A 1953 Light-From-Electrons Effect Was Hiding a Twist: Plain Silicon Now Reveals It
Singapore and Chinese physicists found that Smith-Purcell radiation is inherently chiral, and measured a record handedness of 0.87 in Nature Communications.
A famous piece of 1950s physics has turned out to have a property nobody thought to check. Researchers in Singapore and China report that the light emitted when an electron beam skims past a grating is inherently chiral, meaning it has a handedness, and that a plain silicon grating is enough to bring it out.
In 1953, two physicists showed that an electron beam passing close to a grating gives off light whose color depends only on the electron's speed and the grating's spacing. The effect, known as Smith-Purcell radiation, has since become a tool for building compact, tunable light sources and for tracking particle beams without touching them.
The new work was published in Nature Communications under the title "Observation of inherent chiral Smith-Purcell effect via symmetry breaking." Associate professors Wu Lin and Dong Zhaogang of the Singapore University of Technology and Design led the study. They found that the handedness was always there but was hidden by symmetry. "The left- and right-handed components are distributed symmetrically in space," Wu said. "In conventional measurements, we usually look only along the symmetry plane, so the two opposite contributions cancel each other."
A moving electron drags a twisting electromagnetic field around it, and a grating can convert that field into circularly polarized light. In three-dimensional simulations, the emission is left-handed on one side of the beam, right-handed on the other, and linearly polarized in the middle. The middle is the only place earlier two-dimensional models looked.
To tip the balance, the team broke the symmetry in two ways. In one, they rotated the grating relative to the electron beam. In the other, they kept the grating fixed and moved the beam from the center toward one edge. Both were tested on silicon gratings inside a scanning electron microscope, with the emitted light sorted into left- and right-handed parts. "Twisting lets us tune the chirality, while shifting lets us select a stable handedness over a broad wavelength range," Dong said.
On a scale where 1 means purely one-handed light and 0 means none, the measured value climbed with twist angle to 0.726 at 30 degrees, and reversing the twist flipped the handedness. Shrinking the grating pitch from 120 nanometers to 100 pushed it to 0.87, the highest yet reported for Smith-Purcell radiation and more than double the previous best. Moving the beam to one edge gave one handedness along that edge and the opposite along the other, and simulations predict values as high as 0.98.
The shifting experiments held a surprise. The emission was strongest with the electron beside the grating rather than above it, because the grating's sharp corners scatter the electron's field into light so efficiently. Because neither method relies on a delicate optical resonance, the effect is broadband. The light could be tuned from about 450 to 900 nanometers while the chirality changed only slightly. "The key insight was that we did not need to make the silicon grating itself chiral," Dong said. "The chirality was already hidden in the electromagnetic field carried by the moving electrons."
Wu sees chiral light sources as the nearest application, with extreme ultraviolet light, which is prized in chip lithography, an attractive target. Obstacles remain. Some materials glow under electron bombardment and can swamp the signal, and reaching X-ray wavelengths would require much smaller structures or far higher electron energies, where quantum effects may complicate the picture.
