At a glance
Shake one end of a chain of alternating light and heavy atoms and ask what travels down it. This sim shows how the answer depends on the drive frequency and on the two masses, in the simplest model of a vibrating crystal.
What you'll learn
- How the dispersion relation ω(k) links the drive frequency to the wave on the chain.
- How neighbouring atoms move relative to each other, and how that depends on frequency.
- How the masses and the spring constant set the frequencies at which the chain's behaviour changes.
- How to read k, λ and UB/UA off the Drive & marker panel.
What's on screen
- Atom A (blue) is always 4 u.
- Atom B (orange) has the mass set by its slider.
- Displacement trace u(x): how far each atom sits from rest, right now. The shaded right end absorbs the wave.
- Lower curve (teal) of the ω(k) plot: f in THz against k across the zone −π/a to +π/a.
- Upper curve (amber): the second branch.
- Drive line at the shaker's frequency, and a marker where it meets a curve. The marker's k gives the wavelength.
The controls
Press ⌖ on a row to see where that control sits in the sim.
| Control | What it changes | Range | |
|---|---|---|---|
| Drive frequency f | The frequency at which the left end is shaken. It moves the dashed drive line. | 0.2–14 THz opens at 2 THz | |
| Mass of atom B | The mass of every B atom. Atom A stays at 4 u. Both curves reshape. | 1–16 u opens at 12 u | |
| Spring constant K | The stiffness of every spring. | 5–30 N/m opens at 15 N/m | |
| Readouts | Drive & marker: f, which curve the drive meets, k, λ and UB/UA. A second group, Band readout, appears once the drive first lands between the two curves. | panel | |
| ▶ Play / ⏸ Pause | Runs the chain, or freezes it so you can read the trace. | top bar | |
| Speed | Plays the motion faster or slower on screen. The physics is unchanged. | 0.25×–4× | |
| ↻ Reset | Restarts the chain from rest. The sliders keep their values. | top bar |
Fixed: atom A = 4 u; cell length a = 0.5 nm; springs join nearest neighbours only; the right end of the chain absorbs.
Two branches, not one
A unit cell holds two atoms, so each wavevector k allows two ways for the chain to vibrate. They are the two roots of the dispersion relation, one for each sign:
- The lower (acoustic) branch goes to ω = 0 at k = 0. Over a long wavelength the whole cell moves together, which is sound.
- The upper (optical) branch never reaches zero. At k = 0 it sits at ω = √(2K/μ), where μ = mAmB/(mA+mB) is the reduced mass.
- The plot covers one zone, −π/a to +π/a. A wavevector outside it describes atom motion already drawn inside.
What the atoms do on each branch
The Drive & marker panel prints the ratio of the two atoms' displacements. Its sign tells the branches apart:
- Lower branch: the ratio is positive, so the two atoms of a cell move the same way.
- Upper branch: the ratio is negative, so they move against each other. At k = 0 it equals −mA/mB, which means mAUA + mBUB = 0: the centre of mass stays still and the lighter atom swings more.
- That k = 0 mode is a two-body oscillator with stiffness 2K and reduced mass μ, which is smaller than either mass. That is why the upper branch sits high.
Frequencies with no wave
Some drive frequencies meet neither curve: those between the two branches, and those above the top of the upper one. No real k solves the dispersion relation there, so no travelling wave exists. The solution has a complex k:
- The motion shrinks by e−κa from each cell to the next. At 8 THz on the opening chain that is ×0.35 per cell, so it is under 5% of the drive by the third cell.
- The equations are linear, so the decay per cell does not depend on the drive strength. Shaking harder does not make a wave travel.
- Above the upper branch the motion also dies out, here by ×0.23 per cell at 13 THz.
Where the branches end
Both band edges sit at the zone edge k = π/a. There one sublattice stands still, and each moving atom hangs between two motionless neighbours on two springs, so ω = √(2K/m).
- On the lower branch's last point only the heavy atoms move. On the upper branch's first point only the light atoms move.
- At the opening values the edges are 6.18 and 10.70 THz. The upper branch peaks at 12.35 THz, at k = 0.
Mass contrast sets the width
- The edge that belongs to atom A, √(2K/mA), does not move when you change mB. The other edge travels.
- The ratio of the two edges is √(mheavy/mlight). It is 1 only when the masses are equal, and then the band closes: the chain is a single chain of identical atoms, spacing a/2, and the upper branch is the lower one folded back at k = π/a.
- K multiplies every frequency by √K. Doubling it widens the band by √2 but leaves the edge ratio alone: stiffness shifts the band, mass contrast sizes it.
These move the sim's own sliders. Atom A stays at 4 u.
Where it matters
- Sound in solids. The slope of the lower branch at k = 0 is the speed of sound. At the opening values vs = 8.4 km/s, the right order of magnitude for a solid.
- Infrared absorption in ionic crystals. In a crystal like NaCl the two ions carry opposite charge, so the upper branch's antiphase motion is an oscillating dipole that infrared light can drive. A photon's k is tiny on this scale (about 4×10⁻⁵ of π/a at 12.35 THz in vacuum), so light couples only near k = 0.
- Heat. Quantise these same vibrations and each energy step is a phonon. Phonons carry heat in insulators.
Go deeper: what this model leaves out
- It is one-dimensional, with nearest-neighbour springs only. Real crystals have three dimensions and longer-range forces.
- There is no long-range Coulomb force, so it does not show the LO–TO splitting of a real ionic crystal.
- The motion is classical. The quantum statement is only that each mode's energy comes in steps of ħω.
Summary
Key points
- The lower branch goes to 0 at k = 0; the upper starts at √(2K/μ).
- Cell partners move together on the lower branch and against each other on the upper.
- A frequency that meets no branch gives complex k, and a decay of e−κa per cell at any drive strength.
- K scales every frequency by √K; mass contrast sets the edge ratio.
Common mix-ups
- "Frequencies in the band get through, only weaker." No real k exists there. The motion decays within a few cells at any amplitude.
- "More mass contrast closes the band." It widens it. Equal masses close it.
- "Above the upper branch the wave travels again." There is no mode there either, so k is imaginary.
Things to try on your own
- Set K = 30 N/m (mB = 12 u). Divide each band edge by its opening value, 6.18 and 10.70 THz, and compare with √2.
- Set mB = 4 u and sweep f from 0.2 to 14 THz. Note where the marker moves from the lower curve to the upper one.
- Set mB = 1 u and drive at 14 THz. Use the Band readout to explain the trace.