it only becomes negligible on average, or for windows much wider than one oscillation period $L/n$. Here the window width $0.20$ Å is comparable to the oscillation period $L/n = 5.00/12 = 0.417$ Å, so the oscillation correction ($+0.0246$) is not small compared to the classical term ($0.0400$) — a genuinely quantum correction survives. Only when window widths are averaged over many periods, or when $n\to\infty$ with the window fixed, does $P \to (x_2-x_1)/L$ exactly recover the classical uniform density, since the oscillatory term is then bounded by $1/(n\pi)$ and vanishes in that limit for any fixed window.