Quantum Dots, Wires, and Wells

Squeeze an electron and it pushes back

The previous guide showed one way the nanoscale ambushes you: the surface-to-volume ratio runs away, so surface energy starts to rule — faceting, melting-point depression, a slight lattice contraction. That was a story about the atoms at the edge of a nanocrystal. This guide is about a second, completely separate nanoscale surprise, and it happens to the electrons inside. Shrink a crystal small enough and the electrons stop behaving like a bulk sea and start behaving like a handful of notes on a string. The name for this is quantum confinement, and it is one of the cleanest examples in all of materials of the structure-property relationship: change nothing but the size, and a property — here the colour of light the crystal emits — changes with it.

The picture underneath is the oldest trick in quantum mechanics: a particle in a box. An electron is not a tiny billiard ball but a wave, and a wave trapped between two walls a distance L apart can only fit as a standing wave — a whole number of half-wavelengths, exactly like a guitar string or an organ pipe. That restriction forces the energy to come in a ladder of allowed rungs, E_n = h^2 n^2 / (8 m L^2), where n = 1, 2, 3, ... labels the rungs and h is Planck's constant. Read the L in the bottom: the energy of every rung scales as 1 over L squared. Make the box smaller and the rungs fly apart. A big pipe hums a deep, closely spaced bass; a tiny whistle shrieks high notes far apart. An electron in a big crystal has rungs so close they blur into a continuous band; an electron in a 3 nm crystal has rungs you can count.

Zero, one, two: confinement by dimension

You do not have to squeeze all three directions at once, and the three choices give three named structures. The cleanest way to keep them straight is to count the directions the electron is left free to roam. Confine one direction and leave two open and you have a quantum well: a thin flat layer in which electrons still glide freely across the plane but are pinned in thickness — a two-dimensional electron gas, so it is called a 2D structure. Confine two directions and leave one and you have a quantum wire: a thread down which electrons run in a single dimension — 1D. Confine all three and leave none and you have a quantum dot: a box with electrons stuck at discrete energies, free to move in zero dimensions — 0D. The rule is that tidy: the dimensionality of the structure equals the number of directions still open.

These are not just diagrams — each has a real physical form and a real way of being built. A quantum well is quite literally a thin epitaxial layer: a few-nanometre slab of, say, gallium arsenide grown atom-layer by atom-layer between thicker barriers of aluminium gallium arsenide, so tidy that a single extra atomic layer changes the confinement. That method — growing one crystal in perfect registry on another — is epitaxy, and stacking many wells and barriers in a periodic sequence builds a superlattice; both are the subject of the next guide. A quantum wire is a nanowire or an etched thread. And a quantum dot is often nothing other than the nanocrystal you met in guide 1 — grown in a flask as a colloidal particle, or grown by self-assembly as a strained island that pops up on a substrate to relieve misfit. The dot is where this rung's chemistry and its physics meet in one object.

The density of states: hill, staircase, comb

To see why dimensionality matters so much, look at the density of states — the bookkeeping of how many electron states are available at each energy. In a bulk crystal the electrons fill continuous bands; every allowed state is labelled by a wavevector that ranges smoothly across the whole Brillouin zone, and the density of states rises as a smooth square-root hill, proportional to sqrt(E). Confinement does something violent to this. The standing-wave condition allows only a comb of wavevectors, roughly k = n times pi / L, so the smooth range of allowed k is sliced into discrete values. Each slice carries its own sub-band, and the shape of the density of states is reshaped accordingly.

Walk down the dimensions and watch the hill deform. In a 2D well the square-root hill collapses into a staircase — flat within each sub-band, jumping up a step at every new one. In a 1D wire it sharpens into a row of spikes that flare as 1 over sqrt(E) at each sub-band edge (the van Hove singularities). And in a 0D dot the density of states is nothing but a set of sharp, isolated lines — discrete energy levels, exactly like the electron shells of a single atom filled by its valence electrons. That is why a quantum dot is nicknamed an artificial atom: it has designer levels you set by choosing its size, not levels fixed by nature's periodic table. The figure below lays the four shapes side by side.

DENSITY OF STATES  D(E) = electron states per unit energy

  3D  bulk  (electron free in 3)     2D  well  (free in 2)
   D |          ___                    D |     ______
     |       __/                         |     |    ___
     |     _/     ~ sqrt(E)               | ____|    |     staircase:
     |   _/                              ||         |     flat within
     +----------------> E                +----------------> E   each step

  1D  wire  (free in 1)              0D  dot  (free in 0)
   D | |                              D |   |   |    |
     | |       ~ 1/sqrt(E)              |   |   |    |    sharp lines,
     | ||_     spikes at               |   |   |    |    like an atom's
     | |  |__   each edge              |   |   |    |    levels
     +----------------> E              +----------------> E

  fewer free directions  ->  the smooth band breaks into rungs

Tuning colour by size

Here is the payoff, and it is genuinely beautiful. In a semiconductor the lowest empty rung and the highest full rung are separated by the bandgap, and the colour of light the crystal absorbs or emits is set by that gap. Confinement pushes the electron rung up and the hole rung down, so the effective gap of a dot is the bulk gap PLUS a confinement energy that grows as 1 over L squared. A bigger dot adds little and stays near the bulk colour; a smaller dot adds a lot and shifts bluer. So a single material, made into dots of different size, glows across the rainbow: CdSe dots around 6 nm fluoresce deep red, around 2 nm they glow blue-green — no change in chemistry, only in size. This is the structural magic behind quantum-dot displays and fluorescent biological tags.

1. Start from the bulk bandgap of your material — for CdSe about 1.74 eV, a deep red near 710 nm wavelength.
2. Estimate the confinement energy from the particle-in-a-box rung, E_1 = h^2 / (8 m L^2), using the electron's effective mass in the crystal — often around a tenth of the free value, which makes the confinement energy roughly ten times bigger than a naive free-electron guess.
3. Put in numbers: with an effective mass about 0.1 of the free value and L = 4 nm, that lowest rung sits roughly 0.24 eV above the band edge; halve the box to 2 nm and, because energy goes as 1 over L squared, it leaps about fourfold to near 0.94 eV.
4. Add the confinement energy to the bulk gap: the smaller the dot, the larger the added energy, the wider the effective gap.
5. Turn the gap into colour with photon energy equals h times c divided by wavelength — a wider gap means a shorter wavelength, so a smaller dot glows bluer and a larger dot redder. Size alone paints the spectrum.

What is, and isn't, 'two-dimensional'

One trap is worth disarming before you leave this guide, because the words collide. A quantum well is a *2D structure*, but it is not a *2D material*. The well is a thick slab of perfectly ordinary three-dimensionally bonded crystal — dozens or hundreds of atomic layers of GaAs — and only its electron gas happens to be squeezed into two dimensions. A genuine two-dimensional material like graphene, which the next guide but one is devoted to, is a different beast entirely: a sheet literally one atom thick, held to its neighbours above and below by weak van der Waals forces rather than strong bonds. 'Two-dimensional electron gas' describes where the electrons can move; 'two-dimensional material' describes how thick the crystal actually is. Keep them apart.

The three confined structures also point straight ahead to the rest of the rung. Stack many wells and barriers in a designed periodic sequence and you get a superlattice — an engineered crystal with a man-made repeat, whose mini-bands you tailor by choosing the layer thicknesses; guide 3 builds it, along with the epitaxy that grows it and the misfit strain that both plagues and, cleverly used, creates self-assembled dots. Guide 4 takes the van der Waals stacking hinted at just now and runs with it into graphene and its cousins. So confinement is not a side-show: it is the reason nano-structuring is worth the trouble.

Step back and the whole idea is one sentence. At the nanoscale, size and shape graduate from being incidental facts about a sample to being structural design variables in their own right. The lattice stays put — same motif, same unit cell, same reciprocal-space fingerprint you learned to read in the diffraction rungs — but the boundary wrapped around it dictates which electron waves are allowed to live there. Choose the box, and you choose the states. That is the promise of the quantum dot, wire, and well, and the reason a materials scientist now designs in three dimensions of arrangement AND one of size.