News Breaks

The first Raman laser that is electrically driven and requires no external optical pump was demonstrated by scientists from Harvard University (Cambridge, MA), Texas A&M University (College Station, TX), and Bell Laboratories, Lucent Technologies (Murray Hill, NJ).

Apr 1st, 2005

Electricity drives Raman laser

The first Raman laser that is electrically driven and requires no external optical pump was demonstrated by scientists from Harvard University (Cambridge, MA), Texas A&M University (College Station, TX), and Bell Laboratories, Lucent Technologies (Murray Hill, NJ). By combining the advantages of nonlinear optical devices and semiconductor injection lasers, the scientists achieved an enhancement of orders of magnitude in the Raman gain, high conversion efficiency, and low threshold for their laser.

Unlike conventional solid-state Raman lasers that rely on phonon energy from an external optical pump, the injection Raman laser achieves lasing through electrically actuated intersubband transitions in a quantum-cascade (QC)-laser cavity. The current generates an internal laser beam or pump within the QC cavity that is then used to generate the Raman laser radiation. The laser has an indium gallium arsenide/indium aluminum arsenide heterostructure; the onset of Raman lasing can be observed above a threshold current of 2.6 A. Though not as powerful as a conventional Raman laser, the injection laser is highly efficient (30% of the pump power is converted into the Raman laser beam). Contact Mariano Troccoli at

320 × 256 detector array is solar-blind

Although the first images from an aluminum gallium nitride (AlGaN) UV focal-plane-array (FPA) camera were published in 2002, quality was lacking and they did not provide a full-frame 320 × 256 image. Now, researchers at Northwestern University (Evanston, IL) have produced, entirely in their laboratory, the first AlGaN-based 320 × 256 UV FPA of high quality and have published several good images at a 280-nm wavelength.

The processed FPA consists of an array of 320 × 256 discrete 25 × 25-µm pixels on a 30-µm period. The current-voltage curve of a representative pixel from the middle of the array showed a sharp turn-on voltage of 4.7 V, with a series resistance of 4.3 kΩ. The individual AlGaN photodiodes are bonded to a read-out integrated circuit via flip-chip bonding. A camera was constructed that consisted of the FPA in a leadless chip carrier, an aperture to block stray light, a 280-nm bandpass filter, and a 32-mm lens to collect the light and form the image on the FPA; recorded camera images showed good uniformity for the pixel response. Contact Manijeh Razeghi at

Metal-to-glass bonding technique allows many glasses to be used

Many optoelectronic devices require hermetic sealing, including some CCDs, CMOS sensors, photodetectors, and optical MEMS devices, with a portion of these requiring very high vacuum-tightness. The best configuration is a glass window sealed to a metal frame. Because epoxy or other bonding agents outgas, leak, and degrade, a hard seal is required, with the standard metal/glass combination being Kovar and Corning 7056 glass, used because the two materials have the same coefficient of thermal expansion (CTE).

But the 7056 glass has an uneven visible-transmission spectrum and can contain bubbles. A non-oxide sealing process developed at Tekna-Seal (Minneapolis, MN) used with metal-injection-molded components from FloMet (DeLand, FL) allows use with a variety of glasses-for example, Ohara S-BAL glasses, which have good optical properties. The CTE of the metal frame is tailored to match that of the chosen glass; the heat-sealing process, which chemically bonds the molten glass to the metal, is done in inert gas. Vacuum-tightness is improved from 10-8 to 10-10 cm3/s of helium. With the optical window made to project beyond the metal frame, both sides of the window can be optically figured after sealing, providing very low-stress joints. Even the use of BK-7 glass is possible, if desired. Contact Doug McCarron at

Sensitive detector sees terahertz radiation

Terahertz radiation requires special detectors with a noise-equivalent power (NEP) two orders of magnitude smaller than that of current semiconductor and superconductor bolometer types. Although detectors have been built that implement a combination of superconducting and single-electron devices to trap and detect photons, operation of these detectors is restricted to temperatures below 1 K and, in the case of double quantum-dot (QD) detectors, demands precise adjustment of electrically controlled potential barriers.

By placing a metallic single-electron transistor (SET) on top of a semiconductor QD with a 1-µm diameter, scientists from the Royal Holloway University of London (Surrey, England), the Japan ­Science and Technology Corporation (Tokyo, Japan), the University of Tokyo (Tokyo, Japan), and the University of Glasgow (Glasgow, Scotland), were able to detect terahertz radiation around 200 µm with a high NEP sensitivity of approximately 10-20 W/Hz1/2. When a photon was absorbed by the QD, it excited electrons that impacted the capacitance of the QD and SET configuration, in turn allowing quantification of the detected signal. Contact Vladimir Antonov at

Fabry-Perot interferometer precisely tracks cantilever-probe deflection

Deflections of the probe-containing cantilevers of scanning-force microscopes and similar devices are often measured optically-typically by optical-beam deflection or an interferometer. But the sensitivity of optical-beam deflection drops for small spot sizes (required for small cantilevers), and the straightforward interferometric approach requires a light source placed within a few microns-not possible with some configurations. Researchers at the University of Basel (Basel, Switzerland), IBM Research ­Division (Rüschlikon, Switzerland), Fisba Optik (St. Gallen, Switzerland) and the Swiss Federal Laboratories for Materials Testing and Research (Dübendorf, Switzerland) have developed a low-noise Fabry-Perot interferometric approach with a large 0.8-mm working distance.

Laser light at 783 nm from a single-mode fiber passes through a hole in a spherical mirror with a 0.8-mm radius and strikes the cantilever, which sits at the mirror’s center of curvature. Light bounces back and forth between the mirror and cantilever and eventually makes its way back through the fiber. A beamsplitter beyond the fiber’s other end sends the return signal to a photodetector. The position of a cantilever 20 × 4 × 0.2 µm in size was measured at 1 MHz with a noise floor of 1 fm/(Hz)0.5. Contact Bart Hoogenboom at

Diode-pumped solid-state laser produces multiple lines

In the past few years, a bevy of small and efficient blue- and green-emitting diode-pumped solid-state lasers (DPSSLs) has arisen, replacing relatively large and power-hungry gas lasers for some applications. But argon-ion lasers have always had the bonus of multiline operation-useful in medicine, microscopy, and other biological research. Now, the small DPSSL (which relies on frequency conversion to reach green and blue) is chipping away at that advantage too.

Cobolt (Stockholm, Sweden) has commercially introduced a dual-line DPSSL emitting at both 491 and 532 nm. In the device, a laser diode pumps two layered solid-state gain media in a stable resonator. The collinear emissions undergo intracavity frequency-­mixing in a periodically poled crystal of potassium titanyl phosphate. In one example of a use in confocal microscopy of biological cells, two fluorescent stains are pumped at 491 and 532 nm, producing red and green emissions that are dichroically separated from the laser light and combined into a single multicolor image. Commercial triple-line versions of the DPSSL (for example, emitting wavelengths of 457, 491, and 532 nm) are due out in 2006, says Håkan Karlsson, Cobolt’s vice president of technology and business development. Contact Håkan Karlsson at

More in Research