Small spiral phase plates are made simply

A spiral phase plate (SPP) introduces a phase shift of a magnitude that usually depends (in polar coordinates with origin at the center of the phase plate) linearly on the angle and not at all on the radius, with a phase discontinuity at its 0°/360° position.

Feb 1st, 2005

A spiral phase plate (SPP) introduces a phase shift of a magnitude that usually depends (in polar coordinates with origin at the center of the phase plate) linearly on the angle and not at all on the radius, with a phase discontinuity at its 0°/360° position. An SPP produces phase dislocations in a normal laser beam passing through it. The properties of the resulting Laguerre-Gaussian beam can include a null on the beam’s axis that enables optical trapping of particles. Researchers at Nanyang Technological University (Singapore) and the University of St. Andrews (Fife, Scotland) have developed a simple and reproducible technique for fabricating micron-sized SPPs (useful to insert near an intermediate focus in an optical system).

Photoresist spun on glass is exposed via electron-beam lithography, soft-baked, and developed. In one ­example, 500-μm-diameter SPPs were fabricated with a maximum height 0f 1.04 μm; their 24 levels approximated a continuous phase gradation. The mode-conversion efficiency of the SPPs was measured to be from 80% to 87%, producing beams annularly uniform to 10%. Simultaneous optical trapping of up to six microspheres in water was demonstrated with a 25-mW HeNe beam. Contact Larry Xiaocong Yuan at excyuan@ntu.edu.sg.

Silicon laser on a chip demonstrated

In a significant step toward development of an all-optical computer that uses photons instead of electrons to process information, scientists at Intel (Santa Clara, CA) have demonstrated the first all-silicon laser on a chip, the device is a waveguide Raman ­laser. Earlier, researchers at the University of California at Los Angeles (see Laser ­Focus World, December 2004, p. 9) reported the first all-silicon laser (also a Raman ­laser), but their version required an external fiber ring cavity to produce lasing.

The Intel device uses a silicon-rib waveguide fabricated on the surface of an undoped silicon-on-insulator substrate, with waveguide patterning done using standard projection photolithography and plasma reactive-ion etching. The rib-waveguide width, rib height, and etch depth are 1.5, 1.55, and 0.7 μm respectively, with the waveguide formed into an S-shaped curve with a bend radius of 400 μm and a total length of 4.8 cm. When pumped with a 1536-nm pulsed source and above the 0.4-mW threshold power, the device lased at 1669.5 nm. Contact Haisheng Rong at haisheng.rong@intel.com.

Deep floating guard ring improves reliability of avalanche photodiode

Avalanche photodiodes (APDs) containing a thin multiplication layer have high sensitivity but can suffer from poor reliability as well. The separated absorption, grading, charge, and multiplication (SAGCM) structure of one type of APD has a so-called floating guard ring (FGR), a surrounding ring of p-doped material that is added in attempts to suppress reliability-impairing prebreakdown caused by a high electric field near the edge of the diffused central junction (bottom schematic). The effectiveness of this method has been increased by researchers at Sejong University (Seoul, Korea), ETRI (Taejon, Korea), Hanyang University (Kyunggi-Do, Korea), and Sunmoon University (Chungnam, Korea), who etch a groove to allow a deep FGR to form during the diffusion fabrication step (top schematic).

A prototype SAGCM APD was fabricated from indium phosphide/indium gallium arsenide. A scan of a 3-μm-­diameter light spot with a 1550-nm wavelength across the 25-μm-diameter prototype device showed that gain was successfully suppressed at the edge of the active area, necessary to prevent damage from high electric fields. Another advantage of the deep FGR is that no additional ring-shaped central junction is required, resulting in a more compact device. Contact Kyung-Sook Hyun at kshyun@sejong.ac.kr.

Rare-earth-doped gallium nitride lases

Because rare-earth ions exhibit inner-shell transitions that produce sharp photoemission lines with wavelengths ranging from the UV to the near-IR, they can be used in solid-state lasers that are optically pumped and have an insulating host. To develop an electrically pumped rare-earth laser, however, semiconductor rather than insulator hosts are required. Scientists from the Nanoelectronics Laboratory of the University of Cincinnati (Cincinnati, OH) have demonstrated rare- earth-based lasing action in a semiconductor host.

Using molecular-beam epitaxy, europium (Eu)-doped gallium nitride was grown on a 10-mm-square sapphire substrate resulting in a 0.6-μm active layer doped with approximately 1 to 3 atomic percent Eu. After pumping the region with a nitrogen laser at 337.1 nm and blocking the surface emission, lasing was observed at the edge of the doped film at the dominant emission-peak wavelength of 620 nm at a threshold of 10 kW/cm2. By increasing the pump power above threshold, the modal gain was measured to be 43 cm-1. This low-threshold pump power and strong modal gain are promising indicators for achieving electrically pumped lasing from this material combination. Contact Andrew J. Steckl at a.steckl@uc.edu.

Polyimide smooths surface of EUV aspheric condensor

The reflective aspheric condensor (illumination) optics used for extreme-UV (EUV) lithographic systems are assailed by intense ionizing radiation, heat, and contaminants from outgassing, resulting in the need for frequent replacement. Because of this, cheaper optics are better, at least as long as quality remains good. Researchers at Lawrence Livermore National Laboratory (Livermore, CA) and Lawrence Berkeley National Laboratory (Berkeley, CA) have figured out how to produce aspheric optics for this purpose using the relatively inexpensive technique of diamond-turning, while keeping surface roughness down (the 2-nm finish of ordinary diamond-turned optics reduces EUV reflectance to only a few percent).

The surface of a diamond-turned aluminum ellipsoidal reflector was smoothed by spinning on a film of polyimide, which was then cured in an oven. A multilayer molybdenum/silicon reflective film was deposited over the polyimide; the multilayer had 80 bilayers and was 556 nm thick. The high-spatial-frequency roughness of the substrate was reduced from 1.76 to 0.27 nm root-mean-square without pushing surface slope beyond specifications. A test of the ellipsoid in an EUV system showed that the surface did not degrade more than a conventional reflective surface. Contact Regina Soufli at regina.soufli@llnl.gov.

Teramobile laser guides lightning through rain and clouds

Researchers in Germany and France at the Teramobile femtosecond-laser project have taken a step closer to the long-term goal of laser-guided lightning by successfully testing their system in simulated rain conditions. In 2001, the Teramobile, a 6-TW Ti:sapphire laser, was used to guide simulated lightning in the laboratory. The laser generated self-guided filaments that triggered and channeled high-voltage pulses across a 2.5-m gap from an electrode located at a pulse generator to a remote electrode connected to ground (see Laser Focus World, January 2002, p. 37).

Guiding real lightning, however, will also require operation through real atmospheric conditions, such as rain. So collaborators at the Freie Universität (Berlin, Germany), the Université Claude Bernard (Lyon, France), the Laboratoire d’Optique Appliquée (Palaiseau, France) and the Institut für Energie- und Automatisierungstechnik (Berlin, Germany) sprayed water droplets between the electrodes in their experiment at a flow rate of 1.4 mm/min to simulate atmospheric conditions of heavy rain, while the Teramobile provided 170-fs, 230-mJ pulses centered at 800 nm. They found that even in dense clouds, the pulse-guiding phenomena persisted despite interaction with large aerosol particles on the order of 0.5 mm in diameter. Contact Jérôme Kasparian at jkaspari@lasim.univ-lyon1.fr.

Terahertz detector shows potential for 2-D imaging

In collaboration, researchers from Royal Holloway University (London, England), Japan Science and Technology Corporation (Tokyo, Japan), the University of Tokyo (Tokyo, Japan), and the University of Glasgow (Glasgow, Scotland) have fabricated a highly sensitive detector that could eventually be used in the development of a 2-D terahertz-wavelength imager. They fabricated the device using a metallic single-electron transistor (SET) and a quantum dot (QD), such that photon absorption in the QD induced a potential distribution sensed by the SET. At wavelengths around 200 μm and operating temperatures near 0.4 K, they approached single-photon counting sensitivity with noise-equivalent power on the order of 10-19 W/√Hz.

Unlike double-QD configurations, the QD-SET structure does not require the precise adjustment of electrically controlled potential barriers that would impede applicability of the device for 2-D imaging. With optimization of QD structure and materials, the researchers believe they can improve the quantum ­efficiency of the device above the ­currently estimated 1%. They also state that an operating temperature of about 4 K, currently limited by the SET, should be achievable. Contact Vladimir Antonov at v.antonov@rhul.ac.uk.

Algorithm takes the guesswork out of PC design

Two-dimensional photonic crystals (PCs) have received a lot of press, and deservedly so, because they accomplish feats difficult for ordinary waveguides, such as sharp 90° bends. Their capabilities can be further extended as computer optimization of PCs becomes increasingly sophisticated. Researchers at Stanford University have developed a next-generation optimization method that allows the design of PCs with novel properties; the method places less reliance on the educated guesses that were required to help previous PC optimizations.

The method is based on a so-called Wannier-basis field expansion, a non-­gradient-based method that can operate with little initial information. The complexity of its simulated-annealing approach scales with the number of lattice sites, not the computational grid points, easing use for design of 3-D PCs. An example 2-D device-a mode separator-was designed with the algorithm. In the device, a multimode input waveguide terminates in a complex structure of posts; three exiting waveguides each channel a different selected mode-the fundamental, first, and second modes, respectively. If used with light at a 1.5-μm wavelength, the mode selector is only 8.2 × 13.3 μm in size. Contact Yang Jiao at jiaoyang@stanford.edu.

Dispersion compensator tunes above and below zero

Scientists at Highwave Optical Technologies (Lannion, France) have developed a fiber Bragg grating (FBG)-based chromatic-dispersion compensator with a wide tuning range both above and below the zero dispersion point, and with low insertion-loss variation.

The compensator is based on cascading two identical chirped FBGs ­using two three-port circulators so that both gratings are tuned simultaneously as previously for a single grating; however, the key point is that the gratings have opposite chirp directions in relation to the temperature gradient applied to the package. Since the dispersion sign of a chirped FBG changes with the ­injection direction, it results in zero dispersion without an applied gradient. When a ­gradient is applied, opposite bandwidth variations for each grating are additive and the dispersion can be varied about the zero dispersion point. In an ­actual device that used two identical 70-mm long linearly chirped FBGs, the tuning range extended from -400 to +400 ps/nm with a temperature difference of 60°C. Simulation results based on experimental spectra showed that the technique is suitable for 40-Gbit/s dispersion compensation. Contact Alain Mugnier at alain.mugnier@highwave-tech.com.

More in Research