NANOTECHNOLOGY: Photonic crystal aids quest for visible plasmon laser

Researchers from the New Jersey Institute of Technology (NJIT; Newark, NJ) have reported the first demonstration of laser threshold, gain, spectral line-narrowing, and feedback in the visible spectrum from a surface-plasmon polariton (SPP) light emitter. The research team’s leader, Haim Grebel, expects fabrication of the first visible SPP laser to be a straightforward process based on refinement of the experimental techniques.¹

Surface-plasmon polaritons are electromagnetic modes that exist at a metal-dielectric interface, where light trapped by the metal’s resonantly oscillating free electrons travels adjacent to the metal surface within the dielectric. Even though the theory of SPPs is more than a century old, the field of plasmonic optics–photonic devices based on practical applications of the theory–has only begun to emerge over the last decade or so, enabled by the concurrent growth and development of nanoscience and technology.^2,3

Waveguides based on SPP modes acting over centimeter-scale distances have been demonstrated, but such propagation distances, along with attempts at producing a laser, have been limited by absorption losses in the metal. One of the primary steps that the New Jersey researchers took in achieving laser gain was to replace the dielectric material at the metal-dielectric interface with a two-dimensional subwavelength-pitch photonic-crystal layer made of anodized aluminum oxide.

Choosing dye

Instead of pumping light into the edge of the dielectric material to interact with free-surface electrons along the route of longitudinal travel, Grebel’s group pumped incident light downward onto the patterned surface, using it as a diffraction grating to disperse the light horizontally along the metal surface in a standing wave, thus creating the laser cavity. Another key step involved the selection and placement of the lasing medium. While earlier efforts tended to focus on quantum dots as a laser medium, Grebel’s team chose the fluorescent dye fluorescein.

“This is a very strong fluorescent dye, in general stronger than quantum dots,” he said. “It’s also more amenable to sitting at the top of the holes in the photonic-crystal structure. The strongest electric field is not at the metal surface, but on top of the photonic-crystal structure. So that’s where the dye will experience the largest effect.” Quantum dots have a tendency to fall into the holes and lie closer to the metal, where their fluorescence would be suppressed, he added.

An experimental device based on these concepts yielded a laser threshold at 5 mW averaged pump intensity when pumped by a frequency-doubled, 532 nm, 10 Hz Nd:YAG laser at 10 ns pulse widths. The spectral line of the fluorescence signal narrowed from 35 to 24 nm–a 30% reduction, consistent with a gain times the cavity length factor of three, according to the researchers (this translates to a gain factor of more than 10,000 cm^-1).

The next step is to make better confining structures, and perhaps to use quantum dots as gain media, Grebel said. Quantum dots have a longer lifetime than dyes and may provide enough gain to overcome losses when placed at the top of the holes, as was the fluorescein dye. Improvements in the confining structure may include fabricating it on a silicon substrate using photolithography, he added.

REFERENCES

1. R. Li et al., Optics Exp. 17(3) p. 1622 (Feb. 2, 2009).
2. G. Winter et al., New J. of Physics 8(125) p. 1 (2006).
3. W. L. Barnes et al., Nature 424, p. 824 (Aug. 14, 2003).