The Great Divide: What 'Order' Means
In the previous guide we climbed the ladder of length scales, from the electron cloud up to the microstructure you can hold in your hand. Now plant yourself firmly at the atomic scale and ask one deceptively simple question of any solid: as you step from one atom to the next, is there a rule that tells you where the next atom sits? The answer sorts almost all of matter into a handful of families, and the sorting principle is a single idea — order.
It helps to split 'order' into two ranges. Short-range order is predictability in the immediate neighbourhood — knowing that each silicon atom is bonded to four oxygens at about 1.6 angstrom in a roughly tetrahedral cage. Almost every solid, and even many liquids, has short-range order. Long-range order is far stronger: the predictability survives across thousands of atoms, so that fixing a few atomic positions lets you predict an atom a whole micrometre away. It is the presence or absence of long-range order that draws the great divide.
Crystalline: The Infinite Wallpaper
The crystalline state is long-range order in its purest form: the atoms sit on a pattern that repeats by simple translation, in principle forever. From the earlier rungs you already have the two ingredients — a lattice, the infinite grid of equivalent points (where you stamp), and a motif, the group of atoms attached to each point (the picture on the stamp). Stamp the motif at every lattice point and you have the crystal. Hold onto one honest distinction that trips up beginners: the lattice is not the crystal. The lattice is only points; the crystal is lattice plus motif.
Close packing -- stack spheres so each nests in the hollows below:
layer A o o o o
layer B o o o FCC = A B C A B C ... (a new slot each time)
layer C o o o o HCP = A B A B A B ... (straight back to A)
layer A o o o o
FCC: atomic packing factor 0.74 (74% of space is filled by atoms)
coordination number 12 (each atom touches 12 others)Periodicity carries an iron consequence — an actual theorem, not a rule of thumb. A pattern that repeats by translation can only host rotational symmetry of order 1, 2, 3, 4, or 6. It can never have 5-fold symmetry, for the same reason regular pentagons cannot tile a kitchen floor without leaving gaps. This is the crystallographic restriction theorem, and it is exact. It underlies the famous closed counts you will meet later — exactly 14 Bravais lattices, 32 point groups, and 230 space groups. These are complete enumerations, not tallies that might grow with 'so far'; the lists are finished. File away that 'no 5-fold' ban, because it is what makes section four so startling.
Amorphous and Glassy: Order Without Repetition
Now melt that same silica and cool it too fast for the atoms to find their crystalline slots. They freeze mid-shuffle, and you have the amorphous state — ordinary window glass. Every silicon is still bonded to four oxygens (the short-range order is intact and remarkably sharp), but the linked tetrahedra twist at slightly random angles, so the pattern forgets itself within a nanometre or two. Structurally it is a liquid caught in the act of flowing, its disorder frozen in place — what crystallographers call a continuous random network.
Here is the correction that matters most, because the word misleads almost everyone: amorphous does not mean 'no order'. A glass has crisp short-range order — it simply has no long-range periodicity. The proof is quantitative. A radial distribution function, which counts how many atoms sit at each distance from a chosen one, shows a glass with a tall, sharp first-neighbour peak (a well-defined nearest-neighbour spacing), then broader bumps that smear out and vanish by the third or fourth shell. That is order that fades with distance, not order that was never there. Even metals can be frozen this way into a metallic glass if you quench them fast enough — millions of degrees per second — to outrun crystallisation.
Quasicrystals: The Impossible Middle
For a century the divide looked complete: a solid was either ordered-and-periodic (a crystal) or disordered (a glass). Then in 1982 Dan Shechtman aimed an electron beam at a rapidly cooled aluminium-manganese alloy and recorded a diffraction pattern with sharp spots — the unmistakable signature of long-range order — arranged in a perfect ten-fold star. Sharp spots demand long-range order; ten-fold symmetry (and its five-fold parent) is exactly what the crystallographic restriction theorem forbids to anything periodic. The pattern was, on its face, impossible. Colleagues ridiculed the result for years; Shechtman received the 2011 Nobel Prize in Chemistry for it. The solid was the first known quasicrystal.
The resolution dissolves a hidden assumption we had all made — that long-range order requires a repeating unit. A quasicrystal's atoms are genuinely ordered, their positions fixed by a strict rule, but the rule is not periodic: there is no unit cell you can copy-and-shift to rebuild the whole. The everyday model is the Penrose tiling — two tile shapes (kites and darts, or a fat and a thin rhombus) that cover the floor completely under matching rules, showing gorgeous five-fold symmetry, yet forming a pattern that never once repeats. Slide it any distance and it will never lie back on top of itself. That is aperiodic order: as rigidly ordered as a crystal, as non-repeating as a glass, and belonging fully to neither.
Quasicrystals turned out to be more than a laboratory freak — a natural one, icosahedrite, was later found inside a meteorite. The deeper lesson reshaped the field. In 1992 the International Union of Crystallography quietly rewrote the definition of 'crystal' to mean any solid with an essentially sharp diffraction pattern, deliberately dropping the requirement of periodicity. So the true dividing line was never periodicity at all — it is long-range order, which we read directly off the sharpness of the diffraction. Three families fall out cleanly: periodic long-range order (the ordinary crystal), aperiodic long-range order (the quasicrystal), and no long-range order (the glass).
Reading the Fingerprint: Diffraction Sorts Them
We keep leaning on diffraction, so here is a first glimpse of why it reads out order (the next guide develops the machinery). Send a wave whose wavelength is near the atomic spacing — X-rays, electrons, or neutrons at roughly 1 angstrom — into the solid. Evenly spaced planes of atoms behave like a row of equally spaced cliff walls throwing back echoes; the echoes reinforce only at special angles where the extra path is a whole number of wavelengths. That condition is Bragg's law. A quick number: for planes spaced d = 2 angstrom probed with copper K-alpha radiation, lambda = 1.54 angstrom, the first peak sits at theta = arcsin(1.54 / (2 x 2)) = arcsin(0.385) = 22.6 degrees.
- Send a single-wavelength (monochromatic) beam through the sample and record where the scattered intensity lands.
- Sharp, well-defined spots or rings? Then long-range order is present — you are looking at a crystal or a quasicrystal, not a glass.
- Only a broad, diffuse halo and no sharp peaks? There is no long-range order — the solid is amorphous or glassy.
- Sharp spots, but arranged with 5-, 8-, 10-, or 12-fold symmetry that no periodic lattice is allowed? That is the quasicrystal fingerprint.
- Sharp spots sitting on a periodic grid you can index with a unit cell? An ordinary crystal — and the peak positions give you the cell size, while the peak intensities encode the motif inside it.
One honest caveat before we move on. Diffraction measures the intensity of each scattered beam — the square of a quantity called the structure factor, |F|^2 — but it throws away that wave's phase. The lost phase is precisely the information you would need to rebuild the atomic positions in one step, which is why turning intensities back into a structure is a genuine puzzle with its own name, the phase problem. And diffraction is only half the toolkit: microscopy lets us look at grains, phases, and defects directly, which the final guide of this rung takes up. For now, hold the summary: the positions of the peaks tell you the size and shape of the repeating cell; the intensities of the peaks tell you the motif sitting inside it; and whether the peaks are sharp at all tells you which of the three great families you are holding.