Microlaser Generates Microwatts

Breck Hitz

Researchers at Riken, the Institute of Physical and Chemical Research in Wako, Japan, have demonstrated an approach to the construction of tiny lasers that is likely to find application in miniature photonic devices such as lab-on-a-chip chemical and biological analytical instrumentation. They believe that this development opens the possibility of multiwavelength arrays of miniature lasers for simultaneous analysis in different spectral regions.

The researchers etched an ~800-µm chamber between two retroreflectors in a glass chip, filled it with rhodamine 6G dye and pumped it with a frequency-doubled Nd:YAG laser. After observing laser emission from the dye, they conducted a second experiment with two chambers, each containing a different dye. They pumped both chambers with a single laser and obtained laser radiation at two wavelengths.

Figure 1. A tiny ring-laser resonator was created by the two retroreflectors.

The geometric arrangement of the single-chamber experiment is shown diagrammatically (Figure 1) and in a photograph (Figure 2). It is a ring-laser configuration, with power circulating in both directions around the ring. The retroreflectors are hollow, and the circulating power is totally internally reflected from their surfaces. Because the reflectors' surfaces are not perfectly flat, however, some rays are incident at less than the critical angle and are partially refracted, entering the hollow structure at near-tangential angles. These refracted rays are the laser's output. In the configuration shown, there are eight output beams.

Figure 2. The hollow chamber was filled with rhodamine 6G dye through a small channel (dark hole in middle of chamber).

The process of creating the microstructures in the glass involved four steps. First, the researchers wrote latent images in the glass with tightly focused femtosecond laser pulses. Then they baked the glass in a programmable oven to form the regions written by the pulses. Next, they etched out the regions with hydrofluoric acid in an ultrasonic bath and, lastly, baked the glass again to smooth the surfaces of the internal microstructures.

They confirmed that they were seeing laser radiation, and not spontaneous output, by observing the spectrum of the emitted light. At pump intensities of less than ~1 mJ/cm², they saw only a broad spectrum of spontaneous emission. When they increased the pump intensity to above 1.5 mJ/cm², the first signs of laser emission appeared (Figure 3, top). As they increased the pump intensity further, to 4.5 mJ/cm², the laser emission increased in intensity and narrowed spectrally (Figure 3, bottom). They measured an average power of ~10 µW in one of the eight output beams.

Figure 3. (top) The onset of lasing appears as a broad emission centered around 570 nm. (bottom) At higher pump power, the laser emission narrows and intensifies. The peak at 532 nm is scattered pump power.

In a more-complex experiment, the researchers etched a pair of tiny chambers in the glass, one above the other, and pumped both dye-filled chambers with the same frequency-doubled Nd:YAG laser. They observed laser emission from both chambers.

One area of possible improvement in the future involves the rather awkward, eight-beam output coupling. The researchers note that buried Bragg gratings have been fabricated in glass with femtosecond lasers and that a pair of such reflectors, one on either side of the dye chamber, would result in a linear resonator with a single output beam.