KR100459579B1 - Be-containing ⅱ-ⅵ blue-green laser diodes - Google Patents

Abstract

The II-VI compound semiconductor laser diode 10 includes a plurality of II-VI semiconductor layers forming a pn junction supported by a single crystal GaAs semiconductor substrate 12. The layers formed in the pn junction may comprise a first cladding layer 20 of a first conductivity type, a second cladding layer 22 of a second conductivity type and the first and second cladding layers 20, 22. At least a first guiding layer 14 between. A quantum well active layer 18 is located within the pn junction. Electrical energy is connected to the laser diode 10 by the first and second electrodes 40, 41. Various layers 14, 16, 20, 22, 36, 38 are formed in the laser diode using Be.

Description

BE-CONTAINING II-VI BLUE-GREEN LASER DIODES

The present invention relates generally to semiconductor laser diodes, and more particularly to blue-green laser diodes made from II-VI semiconductor materials.

Conventional laser diodes generate infrared and red light. However, laser diodes that emit short wavelengths, such as green and blue spectral portions (ie wavelengths of 590-430 nm), are widely used in many commercial applications. In addition, these short wavelength laser diodes increase the performance and capabilities of many existing systems that currently use infrared and red laser diodes.

Efforts to manufacture blue-green laser diodes have continued. For example, US Pat. No. 5,513,199, entitled "Blue-Green Laser Diode," issued April 30, 1996 to Haase et al., Is one such example. This reference discloses a blue-green laser using a group II-VI isolation confinement heterostructure (SCH). Typically, the laser diode is fabricated on a GaAs substrate and all layers are nominally lattice matched to the substrate, except for p-type contacts, which typically include pseudomorphic quantum wells and relaxed ZnTe. In such a laser, the light output is 515 nm, the guide layer and the cladding layer are nominally lattice matched with the GaAs substrate and have bandgaps of 2.72 and 2.85 eV, respectively. The laser diode uses a quantum well and a ZnTe layer subjected to high compression distortion.

This particular structure is promising, and the room temperature continuous oscillation laser life is recorded for 10 hours or more. However, it is desirable to increase the power, lifetime and reliability of blue-green laser diodes.

As disclosed in US Pat. No. 5,422,902, it has been proposed to form resistive contacts using beryllium with Te. It also describes the growth of BeTe and BeMgZnSe alloys and the structure of a basic LES that emits blue light. However, the technique lacks laser diodes manufactured by the Be-containing method to form useful devices.

BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a cross-sectional view (not to scale) showing the structure of a II-VI semiconductor laser diode according to the present invention.

2 is a graphical representation of bandgaps and lattice constants for various compositions.

3 is a relationship diagram of light emission-current-voltage of the element shown in FIG. 1;

4 is a cross-sectional view (the scale factor is not constant) showing the structure of a group II-VI semiconductor laser diode having an n-type contact upwardly according to another embodiment of the present invention.

The present invention relates to a group II-VI compound isolation isolation heterostructure semiconductor laser diode comprising a plurality of layers of group II-VI semiconductors forming a pn junction. The layers comprise a first cladding layer of a first conductivity type II-VI semiconductor comprising Be. The first guiding layer of the II-VI semiconductor is stacked on the first cladding layer. The active layer of the II-VI semiconductor is stacked on the first guiding layer, and the second guiding layer of the II-VI semiconductor is stacked on the active layer. The active layer may include one quantum well active layer. A second cladding layer of II-VI semiconductor containing Be and having a second conductivity type is stacked on the second guiding layer.

The performance of prior art laser diodes is limited for various reasons. In the laser diode of the prior art, since sulfur (S) is present in the structure, it is difficult to control the composition of the ternary element and the quaternary element layer. Similarly, the presence of S in the growth chamber generally results in stacking defect failures that occur at the GaAs / ZnSe interface. In addition, p-type contact life is insufficient in prior art laser diodes, which may be the result of a relatively high (8%) lattice mismatch between ZnTe and GaAs.

One feature of the present invention involves the use of beryllium (Be) in laser diodes, which offers many advantages. In particular, many advantages are provided by replacing sulfur (S) with beryllium (Be) in the guiding and cladding layers.

The structure of the laser diode 10 according to the invention is shown in FIG. 1. The laser diode 10 is a group II-VI device with a wide bandgap fabricated by heteroepitaxial growth on a GaAs substrate by a molecular beam epoxy (MBE). The laser diode 10 is fabricated on a GaAs substrate 12 and includes lower (first) and upper (second) BeZnSe light guide layers 14, 16, each separated by a CdZnSe quantum well active layer 18, respectively. do. The surfaces of the light guide layers 14, 16 opposite the active layer 18 are in contact with the lower (first) and upper (second) BeMgZnSe cladding layers 20, 22, respectively. The lower ZnSe: Cl buffer layer 24 is disposed on the surface of the lower cladding layer 20 opposite to the light guide layer 14. The upper BeTn: N / ZnSe resistive contact 34 is located on the surface of the upper cladding layer 22 opposite the light guide layer 16.

The GaAs buffer layer 28 separates the substrate 12 from the lower ZnSe: Cl buffer layer 24 to ensure high crystal quality of the contact layer and subsequent growth layer. The P-type contact 34 is formed by the ZnSe: N layer 26, the ZnSe / BeTe: N grading layer 36 and the BeTe: N layer 38. In addition, electrode 41 is in electrical contact with layer 38. In addition, an electrode 40 is provided on the opposite side of the lower buffer layer 24 to contact the GaAs substrate 12. Layers 20 and 24 are both n-type doped with Cl (ie, first conductivity type). Also, layers 22 and 26 are p-type doped with N (ie, second conductivity type). Active layer 18 is an undoped quantum well layer of a CdZnSe semiconductor.

One feature of the present invention is that Be is used in the guiding, cladding and top contact layers.

Beryllium provides a sticking coefficient of about 1 over a wide range of substrate temperatures. The adhesion rate of about 1 greatly improves composition control. Moreover, since the Be-containing quaternary alloy does not separate laterally, it has a relatively constant transverse composition. In addition, since Be can be easily controlled in the MBE system used during manufacturing, it is less likely to leak around the shutter of the MBE system than S typically used in the prior art. Thus, the use of Be reduces the stacking faults in the structure. Also, because BeTe has higher stacking defect energy than prior art ZnSe, BeTe buffer layers grown on GaAs substrates provide lower stacking defect densities than conventional prior art designs.

Moreover, the present invention improves contact life for the p-type contact 34. In particular, BeTe is used in p-type contact 34 in place of ZnTe. Since BeTe and GaAs are only 0.35% mismatched, approximately 1000 μs of BeTe layers 36 and 38 are in an amorphous state. Rather than attempting to maintain crystal quality at the same time, the p-type contact 34 is digitally grated to assist in hole injection so that the ZnSe / BeTe layer 36 serves this purpose. Can be adjusted exclusively. In addition, since the lattice constant of BeTe is smaller than GaAs, it is also possible to nominally grow the graded region of the contact 34 to a desired thickness while maintaining an irregular shape.

2 is a graph of bandgap versus lattice constant for various materials ignoring bowing effects. Fig. 2 is used for the 3rd and 4th guide layers and is based on the conventional Mg _x Zn _1-x S _y Se _1-y (0≤y≤1) system and Be based on the Be of the present invention. The energy gap of the component compound of the system of _x Mg _y Zn _1-xy Se (0 ≦ x ≦ 1, 0 ≦ y ≦ 1) is shown. In these two systems, the guide layer and cladding layer may be lattice matched with a GaAs substrate. A system based on S theoretically has a higher upper bound for the bandgap, but lateral composition modulation limits the useful bandgap to approximately 3.0 eV. Therefore, because of the improved lateral consistency of the present invention, the Be based quaternary elements can greatly enhance the effect on the same bandgap region. In FIG. 2 all possible bandgap and lattice integer pairs are located in the diamond-shaped region for the S-containing quaternary element and in the triangular-shaped region for the Be-containing quaternary element. However, in order to have a lattice match in GaAs, the composition must be on the vertical line shown in FIG. The resulting bandgap region in the case of the Be and S systems are likewise shown in FIG. 2 as well. The cladding and guide layer composition for the 515 nm laser is also shown in FIG. 2.

The laser structure shown in FIG. 1 includes three-element element guide layers 14 and 16 containing Be and four-element cladding layers 20 and 22, and a p-type contact in which BeTe / ZnSe is digitally clad. . The quaternary elements (cladding layers 20, 22) each grow to a thickness of 1 mu m, and the guide layers 14 and 16 each grow to a thickness of 0.13 mu m. The n-type and p-type quaternary element layers 20, 22 are doped at concentrations of 2x10 ¹⁵ cm ^-3 and 2x10 ¹⁷ cm ^-3 , respectively. The bandgap of the quaternary cladding layers 20, 22 is 2.84 eV. The bandgap of the tertiary element guide layers 14 and 16 is 2.75 eV. The X-ray peaks for the quaternary and tertiary elements are -60 arcsec and +100 arcsec, respectively. The thickness of the quantum well 18 was calculated to be 55 kPa based on the growth rate with a Cd composition of about 25% (ie, Cd _0.25 Zn _0.75 Se).

The device made from this sample exhibits a critical current density of 530 A / cm ² of 4.5 V voltage (minimum 3.8 V). A plot of voltage and optical power as a function of current in a typical gain-guided device of the present invention is shown in FIG. 3. The CW laser wavelength is 511 nm.

4 shows a laser diode 50 in the form of an embodiment, where the n-type contact is over the substrate. That is, the substrate is p-type. The laser diode 50 is similar to the laser diode 10 of FIG. 1 except that the conduction type (p or n) of the laser is reversed. Diode 50 comprises Be in accordance with the invention as described above.

Variations of the specific embodiments described herein are intended to be included within the scope of the present invention. For example, the laser can include a p-type or n-type substrate. For p-type substrates, graded ZnSe-BeTe contacts may be used to facilitate hole transport between the substrate and the II-VI layers. Alternative substrate materials include GaAs, GaP or ZnSe. Alternative alloys that can be used in the guide layer or cladding layer include BeCdTe or BeMgTe, which can be grown lattice matched to GaAs or ZnSe substrates. Active layers that are quantum wells or multiple quantum wells may include alternative II-VI semiconductors such as CdZnSSe, ZnSe, BeCdSe or BeCdZnSe.

Details on MBE Growth

The molecular beam epoxy (MBE) system used in one embodiment of the present invention consists of two MBE chambers connected by ultra high vacuum transfer (UHV) transport tubes. Each chamber can be isolated from the rest of the system by a gate valve. The first chamber is equipped with at least Ga, Si and Be Knudsen effusion cells, a valve-type cracker cell for As and a substrate heater. In addition, the first chamber is further equipped with a reflection high-energy electrom diffraction (RHEED) system for surface characteristics during growth. The substrate temperature is measured by two different means depending on the temperature itself. The temperature is measured by an infrared optical pyrometer from the substrate heater at a temperature above 500 ° C., and by measuring the GaAs band-gap absorption-edge of infrared rays from the substrate heater at a temperature below 500 ° C. Is measured indirectly. The measured absorption edge energy is then obtained by comparing the bandgap versus temperature data for GaAs.

The second MBE chamber contains a Knudsen effluent cell for Mg, Zn, Be, Te, Cd, ZnCl ₂ , a valve cracker cell for Se, an RF plasma source for N (UK, Cambridge, Oxford Applied) Research company) and a substrate heater are provided. The chamber is also equipped with a RHEED system. The substrate temperature is measured by the absorption end measurement described above.

A GaAs substrate 12 doped with Si (001) is placed on the substrate heater in the first MBE chamber to grow the structure 10 with the n-type below in FIG. In preparation for growth, the substrate 12 is heated to a temperature of approximately 580 ° C. under an overpressure of As to remove the oxide film adsorbed onto the substrate 12. After the oxide is removed from the surface, as determined from the RHEED pattern, the substrate 12 is further heated to about 630 ° C. to completely remove the gas and completely remove the oxide from the wafer and wafer holder.

After a few minutes, the substrate 12 is cooled to 580 ° C, and when the temperature is stabilized, the GaAs buffer layer 28 doped with Si grows. Growth of the buffer layer 28 is initiated by simultaneously opening Ga and Si shutters (As valve and shutter are already open). The buffer layer 28 grows at a rate of 1 μm per hour to a thickness of approximately 0.2 μm. During growth, the RHEED pattern is (2 × 4), indicating that As is a rich surface.

After the buffer 28 has grown, growth is terminated by simultaneously closing the shutters of Ga and Si with the As valve and the shutter open. The surface is preferably annealed to the growth temperature for 2 to 10 minutes or longer while As flows. At the end of the annealing period, the substrate temperature drops from 580 ° C. to about 300 ° C. in preparation for moving from the first MBE chamber to the second MBE chamber. When the substrate temperature reaches about 550 ° C., the As valve closes, resulting in a surface reconstruction to instantaneously change from As stabilization (2 × 4) to Ga stabilization (3 × 1). The substrate temperature is still dropping at this time and the temperature drop stops momentarily when the temperature is in the range of 500 ° C. to 520 ° C., but the surface reconstructs its As stabilization (2 × 4) from the As source in the absence of As flux. To recover. When the (2 × 4) surface is fully recovered (after a few minutes), it is better to lower the substrate temperature. The RHEED pattern is the pattern of the As stabilized (2 × 4) surface for the remaining period of dropping substrate temperature.

When the sample is cooled to about 300 ° C., the transfer is ready to the second MBE chamber to grow the II-VI layer. Transfer at high temperatures is recommended to reduce the possibility of condensation on the surface. However, cooling the sample prior to transfer is a limitation of the system used.

When transferring the sample into the second MBE chamber, the sample is placed in a substrate heater and immediately heated to a temperature suitable for growth of the first II-VI layer 24 (ZnSe: Cl). During this time, the specimen and heater are away from the hot source to reduce line-of sight impurities. When the sample temperature is stabilized at the growth temperature (200 ° C.) of the first layer, the sample is rotated to face the source. The RHEED pattern is inspected to ensure that the surface is still As stabilized (2 × 4). When Zn is exposed on the surface, growth of the first layer 24 is initiated. Within 10-20 seconds, the surface reconstruction changes from (2 × 4) to (1 × 4), indicating a complete Zn treatment. The time required for this complete strain was recorded and the Zn treatment continued for additional time, such as twice the strain time. The Zn shutter is then closed and the substrate is heated to 250 ° C. Once the temperature has stabilized, ZnSe: Cl growth is initiated by migration enhanced epitaxy (MEE). After 25 cycles of MEE ZnSe: Cl, the specimen is further heated to about 300 ° C. to complete growth of the ZnSe: Cl buffer layer 24 by conventional MBE. The total thickness of the ZnSe: Cl layer 24 is about 500 GPa.

After ZnSe: Cl buffer 24 growth, growth of lower cladding 20 is started without interruption by opening the Mg and Be shutters. Layer 20 grows at a growth rate of about 1 μm per hour to a thickness of 1 μm. The composition of the Be _x Mg _y Zn _1-xy Se layer 20 is approximately x = 0.075 and y = 0.12, which generates a room temperature bandgap energy of about 2.85 eV, which nominally lattice match to the GaAs substrate 12 (amorphous). )do. Layer 20 is doped to an n-type concentration of approximately 2 × 10 ¹⁸ / cm ³ . In another embodiment, the bandgap of the BeMgZnSe cladding layer 20 is 2.75 eV or greater. In order to provide good optical and carrier confinement, the bandgap of the cladding layers 20, 22 should be at least 50 meV greater than the bandgap of the guide layers 14, 16, and the guide layer 14, A minimum of 100 meV greater than the bandgap of 16) is preferred.

After the lower cladding layer 20 has grown, growth is stopped (closed by closing the Be, Mg, Zn and Se shutters and closing the Se valve). Be cell temperature is lowered for growth of the Be _x Zn _1-x Se waveguide layer 14. The composition of layer 14 is x = 0.03 lattice matched to GaAs substrate 12. When the temperature of the Be cell stabilizes, growth of the bottom guide layer 14 is initiated by first opening the Se valve and then opening the Be, Zn and Se shutters. The growth rate of layer 14 is about 0.9 μm per hour and the layer grows to about 0.15 μm in thickness. After layer 14 has grown, CdZnSe quantum wells grow without interruption by simultaneously closing the shutters of Be and opening the Cd shutters. The Cd composition of layer 18 is about 25% and the thickness is about 5 nm (50 Hz), which together produces an emission wavelength of 515 nm. In yet another embodiment, the emission wavelength can be selected in the range of 550-465 nm by adjusting the Cd composition and the width of the quantum well. In order to provide a better carrier septum during room temperature operation, the active layer (typically quantum well) optical transition energy should be at least 200 meV less than the guide layer bandgap and at least 300 meV less than the guide layer bandgap.

After the quantum well 18 has grown, growth of the top waveguide layer 16 is initiated without interruption by simultaneously closing the Cd shutter and opening the Be shutter. The top waveguide layer 16 grows to a thickness of 0.15 mu m. Both top and bottom waveguide layers 14 and 16 are undoped.

Prior to growth of the upper cladding layer 22 (which has the same composition and thickness as the lower cladding layer 20), growth is stopped to raise the Be cell temperature and start the N plasma. As soon as the temperature stabilizes, growth of the upper cladding layer 22 is initiated by opening the Se valve and simultaneously opening the Mg, Be, Zn, Se and N shutters. Doping (N _A -N _D ) in layer 22 is approximately 2 × 10 ¹⁷ cm ^-3 .

In one embodiment, an etch stop layer is included to improve the continuous organization of the index guided laser by using selective etching. This method is described in a co-pending application filed on the same day as the present application and entitled "Selective Etching for II-VI Semiconductor." Typically, a 50 mn (500 m 3) first BeMgZnSe cladding layer is grown, followed by an etch stop layer (usually 20 nm (200 m 3) thick ZnSe or BeZnSe) on top. The remainder of the BeMgZnSe cladding layer is then grown.

After the top cladding layer 22 grows, the top p-type contact 34 grows. Contact growth begins at 100 nm (1000 μs) layer 26 of ZnSe: N, followed by a digitally graded layer 36 in which BeTe: N and ZnSe alternate. In this grading area, only the BeTe layer is doped with N, and the shutter of N is blocked during the growth of the ZnSe layer. The digitally graded layer 36 consists of a pair of 16 BeTe / ZnSe layers with a thickness of 2 nm to 2.3 nm (20 to 23 microns). The first BeTe layer of the first pair has a thickness of 0.1 nm to 0.2 nm (1 to 2 dB) [ZnSe of 2 nm to 2.3 nm (20 to 23 mm)], and the thickness of the BeTe layer is approximately linear over the remaining pairs. Increase by enemy. Likewise, the thickness of the ZnSe layer is reduced approximately linearly, resulting in a thickness of 0.1 nm to 0.2 nm (1 to 2 mm 3). After the digitally graded layer 36 grows, a 50 nm (500 kV) layer 38 of BeTe: N grows. There is no growth interruption between the p-type contact layers.

As shown in Fig. 4, the structure having the p-type contact below may be grown. In this case, a p-type GaAs substrate is used. Initially, a GaAs buffer layer doped with Be is grown in the above-described growth sequence.

When the p-type buffer is transferred to the second chamber, the sample is placed on the substrate heater and immediately heated to a temperature suitable for the growth of the first II-VI layer. The first layer grown is undoped ZnSe grown by MEE in the same manner as described above, which is initiated by Zn treating the GaAs surface at 200 ° C. In this case, however, the layer is only 4 or 5 single layers thick. After this layer grows, a 50 nm (500 mW) layer of BeTe: N grows, followed immediately by a digitally graded p-type contact layer of BeTe: N / ZnSe. The digitally graded layer is followed by 100 nm (1000 μs) of ZnSe: N. This p-type contact structure is the same as the n-type down structure, but the first layer grown on the GaAs buffer is a 50 nm (500 kV) BeTe: N layer with or without a first Zn treatment of the GaAs surface. to be.

After the p-type contact structure grows, the cladding layer, waveguide layer, and quantum wells are grown at the same temperature, composition, and timing (but the order of the layers are the same) as the structure below the n-type. After the top n-type cladding layer is grown, 100 nm (1000 :) of ZnSe: Cl is grown to help electrical contact. In addition, the Cl doping in this layer is approximately 2 × 10 ¹⁸ / cm ³ to 10 ¹⁹ / cm ³ .

Design guidelines

In order to effectively provide very short wavelength operation (500 nm or less), it is necessary to increase the bandgap energy of the guide layers 14 and 16 and the cladding layers 20 and 22 of the device 10. This is achieved by using BeMgZnSe of two different compositions for the guide layers 14, 16 and the cladding layers 20, 22. If the bandgap of the cladding layers 20, 22 is greater than 100 meV or the bandgap of the guide layers 14, 16, good optical isolation can be achieved. The active layer 18 may be, for example, CdZnSe, ZnSe, BeCdSe, BeMgZnSe or BeZnCdSe. The optical transition energy of the active layer 18 should be at least 200 meV below the bandgap energy of the guide layers 20, 22, and preferably at least 300 meV below the bandgap energy of the guide layers 20, 22.

The appropriate thickness of the cladding layer and the guide layer is mainly determined by the refractive index difference of the layer, which is related to the bandgap. If the bandgap of the guide layer is smaller by 100 meV than the bandgap of the cladding layer, the total thickness of the two guide layers should be between 0.1 μm and 1.0 μm, preferably 0.3 μm. In order to minimize the absorption of the laser light by the substrate and the upper electrode, the thickness of the cladding layers should be at least 0.5 [mu] m each, preferably at least 0.1 [mu] m.

To provide very short wavelength radiation, GaP substrates can be used. This allows the larger bandgap of BeMgZnSe to be lattice matched to the substrate.

Configuration details

In order to construct a laser diode from these epitaxial layers, resistive contact with the n-type GaAs substrate 12 is achieved by etching the lower surface of the GaAs substrate 12 for a short time to remove the micrometer from the back side of the substrate 12. Is done. Conventional GaAs etchant (eg 5H ₂ O-1 H ₂ O ₂ -1 NH ₄ OH) can be used for this purpose. 5 nm Pd, 25 nm Ge and 200 nm Au are sequentially deposited on the etched lower surface of the substrate. The wafer comprising Pd, Ge and Au layers is then annealed in nitrogen or forming gas at 180 ° C. for about 2 minutes. In yet another embodiment, the substrate surface is overlapped to a thickness of about 150 μm and smoothly finished before etching.

The gain guided laser diode element is constructed from an epitaxially grown structure by first providing a metal contact to the p-type semiconductor contact 34. 5 nm of palladium, 50 nm of gold and 1 nm of Ti (Ti improves the photoresist adhesion) are sequentially vacuum deposited onto the surface of the II-VI contact layer.

Subsequently, general microelectronic lithography is performed to draw a line 20 mu m wide on the surface of the deposited metal. The contact 34 and the upper metal layer outside the 20 μm line region are then removed by Xe + ion milling. Subsequently, 250 nm (2500 ns) of Zns is thermally deposited on the substrate. This coated polycrystal Zns is a separator. The Zns applied on top of the 20 μm photoresist line is then removed with the photoresist by acetone or any suitable photoresist solvent. Finally, a metal electrode (typically 100 nm Ti, 200 nm Au) is deposited on the surface. These electrodes are usually formed with strips of 200 mu m width with a center of 20 mu m line. The laser resonator is formed using a cleaved-facet mirror.

In another embodiment, US Patent No. 5,404,027, "BURIED RIDGE II-VI LASER DIODE," and the technology disclosed in the simultaneous continuing application filed on the same day as the present application and entitled "SELECTIVE ETCH FOR II-VI SEMICONDUCTORS" Using this, a buried-ridge index-guided laser is constructed. This fabrication process is initiated by first vacuum depositing 5 nm Pd, 50 nm Au and 1 nm Ti on a II-VI contact layer. Using general techniques, a 4 μm wide photoresist stripe is determined. Xe + ion etching is used to etch the metal and partially etch the top BeMgZnSe cladding layer. The sample is then etched in a high concentration of HCL solution. Since the HCL solution etches the BeMgZnSe cladding layer faster than the BeZnSe guiding layer, the etching is selectively stopped on top of the guiding layer. Optionally, it is desirable to grow an additional etch stop layer in the upper BeMgZnSe cladding layer, usually 50 nm above the guide layer. In this case, after the selective HCL etching, a 50 nm BeMgZnSe layer and an etch-stop layer remain. This embodiment provides fine tuning of the optical mode and also minimizes surface readjustment during operation of the device. The etch stop layer may be any II-VI material that etches significantly later than the BeMgZnSe cladding layer, such as ZnSe, BeZnSe or BeMgZnSe, which contains less Mg than the cladding layer. Alternative HCl, HBr solutions may be used to provide later selective etching.

After the selective etching, polycrystal Zns is vacuum deposited onto the sample. The thickness of Zns is sufficient to fill the area where the cladding has been etched away. The Zns applied on top of the 4 μm photoresist line is then removed along the photoresist using acetone or any suitable photoresist solution and ultrasound. Subsequently, a final metal electrode (typically 100 nm Ti, 200 nm Au) is deposited on the surface. These electrodes are usually formed in a stripe having a width of 200 mu m around a buried ridge of 4 mu m. The laser resonator is formed using a cleaved-facet mirror.

Heat is removed from the device by soldering the fabricated laser diode to a heat sink using an indium solder. The laser may be mounted to an upper (epitaxial side) electrode or substrate electrode soldered to the substrate electrode or heat sink.

While the invention has been described with reference to preferred embodiments, those skilled in the art will be able to make changes in form and detail without departing from the scope of the invention. It can be seen that the II-VI semiconductor diode containing Be of the present invention is used in electronic devices, electronic systems, optical data storage systems, communication systems, electronic display systems, laser pointers, and the like.

[Government Rights]

In accordance with Contract No. DARPA / ARO DAAH 04-94-C0049, approved by the United States Department of Defense Research Projects and the US Army Research Institute, the United States government has certain rights in the present invention.

Claims (3)

The first cladding layer 20 of the II-VI semiconductor containing Be of the first conductivity type, A first guide layer 14 of group II-VI semiconductor overlying the first cladding layer, An active layer 18 of the II-VI semiconductor overlying the first guide layer, A second guide layer 16 of the II-VI semiconductor overlying the active layer, A second cladding layer 22 of a II-VI semiconductor containing Be of a second conductivity type overlying the second guide layer, At least one of the guide layers contains Be, The active layer (18) is a group II-VI compound semiconductor laser diode having an optical transition energy of at least about 200 meV less than the bandgap energy of the guide layer (14, 16). The method of claim 1, wherein the first cladding layer contains Be _x Mg _y Zn _1-xy Se (0 <x ≦ 1, 0 ≦ y ≦ 1), and the second cladding layer is Be _x Mg _y Zn _{1 A} II-VI compound semiconductor laser diode containing _-xy Se (0 <x ≦ 1, 0 ≦ y ≦ 1). 3. The method of claim 1, wherein the first guiding layer contains Be _x Zn _1-x Se [0 <xSe (0 <x ≦ 1)] and the second guide layer is Be _x Zn _{1. A} II-VI compound semiconductor laser diode containing _-x Se (0 <x≤1).