the screw axis

Picture a spiral staircase or the thread on a screw. To climb the stairs you must both turn AND rise at the same time — a plain rotation would just leave you spinning on the spot. A screw axis is exactly this kind of combined move: it rotates the pattern by a set angle and, in the same breath, slides it along the axis by a fixed step. It is a symmetry that literally cannot exist in a finite object like a molecule; it only makes sense in an endless, repeating crystal where there is always more pattern ahead to land on.

The symbol is written n_m (read 'n sub m'). The n tells you the rotation: turn by 360/n degrees. The subscript m tells you how far to slide: a fraction m/n of the lattice repeat along the axis. Take a 2_1 screw along the b direction: rotate by 180 degrees and slide half a cell. In coordinates a point at (x, y, z) is carried to (-x, y+1/2, -z). Do it a second time and you get (x, y+1, z) — the same orientation you started with, moved one full cell along b, which is just an ordinary lattice translation. That self-consistency is exactly why only certain screws (2_1; 3_1, 3_2; 4_1, 4_2, 4_3; 6_1 through 6_5) are allowed. The pairs like 3_1 and 3_2, or 4_1 and 4_3, are left- and right-handed versions of each other, like the two twists of a corkscrew.

Screw axes are everywhere in real materials. Hexagonal close-packed metals (magnesium, titanium, zinc) sit in the space group P6_3/mmc, whose 6_3 screw is what stitches the ABAB stacking together. The single most common space group for organic molecular crystals, P2_1/c, is named for its 2_1 screw. And because a screw carries a fractional translation, it leaves a fingerprint in a diffraction pattern: a 2_1 axis along b wipes out every 0k0 reflection with k odd — a systematic absence explored fully in the diffraction fields.