In the limelight. Interview by Kosmas Tsakmakidis.

■ ■ In 1987 you started working on the idea of a photonic crystal. What made you consider such a problem? At that time, some papers had appeared suggesting it was possible to suppress spontaneous emission (SE). I think the first paper was by Dan Kleppner, an atomic physicist from MIT, who suggested putting an atom between two metal plates1. However, that covers only 50% of the SE, that is, it suppresses one polarization but not the other. Then a series of papers appeared where instead of using a metallic reflector, use was made of one-dimensional (1D) Bragg gratings. I looked at that and said: “That can’t possibly be right” because you cannot suppress SE in that way — it will simply go sideways, propagating parallel to the planes, and therefore there would be little benefit by using a 1D Bragg reflector. I asked myself how to convert a 1D reflector into a 3D Bragg reflector, and started drawing lines on a page. They were just cross-hatched lines. Staring at them, the lines ‘turned’ into a checkerboard. I visualized a checkerboard in two dimensions. I then realized that a checkerboard extended in the third dimension is a face-centred cubic crystal — and that’s what started it.