Changes in red were made as a group by Pete Latham – course director, Yoriko Morita, Nick Massa andTung Jeong.

(1) Semiconductor lasers are diodes that emit coherent light by stimulated emission. They consist of a p-n junction inside a slab of semiconductor that is typically much less than a millimeter in any dimension. Excitation is provided by current flow through the device, and the cleaved ends of the diode provide the feedback mirrors.

(2)The output characteristics of diode lasers differ from those of other laser types in two important ways. Because of their small size they have large beam divergence angles of as much as 20° . The nature of the active medium also allows lasing over a broad wavelength range and produces an output that is far less monochromatic than other laser types. Nevertheless, semiconductor lasers have important applications in communications and range finding. Diode lasers are also being developed for medical and industrial applications.

(3) This module discusses in detail the construction and operation of a gallium arsenide semiconductor laser, the most common type. In the laboratory the student will operate a gallium arsenide laser and measure its output characteristics.

Upon completion of this module, the student should be able to:

1. Draw and label a diagram of a simple gallium arsenide laser.

2. Explain the charge flow through a semiconductor in the n-region, the p-region, and the junction.

3. Draw and label the energy-level diagram of the gallium arsenide laser; use the diagram to explain the energy transfer during the lasing process.

4. Explain the design of four junction configurations used in gallium arsenide lasers and the characteristics of each.

5. Draw a diagram showing the cross section of a stripe geometry laser diode, and explain how this configuration improves laser performance.

6. Explain how the laser wavelength and threshold current vary with operating temperature.

7. Describe the spectral output of a gallium arsenide laser.

8. Describe the beam divergence of a laser diode, and explain the factors leading to this divergence.

9. Describe two processes that produce damage in laser diodes.

10. Explain two important applications of laser diodes.

11. Operate a gallium arsenide laser in the laboratory and measure its output characteristics.

Energy Transfer In Semiconductor Lasers

(4) Semiconductors are materials that have electrical conductivity intermediate between the high conductivity of metals and the low conductivity of insulators. In a good electrical conductor, such as a metal, so that conduction of electricity is easy. In an insulating material (for example, common table salt), the electrons are tightly bound to their parent atoms and are not free to move through the material when a voltage is applied to it. Therefore, the electrical conductivity is low. In a semiconductor the outer electrons are usually weakly bound to their parent atoms, but a small fraction of them can migrate through the material, so that there is a small amount of electrical conductivity.

(5) The most commonly known semiconductors are silicon and germanium, which have been used for electronic applications such as rectifiers and transistors. However, silicon and germanium have not yet been used in lasers. The most common semiconductor material that has been used in lasers is gallium arsenide, which is a compound of chemical element 31, gallium, and chemical element 33, arsenic. Its chemical designation is GaAs. Other semiconductor materials that have been employed as lasers are discussed later in this module, but the most common semiconductor laser, and the only one that is widely available commercially, is gallium arsenide.

Current Flow Through A Semiconductor

(6) Semiconductors such as silicon are characterized by having four electrons in their outer atomic shell. In the semiconductor crystal each atom forms electron pair bonds with four other atoms. The configuration of atoms in a semiconductor crystal is called a crystaline lattice. Each of these contributes an additional electron, resulting in a total of eight electrons which completely fills the outer atomic shell. By sharing each of its four outer electrons with four neighboring atoms, each atom "fills" its outer shell. This leaves no free electrons for conduction in a pure semiconductor. Some conduction occurs because the outer electrons are not tightly bound to the atoms and may be freed by thermal or electrical energy. Thus pure semiconductors have a high electrical resistance at low temperature and lower resistance at higher temperature.

(7) The electrical conductivity of a semiconductor can be increased by adding doping elements, or small percentages of impurity elements, to the semiconductor. The presence of the small traces of impurity elements can yield extra charge carriers which are free to move through the material.

(8) In the compound gallium arsenide, each gallium atom has three electrons in its outermost shell of electrons and each arsenic atom has five. This gives an average of four electrons per atom in the compound. When a trace of an impurity element with two outer electrons, such as zinc, is added to the crystal, the result is the shortage of one electron from one of the pairs. This shortage sets up an imbalance in which there is a place in the crystal for an electron but there is no electron available. This is commonly called a "hole." This forms the so-called p-type semiconductor in which the conduction of electricity is by motion of the hole from one atom to another. Here p stands for positive, because the hole, or lack of an electron, looks like an extra positive charge.

(9) When a trace of an impurity element with six outer electrons, such as selenium, is added to a crystal of GaAs, it provides on additional electron which is not needed for the bonding. This electron can be free to move through the crystal. Thus, it provides a mechanism for electrical conductivity. This type is called an n-type semiconductor, where n stands for negative because the carrier of electricity is the negatively charged electron.

(10) When p-type and n-type regions are grown side by side in a semiconductor material the result is a p-n junction. On one side of the junction conduction is by electrons, and on the other side it is by positive holes. Such a device is called a diode, and it allows current flow in only one direction.

(11) When a forward voltage is applied to the junction, i.e., the positive side of a battery is connected to the p-side and the negative to the n-side, the density of the carriers, both p-type and n-type, increases around the junction. Electrons move into the junction from the n-type side, and holes move into the junction from the p-type side. The battery applies energy to the electrons and holes, so that they are freed into a higher energy state. Because a hole is essentially the lack of electron in a bond, the hole may be filled with one of the extra electrons. When a reverse voltage is applied to the junction, the charge carriers move away from the junction to create a depletion region with no charge carriers. Thus, only a very small current flows when the diode is reverse biased.

Emission of Light by Semiconductor Diodes

(12) Figure 1 shows the energy-level diagram of a GaAs semiconductor diode. In semiconductor materials, electrons may have an energy within certain bands. In the figure the lower region is called the valence band and represents the energy states of bound electrons. The upper region is called the conduction band and represents the energy states of free or conduction electrons. The conduction band is drawn above the valence beam because it indicates a higher energy level. Electrons may have energies in either of these bands, but not in the gap between the bands.

Fig. 1
Energy-level diagram of GaAs semiconductor diode.

(13) Figure 1a shows the relative populations of the energy bands or both sides of a p-n junction with no voltage applied to the diode. The n-type material contains electrons which behave as the current carriers in its conduction band, whereas the p-type material has holes for carriers in its valence band. When a forward voltage is applied to the diode, the energy levels are caused to shift as illustrated in Figure 1b. Under these conditions there is a significant increase in the concentration of electrons in the conduction band near the junction on the n-side and the concentration of holes in the valence band near the junction on the p-side. The electrons and holes recombine (conduction band electrons move into empty valence band states) and energy is given off in the form of photons. As the electron recombines with a hole it moves from a higher energy state in the conduction band back to the lower energy state in the valence band. The energy of the photon resulting from this recombination is equal to that associated with the energy gap. That is the difference in energy between the conduction and valence bands as shown by Fig. 1b. This energy is often in the form of electromagnetic radiation. In light-emitting diodes (LED) this light energy is transmitted out through the sides of the junction region. For the LED, all of the light is created by spontaneous emission due to electron and hole recombination. In semiconductor lasers the junction forms the active medium, and the reflective ends of the laser material provide feedback. Because of this feedback in diode lasers, most of the light is created by stimulated emission.

Gallium-Arsenide Laser Construction

(14) GaAs lasers are prepared from ingots of gallium arsenide crystals that are mounted on a substrate glass plate with wax and cut into slices, usually about 0.5 mm thick. After the slice is formed, the junction is prepared by diffusing impurities in from the surface. Thus, if the original ingot is n-type, a p-type impurity (for example, zinc) is diffused in. The diffusion is carried out by sealing the gallium arsenide slice into an evacuated ampoule, together with a weighed amount of zinc. The ampoule is then heated in a furnace. Zinc vapor diffuses a short distance (a few microns) into the surface of the slice and forms a p-n junction. The slice then is cut or cleaved into blocks of suitable size to form the individual lasers. The blocks are small and fragile and are mounted by sandwiching them between two gold-clad metal disks. Typical dimensions of the block are of the order of one or two millimeters. The light-emitting region—that is the junction—from which the radiation originates, is a thin layer only a few microns thick. The completed laser diodes are then attached to a copper heat sink on one side and a small electrical contact on the other.

Basic Semiconductor Laser Design

(15) Figure 2 is a diagram of the simplest (and earliest) type of gallium arsenide laser. The laser is a brick-shaped piece of GaAs prepared by the process outlined above. GaAs cleaves easily along certain crystal planes, leaving flat parallel surfaces.

Fig. 2
Gallium arsenide laser diode.

(16) Usually, the mirrors for feedback and output coupling are formed by the cleaved ends of the laser diode, with no further coating. The reflectivity at the interface between gallium arsenide and air is approximately 36%. If output is desired from only one end of the device, or if mirrors of higher reflectivity are desired to reduce the threshold for laser operation, the reflectivity may be increased by coating the surfaces with metal films. Optical standing waves may exist between any two of the parallel surfaces of the diode. Two sides are purposely roughened to reduce reflection and prevent lasing "across" the diode cavity.

(17) Because the diode laser is not 100% efficient, not all of the electrical energy input goes into creating photons. Some input energy into the semiconductor crystaline lattice itself. This elevates the bulk temperature of the semiconductor, but does not contribute to lasing.

(18) The output power available from this laser is limited by the loop gain available within the laser cavity. The amplifier gain of the active medium is dependent on the current density through the junction. Higher currents produce greater power, but higher currents also increase heating effects that can damage the device.

(19) Loss in the laser cavity has two primary contributors. The first of these is diffraction loss. The active region has a width of only about a micron. Thus, light quickly diverges out of the active region. This loss may be reduced by making the junction wider and by better confining the light to the active region. The second loss factor is absorption of the laser light by free carriers in the junction region. This loss may be reduced by reducing the number of free carriers. This reduction is accomplished by lowering the temperature of the device.

Heterojunction Laser Design

(20)Heterojunction lasers are GaAs laser diodes with an advanced junction design to reduce diffraction loss in the optical cavity. This is accomplished by modification of the laser material to control the index of refraction of the cavity and the width of the junction.

(21)The p-n junction of the basic GaAs laser described above is called a homojunction because only one type of semiconductor material is used in the junction with different dopants to produce the junction itself. The index of refraction of the material depends upon the impurity used and the doping level. Figure 3a shows the variation of index of refraction across such a junction. The junction region is actually lightly doped with p-type material and has the highest index of refraction. The n-type material and the more heavily doped p-type material both have lower indices of refraction. This produces a light pipe effect that helps to confine the laser light to the active junction region. In the homojunction, however, this index difference is low and much light is lost.

(22)Figure 3b shows a gallium arsenide-aluminum gallium arsenide heterojunction. A fraction of the gallium in the p-type layer has been replaced by aluminum. This reduces the index of refraction of this layer and results in better confinement of the laser light to the optical cavity. By varying the amount of aluminum, changing x in Fig. 3b, the wavelength of the laser output can be varied.

Fig. 3
Structure and index of refraction n for various types of junctions in gallium arsenide with a junction width d.
(a) is for a homojunction. (b) is for a gallium arsenide-aluminum gallium arsenide single heterojunction. (c) is for a gallium arsenide-aluminum gallium arsenide double heterojunction with improved optical confinement. (d) is for a double heterojunction with a large optical cavity of width w.

(23)A further improvement is shown in Figure 3c. In this case, only the junction region is composed of gallium arsenide. Both the p and n regions are of aluminum gallium arsenide. The result is better confinement of the optical standing wave on both sides of the optical cavity. This confinement greatly reduces the optical loss, but also leads to two additional difficulties. The optical radiation in the cavity is so well confined that the irradiance at the diode optical surfaces may easily reach the damage threshold, increasing the likelihood of catastrophic failure. The tight confinement of the beam also reduces the effective width of the output aperture of the laser. This increases the divergence angle in the direction perpendicular to the junction. As in Fig. 3b, the wavelength can be changed by varying the aluminum concentration, x.

(24)These difficulties are overcome by a further development shown in Figure 3d. This device is called a large-optical-cavity (LOC) laser and uses regions of aluminum gallium arsenide of varying composition. The active region consists of the GaAs layer at the center of the structure. High gain and lower-current operation are achieved by confining the charge carriers to a smaller region. The typical width of this region is from 0.1 to 1.0 micron. The optical cavity of the system is the width w of the diagram and is typically several microns. This reduces the diffraction loss of the system.

(25)Each of the advances described has lowered the operating threshold of GaAs lasers. Table 1 lists the typical current densities necessary to achieve the lasing threshold of the various junction types. In this case, there is considerable flexibility in laser wavelength by varying the amount of aluminum in the layers, changing x and y.

Table 1. Typical Threshold Current Density for
GaAs Laser Operation at 300° K.

Stripe Geometry

(26)In the laser shown in Figure 2, the top electrode covers the entire top surface of the p-type material. This allows current flow across the full width of the diode, which is typically several hundred microns. The current density, and thus the gain, can be greatly increased if current flow is confined to a narrow strip of the junction. This does not greatly reduce the maximum current that can be used, as the current limitation is the heating effect in the material.

(27)Figure 4 shows a cross section of a large-optical-cavity heterojunction laser diode with stripe geometry for current confinement. The diode junction is constructed in the usual fashion. A layer of silicon dioxide insulating material is then deposited on the top of the diode, except for a strip about 13 microns wide down the center of the diode. The metal electrode is then applied. The result is a laser diode in which the current is confined to a single line along the length of the laser.

Fig. 4
Configuration of stripe geometry
heterojunction semiconductor laser.

Output Characteristics of GaAs Lasers

(29)The spectral width of the radiation from the gallium arsenide laser is typically around two or three nanometers, much larger than the spectral width of most other lasers. Therefore, a gallium arsenide laser can be regarded as a small bright area source of radiation. Although gallium arsenide lasers do not possess the properties of directionality and monochromaticity to the same degree as other lasers, they do have many important properties that make them attractive for some practical applications. They can be modulated easily at high frequencies by modulating the current through the junction. They are efficient, small, and rugged, and are much less expensive than other types of lasers. Their small size makes semiconductor diode lasers readily packageable for multiple applications.

Temperature Dependence of Laser Output

(30)The current threshold for lasing in GaAs is strongly temperature-dependent, as shown in Figure 5. At low temperature (up to approximately 30 ° K) the threshold is fairly constant. Above 100 ° K, the threshold current density for laser operation increases rapidly with increasing temperature. At the cryogenic temperature of 77 ° K, the threshold current in a gallium arsenide laser is about one tenth that of the room temperature value. This means that cooling to cryogenic temperatures changes the operating and performance characteristics of the laser.

Fig. 5
Temperature dependence of threshold current.

(31)Gallium arsenide lasers emit radiation in the near infrared portion of the spectrum. The exact wavelength depends on the temperature at which the laser is operated. This is shown in Figure 6 which gives the wavelength of a gallium arsenide laser as a function of temperature. At temperatures above room temperature, the threshold for laser operation rises and it becomes difficult to operate the laser. However, gallium arsenide lasers have been operated over the range of temperatures from liquid helium temperature to room temperature.

Fig. 6
Temperature dependence of lasing wavelength.