Some types of lasers

Gas Lasers are lasers with a gas (or plasma) as gain medium

A variety of lasers is based on gases as gain media. The laser-active entities are either single atoms or molecules, and are often used in a mixture with other substances having auxiliary functions. A population inversion as the prerequisite for gain via stimulated emission is in most cases achieved by pumping the gas with an electric discharge, but there are also gas lasers using a chemical reaction, optically pumped devices, and Raman lasers. During operation, the gas is often in the state of a plasma, containing a significant concentration of electrically charged particles.

Most gas lasers emit with a high beam quality, often close to diffraction-limited, since the gas introduces only weak optical distortions, despite considerable temperature gradients. Their operation usually requires a high-voltage supply, often with a high electrical power. Some high-power gas lasers use a system for quickly circulating the gas (forced convection, fast flow).

Types of Gas Lasers

There are very different kinds of gas lasers, operating in entirely different regimes concerning emission wavelength and output power:

Helium–neon lasers (He–Ne lasers) often emit red light at 632.8 nm, but can also be made for other wavelengths such as 543.5 nm (green), 594.1 nm (yellow), 611.9 nm (orange), 3.39 μm, or 1.15 μm. Typical He–Ne lasers have a gas cell with a length of the order of 20 cm and generate a few milliwatts of output power in continuous-wave operation at 632.8 nm, using several watts of electrical power. Helium–neon lasers are often used for alignment and in interferometers, and compete with laser diodes, which are more compact and efficient. Some HeNe lasers serve in optical frequency standards.

Argon ion lasers use a typically larger (e.g. 1 m long) water-cooled tube with an argon plasma, made with an electrical discharge with high current density in order to achieve a high degree of ionization. They can generate more than 20 W of output power in green light at 514.5 nm, and less at some other wavelengths such as 457.9, 488.0, or 351 nm. Their power efficiency is fairly low, so that tens of kilowatts of electrical power are required for multi-watt green output, and the cooling system has corresponding dimensions. There are smaller tubes for air-cooled argon lasers, requiring hundreds of watts for generating some tens of milliwatts. Argon ion lasers can be used e.g. for pumping titanium–sapphire lasers and dye lasers, and are rivaled by frequency-doubled diode-pumped solid-state lasers.

Krypton ion lasers are similar to argon ion lasers and can emit high powers at 647.1 nm and some other wavelengths.

Carbon dioxide lasers (CO2 lasers) use a gas mixture of CO2, helium (He), nitrogen (N2), and possibly some hydrogen (H2), water vapor, and/or xenon (Xe) for generating laser radiation

mostly at 10.6 μm. They have wall-plug efficiencies above 10% and are suitable for output powers of multiple kilowatts with fairly high beam quality. They are widely used for material processing, e.g. cutting, welding and marking, but also in laser surgery.

Excimer lasers (rare gas halide lasers, exciplex lasers) are also pumped with an electrical discharge, but in that case the pumping energy is used to form unstable molecules which can emit photons when disassociating. Most excimer lasers are ultraviolet lasers and are operated with current pulses, leading to the emission of intense nanosecond pulses. They are used for various types of material processing, includingpulsed laser deposition, laser marking, and the fabrication of fiber Bragg gratings. There are also medical applications e.g. in ophthalmology.

Nitrogen lasers are another type of pulsed ultraviolet laser, based on pure nitrogen, a nitrogen–helium mixture, and sometimes even simply air (with lower performance). Emission typically occurs at 337.1 nm. The high gain leads to relatively efficient superluminescent emissioneven without a laser resonator. Nitrogen lasers are relatively easy to build and operate, and have been made by many hobbyists without refined laboratory equipment.

Hydrogen lasers can be used to access even shorter wavelengths around 160 nm, 123 nm or 116 nm.

Various metal vapor lasers use a metal vapor, excited and heated by an electric discharge. Copper vapor lasers are excited with intense current pulses and generate nanosecond pulses at 510.6 nm (green) or 578.2 nm (yellow). The average output power can exceed 100 W.Helium–cadmium lasers are more similar to helium–neon lasers, emitting continuously at 442 nm (blue) or 325 nm (ultraviolet), with optical powers of the order of 100 mW. The laser transition occurs in Cd+ ions, which become excited in collisions with excited helium atoms.

There are alkali vapor lasers, using e.g. a cesium or rubidium cell in an oven as the gain medium. Such lasers can be pumped with laser diodes. The power efficiency can be fairly high; note that the quantum defect can be small for a typical pumping scheme where one pumps from the ground state 6S1/2 to 6P3/2 and uses a transfer via a buffer gas (e.g. ethane) to the nearby 6P1/2 as the upper laser level. At the same time, the beam quality can be much higher than that of the pump source, so that such a laser acts as an efficient brightness converter.

Chemical lasers convert chemical energy of gases into laser light (typically in the mid- or near-infrared region) with powers up to the megawatt level. There are e.g. hydrogen-fluoride (HF) lasers, fueled with H2 and F2, which is converted to HF, and oxygen-iodine lasers (COIL). Chemical lasers are mainly studied for military purposes, e.g. as anti-missile weapons, to be operated even on board large airplanes.

Raman gas lasers are Raman lasers, based on optical amplification via stimulated Raman scattering rather than on stimulated emission of excited ions. They can e.g. use a hydrogen cell, and need to be optically pumped.

Gas lasers can also be grouped according to the nature of their laser-active species:

Neutral atom gas lasers include helium–neon lasers and copper vapor lasers.

Ion lasers use free ions; examples are helium–cadmium lasers, argon ion lasers and krypton lasers. Typically, ion lasers generate shorter wavelengths, but with moderate power efficiency.

Molecular gas lasers use gas molecules. Examples are carbon dioxide and carbon monoxide lasers, nitrogen lasers, and excimer lasers.

Many gas lasers have self-terminating laser transitions, where the lower state has a long lifetime. Lasing stops once the lower-state population becomes too high. Examples for such gas lasers are nitrogen lasers and copper vapor lasers. Excimer lasers can also only be operated in pulsed mode, although for different reasons [3].

Application of Gas Lasers

A primary reason for using certain gas lasers instead of solid-state lasers is that they offer special wavelengths, which are otherwise difficult to obtain. Another interesting aspect is that relatively high optical powers can be obtained with gas lasers; compared to solid-state lasers, particularly diode-pumped ones, the price depends less strongly on the required output power level. A good example for both aspects is thecarbon dioxide laser, which is quite a unique long-wavelength source with high output power. Similarly, excimer lasers provide high powers in theultraviolet spectral region.

The helium–neon laser has been widely used for the generation of red laser light, but is now increasingly replaced by cheaper and more compactlaser diodes. Similarly, argon ion lasers were often used for pumping titanium-sapphire lasers, but are now often replaced by frequency-doubled solid-state lasers .

Helium–neon (He–Ne) lasers, gas lasers based on a helium–neon mixture, and are a frequently used type of continuously operating gas lasers, most often emitting red light at 632.8 nm at a power level of a few milliwatts and with excellent beam quality. The gain medium is a mixture of helium and neon gas in a glass tube, which normally has a length of the order of 15–50 cm.

Figure 1: Setup of a helium–neon laser.

A DC current, which is applied via two electrodes with a voltage of the order of 1 kV, maintains an electric glow discharge with a moderate current density. In the simplest case, a ballast resistor stabilizes the electric current. The current is e.g. 10 mA, leading to an electrical power of the order of 10 W. The glass tube as shown in Figure 1 has Brewster windows, and the laser mirrors must form a laser resonator with a small round-trip loss of typically below 1%. Due to the polarization-dependent loss at the Brewster windows, a stable linear polarization is obtained.

Some He–Ne lasers have a tube with internal resonator mirrors, which can not be exchanged. Brewster windows are then not required.

In the gas discharge, helium atoms are excited into a metastable state. During collisions, the helium atoms can efficiently transfer energy to neon atoms, which have an excited state with similar excitation energy. Neon atoms have a number of energy levels below that pump level, so that there are several possible laser transitions. The transition at 632.8 nm is the most common, but other transitions allow the operation of such lasers at 1.15 μm, 543.5 nm (green), 594 nm (yellow), 612 nm (orange), or 3.39 μm. The emission wavelength is selected by using resonator mirrors which introduce high enough losses at the wavelengths of all competing transitions.

Due to the narrow gain bandwidth, He–Ne lasers typically exhibit stable single-frequency operation, even though mode hopping is possible in some temperature ranges where two longitudinal resonator modes have similar gain.

Applications

Helium–neon lasers, particularly the standard devices emitting at 632.8 nm, are often used for alignment and in interferometers. They compete with laser diodes, which are more compact and efficient, but have less convenient beam profiles.

Some He–Ne lasers are serving in optical frequency standards. For example, there are methane-stabilized 3.39-μm He–Ne lasers, and 633-nm iodine-stabilized versions.

Argon ion gas lasers based on light amplification in ionized argon in a gas discharge, these lasers are powerful gas lasers, which typically generate multiple watts of optical power in a green or blue output beam with high beam quality.

The core component of an argon ion laser is an argon-filled tube, made e.g. of beryllium oxide ceramics, in which an intense electrical discharge between two hollow electrodes generates a plasma with a high density of argon (Ar+) ions. A solenoid around the tube (not shown in Figure 1) can be used for generating a magnetic field, which increases the output power by better confining the plasma.

Figure 1: Setup of a 20-W argon ion laser. The gas discharge with high current density occurs between the hollow anode and cathode. The intracavity prism can be rotated to select the operation wavelength.

A typical device, containing a tube with a length of the order of 1 m, can generate 10 W or 20 W of output power in the green spectral region at 514.5 nm, using several tens of kilowatts of electric power. (The voltage drop across the tube may be 100 V or a few hundred volts, whereas the current can be several tens of amperes.) The dissipated heat must be removed with a water flow around the tube; a closed-circle cooling system often contains a chiller, which further adds to the power consumption. The total wall-plug efficiency is thus very low, usually below 0.1%. There are smaller air-cooled argon ion lasers, generating some tens of milliwatts of output power from several hundred watts of electric power.

The laser can be switched to other wavelengths such as 457.9 nm (blue), 488.0 nm (blue–green), or 351 nm (ultraviolet) by rotating the intracavity prism (on the right-hand side). The highest output power is achieved on the standard 514.5-nm line. Without an intracavity prism, argon ion lasers have a tendency for multi-line operation with simultaneous output at various wavelengths.

There are similar noble gas ion lasers based on krypton instead of argon. Krypton ion lasers typically emit at 647.1 nm, 413.1 nm, or 530.9 nm, but various other lines in the visible, ultraviolet and infrared spectral region are accessible.

Other types of ion lasers are mentioned in the article on gas lasers.

Applications

Multi-watt argon ion lasers can be used e.g. for pumping titanium–sapphire lasers and dye lasers, or for laser light shows. They are rivaled byfrequency-doubled diode-pumped solid-state lasers. The latter are far more power efficient and have longer lifetimes, but are more expensive. Argon tubes have a limited lifetime of the order of a few thousand hours. An argon laser may thus be preferable if it is used only during a limited number of hours, whereas a diode-pumped solid-state laser is the better solution for reliable and efficient long-term operation.

Laser safety issues arise both from the high output power of typical ion lasers and from the high voltage applied to the tube.

Excimer lasers where optical amplification occurs in a plasma containing excited dimers (or other molecules) with an anti-binding electronic ground state

An excimer laser is a powerful kind of laser which is nearly always operated in the ultraviolet (UV) spectral region (→ ultraviolet lasers) and generates nanosecond pulses. The excimer gain medium is a gas mixture, typically containing a noble gas (rare gas) (e.g. argon, krypton, or xenon) and a halogen (e.g. fluorine or chlorine, e.g. as HCl), apart from helium and/or neon as buffer gas. An excimer gain medium is pumped with short (nanosecond) current pulses in a high-voltage electric discharge (or sometimes with an electron beam), which create so-called excimers (excited dimers) – molecules which represent a bound state of their constituents only in the excited electronic state, but not in the electronic ground state. (More precisely, a dimer is a molecule consisting of two equal atoms, but the term excimer is normally understood to include asymmetric molecules such as XeCl as well. The term rare gas halide lasers would actually be more appropriate, and the term exciplex laser is sometimes used.) After stimulated or spontaneous emission, the excimer rapidly dissociates, so that reabsorption of the generated radiation is avoided. This makes it possible to achieve a fairly high gain even for a moderate concentration of excimers.

Different types of excimer lasers typically emit at wavelengths between 157 and 351 nm:

Excimer Wavelength

F2 (fluorine) 157 nm

ArF (argon fluoride) 193 nm

KrF (krypton fluoride) 248 nm

XeBr (xenon bromide) 282 nm

XeCl (xenon chloride) 308 nm

XeF (xenon fluoride) 351 nm

Typical excimer lasers emit pulses with a repetition rate up to a few kilohertz and average output powers between a few watts and hundreds of watts, which makes them the most powerful laser sources in the ultraviolet region, particularly for wavelengths below 300 nm. The power efficiency varies between 0.2 and 2%.

Device Lifetime

Early excimer lasers had limited lifetimes due to a variety of problems, arising e.g. from the corrosive nature of the gases used and from contamination of the gas with chemical byproducts and dust created by the electric discharge. Other challenges are the ablation of material from the electrodes and the high peak power of the required current pulses, which often allowed the thyratron switches to last only for a couple of weeks or months. However, a lot of engineering, involving e.g. the use of corrosion-resistant materials, advanced gas recirculating and purification systems, and solid-state high-voltage switches, has mitigated

challenges of the excimer laser concept to a significant extent. The lifetime of modern excimer lasers is now limited by that of the UV optics, which have to withstand high fluxes of short-wavelength radiation, to something of the order of a few billion pulses.

Applications

The short wavelengths in the ultraviolet spectral region make possible a number of applications:

the generation of very fine patterns with photolithographic methods (microlithography), for example in semiconductor chip production

material processing with laser ablation, exploiting the very short absorption lengths of the order of a few micrometers in many materials, so that a moderate pulse fluence of a few joules per square centimeter is sufficient for ablation

pulsed laser deposition

laser marking and microstructuring of glasses and plastics

fabrication of fiber Bragg gratings

ophthalmology (eye surgery), particularly for vision correction by corneal reshaping with ArF lasers at 193 nm; common methods are laser in-situ keratomileusis (LASIK) and photorefractive keratectomy (PRK)

psoriasis treatment with XeCl lasers at 308 nm

pumping other lasers, e.g. certain dye lasers

Note that excimer lasers raise a variety of safety issues, related to the use of high voltages, the handling of poisonous gases (halogens), and the risk of causing skin cancer and eye damage by irradiation with ultraviolet light.

CO2 lasers are based on a gas mixture in which light is amplified by carbon dioxide molecules. The CO2 laser (carbon dioxide laser) is a laser based on a gas mixture as the gain medium, which contains carbon dioxide (CO2), helium (He), nitrogen (N2), and possibly some hydrogen (H2), water vapor and/or xenon (Xe). Such a laser is electrically pumped via a gas discharge, which can be operated with DC current, with AC current (e.g. 20–50 kHz) or in the radio frequency (RF) domain. Nitrogen molecules are excited by the discharge into a metastable vibrational level and transfer their excitation energy to the CO2 molecules when colliding with them. Helium serves to depopulate the lower laser level and to remove the heat. Other constituents such as hydrogen or water vapor can help (particularly in sealed-tube lasers) to reoxidize carbon monoxide (formed in the discharge) to carbon dioxide.

Figure 1: Schematic setup of a sealed-tube carbon dioxide laser. The gas tube has Brewster windows and is water-cooled.

CO2 lasers typically emit at a wavelength of 10.6 μm, but there are other lines in the region of 9–11 μm (particularly at 9.6 μm). In most cases, average powers are between some tens of watts and many kilowatts. The power conversion efficiency can be well above 10%, i.e., it is higher than for most lamp-pumped solid-state lasers, but lower than for many diode-pumped lasers.

Laser Types

The family of CO2 lasers is very diverse:

For laser powers between a few watts and a several hundred watts, it is common to use sealed-tube or no-flow lasers, where the laser bore and gas supply are contained in a sealed tube. Such lasers are compact and rugged, and reach operation lifetimes of several thousands of hours.

High-power diffusion-cooled slab lasers (not to be confused with solid-state slab lasers) have the gas in a gap between a pair of planar water-cooled RF electrodes. The excess heat is efficiently transferred to the electrodes by diffusion, if the electrode spacing is made small compared with the electrode width. Several kilowatts of output are possible.

Fast axial flow lasers and fast transverse flow lasers are also suitable for multi-kilowatt continuous-wave output powers. The excess heat is removed by the fast-flowing gas mixture, which passes an external cooler before being used again in the discharge.

Transverse excited atmosphere (TEA) lasers have a very high (about atmospheric) gas pressure. As the voltage required for a longitudinal discharge would be too high, transverse excitation is done with a series of electrodes along the tube. TEA lasers are operated in pulsed mode only, as the gas discharge would not be stable at high pressures, and are suitable for average powers of tens of kilowatts.

There are gas dynamic CO2 lasers for multi-megawatt powers (e.g. for anti-missile weapons), where the energy is not provided by a gas discharge but by a chemical reaction in a kind of rocket engine.

The concepts differ mainly in the technique of heat extraction, but also in the gas pressure and electrode geometry used. In low-power sealed-tube lasers (used e.g. for laser marking), waste heat is transported to the tube walls by diffusion or a slow gas flow. The beam

quality can be very high. High-power CO2 lasers utilize a fast forced gas convection, which may be in the axial direction (i.e., along the beam direction) or in the transverse direction (for the highest powers).

Applications

CO2 lasers are widely used for material processing, in particular for

cutting plastic materials, wood, die boards, etc., exhibiting high absorption at 10.6 μm, and requiring moderate power levels of 20–200 W

cutting and welding metals such as stainless steel, aluminum or copper, applying multi-kilowatt powers

laser marking of various materials.

Other applications include laser surgery (including ophthalmology) and range finding.

CO2 lasers used for material processing (e.g. welding and cutting of metals, or laser marking) are in competition with solid-state lasers(particularly YAG lasers and fiber lasers) operating in the 1-μm wavelength regime. These shorter wavelengths have the advantages of more efficient absorption in a metallic workpiece, and the potential for beam delivery via fiber cables (there are no optical fibers for high-power 10-μm laser beams). The potentially smaller beam parameter product of 1-μm lasers can also be advantageous. However, the latter potential normally cannot be realized with high-power lamp-pumped lasers, and diode-pumped lasers tend to be more expensive. For these reasons, CO2 lasers still dominate the cutting and welding business, particularly for parts with a thickness greater than a few millimeters. This may change in the future due to the development of high-power thin-disk lasers and advanced fiber cables in combination with techniques which exploit the high beam quality of such lasers.

Due to their high powers and high drive voltages, CO2 lasers raise serious issues of laser safety. However, their long operation wavelength makes them relatively eye-safe at low intensities.

Bibliography

[1] C. K. N. Patel, “Continuous-wave laser action on vibrational–rotational transitions of CO2”, Phys. Rev. 136 (5A), A1187 (1964)

[2] C. K. N. Patel, “Interpretation of CO2 optical maser experiments”, Phys. Rev. Lett. 12 (21), 588 (1964)

[3] A. Robinson and D. Johnson, “A carbon dioxide laser bibliography, 1964–1969”, IEEE J. Quantum Electron. 6 (10), 590 (1970)

[4] P. T. Woods et al., “Stable single-frequency carbon dioxide lasers”, J. Phys. E: Sci. Instrum. 9, 395 (1976)

[5] A. L. S. Smith and J. Mellis, “Operating efficiencies in pulsed carbon dioxide lasers”, Appl. Phys. Lett. 41, 1037 (1982)

[6] K. M. Abramski et al., “Power scaling of large-area transverse radiofrequency discharge CO2 lasers”, Appl. Phys. Lett. 54, 1833 (1989)

[7] O. Svelto, Principles of Lasers, Plenum Press, New York (1998)

Bibliography

[1]F. G. Houtermans, “Über Massen-Wirkung im optischen Spektralgebiet und die Möglichkeit absolut negativer Absorption für einige Fälle von Molekülspektren (Licht-Lawine)”, Helv. Phys. Acta 33, 933 (1960)

[2]I. S. Lakoba and S. I. Yakovlenko, “Active media of exciplex lasers (review)”, Sov. J. Quantum Electron. 10 (4), 389 (1980)

[3]J. J. Ewing, “Excimer laser technology development”, IEEE J. Sel. Top. Quantum Electron. 6 (6), 1061 (2000)

[4] Ch. K. Rhodes (Editor), Excimer Lasers, 2nd edition, Springer, Berlin (1998)

[5] D. Basting and G. Marowski (Editors), Excimer Laser Technology, Springer, Berlin (2004)

Laser Diodes

semiconductor lasers with a current-carrying p–n junction as the gain medium. Laser diodes (= diode lasers) are electrically pumped semiconductor lasers in which the gain is generated by an electrical current flowing through a p–n junction or (more frequently) a p–i–n structure. In such a heterostructure, electrons and holes can recombine, releasing the energy portions as photons. This process can be spontaneous, but can also be stimulated by incident photons, in effect leading to optical amplification, and with optical feedback in a laser resonator to laser oscillation. The article on semiconductor lasers describes more in detail how the laser amplification process in a semiconductor works.

Most semiconductor lasers are diode lasers, but there are also optically pumped semiconductor lasers which do not require a diode structure and thus do not belong to the category of diode lasers.

Types of Laser Diodes

Laser diodes are normally built as edge-emitting lasers, where the laser resonator is formed by coated or uncoated end facets (cleaved edges) of the semiconductor wafer. They are often based on a double heterostructure, which restricts the generated carriers to a narrow region and at the same time serves as a waveguide for the optical field (double confinement). The current flow is restricted to the same region, sometimes using isolating barriers. Such arrangements lead to a relatively low threshold pump power and high efficiency. The active

region is usually quite thin – often so thin that it acts as a quantum well. In some cases, quantum dots are used.

Some modern kinds of LDs are of the surface-emitting type (see below), where the emission direction is perpendicular to the wafer surface, and the gain is provided by multiple quantum wells.

There are very different kinds of LDs, operating in very different regimes of optical output power, wavelength, bandwidth, and other properties:

Small edge-emitting LDs generate between a few milliwatts and up to roughly half a watt of output power in a beam with high beam quality. The output may be emitted into free space or coupled into a single-mode fiber. Such lasers can be designed to be either index guiding (with a waveguide structure guiding the laser light within the LD) or gain guiding (where the beam profile is kept narrow via preferential amplification on the beam axis).

Small LDs made as distributed feedback lasers (DFB lasers) or distributed Bragg reflector lasers (DBR lasers) with short resonators can achieve single-frequency operation, sometimes combined with wavelength tunability.

External cavity diode lasers contain a laser diode as the gain medium of a longer laser resonator, completed with additional optical elements such as laser mirrors or a diffraction grating. They are often wavelength-tunable and exhibit a small emission linewidth.

Broad area laser diodes (also often called broad stripe laser diodes, wide stripe lasers, or high brightness diode lasers) generate up to a few watts of output power. The beam quality is significantly poorer than that of lower-power LDs, but better than that of diode bars (see below). Tapered broad-area lasers can exhibit an improved beam quality and brightness.

Slab-coupled optical waveguide lasers (SCOWLs), containing a multi-quantum well gain region in a relatively large waveguide, can generate a watt-level output in a diffraction-limited beam with a nearly circular profile.

High-power diode bars contain an array of broad-area emitters, generating tens of watts with poor beam quality. Despite the higher power, the brightness is lower than that of a broad area LD.

High-power stacked diode bars (→ diode stacks) are stacks of multiple diode bars for the generation of extremely high powers of hundreds or thousands of watts.

Monolithic surface-emitting semiconductor lasers (VCSELs) typically generate a few milliwatts with high beam quality. There are also external-cavity versions of such lasers

(VECSELs) which can generate much higher powers with still excellent beam quality.

Laser diodes may emit a beam into free space, but many LDs are also available in fiber-coupled form. The latter makes it particularly convenient to use them, e.g., as pump sources for fiber lasers and fiber amplifiers.

Nearly all electrically pumped semiconductor lasers are laser diodes; quantum cascade lasers are an exception. Other semiconductor lasers rely on optical pumping and therefore do not require a p–n junction; they can be made of undoped semiconductor materials.

Emission Wavelengths

The emission wavelength of a laser diode is essentially determined by the bandgap of the laser-active semiconductor material: the photon energy is close to the bandgap energy. In quantum well lasers, there is also some influence of the quantum well thickness. A variety of semiconductor materials makes it possible to cover wide spectral regions. In particular, there are many ternary and quaternary semiconductor compounds, where the bandgap energy can be adjusted in a wide range simply by varying the composition details. For example, an increased aluminum content (increased x) in AlxGa1−xAs causes an increase in the bandgap energy and thus a shorter emission wavelength. Table 1 gives an overview on typical material systems.

Laser diode material

(active region / substrate) Typical emission wavelengths Typical application

InGaN / GaN, SiC 380, 405, 450, 470 nm data storage

AlGaInP / GaAs 635, 650, 670 nm laser pointers, DVD players

AlGaAs / GaAs 720–850 nm CD players, laser printers, pumping solid-state lasers

InGaAs / GaAs 900–1100 nm pumping EDFAs and other fiber amplifiers; high-power VECSELs

InGaAsP / InP 1000–1650 nm optical fiber communications

Table 1: Emission wavelengths of various types of laser diodes.

Most laser diodes emit in the near-infrared spectral region, but others can emit visible (particularly red or blue) light or mid-infrared light.

Emission Bandwidth and Wavelength Tuning

Most LDs emit a beam with an optical bandwidth of a few nanometers. This bandwidth results from the simultaneous oscillation of multiple longitudinal (and possibly transverse) resonator modes (multimode laser diodes). Some other kinds of LDs, particularly distributed feedback lasers, operate on a single resonator mode (→ single-frequency operation), so that the emission bandwidth is much narrower, typically with a linewidth in the megahertz region. Further linewidth narrowing is possible with external cavities and particularly with narrowband optical feedback from a reference cavity (→ stabilization of lasers).

The emission wavelength (center of the spectrum) of multimode LDs is usually temperature sensitive, typically with an increase of ≈ 0.3 nm per 1 K temperature rise, resulting from the temperature dependence of the gain maximum. (The temperature influences the thermal population distributions in the valence and conduction band.) For that reason, the junction temperature of LDs for diode pumping of solid-state bulk lasers has to be stabilized, if the absorption bandwidth of the laser crystal is narrow (e.g. only a few nanometers wide). It is also possible to tune the emission wavelength via the junction temperature.

Single-mode diodes can have a significantly smaller temperature coefficient of the emission wavelength. For applications in scanning spectroscopy, the wavelength is sometimes scanned by operating the laser intermittently. The temperature then rises during each current pulse and causes the optical frequency to fall. The wavelength of external-cavity lasers can also be tuned, e.g. by rotating the diffraction grating in the laser cavity.

Power Conversion Efficiency

Diode lasers can reach high electrical-to-optical efficiencies – typically of the order of 50%, sometimes even above 60%. (There are development programs on the way to push efficiencies of high-power LDs above 70%.) The efficiency is usually limited by factors such as the electrical resistance, carrier leakage, scattering, absorption (particularly in doped regions), and spontaneous emission. Particularly high efficiencies are achieved with laser diodes emitting e.g. around 940–980 nm (as used e.g. for pumping ytterbium-doped high-power fiber devices), whereas 808-nm diodes are somewhat less efficient.

Beam Quality and Beam Shaping

Some low-power LDs can emit beams with relatively high beam quality (even though the high beam divergence requires some care to preserve that during collimation). Most higher-

power LDs, however, exhibit a relatively poor beam quality, combined with other non-favorable properties, such as a large beam divergence, high asymmetry of beam radius and beam quality between two perpendicular directions, and astigmatism. It is not always trivial to find the best design for beam shaping optics, being compact, easy to manufacture and align, preserving the beam quality and avoiding interference fringes, removing astigmatism, having low losses, etc. Typical parts of such diode laser beam shaping optics are collimating lenses (spherical or cylindrical), apertures, and anamorphic prisms.

Beam Combining

As the light emitted by a laser diode is linearly polarized, it is possible to combine the outputs of two diodes with a polarizing beam splitter, so that an unpolarized beam with twice the power of a single diode but the same beam quality can be obtained (polarization multiplexing). Alternatively, it is possible to combine the beams of LDs with slightly different wavelengths using dichroic mirrors (→ spectral beam combining). More systematic approaches of beam combining allow combining a larger numbers of emitters with a good output beam quality.

Pulse Generation

Although the most common mode of operation of LDs is continuous-wave operation, many LDs can also generate optical pulses. In most cases, the principle of pulse generation is gain switching, i.e. modulating the optical gain by switching the pump current. Small diodes can also be mode-locked for generating picosecond or even femtosecond pulses. Mode-locked laser diodes can be external-cavity devices or monolithic, in the latter cases often containing different sections operated with different current.

Noise Properties

Different types of diodes have very different noise properties. The intensity noise is typically close to quantum-limited only well above the relaxation oscillation frequency, which is very high (often several gigahertz). However, some low-power LDs operated at cryogenic temperatures have been demonstrated to exhibit even significant amplitude squeezing, i.e., intensity noise well below the shot noise limit. In all semiconductor lasers, intensity noise is generally coupled to phase noise, making these noise properties strongly correlated.

As mentioned above, linewidth values are very different. Multimode LDs exhibit a lot of excess noise associated with mode hops. Noise in different modes can be strongly anti-correlated, so that the intensity noise in single modes can be much stronger than the noise of

the combined power. This has the important consequence that the intensity noise can be increased when the beam e.g. of a diode bar is truncated at an aperture or spectrally filtered.

The diode driver can also contribute a lot to the laser noise, because even very fast current fluctuations can be transformed into intensity and phase fluctuations of the generated light.

Device Lifetime

When operated under proper conditions, diode lasers can be very reliable during lifetimes of tens of thousands of hours. However, much shorter lifetimes can result from a number of factors, such as operation at too high temperatures (e.g. caused by insufficient cooling) and current or voltage spikes, e.g. from electrostatic discharge or ill-designed laser drivers.

There are different failure modes, including catastrophic optical damage (COD) (with complete device destruction within milliseconds or less) and steady degradation. Apart from the operation conditions, various design factors strongly influence the lifetime. For example, designs with aluminum-free active regions have been found to have superior reliability and lifetimes, and certain coatings (or just additional semiconductor layers) on optical surface can also be very helpful. The details of some advanced diode designs have not been disclosed by manufacturers in order to maintain a competitive advantage.

In order to improve device lifetimes, LDs are often operated at reduced current levels (and thus output powers). Moderate power reductions can at the same time increase the wall-plug efficiency due to the lower junction voltage, whereas stronger reductions reduce the efficiency.

Applications

Laser diodes are used in a very wide range of applications. The following list gives some important examples:

Low-power single-mode LDs with high beam quality are used for data recording and reading on CD-ROMs, DVDs, Blu-ray Discs, and holographic data storage media. Such lasers can operate in different spectral domains from the infrared to the blue and violet region, with the shorter wavelengths allowing for higher recording densities.

Single-mode LDs are widely used in optical fiber communications, particularly in data transmitters. In some cases, the data modulation is done directly via the drive current.

Single-mode LDs are also applied in spectroscopy with very compact low-power measurement devices.

Small red laser diodes (→ red lasers) are used as laser pointers.

Distance measurements are often done with modulated low-power diode lasers. Similar lasers are used in laser printers, scanners and bar code readers.

Broad area laser diodes, diode bars and diode stacks are often used for diode pumping of solid-state lasers. Fiber-coupled broad area LDs also serve as pump sources of fiber amplifiers.

Some kinds of surgery (e.g. treatment of enlarged prostates) and dermatological therapies can be done with radiation from diode bars.

High-power diode stacks are directly used in material processing in cases where a high beam quality is not required, e.g. for surface hardening, welding and soldering. Compared with other high-power lasers, they are simpler and have a much better wall-plug efficiency.

In terms of sales volumes, the applications in optical data storage and telecommunications are very dominating. The third most important application, which is pumping of solid-state lasers, already has sales volumes which are nearly an order of magnitude lower than the previously mentioned sectors.

Related Devices

LDs are often used in the form of laser diode modules, containing a variety of additional components e.g. for beam shaping and cooling and protection of the LD, and wavelength conversion.

A semiconductor optical amplifier (SOA) has a setup which is similar to an LD, but the end reflections are suppressed. Without an input signal, such a device can act as a superluminescent diode (SLD), generating light via amplified spontaneous emission. The optical spectrum is then smooth and normally much broader.

Light-emitting diodes (LEDs) use the same mechanism of generating photons as LDs, but they usually do not exhibit significant optical amplification (laser gain). So-called resonant-cavity LEDs do exploit some degree of stimulated emission, and are in this sense intermediate between ordinary LEDs and SLDs.

Bibliography

[1] R. N. Hall et al., “Coherent light emission from GaAs junctions”, Phys. Rev. Lett. 9 (9), 366 (1962)

[2] N. Holonyak and S. F. Bevacqua, “Coherent (visible) light emission from Ga(AS1−xPx) junctions”, Appl. Phys. Lett. 1, 82 (1962)

[3] J. F. Butler et al., “Properties of the PbSe diode laser”, IEEE J. Quantum Electron. 1 (1), 4 (1965)

[4] C. A. Wang and S. H. Groves, “New materials for diode laser pumping of solid-state lasers”, IEEE J. Quantum Electron. 28 (4), 942 (1992)

[5] P. J. Delfyett et al., “High-power ultrafast laser diodes”, IEEE J. Quantum Electron. 28 (10), 2203 (1992)

[6] J. G. Endriz et al., “High power diode laser arrays”, IEEE J. Quantum Electron. 28 (4), 952 (1992)

[7] J. V. Moloney et al., “Quantum design of semiconductor active materials: laser and amplifier applications”, Laser & Photon. Rev. 1 (1), 24 (2007)

[8] W. W. Chow and S. W. Koch, Semiconductor-Laser Fundamentals, Springer, Berlin (1999)

[9] L. A. Coldren and S. W. Corzine, Diode Lasers and Photonic Integrated Circuits, John Wiley & Sons, New York (1995)