Performance advances in a wide variety of laser-based devices have been awaiting commercial availability of blue-emitting diode lasers. Replacing the red laser in a DVD player with a blue-emitting one, for example, increases storage capacity by about 2.5 times-but besides optical data storage, blue-violet laser diodes ultimately will benefit many other fields.
In January, Nichia Chemical Industries (Anan, Japan) became the first company to introduce a blue-violet-output semiconductor laser to the market. At the same time-and in collaboration with Nichia-TuiOptics (Munich, Germany) introduced a tunable-output spectroscopic platform based on an external-cavity laser diode configuration that incorporates the Nichia device (see photo). This instrument provides ultranarrow linewidths with higher stability than alternative methods at similar output powers. Because it is less expensive than conventional approaches for obtaining this wavelength, it is expected to open up new opportunities in spectroscopy, interferometry, and microscopy.
Blue-violet diodes arrive
Nichia was the first company to demonstrate a blue-emitting light-emitting diode (LED) based on gallium nitride (GaN). The performance of the device, which was demonstrated in the early 1990s, was impressive, and many researchers then working with alternative materials, such as zinc selenide (ZnSe) and silicon carbide (SiC), shifted their efforts in this promising new direction. A commercial LED soon followed from Nichia, while work continued on a stable, long-lived room-temperature semiconductor laser based on gallium nitride.
Fig. 1 Existing TuiOptics external-cavity laser diode platform has been extended to incorporate the Nichia blue-output laser diodes by using different optics and gratings.
The resulting Nichia product emits at around 400 nm. The devices produce 5 mW of continuous-wave (CW) output at room temperature and are rated for a 10,000-hour lifetime-the level demanded by systems manufacturers for commercial viability. The new laser is fabricated by first growing an epitaxial laterally overgrown (ELOG) GaN layer on top of a sapphire substrate. The laser diode structure is then fabricated on top of the ELOG layer (see Fig. 1). The sapphire substrate is polished away leaving only the GaN device. This step improves thermal heat sinking to the laser and makes it easier to cleave individual laser devices. Obtaining improved heat sinking is particularly important because excessive thermal effects have a dramatic impact on device lifetime.
The maximum output of the lasers is currently 30 mW, but the lifetime can be compromised when operating above the recommended 5 mW output level. The threshold current density of the device is 3.9 kA/cm2, while 50 mA and 5 V are required to achieve the 5 mW output level. The laser comes in a 5.6-mm-diameter package.
External cavity key to spectroscopy
At Photonics West (San Jose, CA) in January Polytec PI (Auburn, MA), the North American distributor of TuiOptics laser products, demonstrated the world`s first spectroscopic instrument that incorporates the Nichia blue-violet laser. The rapid availability of this product was possible because of an existing TuiOptics external-cavity laser diode (ECLD) platform, the DL100, which currently uses a series of semiconductor lasers with outputs ranging from about 632 up to 1700 nm. It was adapted to the blue laser diodes by using different optics and gratings to assure reliable single-frequency operation at any wavelength within the gain profile of the laser diode.
FIGURE 1. Commercially available blue-violet-emitting laser diode from Nichia Chemical Industries is based on a gallium nitride material structure.
For many spectroscopic and scientific applications, even the relatively narrow linewidth of a semiconductor laser is too broad, and tuning by temperature or current results in poor performance with a limited tuning range because of unpredictable modal behavior. One way to achieve better spectral purity and tuning of the laser diode is to operate it in the ECLD configuration. This arrangement uses the same diode source, but collimating optics are added with a tiltable grating for wavelength selection-a more-reliable tuning method than drive current or temperature techniques.
A free-running laser diode is a high-gain device with a very low cavity finesse. It produces a relatively large fundamental linewidth because the linewidth of the laser diode source is inversely proportional to the photon lifetime in the cavity. By using the laser diode as the gain medium within a larger external resonator cavity, the short cavity length formed by the laser`s cleaved facets is extended to the length of the external cavity. Additional higher reflective resonator mirrors increase the photon lifetime, thus reducing the linewidth of the laser.
The DL 100 platform uses the highly reflecting back facet of the Nichia laser diode as one mirror in the external cavity. The grating is mounted in the Littrow configuration and controls both the emission wavelength and the selection of one longitudinal mode of the laser diode (see Fig. 2). This assures real single-frequency operation, which is required for high-resolution spectroscopy. The grating doubles as the second resonator mirror and output coupler and has a reflection coefficient higher than the cleaved facet of the laser diode. Long-term wavelength drifts are minimized by incorporating a thermoelectric element for active temperature control of the complete laser set-up.
Typically, free-running laser diode sources emit over a band of perhaps 10 to 15 MHz, but with this ECLD configuration, a linewidth as small as 1 MHz can be achieved, more than sufficient to resolve the natural linewidth of an atomic transition.
This design was conceived by Theodor H?nsch`s group at the University of Munich for applications in laser cooling of atoms, a discipline that has made remarkable progress in the last decade, ostensibly due to the availability of ECLDs for manipulating lithium, rubidium, and cesium atoms. In 1997, the awarding of a Nobel Prize recognized these advances. Interest in this type of research is currently growing worldwide, especially now that researchers have discovered that the ECLD platform can be used to perform Bose-Einstein condensate experiments on laser-cooled rubidium gas. Key to this fast progress has been the ruggedness, long-term stability, modularity, and ease of use of the ECLD platform that is found in nearly every laboratory in this field today.
FIGURE 2. Tunable laser diodes can be configured as a Littrow design, which uses the first-order reflection of a high-quality grating to couple back light directly into the laser diode, resulting in optical stabilization and frequency control (top). The alternative Metcalf-Littman design, with a second mirror and more-complex tuning mechanics, is more sensitive to acoustic and thermal changes (bottom). Both designs offer similar wavelength-tuning capabilities, but the DL 100 platform promises more output power, improved ruggedness, and long-term stability in a rough environment.
Based upon the initial production runs at Nichia, TuiOptics expects lasers will be available to cover at least the spectral range from 390 to 410 nm. For any individual blue-violet laser diode, available output power will be around 3 mW, reduced from a nominal 5 mW as a result of losses in the external cavity. The narrowband, 1-MHz laser linewidth can be coarsely adjusted over a tuning range of 2.5 nm and then fine-tuned by a piezo stack over a more than 20-GHz range without mode hops. Most important, no additional antireflection coatings need be applied to the laser diode. Under development are im proved optics for beam collimation and focusing.
Alternative tunable blue sources - A tunable blue-output laser diode source offers new performance-cost opportunities for spectroscopy applications. Previously, researchers wanting to perform experiments in atomic physics and chemistry have had limited options in achieving a tunable CW blue-light source. Usually, they resort to one of three possibilities-a frequency-doubled dye laser, a Ti:sapphire laser, or, more recently, frequency-doubled semiconductor lasers.
Frequency-doubled dye and Ti:sapphire systems are bulky and require an expensive high-power pump-laser source. Besides cost and size, these systems have additional problems with output power levels and stability. They require nonlinear crystals for frequency doubling, which are inefficient unless an intracavity approach is used-but this often leads to complicated solutions suitable only for highly trained people. While they can generate several hundred milliwatts of tunable blue laser light, these systems typically cost more than $100,000.
As an alternative approach for limited power requirements, TuiOptics already offers much smaller CW frequency-doubling systems based on near-infrared laser diode sources with selected nonlinear crystals for frequency conversion of the output to the green and blue. A tunable output of 15 mW at 490 nm or 5 mW at 430 nm, for example, can be achieved with these systems. The new blue-violet ECLD systems will, however, be priced about 75% less than these frequency-doubled laser diode systems.
In addition to the price advantage, the new blue-violet ECLD platforms offer the best amplitude stability of any laser source at this wavelength, including fixed-frequency gas lasers. The near-term prospect of higher output power and wider wavelength coverage also makes for a bright future for the ECLD platform.
Applications should multiply - While $100,000 laser sources may be acceptable in some research environments, they are much too expensive for high-volume, price-sensitive markets, although these applications require frequency-stabilized and frequency-tunable sources. In chemical trace analysis with atomic absorption spectroscopy, for example, blue laser light is useful for detecting or manipulating chemical elements that have resonant lines in the blue region of the spectrum. These include aluminum, erbium, gallium, gadolinium, holmium, indium, lead, manganese, titanium, tungsten, and ytterbium.
Blue light is also ideal for working with organic dyes, which are used in fluorescence detection. Raman spectroscopy, microscopy, interferometry, and holography are also eagerly awaiting cheaper sources. Add in the ability to provide faster modulation of the laser light-up to 1 GHz can be reached with the DL 100 platform-and new opportunities emerge not only in spectroscopy, but also in multiple-wavelength printing and telecommunications.
With the availability of the first violet-blue laser diodes, many exciting new products are just around the corner. First to market, however, will be more-affordable, higher-performance spectroscopic systems.
SHUJI NAKAMURA is senior researcher, R&D department at Nichia Chemical Industries Ltd., Anan, Japan; e-mail: [email protected]. WILHELM KAENDERS is president of TuiOptics GmbH, Munich, Germany; e-mail [email protected].