JP2004274042A - Nitride semiconductor device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a nitride semiconductor device having a highly reliable electrode structure with low resistance. <P>SOLUTION: This nitride semiconductor device has a semiconductor layer, a first electrode which is formed on the semiconductor layer and makes ohmic contact, and a second electrode which is formed on the first electrode and has a different shape from that of the first electrode. The first and second electrodes has a junction layer region which is made of a platinum group element and is composed of the upper layer of the first electrode, which is the surface of the first electrode, and the lower layer of the second electrode, which is deposited on the heat-treated first electrode. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

  The present invention relates to a semiconductor element in which an electrode having a small area (width) is provided on a semiconductor layer using a nitride semiconductor, and particularly to a large current driving element (an electronic element such as a laser diode, a high power LED, an FET, and a high frequency). Element). Specific examples of the composition of the semiconductor element include Group III-V nitride semiconductors including GaN, AlN, or InN, or an AlGaN-based, InGaN-based, or AlInGaN-based mixed crystal thereof.

  A nitride semiconductor device emits light in a wide wavelength range from a relatively short wavelength ultraviolet region to a visible light region including red, and is a material constituting a semiconductor laser diode (LD) or a light emitting diode (LED). Widely used as. In recent years, miniaturization, long life, high reliability, and high output have been promoted, and they are mainly used for electronic devices such as personal computers and DVDs, medical devices, processing devices, and light sources for optical fiber communication.

  Such a nitride semiconductor device mainly includes a buffer layer, an n-type contact layer, a crack prevention layer, an n-type cladding layer, an n-type light guide layer, an active layer, a p-type electron confinement layer, and a p-type light guide on a sapphire substrate. It has a laminated structure in which a layer, a p-type cladding layer, a p-type contact layer and the like are sequentially laminated. In an LED, the light guide layer and the like can be omitted. An electrode is provided on such a laminated structure, and the active layer emits light when electricity is supplied.

  As the electrodes provided in the nitride semiconductor element as described above, an electrode (ohmic electrode) for making ohmic contact with the semiconductor layer is important, and a metal having a large work function is mainly used. A film or an alloy is used. Depending on the material, ohmic contact with the semiconductor layer is possible only by forming a metal film on the semiconductor layer. As such a material, a p-electrode made of a multilayer film of Pd / Pt / Au is exemplified. In addition, an electrode material that is hardly in ohmic contact with the semiconductor layer only by film formation can be brought into ohmic contact by performing a heat treatment step. For example, in the case of a p-electrode formed of a Ni / Au multilayer film, a transparent ohmic electrode can be formed by alloying after film formation (lamination).

  An extraction electrode (pad electrode) for bonding a wire is provided separately from the ohmic electrode in contact with the semiconductor layer. When an insulating substrate is used, since the p-electrode and the n-electrode are provided on the same surface side, pad electrodes are provided on both of them. In the case of an n-electrode, since it is relatively easy to make ohmic contact, it is also possible to provide a wire as an electrode for taking out the ohmic electrode. Further, by forming a metallized layer for connection with an external electrode or the like instead of a wire on the extraction electrode, the metallized layer can be used face down.

Further, an insulating film for preventing a short circuit is provided between the n-electrode and the p-electrode. The insulating film is formed of a single layer or a multilayer film using an oxide or the like. In the LD, the insulating film is also used as a current confinement layer for controlling a current injection region, or as a functional film having another function, such as being provided on a resonator surface and used as a reflection film.
JP 2000-299528 A

  However, the surfaces of the above-mentioned Ni / Au electrodes are easily roughened during heat treatment. Therefore, the resistance may increase at the interface with the pad electrode. In addition, when an insulating film is in contact with an electrode, the insulating film deteriorates during heat treatment, causing problems such as poor adhesion to the electrode. In the case of the Pd / Pt / Au electrode, the above-described problem hardly occurs because no heat treatment is performed. However, when the element temperature rises during element driving, the characteristics change due to the heat, which may cause an increase in operating voltage. There is. In addition, when the film thickness is reduced or when the film is formed in a large area, the adhesion and the mechanical strength are inferior.

  In view of the above problems, the present invention realizes an electrode having low contact resistance with a semiconductor layer and low interface resistance between a pad electrode and an ohmic electrode and excellent adhesion to a semiconductor layer and an insulating film. Accordingly, it is an object of the present invention to provide a nitride semiconductor device having a low threshold current and low operating voltage, and further having excellent device characteristics.

  The nitride semiconductor device according to the present invention is a nitride semiconductor device having a first electrode in ohmic contact on a semiconductor layer, and a second electrode in contact with the first electrode and having a different shape from the first electrode. The first electrode and the second electrode have a bonding layer region including an upper layer of the first electrode forming the first electrode surface and a lower layer of the second electrode deposited on the heat-treated first electrode. Consists of a platinum group element. With such a structure, a nitride semiconductor device having an electrode with excellent adhesion and a low operating voltage can be obtained.

  The nitride semiconductor device according to claim 2, wherein the nitride semiconductor includes a first electrode in ohmic contact with the semiconductor layer, and a second electrode in contact with the first electrode and having a shape different from the first electrode. In the device, the first electrode and the second electrode form a bonding layer region including an upper layer of the first electrode forming a surface of the first electrode and a lower layer of the second electrode deposited on the heat-treated first electrode. Wherein the upper layer of the first electrode and the lower layer of the second electrode are made of the same element or the same material. With such a structure, a nitride semiconductor device having an electrode with excellent adhesion and a low operating voltage can be obtained.

  The nitride semiconductor device according to claim 3 of the present invention is characterized in that the first electrode has a lower layer made of a heat-treated alloying material. The lower layer of the first electrode made of the heat-treated alloyed material is a layer that undergoes some thermal change due to the heat treatment, and has a different internal structure from that at the time of film formation (before heat treatment). For example, even if it has a multilayer structure at the time of film formation, it does not maintain the multilayer structure after the heat treatment, and becomes an alloyed layer in which alloying materials are mixed. As in the present invention, by forming the lower layer in contact with the semiconductor layer as a layer made of a material alloyed by heat treatment regardless of the material, compared to a laminated film simply deposited on the semiconductor layer, The first electrode having good adhesion to the semiconductor layer can be obtained. Ohmic contact may be possible without heat treatment depending on the composition of the semiconductor layer and the electrode material. However, in a semiconductor device having a low internal quantum efficiency and a low external quantum efficiency, some heat is generated at the time of driving the device, and the characteristics may be changed by the heat. Therefore, even when the lower layer of the first electrode is made of a single element, the heat treatment improves the adhesion to the semiconductor layer, and the characteristics can be hardly changed by heat during element driving. The upper layer of the first electrode having such a lower layer and the lower layer of the second electrode form a bonding layer region, and the bonding layer region is made of a platinum group element. Therefore, the bonding from the semiconductor layer to the second electrode can be performed with extremely excellent adhesion. This makes it possible to realize a highly reliable device which has a low operating voltage and hardly raises the voltage at the time of high output, and thus has little change with time.

  Further, when the lower layer of the first electrode is made of the heat-treatable alloying material, and the upper layer is made of a material other than the platinum group element, the lower layer of the second electrode and the upper layer of the first electrode are the same element or the same. As long as the material is excellent in conductivity and can maintain stable characteristics, a bonding layer region having excellent adhesion can be formed. However, it cannot be applied to a material in which an upper layer and a lower layer of the first electrode react with each other during the heat treatment and are alloyed to form an insulating oxide.

  Therefore, both the upper layer of the first electrode and the lower layer of the second electrode are made of a platinum group element, and the same element, an alloy of the platinum group element, or a conductive oxide of the platinum group element is used. By forming the upper layer of one electrode and the lower layer of the second electrode, it is possible to obtain an electrode structure having extremely low resistance, excellent adhesion, and excellent in reliability with little change over time.

  In the nitride semiconductor device according to claim 4 of the present invention, the upper layer of the first electrode is a platinum group single layer (Pt, Pd, Rh, Ir, Ru, Os layer) made of a single platinum group element, or And an alloying layer (Ru-Os layer, Rh-Ir layer, Pd-Pt layer) composed of a homologous element of a platinum group element. An upper layer made of such a material is relatively stable to heat. Therefore, even when the heat-treated alloyed material is formed in the lower layer, it is hardly alloyed with the lower layer, and although some reaction occurs near the interface with the lower layer, the reaction does not proceed to the inside of the layer and the interface does not proceed. Can be in a stable state. By using the same material for the upper layer of the first electrode and the lower layer of the second electrode, an electrode having lower resistance and stable operating characteristics can be obtained. In the case where the alloy layer is formed of an alloyed layer instead of a platinum group single layer, the material in the alloyed layer is alloyed by a heat treatment, and the bonding force is increased as compared with a simply deposited alloy layer, so that the alloy is hardened. Has become. The lower layer is not alloyed with the heat-treated alloyed material, and the laminated form of the upper layer and the lower layer is maintained even after the heat treatment.

  Further, since the surface of the upper layer of the first electrode is exposed to the outside air during the heat treatment, it also reacts with the outside air. However, if the layer is a platinum group element-based layer as described above, it is difficult to react with the outside air, and it is particularly difficult to form an oxide having high insulating properties. However, it is considered that oxygen is adsorbed on the surface of the first electrode upper layer in some state as long as oxygen is present. In particular, since the platinum group element is an element having a catalytic action, it is considered that oxygen exists in a state where oxygen is coordinated around the atom. The reaction here is assumed to be as follows.

  FIG. 6 shows that a first insulating film 609 is formed on a side surface of a ridge formed on a p-type semiconductor layer of a nitride semiconductor and on a plane of a p-type semiconductor layer on both sides of the ridge, and a nitride semiconductor layer on the ridge, The lower layer 605 (b) and the upper layer 605 (a) of the first electrode on the first insulating film, and the lower layer 606 (b) and the upper layer 606 (a) of the second electrode on the first insulating film via the bonding layer region 613. FIG. 4 is a schematic view showing a peripheral portion of a ridge of a nitride semiconductor laser device in which is formed. After the heat treatment step of the first electrode, as shown in FIG. 6A, it is considered that oxygen is adsorbed (coordinated) on the surface of the upper layer of the platinum group element. Then, at the time of forming the second electrode, the raw material of the lower layer of the second electrode accelerated by a sputtering method or the like collides with the surface of the first electrode. At that time, as shown in FIG. 6B, the adsorbed oxygen is repelled, and instead, the accelerated platinum group material enters the inside of the first electrode, and the bonding layer region 613 starts to be formed. Finally, as shown in FIGS. 6B and 6C, a state where a part of the upper layer 605 (a) of the first layer and the lower layer 606 (b) of the second electrode are shared. Thus, a bonding layer region 613 is formed.

  The reason that the above reaction is considered to be proceeding is as follows. Through the heat treatment alloying step of the first electrode, the platinum group element and the oxygen in the state where oxygen is adsorbed or coordinated in the first electrode as shown in FIG. 6A are temporarily stable. However, it is considered that they are connected by a weak connection that is easily broken by the application of some external force. Such a temporary weak bond can be easily caused, for example, by forming the second electrode by a sputtering method or the like, by an impact when the accelerated platinum group element material reaches the surface of the first electrode (the surface of the upper layer). Can be cut. In addition to the mechanical force such as sputtering, the bond with oxygen can be cut by, for example, heating.

  When the bond with the platinum group element is broken to remove the coordinated (adsorbed) oxygen, the platinum group element present on the first electrode surface is in an unstable state (active state). Conceivable. Therefore, the initial deposition region of the second electrode is formed in a state where the interface between the first electrode and the second electrode is broken or lost, and the bonding layer region 613 in which both are mixed is easily formed. In other words, there is no clear discontinuous surface (interface) between the heat-treated first electrode and the second electrode deposited thereon, and the state is as if it were formed continuously. When a stable oxide is formed on the surface of the first electrode, it is difficult to cut the bond of the oxide on the surface by sputtering, even if no oxide is formed inside, If oxygen is only coordinated, the bond can be easily broken, so that the platinum group element in the upper layer of the first electrode and the platinum group element in the lower layer of the second electrode are in a state where oxygen is not interposed, or The platinum group elements can be combined with each other in a state where the thin oxygen-containing region of the surface layer portion has disappeared or been separated by the impact at the time of film formation (deposition).

  Furthermore, when the upper layer of the first electrode is made of the above-described material, the inside has ductility and malleability, so that the accelerated raw material easily enters the inside of the upper layer of the first electrode. Such bonding is possible because a material made of a platinum group element and a homologous alloy of the platinum group element, which is a material having ductility, is unlikely to form an oxide on the surface even after heat treatment, This is because it is formed on the first electrode.

  When Au having the largest malleability among the metal elements is used for the first electrode, it is difficult to establish ohmic contact with the nitride semiconductor by using Au alone. Can be used for Au. However, when Au is used for the upper layer of the first electrode, it is extremely easily diffused during the electrode heat treatment, so that the alloying reaction with other elements used for the lower layer proceeds throughout the layer while taking in oxygen and the like from the outside. Cheap. Therefore, Au, which was the uppermost layer at the time of film formation, moved into the inside of the layer, and as a result, Au in the uppermost layer was reduced, or a metal other than Au used for the lower layer was exposed to generate an oxide. It will be easier. As described above, the first electrode containing Au in the surface layer (upper layer) has a large electrical barrier due to the interposition of the insulating oxide film and the like at the interface with the second electrode after the first electrode heat treatment, so that the second electrode has the second electrode. Adhesion with electrodes is reduced. In addition, the operation characteristics become unstable. When the upper layer of the first electrode is a layer made of a platinum group element as in the present invention, a low-resistance electrode can be obtained.

  If the upper layer of the first electrode is an alloy other than the same group as the platinum group element, for example, an alloy of Pt and a platinum group element other than Pt such as a Pt-Ir alloy, oxygen coordinated on the surface is sputtered. To bond with the platinum group element in the lower layer of the second electrode in a state where oxygen is hardly present at the interface. However, since this alloy is hard and has low ductility, it is difficult for the accelerated raw material to enter the inside of the alloy, so that the thickness of the bonding layer region is almost nil. Therefore, it is conceivable that the bonding becomes slightly weaker (the bonding layer region becomes smaller) as compared with the case where a material having high malleability is used. However, when such an alloy is formed as a lower layer of the second electrode without considering ductility by simultaneously sputtering different kinds of raw materials, there is no problem, and preferable adhesion to the upper layer of the first electrode is realized.

  In addition, other than a platinum group single layer or an alloyed layer made of the platinum group element, a compound of a platinum group element and oxygen, such as Rh oxide, Pd oxide, and Ru oxide, having conductivity. An oxide can be used as an upper layer of the first electrode. Although a compound of these platinum group elements and oxygen is an oxide, the first electrode having low resistance and excellent adhesion to the second electrode can be formed by using the compound in the upper layer of the first electrode. . A compound of a platinum group element and oxygen forms an oxide (an oxide having a weak bond) only on its surface, as in the above-described single platinum group element layer or an alloy layer of the platinum group element, or oxygen The bond strength with oxygen is higher than in the coordinated (adsorbed) state. Therefore, when an oxide of a platinum group element is used, by forming the bonding region with a single layer of the platinum group element or an alloy of the platinum group element, it is not the mechanism described above that can realize excellent adhesion with low resistance. It is speculated that another mechanism holds.

  When an oxide of a platinum group element is formed as an upper layer of the first electrode, and then a heat treatment alloying treatment is performed, a thermal reaction occurs inside the layer. It is considered that a stable interface is maintained with the lower layer of the first electrode as in the case of the single platinum group element layer. The difference is in the vicinity of the upper surface of the first electrode in contact with the outside air. In the case of a single platinum group element, oxygen is temporarily adsorbed to be in a pseudo-stable state. In the case of a compound of oxygen and oxygen, the ratio of oxygen to the platinum group element is not constant in a region near the surface and in a region separated from the surface (layer internal region), and the depth from the surface of the first electrode It is considered that the internal change occurs due to the composition gradient. Depending on the condition of the outside air during the heat treatment, the composition gradient near the surface becomes a platinum group element-rich layer or an oxygen-rich layer, and the composition ratio that was stable at the time of film formation was collapsed to be activated. It is considered that the electrode surface is bonded to the platinum group element in the lower layer of the second electrode so that a substantially clear interface is not formed, thereby realizing a low-resistance electrode. This is because the oxide is peculiar to the platinum group element having conductivity and is considered to be caused by the catalytic action and the like.

  In any case, since the upper layer of the first electrode is activated or quasi-activated after the heat treatment, when the lower layer of the second electrode is formed, it is strongly bonded to the active species on the surface thereof, It is possible to form a bonding layer region in which almost no interface level exists or in which the interface level has a gentle slope. Such a bond is formed mainly by using a platinum group element which itself hardly forms an oxide or a platinum group element which can maintain conductivity even when an oxide is formed by using a first electrode and a second electrode. By forming the above-described bonding layer region at each of the interfaces (contact surfaces), it is possible to realize an electrode structure with extremely low resistance and excellent reliability.

  The nitride semiconductor device according to claim 5 of the present invention is characterized in that the lower layer of the second electrode is made of a platinum group single layer made of a single platinum group element or a platinum group element alloy layer. . Here, the platinum group element is not an alloyed layer (a layer that has been heat-treated into an alloy), but a mixed layer (alloy layer) using two or more kinds of platinum group elements at the time of film formation. The lower layer of the second electrode deposited on the heat-treated first electrode forms a reaction product generated by the heat treatment or the second layer on the surface of the first electrode at an initial stage of deposition (early film formation). It is susceptible to impurities and the like remaining in the film forming apparatus for forming electrodes.

  If the first electrode and the second electrode, which determine the operation area, can be continuously formed without involving the movement between the devices, the formation of the discontinuous interface due to the movement between the devices can be suppressed. However, a non-heat treatment or a heat treatment of the first electrode is used to make an ohmic contact and to use as a lead electrode with respect to the first electrode which requires a heat treatment step to improve the reliability of the element. The second electrode, which is subjected to a heat treatment under milder conditions than the above conditions, often has a different shape due to a different purpose, and thus is preferably formed in a separate step. Therefore, by forming the lower layer of the second electrode formed at the initial stage of the film formation after the movement between the apparatuses as the above-described platinum group single layer or the alloy layer of the platinum group element, the first electrode surface is formed. It is difficult to react with the adhered impurities or reaction products, and also residual components such as oxygen remaining in the film forming apparatus for the second electrode, and it is possible to suppress the increase in resistance at the interface with the first electrode. it can.

  If a material other than the above, for example, Ti is used as the lower layer of the second electrode and Pt is used as the upper layer of the first electrode, the second electrode can be bonded to the first electrode with good adhesion. However, since Ti is easily bonded to oxygen, it reacts with oxygen coordinated (adsorbed) on the surface of the first electrode to immediately form an oxide in an early stage of film formation. Therefore, oxygen (oxide) is interposed between the first electrode and the second electrode. Since the oxide of Ti is insulative, it is not preferable to use it as a lower layer of the second electrode because the interface with the first electrode becomes high in resistance. As described above, if a material having good adhesion (such as Ti) is simply selected, it is possible to suppress the increase in resistance due to peeling of the second electrode, but it is possible to suppress the increase in resistance by forming an insulating intermediate layer. It is not possible to suppress up to conversion.

  As described above, by forming the upper layer of the first electrode and the lower layer of the second electrode as the layers made of the platinum group element-based material as described above, the interface region (bonding layer region) having a very low oxygen existence probability. Alternatively, a bonding layer region which does not include an oxide which increases resistance can be formed. By forming such a material on either the upper layer of the first electrode or the lower layer of the second electrode, the characteristics of the entire electrode (the first electrode and the second electrode) can be improved. By arranging the above-described material at the joint between the first electrode and the second electrode as in the present invention, an extremely excellent electrode can be obtained. In particular, when the upper layer of the first electrode and the lower layer of the second electrode are made of the same platinum group element layer, excellent adhesion can be obtained. Particularly preferred is Pt, which makes it possible to realize an element having a low operating voltage, little change over time even during high-output driving, and extremely excellent reliability.

  In the nitride semiconductor device according to claim 8 of the present invention, the surface of the semiconductor layer provided with the first electrode has an electrode formation region and an insulating film formation region, and the second electrode is insulated from the electrode formation region. The film forming region is covered. The first electrode is formed in contact with the semiconductor layer, and the contact portion forms a conduction path. In an LD, a current conduction path is controlled in order to efficiently inject a current into a waveguide region, and also in an LED, in order to effectively and effectively inject a current into a light emitting layer. In such a case, the conductive path is easily controlled by forming an insulating film on the surface of the semiconductor layer to form a non-conductive region and forming the first electrode, instead of controlling the formation region of the first electrode. it can. Then, the second electrode formed so as to be bonded to the first electrode is formed so as to cover both the insulating film forming region and the electrode forming region as described above, so that the first electrode is efficiently formed. It can be controlled so that a current flows.

  In the nitride semiconductor device according to the ninth aspect of the present invention, the insulating film formation region may be a plurality of regions sandwiching the stripe-shaped electrode formation region, or a plurality of regions separated by the electrode formation region. Features. In an LD as shown in FIG. 1, an insulating film is formed on both sides of a stripe-shaped electrode formation region, and a second electrode is formed so as to cover both regions. Thus, a current path to the semiconductor layer can be stably formed at a desired position. In the case where the LED also has a grid-like or linear first electrode, an insulating film is formed on an exposed surface of the semiconductor layer sandwiched between the first electrode forming regions, so that the insulating film forming region is formed. It is possible to adopt a form in which the electrode is divided into a plurality of regions by an electrode formation region. As described above, by separating the insulating film region, light extraction efficiency can be improved.

  A nitride semiconductor device according to a tenth aspect of the present invention is a laser device having a ridge formed of a convex portion, wherein the first electrode is formed in contact with an upper surface of the ridge. The LD ridge is an important part in which a waveguide region (operating region) is formed immediately below. Then, a large current flows through the ridge having a small width when the element is driven. Therefore, by forming an electrode having the structure of the present invention as an electrode formed in this region, an extremely reliable LD element can be obtained.

  The nitride semiconductor device according to claim 11 of the present invention has a first insulating film on both sides of a ridge and a plane of a semiconductor layer continuous from the side surfaces thereof, and a semiconductor layer formed on the first insulating film. It has a second insulating film continuous over the side surface, and the first electrode is formed so as to be separated from the second insulating film.

  In FIG. 7A, a ridge is formed in a semiconductor layer, a first insulating film 709 is provided on a side surface of the ridge and on both sides of the ridge, and a first insulating film 709 is formed so as to cover the upper portion of the ridge and the upper portion of the first insulating film. It is a schematic diagram which shows that an electrode is formed. When the lower layer of the first electrode is formed with a laminated structure of a metal film alloyed by a heat treatment such as Ni / Au, the layer structure is changed by the heat treatment. At this time, the reaction proceeds not only inside the lower layer but also at the interface between the lower layer and the semiconductor layer and the interface between the lower layer and the upper layer (platinum group element layer) to form an active interface region. Here, since the upper layer is a layer made of a platinum group element, oxygen can be moved out of the system near the interface between the upper layer and the lower layer by the catalytic action. This makes it possible to control the amount of oxygen (outside air) involved in the reaction inside the lower layer and the reaction between the lower layer and the semiconductor layer to an appropriate amount, thereby stabilizing the interface between the upper layer and the lower layer (indicated by a thick line). . As described above, the layer (upper layer) of the platinum group element formed on the surface functions as a cap layer for stably alloying the lower layer during heat treatment.

  In addition, since the constituent elements of the lower layer of the first electrode do not move to the surface of the first electrode through the upper layer of the platinum group element, the surface can be maintained in a stable state. Therefore, an insulating oxide based on the constituent element of the lower layer is not formed on the surface of the first electrode, and the bonding layer region 713 is formed between the first electrode and the second electrode, thereby improving the adhesion. It can be.

  A nitride semiconductor device according to a twelfth aspect of the present invention is characterized in that the first insulating film and / or the second insulating film is covered with a single-layer or multi-layer adhesive layer. The electrode is formed so as to be in contact with the semiconductor layer in order to inject a current into the semiconductor layer. However, an electrode formed on the semiconductor layer for the purpose of restricting the injection region or preventing a short circuit. It is formed so as to be in contact with the film. The electrode material does not always have good adhesion to the insulating film. For this reason, even if there is no problem in the adhesion between the first electrode and the second electrode, the insulating film has poor adhesion, so that the first electrode and the second electrode are easily peeled off, which may cause deterioration of device characteristics such as high resistance. In such a case, by providing a reinforcing layer (adhesion layer) for reinforcing the adhesion between the electrode and the insulating layer, peeling of the second electrode can be suppressed, and deterioration of element characteristics can be suppressed.

  The nitride semiconductor device according to claim 13 of the present invention is characterized in that the uppermost layer of the adhesion layer is a layer containing a platinum group element. By forming the upper layer made of a platinum group element as the upper surface, an electrode having excellent adhesion to the second electrode can be obtained. However, the adhesion to the insulating film may not always be good. In particular, the film has low adhesion to an oxide-based insulating film and is easily peeled. On the other hand, there is a metal material which is not suitable as an electrode material but has high adhesion to an insulating film. FIG. 7C is a schematic diagram showing that an adhesion layer having a multilayer structure is formed. In the present invention, the first insulating film 709 and the second insulating film 710 are covered with an adhesive layer 711 in which a material (metal material) having excellent adhesion to an insulating film is a lower layer and a platinum group element is an upper layer. In addition, the surface on which the second electrode is formed can be a surface with a small area occupied by the insulating film, and the adhesion of the second electrode can be reinforced. In particular, as shown in FIG. 7C, by covering both the first and second insulating films with an adhesive layer, the formation region W2 of the second electrode can be separated from the region W1 in contact with the first electrode 705 and the adhesive layer. All of the region W3 in contact with the region 711 and the region W3 can be a region made of a platinum group element. That is, since the entire formation interface of the second electrode is formed by metal bonding, and the bonding layer region can be formed over a wide region, extremely excellent adhesion can be realized.

  The nitride semiconductor device according to claim 14 of the present invention is characterized in that the uppermost layer of the adhesion layer is made of the same element or the same material as the upper layer of the first electrode. With such a configuration, it is possible to make it difficult to form a high-resistance region at the contact interface between the first electrode upper layer and the second electrode lower layer formed in contact with the uppermost layer of the adhesion layer, In addition, an electrode structure having excellent adhesion can be obtained by using the same material.

  The nitride semiconductor device according to claim 15 of the present invention is characterized in that the upper layer of the adhesion layer is Pt. Thus, an electrode structure having excellent adhesion to the second electrode can be obtained.

  The nitride semiconductor device according to claim 16 of the present invention is characterized in that the adhesion layer is formed so as to be in contact with above or below the first electrode. Although the adhesion layer covers the first insulating film and / or the second insulating film, it is preferable that the insulating film made of an oxide or the like is not exposed when the second electrode is formed. Therefore, by forming an adhesion layer having an upper layer of a platinum group element, the second electrode can be more firmly adhered. In FIG. 3, the adhesion layer is formed on the first electrode. However, when the adhesion layer is formed before the first electrode, the adhesion layer is formed below the first electrode. Since this adhesion layer is formed as a layer not involved in current injection into the semiconductor layer, it can be formed, for example, near the ridge of the LD to function as a layer for controlling optical characteristics. In particular, by forming Ti with excellent adhesion to the insulating film as a lower layer of the adhesion layer and forming this layer near the ridge, it can function as a light absorption area and control light confinement. It is.

  With the above-described configuration, the semiconductor element of the present invention has a first electrode formed of a heat treatment layer capable of forming an ohmic contact with the semiconductor layer by performing a heat treatment, and a lead electrode formed on the first electrode. By forming a bonding layer region between the second electrode and the second electrode, it has excellent adhesion, low operating voltage, little change over time even during high-output driving, and extremely stable operating characteristics. Thus, a nitride semiconductor device having excellent reliability can be obtained. In addition, by providing an adhesion layer between the electrode and the insulating film, the electrode material and the insulating film material can be selected without considering their mutual adhesion, so that a better operating voltage can be realized. In addition, a nitride semiconductor device having excellent adhesion can be obtained.

  Hereinafter, the present invention will be described, but the nitride semiconductor device of the present invention is not limited to the device structure shown in the embodiment.

  In the nitride semiconductor device of the present invention, the first electrode mainly provided on the semiconductor layer and brought into ohmic contact with the semiconductor layer and the second electrode mainly used as an extraction electrode have a specific interface. Thus, the electrode structure has excellent adhesion, low resistance at the interface, and extremely stable operation characteristics.

  Since the first electrode and the second electrode have different functions, the size (width / length) and shape of each can be set to a preferable shape depending on the purpose or in consideration of the process and the like. The first electrode and the second electrode need only be formed so as to be in contact with each other in the operation section, and need not be connected on the entire surface. In the case of an LED, the first electrode and the second electrode are formed in consideration of the film thickness, the shape, and the like so that a current can flow uniformly in a wide area of the light emitting layer. Further, it is preferable that the junction between the first electrode and the second electrode allows effective current injection into the light emitting layer in consideration of the arrangement of the p-side electrode and the n-side electrode. In the case of an LD, the first electrode and the second electrode are joined at the upper part of the ridge, whereby an increase in operating voltage due to interface resistance can be suppressed. The second electrode provided in a step subsequent to the first electrode may be formed so that the entire bottom surface thereof is in contact with the first electrode, or a part of the second electrode is formed on the first electrode and other The portion may be provided so as to be in contact with the semiconductor layer or the insulating film.

  In the LD, in the stripe direction of the ridge formed in the semiconductor layer, the first electrode is preferably formed in a stripe shape so as to be parallel to the stripe-shaped waveguide region, but is not limited thereto. That is, the first electrode need not be formed in a stripe shape, but may be formed so that the contact region with the semiconductor layer becomes a stripe. Further, it is preferable to provide so as to cover the entire region of the waveguide region in the direction parallel to the stripe, but it is separated from the end portion in consideration of a photolithography process at the time of electrode formation, a chip formation process in a later process, and the like. For example, a preferable size and shape can be selected.

  The connection region between the first electrode and the second electrode is preferably connected to a region corresponding to the entire region of the waveguide region, so that the operating voltage can be stabilized. It is preferable to make the length short so that the second electrode is not formed above the divided region. This is because, when the metal formed on the second electrode, particularly on the uppermost layer, is Au, the division becomes very difficult due to its ductility. Further, by using the upper portion of the ridge as the connection portion, the current injected into the second electrode can efficiently flow through the semiconductor layer via the first current, so that light confinement in the waveguide region can be stabilized. As a result, the beam shape of the laser beam can be maintained satisfactorily, the threshold value can be stabilized, and the operating voltage can be stabilized.

  In the direction perpendicular to the ridge, the first electrode needs to be formed above the ridge. Further, the first electrode does not need to be formed even in a region far away from the ridge, and is formed to have a width equal to or larger than the width of the ridge. Further, it is preferable that the ridges are formed so as to have the same length on the left and right sides. Further, it is preferable that the stripes (resonators) be formed with the same width over the entire region.

  Further, it is preferable that the junction layer region of the first electrode and the second electrode in the direction perpendicular to the ridge is formed to have the same width as the ridge width or a width larger than that, thereby stabilizing the operation voltage. If the connection is made in a range narrower than the ridge width, the current injection region becomes narrower and the operating voltage may increase, which is not preferable. In particular, it is preferable to form a bonding surface on the ridge. The second electrode has such a width that a wire can be bonded to a region other than the upper portion of the ridge for wire bonding. Since this region may be only a part of the entire region of the stripe, the width of the second electrode may not be the same over the entire region of the stripe.

  As shown in FIG. 7A, it is preferable that the formation region (width) W2 of the second electrode be larger than the formation region W1 of the first electrode. Therefore, an element having excellent reliability at the time of high output can be obtained. Further, by making the formation region (width) W2 of the second electrode smaller than the formation region W1 of the first electrode, an exposed surface of the insulating film can be formed around the electrode. With the exposed insulating film region, a short circuit due to a shape change due to heat of the second electrode at the time of face down can be reduced, and an element can be obtained with a high yield.

  The structure of the electrode of the present invention may be provided on both the p-side electrode and the n-side electrode of the nitride semiconductor element, or on either one of them. In an LD, however, it is particularly preferable to provide the structure on the p-side electrode. Among them, it is extremely effective for an LD having a ridge. When used for both electrodes, the platinum group element at the junction between the first electrode and the second electrode of the p-side electrode and the n-side electrode may be the same or different. Preferably, Pt is used on both the p-side and n-side bonding surfaces, thereby achieving extremely excellent adhesion. Further, the second electrode of the p-side electrode and the n-side electrode is formed of the same material. Thus, the second electrodes of both electrodes can be formed at the same time, which is advantageous also in the process.

  The first electrode can achieve good ohmic properties by heat treatment. The heat treatment temperature is preferably in the range of 350 ° C to 1200 ° C, more preferably 400 ° C to 750 ° C, and particularly preferably 450 ° C to 600 ° C.

  The upper layer of the first electrode may have an intermediate layer sandwiched between layers made of a platinum group element. When the upper layer of the first electrode is a single layer of a single element of the platinum group element or an alloy layer of the same group element of the platinum group element, the upper and lower surfaces of the layer have different functions. I have. The upper surface does not easily react with the outside air and improves the adhesion to the second electrode. Further, the lower layer forms a stable interface with the alloyed layer formed thereunder, and stabilizes the alloying reaction. The two functions can be separated. For example, a first electrode including a lower layer made of Ni / Au and an upper layer made of Pt / Ti / Pt is formed on a semiconductor layer. The upper layer may have a single element layer of a platinum group element on the upper surface and the lower surface, and an intermediate layer made of Ti (a layer made of an element other than the platinum group element) may be sandwiched between the layers. Thus, the lower layer of the platinum group element that is in contact with the lower layer of the first electrode can reduce the thermal deterioration of the first layer and the variation in the heat treatment process, and can provide a highly reliable device. By the upper platinum group element layer provided on the intermediate layer, an interface with a very low probability of oxygen inclusion is formed between the second electrode and the lower electrode, so that a low-resistance electrode structure can be obtained. As described above, the function can be separated by separating the layer of the platinum group element into the upper layer and the lower layer.

  In addition to the function of separating the functions by arranging an optimal material at the interface between the upper surface and the lower surface of the upper layer made of a platinum group element, the intermediate layer on the first electrode has only a platinum group element-based layer. Can function as a layer that compensates for insufficient properties. For example, the heat dissipation can be improved by increasing the thickness of the first electrode by interposing the intermediate layer. Further, as a multilayer film having three or more layers, the stress can be reduced as compared with the case where the film thickness is increased by a single composition layer. In particular, the first electrode formed on the ridge of the LD is formed in a very narrow region, and the load on the ridge is greatly affected by its film quality. As a result, LD characteristics with excellent reliability can be obtained. Further, since the light absorption coefficient can be changed, the optical characteristics can be controlled.

  As described above, since the upper layer of the first electrode and the lower layer of the second electrode are both platinum group elements, a bonding layer region having excellent adhesion can be formed. In such a case, the following materials can be used as the electrode material of the lower layer of the first electrode or the upper layer of the second electrode. The following materials can be used as the upper layer of the first electrode and the lower layer of the second electrode when the upper layer of the first electrode and the lower layer of the second electrode are made of the same element or the same material.

  As the first electrode (lower layer of the first electrode) provided on the n-type nitride semiconductor layer, a material having high ohmic properties and adhesion to the n-type nitride semiconductor layer can be selected. , Co, Fe, Ti, Cu, Au, W, Zr, Mo, Ta, Al, Ag, Pt, Pd, Rh, Ir, Ru, Os and the like. Can be used. Preferably, it has a multilayer structure in which Ti and Al are sequentially laminated. After the formation of the first electrode, it may be preferable to perform heat treatment depending on the material in order to improve ohmic contact with the semiconductor layer. The total thickness of the n-side first electrode is preferably about 100 to 30000, more preferably about 3000 to 15000, and particularly preferably 5000 to 10000. Forming in this range is preferable because an electrode having low contact resistance can be obtained.

  The electrode material of the n-side second electrode (upper layer of the second electrode) formed in contact with the n-side first electrode includes Ni, Co, Fe, Ti, Cu, Au, W, Zr, and Mo. , Ta, Al, Ag, Pt, Pd, Rh, Ir, Ru, Os, etc., and a single layer, alloy, or multilayer film thereof can be used. It is preferable to use Au since a multilayer film is preferably used and a wire or the like is connected to the uppermost layer. It is preferable to use a material having a relatively high melting point that functions as a diffusion preventing layer as a lower layer so that the Au does not diffuse. For example, Ti, Pt, W, Mo, TiN and the like can be mentioned. The total thickness is preferably from 3000 to 20000, more preferably from 7000 to 13000.

  The n-side electrode does not provide the first electrode and the second electrode in separate steps as described above, but is formed continuously and has both functions of the first electrode and the second electrode. An n-electrode that is an ohmic electrode in ohmic contact with the layer and also serves as an extraction electrode (pad electrode) for forming a wire can be used. This is a region where the ohmic contact with the n-type semiconductor layer is relatively easy as compared with the p-side electrode, and since it is a region slightly separated from the waveguide region, it is not necessary to consider optical characteristics so much. This is because the degree of freedom is large. The thickness of such an n-electrode is preferably in the range of 3,000 to 20,000, and more preferably in the range of 7000 to 13,000. When the n-side electrode is not separated into the first electrode and the second electrode, the configuration of the present invention is applied to the p-side electrode.

  Next, as the electrode material of the p-side first electrode (lower layer of the first electrode) provided in the p-type nitride semiconductor layer, a material having high ohmic properties and adhesion to the p-type nitride semiconductor layer is selected. Specifically, Ni, Co, Fe, Cr, Al, Cu, Au, W, Mo, Ta, Ag, Pt, Pd, Rh, Ir, Ru, Os, and oxides and nitrides thereof And a single layer, an alloy, or a multilayer film thereof can be used. Preferably, at least one selected from Ni, Co, Fe, Cu, Au, and Al, and oxides and nitrides thereof are used.

  The electrode material of the p-side second electrode (upper layer of the second electrode) includes Ni, Co, Fe, Ti, Cu, Au, W, Zr, Mo, Ta, Ag, Pt, Pd, Rh, and Ir. , Ru, Os and their oxides and nitrides, and a single layer, alloy, or multilayer film thereof can be used. The uppermost layer is preferably made of Au because it connects wires and the like. It is preferable to use a material having a relatively high melting point that functions as a diffusion preventing layer as a lower layer so that the Au does not diffuse. For example, Ti, Pt, W, Ta, Mo, TiN and the like can be mentioned, and particularly preferable material is Ti. The total thickness is preferably from 3000 to 20000, more preferably from 7000 to 13000.

  The upper layer of the first electrode may have an intermediate layer sandwiched between layers made of a platinum group element. The intermediate layer may be a layer made of a single element, or may be a multilayer film or an alloy. Further, since the upper layer and the lower layer sandwiching the intermediate layer only need to be formed of a platinum group element layer, even if they are the same platinum group element, they are different platinum group elements. No problem. In consideration of the adhesion to the lower layer of the first electrode, the adhesion to the intermediate layer, or the adhesion to the extraction electrode provided on the upper layer of the first electrode after the alloying treatment, respectively. Material can be selected. Alternatively, when the mesa portion (ridge) is formed by a self-alignment method in which etching is performed using the first electrode as a mask, it is preferable to select a material in consideration of the type of an etching gas and the like of the platinum group element in the uppermost layer. .

  Since the upper and lower intermediate layers of the first electrode are sandwiched between stable white metal element layers, the material of the alloying layer used as the lower layer of the first electrode can also be used. The same material as the lower layer of the first electrode may be used, or a different material may be used. Alternatively, a material that cannot be used for the first layer can be used. The preferred materials for the intermediate layer include the preferred materials for the first layer (Ni, Co, Fe, Cu, Au, W, Mo, Ti, Ta, Ag, Al, Cr, Pt, Pd, Ph). , Ir, Ru, Os, and oxides and nitrides thereof, Hf and the like can be used.

Embodiment 1
FIG. 1 shows a configuration of a nitride semiconductor device according to a first embodiment of the present invention, in which an n-type nitride semiconductor layer 102, an active layer 104, and a p-type nitride semiconductor layer 103 are formed on a substrate 101. Are stacked and a p-type nitride semiconductor layer is provided with a stripe-shaped ridge to provide a semiconductor laser (LD). The ridge can be formed by removing a part of the p-type nitride semiconductor layer by means such as etching or the like, whereby an effective refractive index type waveguide can be formed. Further, the ridge may be formed by etching a part of the region from the p-type nitride semiconductor layer to the n-type nitride semiconductor layer to form a complete refractive index type waveguide, or the ridge may be formed by selective growth. May be. The ridge is not limited to a forward mesa shape in which the width of the bottom side is wide and the stripe width decreases as approaching the top surface, but may be an inverted mesa shape in which the stripe width decreases as approaching the ridge bottom, and may be perpendicular to the lamination surface. It may be a stripe having various side surfaces, or a shape in which these are combined. Further, the stripe-shaped waveguides do not need to have substantially the same width. Also, a buried laser device in which a semiconductor layer is regrown on the ridge surface or on both sides of the ridge after forming such a ridge may be used. Further, a gain-guided waveguide having no ridge may be used.

  Alternatively, a current confinement layer made of a high resistance layer (insulating layer) may be formed inside the element. The current confinement layer may be formed either in the n-type semiconductor layer or in the p-type semiconductor layer. Preferably, it is formed in a p-type semiconductor layer. Further, in the n-type semiconductor layer and the p-type semiconductor layer, a current confinement layer may be provided at a boundary between respective layers such as a contact layer, a cladding layer, a guide layer, a cap layer, and an active layer. , A cladding layer, a guide layer, etc., in each functional layer. In order to form the current confinement layer, the reaction is temporarily interrupted. At that time, an interface state is generated due to formation of an insulating oxide or the like, and the current injection efficiency is reduced. It is preferred to select a layer with a low composition. Further, at the start of the regrowth, the regrown surface may be slightly removed by etchback or the like to remove the layer (surface film) that causes the increase in resistance. As the current confinement layer, for example, AlN, AlGaN having a high Al mixed crystal ratio, or the like can be used. Preferably, AlN is used because it has a high insulating property and can be continuously grown in the same device following an earlier layer. Furthermore, when AlN is removed as a conductive region, In addition, since it can be easily and selectively removed with an acid or the like, other parts of the element are not easily damaged. Further, since the refractive index is low, it is suitable for confining light. The current constriction layer composed of these high resistance layers may have any thickness as long as it can block current. The conductive portion serving as the waveguide region may be formed by selectively growing portions other than the conductive portion, or by forming an opening in a continuously grown layer to form a conductive portion.

  A first insulating film 109 is formed from the side surface of the ridge and the upper surface of the continuous p-type nitride semiconductor layer to the ridge. A p-side first electrode 105 is provided on the ridge upper surface and the upper surface of the first insulating film, and an n-side first electrode 107 is provided on the upper surface of the n-type nitride semiconductor layer. A second insulating film 110 having an opening above the n-side first electrode is provided so as to be continuous up to the upper portion of the first insulating film. Above the p-type nitride semiconductor layer, a p-side second electrode 106 that is in contact with the second insulating film and the p-side first electrode is provided. The n-side second electrode 108 is also provided on the n-side first electrode.

  Further, in order to make the stripe direction of the ridge the resonator direction, the pair of resonator surfaces provided on the end faces can be formed by cleavage or etching. In the case of forming by cleavage, it is necessary that the substrate and the semiconductor layer have a cleavage property, and by using the cleavage property, an excellent mirror surface can be easily obtained. In addition, even if there is no cleavage, the resonator surface can be formed by etching. In this case, by performing this step simultaneously with exposing the n-electrode formation surface, it can be obtained with a small number of steps. Also, it can be formed simultaneously with the formation of the ridge. As described above, the number of steps can be reduced by forming them at the same time as each step. However, in order to obtain a better resonator surface, it is preferable to provide another step. Further, on the resonator surface formed by cleavage or etching as described above, a reflection film composed of a single film or a multilayer film can be formed in order to efficiently reflect the emission wavelength of the active layer. One of the resonator surfaces is composed of a surface having a relatively high reflectivity and serves as a light reflection-side resonator surface for mainly reflecting light into the waveguide region, and the other is composed of a surface having a relatively low reflectivity and mainly serves to emit light to the outside. It functions as a light emitting side resonator surface to emit light.

  In the first embodiment, the second insulating film 110 and the p-side first electrode 105 are provided so as to be separated from each other. The p-side first electrode formation region can be formed so as to cover the upper surface of the p-type nitride semiconductor layer over a wide range. At this time, there is no problem if the adhesion between the first insulating film and the p-side first electrode is good, but if the adhesion is poor, the area for forming the p-side first electrode should be increased. Thus, there is a case where a problem that the film is easily peeled off occurs. In the first embodiment, the formation region of the p-side first electrode is formed so as to be separated from the end of the p-type semiconductor layer, and is provided at least near both sides of the ridge. This makes it possible to reduce the contact area between the first insulating film and the first insulating film as compared with the case where the first electrode is formed on the entire surface, and to prevent the first insulating film from peeling off even when the adhesion to the first insulating film is weak. Since the second insulating film and the p-side first electrode are formed so as to be separated from each other, the structure is such that the p-side second electrode and the first insulating film are in contact with each other. Also, since a relatively deep concave portion is formed between the ridge and the second insulating film formed with a relatively large thickness in order to prevent a short circuit between the p-side electrode and the n-side electrode, The bonding surface of the p-side second electrode is a surface having a large difference in unevenness. Due to the unevenness, the bonding area is increased, and the second electrode can be hardly peeled off.

  Although the first insulating film is provided to limit the current injection region to the upper surface of the ridge, the first insulating film is provided close to the waveguide region and also acts on light confinement efficiency. Therefore, the film thickness cannot be unnecessarily increased. It is necessary to reduce the film thickness depending on the insulating film material used. When the first insulating film is formed to be thin, there may be a portion where the insulating property is slightly weakened. However, even in such a case, the p-side second electrode and the first By interposing the second insulating film to the region relatively close to the ridge, the current injection region can be controlled near the ridge.

  The first insulating film can be formed so as to have substantially the same width as the nitride semiconductor layer as shown in FIG. The first insulating film formed before the first electrode is heat-treated together with the heat treatment of the first electrode. By performing the heat treatment, the strength of the film (bonding strength at the atomic level in the film) is increased as compared to the simply deposited film, and the bonding strength at the interface with the semiconductor layer is also improved. By forming such a first insulating film up to the edge of the upper surface of the semiconductor layer where the second insulating film is formed, the adhesion of the second insulating film can be improved.

  Further, as shown in FIG. 1B, the second electrode can be formed so as not to be in contact with the second insulating film. In particular, when used face-down, heat is applied to the second electrode. At this time, the volume is increased due to thermal expansion, and the second electrode is likely to flow out in the side direction of the element (the end direction of the p-type semiconductor layer). In addition, since not only heat but also pressure is applied, the electrode material is also likely to flow out in the lateral direction. Therefore, as shown in FIG. 1B, by separating the second insulating film from the second insulating film, it is possible to prevent the p-side second electrode material from flowing out in the side direction and causing a short circuit.

  Not only in the first embodiment but also in the following embodiments, as a material of the first insulating film, at least one element selected from the group consisting of Si, Ti, V, Zr, Nb, Hf, and Ta is used. It is desirable to form at least one of oxides containing, SiN, BN, SiC, AlN and AlGaN, and among them, it is particularly preferable to use oxides of Zr, Hf and Si, BN, AlN and AlGaN.

  Further, the thickness of the first insulating film is specifically in the range of 10 ° to 10,000 °, preferably in the range of 100 ° to 5000 °. This is because if it is less than 10 °, it is difficult to secure sufficient insulation during the formation of the electrode, and if it is more than 10,000 °, the uniformity of the protective film is rather lost and a good insulating film is not obtained. . Further, by being in the above-mentioned preferred range, a uniform film having a good refractive index difference with the ridge is formed on the side surface of the ridge.

The second insulating film can be provided on the entire surface of the p-side first electrode except the upper part of the ridge, and is provided so as to be continuous with the side end surfaces of the p-type semiconductor layer and the active layer exposed by etching. Is preferred. As a preferable material, an oxide containing at least one element selected from the group consisting of Si, Ti, V, Zr, Nb, Hf, and Ta, and at least one of SiN, BN, SiC, AlN, and AlGaN are used. It is preferable to use a single layer film or a multilayer film such as SiO ₂ , Al ₂ O ₃ , ZrO ₂ , and TiO ₂ as a particularly preferable material.

Embodiment 2
FIG. 2 shows a configuration of the nitride semiconductor device according to the second embodiment of the present invention. In the second embodiment, similarly to the first embodiment, an n-type nitride semiconductor layer 202, an active layer 204, and a p-type nitride semiconductor layer 203 are stacked on a substrate 201, and a p-type nitride semiconductor layer is formed. In a semiconductor laser (LD) provided with a stripe-shaped ridge, a p-side first electrode 205 is formed in a region that widely covers the first insulating film 209 and is separated from portions other than the ridge of the first electrode. A second insulating film 210 is formed so as to cover a part of the upper region. The p-side second electrode 206 is formed over the p-side first electrode and the second insulating film. That is, in the second embodiment, the p-side second electrode is provided so as not to be in contact with the first insulating film. The confinement of light in the waveguide region changes depending on the depth (height) of the ridge and the optical characteristics of the first insulating film. However, if the thickness can be controlled by increasing the thickness of the first insulating film, By increasing the formation area of the first electrode and keeping it in close contact with the second electrode over a wide range, heat concentration is suppressed, and the bonding layer area between the first electrode and the second electrode is increased. Thus, an electrode having excellent adhesion can be obtained.

  In this case, as shown in FIG. 2A, the second electrode 206 may not be in contact with the second insulating film 210. Accordingly, the second electrode having poor adhesion to the insulating film comes into contact with only the first electrode, so that the second electrode can be hardly peeled off. In addition, as described in Embodiment 1, when used face down, it is possible to suppress outflow to the n-type semiconductor layer side when the second electrode is deformed by heating, and to obtain a highly reliable element.

Embodiment 3
FIG. 3 shows a nitride semiconductor device according to the third embodiment of the present invention. In the third embodiment, similarly to the first embodiment, an n-type nitride semiconductor layer 302, an active layer 304, and a p-type nitride semiconductor layer 303 are stacked on a substrate 301, and a stripe is formed on the p-type nitride semiconductor layer. A semiconductor laser (LD) provided with a ridge in the shape of a ridge, wherein an adhesion layer 311 is formed between the second electrode 306 on the p-side and the first insulating film 309 and / or the second insulating film 310. Have been. This adhesion layer is formed at a position where it is not formed on the ridge upper surface, which is a junction region between the first electrode 305 and the second electrode 306 on the p side, and in the vicinity thereof, and the second electrode and the first and / or second Has the function of reinforcing the adhesion to the insulating film.

  As described above, the electrode material and the insulating film material need not only have excellent adhesion with the material on which they are formed, but also have excellent adhesion with the material formed thereon. Need to be. Therefore, both materials can be satisfied by using different materials for the upper layer and the lower layer. However, as in Embodiment 3, by providing an additional adhesive layer, more excellent adhesiveness can be obtained. In FIG. 3, a part of the adhesion layer is interposed between the p-side first electrode and the second electrode, but the adhesion layer may not be in contact with the first electrode.

  The adhesion layer has a low contact resistance with the semiconductor layer like the first electrode, and it is not necessary to select a material having excellent adhesion with the semiconductor layer, and the interface resistance between the first electrode and the second electrode is reduced. There is no need to select a material to lower. Further, unlike the first and second insulating films, it is not necessary to select a material having excellent adhesion to the electrode and having high insulating properties. That is, the adhesion layer may be made of a conductive material or an insulating material, and the only function required is adhesion to the insulating film and the electrode. Therefore, as the electrode material, a material having a problem such as high resistance can be used as long as the material has excellent adhesion to the insulating film. Then, by selecting a material having excellent adhesion to the p-side second electrode as the upper layer as a multilayer structure, the adhesion can be reinforced.

  The adhesion layer can have a single-layer or multilayer structure. In the case of a multilayer structure, a material having excellent adhesion to the insulating film is selected as a lower layer of the adhesion layer, and a material having excellent adhesion to the lower layer of the second electrode is selected as an upper layer of the adhesion layer. More excellent adhesion can be realized.

  Further, as a preferable material of the adhesion layer, as described above, a conductive material or an insulating material can be used. For example, the upper layer of the first electrode is Pt, the lower layer of the second electrode is Rh, In the case where the bonding layer region is made of a platinum group element, a material other than the platinum group element can be used, for example, the upper layer of the adhesion layer is made of Au. Alternatively, even when Pt is used for both the upper layer of the first electrode and the lower layer of the second electrode so that the contact surface is made of the same material, another material such as Au is used for the upper layer of the adhesion layer. May be used. Particularly preferred is a case where a platinum group element or an alloy of the same group of platinum group elements is used as the upper layer of the adhesion layer. For example, when the upper layer of the first electrode is Pt and the lower layer of the second electrode is Rh, and the bonding layer region is made of a platinum group element, the upper layer of the adhesion layer is also made of Pt and the entire bonding layer region is made of a platinum group element. It is preferred to configure. Alternatively, when Pt is used for both the upper layer of the first electrode and the lower layer of the second electrode so that the contact surface is made of the same material, the upper layer of the adhesion layer also uses Pt to form the contact surface. It is preferable that all the elements are composed of the same element. Since the p-side second electrode is formed on the same material having irregularities, extremely excellent adhesion can be realized by adopting the above configuration. Pt is particularly preferable among platinum groups. In addition, a preferable material for the lower layer includes Ti.

  Further, it is preferable that the adhesion layer is formed larger than the second electrode. However, as shown in FIG. 3A, the adhesion layer 311 and the second electrode 306 may have substantially the same size. Further, as shown in FIG. 3B, the adhesion layer 311 may be lower than the second insulating film 310. Thereby, a short circuit at the time of face down can be prevented.

Embodiment 4
FIG. 4 shows a configuration of a nitride semiconductor device according to Embodiment 4 of the present invention. In the fourth embodiment, similarly to the first embodiment, an n-type nitride semiconductor layer 402, an active layer 404, and a p-type nitride semiconductor layer 403 are stacked on a substrate 401, and a stripe is formed on the p-type nitride semiconductor layer. A semiconductor laser (LD) provided with a ridge in the shape of an arrow, in which an adhesion layer 411 is formed between a first electrode 405 on the p-side and a first insulating film 409. Such a form can be obtained by forming the adhesion layer before the first electrode. In this case, since the adhesion layer is disposed closer to the semiconductor layer, light confinement in the waveguide region can be controlled depending on the formation position. When the material of the lower layer of the first electrode is selected, for example, with emphasis on the ohmic property with the semiconductor layer, if the material has an extremely large light absorption coefficient, the optical characteristics may be deteriorated. By forming them below the first electrode, their adverse effects can be reduced, and peeling can be reduced even when the adhesion between the first electrode and the first insulating film is poor. . In addition, since the adhesive layer is formed before the first electrode, the adhesive layer is also subjected to heat treatment. Thereby, depending on the material, the adhesion to the first insulating film can be improved.

  In addition, since such an adhesion layer is formed so as to be continuous on the second insulating film, it is possible to suppress a decrease in element characteristics due to peeling of the second electrode and the second insulating film. it can. In addition, as shown in FIG. 4A, an adhesive layer 411 is provided over the second insulating film 409 so that the second insulating film 410 and the second electrode 406 do not overlap each other or are separated from each other. To be provided. Thereby, a short circuit at the time of face down can be prevented. Further, as shown in FIG. 4B, the adhesion layer 411 can be formed so as to extend over the insulating film 409 on the side surface of the ridge. As a result, the adhesion layer can be disposed closer to the waveguide region, so that ripples can be reduced by absorbing stray light, and an excellent FFP can be obtained.

Embodiment 5
FIG. 5 shows a configuration of a nitride semiconductor device according to Embodiment 5 of the present invention. In the fifth embodiment, similarly to the first embodiment, an n-type nitride semiconductor layer 502, an active layer 504, and a p-type nitride semiconductor layer 503 are stacked on a substrate 501, and a stripe is formed on the p-type nitride semiconductor layer. A semiconductor laser (LD) provided with a ridge in a shape, in which a first electrode 505 is formed only above the ridge. In order to form a first electrode having substantially the same width as the ridge width of the LD having a small width, a first electrode 505 having a desired ridge width is formed on a flat wafer, and the semiconductor layer is formed using the first electrode as a mask. Is etched to form a first electrode having the same width as the ridge on the ridge. In order to etch the semiconductor layer using such a self-alignment method, it is preferable to dry-etch mainly using a chlorine-based etching gas. Then, by using a layer made of a platinum group element as an upper layer of the first electrode which is a mask for performing such etching, it is possible to function as an electrode whose surface is relatively less rough even after etching. In the case where the first electrode is formed after the ridge is formed, a mask having a desired ridge width is formed using SiO ₂ or a resist, and the mask is removed after etching the semiconductor layer. A mask may remain on the surface of the semiconductor layer in order to form the first electrode in contact with the semiconductor layer. There is a possibility that ohmic contact and adhesion between the first electrode and the semiconductor layer may be reduced by these, but in the fifth embodiment, such a problem is unlikely to occur.

  As shown in FIG. 5, when the width of the first electrode and the width of the ridge are substantially the same, the bonding surface between the second electrode and the first electrode is an extremely limited region of the width of the ridge. Therefore, the upper layer of the first electrode and the lower layer of the second electrode (when a metal layer is interposed, between the metal layer and the first electrode and between the metal layer and the second electrode) By forming such a bonding layer region, bonding can be performed with extremely excellent adhesion even in a bonding region having a small width, and an operating voltage is low by suppressing an increase in interface resistance and operating characteristics are low. An extremely stable and highly reliable element can be realized.

When the ridge is formed using the self-alignment method, the upper surface of the first electrode is exposed to a chlorine-based gas at the time of etching a semiconductor layer, a fluorine-based gas at the time of etching a SiO ₂ film, or the like. Therefore, a chloride, a fluoride, or the like is formed instead of the oxide. However, even if the platinum group element layer reacts with these chlorine-based gas or fluorine-based gas, the reaction is limited to the vicinity of the surface. Therefore, as in the case of heat treatment, the inside of the layer is easily maintained at the same composition as that at the time of film formation. If the compound with chlorine or fluorine is stable and exhibits insulating properties, an interface resistance will be generated between the compound and the second electrode. In such a case, by cleaning the surface, the compound inside the layer can be removed. By exposing the generation region and forming the second electrode on the exposed portion, the ohmic property can be hardly impaired.

  Also, when exposing the n-type semiconductor layer, the metal layer can be used as a mask. A layer made of a platinum group element is formed as a metal layer 512 shown in FIG. 5 in a desired shape in a region including a region above the first electrode 505, and etching is performed until the n-type semiconductor layer is exposed. When the metal layer is used as a mask, it can be peeled off later, but it can be left formed after etching as shown in FIG. As described above, when the metal layer is used as an etching mask and then used as a part of an electrode without being removed, an insulating film is first formed on both sides of the ridge, and the insulating film and the upper surface of the ridge are formed. It is preferable to form a metal layer so as to cover the first electrode. Thereby, the configuration as shown in FIG. 5 can be obtained. In such a case, since the metal layer in contact with the first electrode can be regarded as a part of the second electrode, a layer made of a platinum group element, which is a preferable material, is used as a lower layer of the second electrode. Preferably it is Pt.

Further, for etching for exposing the n-type semiconductor layer, SiO ₂ or the like can be used instead of the metal layer. In that case, after exposing the n-type semiconductor layer, the SiO ₂ is removed, an insulating film is formed on both sides of the ridge, and a second electrode is provided in contact with the first electrode on the ridge.

  In addition, as shown in FIG. 5A, by forming the second electrode 506 so as not to be in contact with the second insulating film 510, a short circuit at the time of face down can be prevented. Further, since the second electrode 506 can be in contact with only the metal layer 512, an electrode structure with excellent adhesion can be obtained.

Embodiment 6
FIG. 8 shows a configuration of a nitride semiconductor device according to Embodiment 6 of the present invention. FIG. 8B is an XY cross-sectional view of FIG. 8A, which is a nitride semiconductor in which an n-type semiconductor layer 802, an active layer 804, and a p-type semiconductor layer 803 are stacked over a substrate 801. A p-side first electrode 805 and a second electrode 806 on the surface of the p-type semiconductor layer, and an n-side first electrode 807 on the surface of the n-type semiconductor layer exposed by etching from the p-type semiconductor layer side. The second electrode 808 is a light emitting diode (LED) formed of a nitride semiconductor element on which each is formed. No ridge is formed unlike the LD, and light emission from the active layer is emitted to the outside from the p-type semiconductor layer side or the n-type semiconductor layer side, and further, from the end face. A p-side first electrode is formed on almost the entire upper surface of the p-type semiconductor layer, and is joined to the second electrode in a region indicated by reference numeral 813. The first electrode can be a transparent electrode that can be in ohmic contact with the semiconductor layer by controlling the film thickness so that light from the active layer can be transmitted and performing heat treatment.

  In the LED as in the sixth embodiment, the junction area between the first electrode, which is a transparent electrode, and the second electrode, which is a pad electrode, is small, so that a large current flows through the junction area during conduction. By forming the bonding region with good adhesion and low resistance, an LED with excellent reliability can be obtained.

  Further, in the LED, the first electrode on the p-side is not limited to the embodiment shown in FIG. 8, but may be formed in such a shape that the sheet resistance is reduced by increasing the film thickness, and an opening is formed to take out light from the opening. Electrode. Alternatively, the present invention can be applied to various forms such as forming irregularities on the surface of the p-type semiconductor layer and filling the recesses with an insulating film.

  Hereinafter, a semiconductor laser device using a nitride semiconductor will be described as an example. In the present invention, a device structure of an n-type nitride semiconductor layer, an active layer, and a p-type nitride semiconductor layer constituting the nitride semiconductor layer is described. Is not particularly limited, and various layer structures can be used. As a device structure, for example, a laser device structure described in an embodiment to be described later can be cited, but other laser structures and LEDs can also be applied. Specific examples of the nitride semiconductor include nitride semiconductors such as GaN, AlN, and InN, and III-V group nitride semiconductors that are a mixed crystal thereof, and further include B, P, and the like. One or the like can be used. Nitride semiconductor growth is known for growing nitride semiconductors such as MOVPE, MOCVD (metal organic chemical vapor deposition), HVPE (halide vapor phase epitaxy), MBE (molecular beam vapor phase epitaxy), and the like. All methods that apply are applicable.

(substrate)
As the substrate, a sapphire substrate whose main surface is the C plane is used. The substrate is not particularly limited to this, and various substrates such as a sapphire substrate, a SiC substrate, a Si substrate, a spinel substrate, and a GaN substrate having an R surface and an A surface as main surfaces can be used as necessary. . As a GaN substrate, a nitride semiconductor layer (AlGaInN-based) is grown by a so-called ELOG method of performing selective growth (lateral growth) after providing a protective film on a growth substrate such as sapphire or GaAs to suppress growth. Then, by removing the growth substrate, a GaN substrate (nitride semiconductor substrate) having excellent crystallinity can be obtained. The conductivity can also be adjusted by adding impurities such as Si and oxygen during the growth of the nitride semiconductor layer. Further, a GaN substrate obtained by using such an ELOG method is formed as a substrate in which a region having a high dislocation density and a region having a low dislocation density are ubiquitous depending on the conditions of selective growth, the shape of a protective film, and the like. For example, in the case of a laser element that requires reliability under high current density conditions, a semiconductor laser element having excellent characteristics can be obtained by forming a waveguide region in a region having a low dislocation density. When an insulating substrate such as sapphire is used, the p electrode and the n electrode are formed on the same surface side. In a conductive substrate such as a GaN substrate, the p-electrode and the n-electrode may be formed on the same surface side, or the n-electrode may be formed on the back surface of the GaN substrate (the side where the functional layer is not laminated). .

(Underlayer)
An undoped GaN layer is grown at a temperature of 1050 ° C. to a thickness of 2.5 μm, and a protective film made of SiO ₂ is formed to a thickness of 0.27 μm. This SiO ₂ protective film forms a stripe-shaped opening (non-mask region) by etching. This protective film is formed so that the stripe width is 1.8 μm and is substantially perpendicular to the orientation flat surface, and the ratio of the protective film to the opening is such that their width ratio is 6:14. Next, an undoped GaN layer is grown to a thickness of 15 μm. At this time, the GaN layer grown on the opening is grown laterally on the SiO ₂ , and is finally grown so that GaN is aligned in the upper direction of the SiO ₂ . As this underlayer, AlGaN, InGaN, AlInGaN, or the like can be used in addition to GaN.

(Buffer layer)
Next, a buffer layer made of Al _0.02 Ga _0.98 N doped with Si is grown to a thickness of 1 μm using trimethyl gallium (TMG) and ammonia (NH ₃ ) at a temperature of 500 ° C.

(N-type contact layer)
Subsequently, at 1050 ° C., an N-type contact layer made of Si-doped n-Al _0.02 Ga _0.98 N was formed at a thickness of 3.5 μm using TMG, ammonia gas, and silane gas as the source gas. Let it grow. The thickness of the n-type contact layer may be 2 to 30 μm.

(Crack prevention layer)
Next, a crack preventing layer made of Si-doped n-In _0.05 Ga _0.95 N is grown to a thickness of 0.15 μm using TMG, TMI (trimethyl indium), and ammonia at a temperature of 800 ° C. . The crack prevention layer can be omitted depending on the type of the substrate, the composition of other layers, and the like.

(N-type cladding layer)
Next, the temperature was raised to 1050 ° C., and TMA (trimethylaluminum), TMG and ammonia were used as source gases, and an A layer made of undoped Al _0.05 Ga _0.95 N and a B layer made of GaN doped with Si were used. The layers are each grown to a thickness of 50 °. This operation is repeated 110 times, and the A-layer and the B-layer are alternately laminated to grow an n-type clad layer composed of a multilayer film (superlattice structure) having a total film thickness of 1.1 μm. At this time, if the mixed crystal ratio of Al in the undoped AlGaN is in the range of 0.02 or more and 0.3 or less, it is possible to provide a refractive index difference that sufficiently functions as a cladding layer. For each layer constituting the superlattice structure, a mixed crystal ratio other than the above composition, an InGaN-based composition, or the like can be used, and a composition effective for confining light in the active layer can be selected. The n-type cladding layer does not need to have a superlattice structure, and may be a single layer made of Al _0.05 Ga _0.95 N.

(N-type light guide layer)
Next, at the same temperature, an n-type optical guide layer made of undoped GaN is grown to a thickness of 0.15 μm using TMG and ammonia as source gases. This layer may be doped with n-type impurities. The light guide layer may be a layer of InGaN, AlGaN, AlInGaN or the like depending on the composition of the active layer. Alternatively, it may be omitted depending on the composition of the cladding layer and the like.

(Active layer)
Next, the temperature was set to 800 ° C., TMI (trimethyl indium), TMG, and ammonia were used as raw materials, silane gas was used as an impurity gas, and a barrier layer made of Si-doped In _0.02 Ga _0.98 N was formed at 140 ° C. It grows with the film thickness. Subsequently, the silane gas is stopped, and a well layer made of undoped In _0.1 Ga _0.9 N is grown to a thickness of 70 °. This operation is repeated twice, and finally, a barrier layer made of Si-doped In _0.02 Ga _0.98 N is grown to a thickness of 140 ° to form an active layer of a multiple quantum well structure (MQW) having a total thickness of 560 °. Let it grow. The number of stacked MWQs is preferably about 2 to 30, and the composition can be selected from combinations other than those described above, such as InGaN / GaN, AlGaN / InGaN, InGaN / AlInGaN, and AlGaN / AlInGaN. Further, an SQW structure may be used.

(P-type electron confinement layer)
At a similar temperature, a p-type electron confinement layer made of Mg-doped Al _0.25 Ga _0.75 N is grown to a thickness of 30 ° in an N ₂ atmosphere. Next, a p-type electron confinement layer made of Mg-doped Al _0.25 Ga _0.75 N is grown to a thickness of 70 ° in an H ₂ atmosphere. The p-type electron confinement layer may be a single layer, and is stacked at a temperature similar to that of the active layer. Further, AlGaN-based, AlInGaN-based, GaN, or the like having a composition ratio other than the above can be used, and an InGaN-based material can be used by further increasing the film thickness.

(P-type light guide layer)
Next, at a temperature of 1050 ° C., a p-type optical guide layer made of undoped GaN is grown to a thickness of 0.15 μm using TMG and ammonia as source gases. The p-type light guide layer is grown as undoped, but may be doped with Mg. The light guide layer may be a layer of InGaN, AlGaN, AlInGaN or the like depending on the composition of the active layer.

(P-type cladding layer)
Subsequently, an A layer made of undoped Al _0.08 Ga _0.92 N is grown to a thickness of 80 °, and a B layer made of Mg-doped GaN is grown to a thickness of 80 °. This is repeated 28 times to alternately laminate the A layer and the B layer to grow a p-type cladding layer composed of a multilayer film (superlattice structure) having a total film thickness of 0.45 μm. When the p-type cladding layer is made of a superlattice in which at least one includes a nitride semiconductor layer containing Al and has different band gap energies from each other, the p-type cladding layer is doped with one of the impurities in one of the layers. The so-called modulation doping tends to improve the crystallinity, but both may be doped in the same manner. For each layer constituting the superlattice structure, a mixed crystal ratio other than the above composition, or an InGaN-based composition can be used, and a composition effective for confining light in the active layer can be selected. Further, the p-type cladding layer does not have to have a super lattice structure, and may be a single layer made of Al _0.05 Ga _0.95 N or the like .

(P-type contact layer)
Finally, a p-type contact layer made of Mg-doped GaN is grown at a thickness of 150 ° on the p-type cladding layer at 1050 ° C. p-type contact layer may be composed of a p-type _{In x Al y Ga 1-x} -y N (x ≧ 0, y ≧ 0, x + y ≦ 1), p if preferably a GaN doped with Mg The most favorable ohmic contact with the electrode is obtained. After completion of the reaction, the wafer is annealed at 700 ° C. in a nitrogen atmosphere in a reaction vessel to further reduce the resistance of the p-type layer.

(N-type layer exposure)
After the nitride semiconductor is grown to form a laminated structure as described above, the wafer is taken out of the reaction vessel, a protective film made of SiO ₂ is formed on the surface of the uppermost p-type contact layer, and RIE (reaction) is performed. (Native ion etching) using Cl ₂ gas to expose the surface of the n-type contact layer on which the n-electrode is to be formed. At this time, the resonator surface may be formed by etching. It is preferably performed at the same time as the exposure of the n-type contact layer, but may be performed in a separate step.

(Ridge formation)
Next, in order to form a stripe-shaped waveguide region, a protective film made of Si oxide (mainly SiO ₂ ) having a thickness of 0.5 μm is formed on almost the entire surface of the uppermost p-type contact layer by a CVD apparatus. After the formation, a mask having a predetermined shape is formed on the protective film by a photolithography technique, and a protective film made of a striped Si oxide is formed by etching using a CHF ₃ gas by an RIE apparatus. Using the Si oxide protective film as a mask, the semiconductor layer is etched using SiCl ₄ gas to form a ridge stripe above the active layer. At this time, the width of the ridge is set to 1.6 μm.

(First insulating film)
With the SiO ₂ mask formed, a first insulating film made of ZrO ₂ is formed on the surface of the p-type semiconductor layer. This first insulating film may be provided over the entire surface of the semiconductor layer by masking the n-side first electrode formation surface. In addition, a portion where an insulating film is not formed can be provided so that the film is easily divided later.

After the formation of the first insulating film, the wafer is heat-treated at 600 ° C. As described above, when a material other than SiO ₂ is formed as the first insulating film, after forming the first insulating film, the temperature is higher than or equal to 300 ° C., preferably higher than or equal to 400 ° C. and lower than or equal to the decomposition temperature of the nitride semiconductor (1200 ° C.). The heat treatment can stabilize the insulating film material. In particular, when device processing is mainly performed using SiO ₂ as a mask in a step after the formation of the first insulating film, the SiO ₂ mask is dissolved in a mask dissolving material used for later removal. It can be difficult. The heat treatment step of the first insulating film can be omitted depending on the material and the process of the first insulating film, and the process order and the like can be appropriately selected, such as performing simultaneously with the heat treatment of the ohmic electrode. Can be. After the heat treatment, the substrate is immersed in a buffered solution to dissolve and remove SiO ₂ formed on the upper surface of the ridge stripe, and ZrO ₂ on the p-type contact layer (and further on the n-type contact layer) together with SiO ₂ by a lift-off method. Is removed. As a result, the upper surface of the ridge is exposed, and the side surface of the ridge is covered with ZrO ₂ .

(First electrode: ohmic electrode)
Next, a p-side first electrode is formed by sputtering on the top surface of the ridge on the p-type contact layer and on the first insulating film. The p-side first electrode uses Ni / Au (100 ° / 1500 °) as a lower layer and Pt (1500 °) as an upper layer. An n-side first electrode is also formed on the upper surface of the n-type contact layer. The n-side first electrode is made of Ti / Al (200 ° / 8000 °), and is formed in a stripe shape parallel to the ridge and approximately the same length. After forming these electrodes, heat treatment is performed at 600 ° C. in a mixed atmosphere of oxygen and nitrogen.

(Second insulating film)
Next, a resist is formed to cover the entire surface of the p-side first electrode on the ridge and a part of the upper part of the n-side first electrode. Next, a _second insulating film made of SiO ₂ is formed on almost the entire surface, and lift-off is performed so that the entire upper surface of the p-side first electrode and part of the n-side first electrode are exposed. An insulating film (protective film) is formed. The second insulating film is separated from the p-side first electrode, and the first insulating film is exposed therebetween. In consideration of the later division, the first and second insulating films and the electrodes may not be formed in a stripe-shaped area having a width of about 10 μm across the division position in consideration of a later division. Good.

The second insulating film is provided so as to cover the entire surface excluding the upper portions of the p-side and n-side first electrodes. As a preferable material, an oxide containing at least one element selected from the group consisting of Si, Ti, V, Zr, Nb, Hf, and Ta, and at least one of SiN, BN, SiC, AlN, and AlGaN are used. It is preferable to use a single layer film or a multilayer film of SiO ₂ , Al ₂ O ₃ , ZrO ₂ , TiO _2, etc. as a particularly preferable material.

(Second electrode: pad electrode)
Next, a second electrode is formed so as to cover the first electrode. At this time, the second insulating film is preferably formed so as to cover the second insulating film. The lower layer of the p-side second electrode is Pt (1000 °), and is stacked thereon in the order of Ti / Pt / Au (50 ° / 1000 ° / 6000 °). The n-side second electrode is formed of Ni / Ti / Au (1000/1000/8000) from below. The second electrode is in stripe contact with the p-side first electrode and the n-side first electrode via the second insulating film.

(Cleavage and cavity surface formation)
Next, the substrate is polished and adjusted to have a film thickness of about 150 μm. Then, a scribe groove is formed on the back surface of the substrate, and the substrate is broken from the nitride semiconductor layer side and cleaved to obtain a bar-shaped laser. The cleavage plane of the nitride semiconductor layer is the M plane (1-100 plane) of the nitride semiconductor, and this plane is defined as a resonator plane.

(Formation of edge protection film)
It is preferable to provide a protective film on the surface of the resonator formed as described above in order to efficiently resonate light generated in the active layer. In particular, it is preferable to provide a protective film on the resonator surface on the monitor side in order to provide a refractive index difference from the resonator surface on the emission side. Specific examples of the conductive material include Si, Mg, Al, Hf, Nb, Zr, Sc, Ta, Ga, Zn, Y, B, and Ti, and oxides, nitrides, and fluorides thereof. Any of the compounds selected from the above compounds can be used. These may be used alone, or may be used as a compound or a multilayer film in which a plurality of them are combined. Preferred materials are materials using Si, Mg, Al, Hf, Zr, Y, and Ga. Further, AlN, AlGaN, BN, or the like can be used as the semiconductor material. As the insulator material, compounds such as oxides, nitrides, and fluorides of Si, Mg, Al, Hf, Nb, Zr, Sc, Ta, Ga, Zn, Y, and B can be used.

In this embodiment, specifically, a dielectric multilayer film made of SiO ₂ and ZrO ₂ is formed as an end face protective film. On the resonator surface on the light reflection side (monitor side), a protective film made of ZrO ₂ is formed by using a sputtering device, and then a high reflection film is formed by alternately stacking six pairs of SiO ₂ and ZrO _2. . Here, the thicknesses of the protective film, the SiO ₂ film and the ZrO ₂ film constituting the high reflection film can be set to preferable thicknesses according to the wavelength of light emitted from the active layer. Nothing needs to be provided on the light emitting side resonator surface, and a first low reflection film made of ZrO ₂ , Nb ₂ O ₅ , Al ₂ O ₃ , and ZrO ₂ is formed by using a sputtering apparatus. A _second low reflection film made of SiO ₂ may be formed.

Finally, a groove is formed by scribing so as to be substantially parallel to the ridge stripe, and the bar is cut at the groove to obtain the nitride semiconductor laser device of the present invention. The nitride semiconductor laser device obtained as described above is capable of continuously oscillating at a threshold current density of 2.0 kA / cm ² and a high output of 60 mW at an oscillation wavelength of 405 nm without peeling of the electrodes.

  In the second embodiment, the second insulating film is formed so as to cover a part of the p-side first electrode. A second insulating film is formed in the same manner as in the first embodiment, except that the second insulating film is formed as follows.

(Second insulating film)
After the formation of the first electrode, a _second insulating film made of SiO ₂ is formed on substantially the entire surface, and the entire surface of the p-side first electrode on the ridge and the p-side first electrode on the upper surface of the p-type semiconductor layer on both sides of the ridge are formed. A resist is applied so as to expose a part and a part of the n-side first electrode, and a part of each electrode is exposed by dry etching. Thereby, a part of the p-side and n-side first electrodes and the side surfaces are covered with the second insulating film. Here, the first insulating film is formed so as not to be exposed, but may be exposed.

(Second electrode: pad electrode)
Next, a p-side second electrode is formed by sputtering to cover the second insulating film. The lower layer of the p-side second electrode is Pt (1000 °), on which Ti / Pt / Au (50 ° / 1000 ° / 6000 °) is formed. The n-side second electrode is formed of Ni / Ti / Au (1000/1000/6000) from below. The second electrode is in stripe contact with the p-side first electrode and the n-side first electrode via the second insulating film. The nitride semiconductor laser device obtained in this manner is capable of continuously oscillating at an oscillation wavelength of 405 nm at room temperature with a threshold current density of 2.0 kA / cm ² and a high output of 60 mW without any peeling of the electrodes confirmed.

In the third embodiment, an adhesion layer is formed. In Example 1, the p-side first electrode is formed of Ni / Au / Pt (100 ° / 1500 ° / 1500 °) on the first insulating film. Next, a multilayer film composed of two pairs of SiO ₂ / TiO ₂ (1500 ° / 1000 °) is formed as a second insulating film, and at this time, it is formed so as to be separated from the p-side first electrode by about 225 μm. Next, Ti / Pt (100 ° / 500 °) as an adhesion layer is exposed on a part of the p-side first electrode and between the p-side first electrode and the second insulating film. And over the second insulating film. Then, Pt / Ti / Pt / Au (1000/50/1000/6000) is formed as a p-side second electrode from the p-side first electrode to the adhesion layer. Other steps are performed in the same manner as in Example 1 to obtain the nitride semiconductor laser device of the present invention. In the nitride semiconductor laser device obtained as described above, no peeling of the electrode was confirmed, and the device was capable of continuous oscillation at a threshold current density of 2.0 kA / cm ² and a high output of 60 mW at a room temperature with an oscillation wavelength of 405 nm. .

In the fourth embodiment, in the first embodiment, the lower layer of the p-side first electrode is Ni / Au (100 ° / 1500 °), the upper layer is Pt / Ti / Pt (500 ° / 100 ° / 500 °), and the lower layer of the second electrode is Pt. (1000 °), and the same process as in Example 1 except that Ti / Pt / Au (100 ° / 1000 ° / 6000 °) is formed thereon. The obtained nitride semiconductor laser device is capable of continuous oscillation with a threshold current density of 2.0 kA / cm ² at room temperature and an oscillation wavelength of 405 nm at a high output of 60 mW.

  The present invention is applicable to all devices to which a laser element can be applied, for example, CD players, MD players, various game machines, DVD players, main lines and optical communication systems such as telephone lines and submarine cables, laser scalpels, and laser treatment devices. , Medical equipment such as laser acupressure machines, printing machines such as laser beam printers and displays, various measuring instruments, laser level, laser length measuring machines, laser sensing guns such as laser speed guns, laser thermometers, laser power transport, etc. It can be used in various fields.

1 is a schematic cross-sectional view illustrating a nitride semiconductor device according to a first embodiment of the present invention. FIG. 4 is a schematic cross-sectional view illustrating a nitride semiconductor device according to a second embodiment of the present invention. Schematic sectional view illustrating a nitride semiconductor device according to a third embodiment of the present invention. Schematic sectional view illustrating a nitride semiconductor device according to a fourth embodiment of the present invention. Schematic cross-sectional view illustrating a nitride semiconductor device according to a fifth embodiment of the present invention. FIG. 2 is a schematic cross-sectional view illustrating the formation of a bonding layer region between a first electrode and a second electrode according to the present invention. FIG. 4 is a schematic cross-sectional view illustrating a bonding layer region between a first electrode and a second electrode according to the present invention. A schematic plan view and a schematic cross-sectional view illustrating a nitride semiconductor device according to a sixth embodiment of the present invention.

Explanation of reference numerals

101, 201, 301, 401, 501, 801 ... substrate 102, 202, 302, 402, 502, 802 ... n-type nitride semiconductor layer 103, 203, 303, 403, 503, 803 ... p -Type nitride semiconductor layer 104, 204, 304, 404, 504, 804 ... active layer 105, 205, 305, 405, 505, 705, 805 ... first electrode (p-side ohmic electrode)
605 (a): upper layer of first electrode 605 (b): lower layer of first electrode 106, 206, 306, 406, 506, 706, 806 ... second electrode (p-side pad electrode)
606 (a): upper layer of second electrode 606 (b): lower layer of second electrode 107, 207, 307, 407, 507, 807 ... first electrode (n-side ohmic electrode)
108, 208, 308, 408, 508, 808 ... second electrode (n-side pad electrode)
109, 209, 309, 409, 509, 609, 709 ... first insulating film 110, 210, 310, 410, 510, 710 ... second insulating film 311, 411, 711 ... adhesion layer 512: metal layer 613, 713, 813: bonding layer region 814: insulating film

Claims (23)

A nitride semiconductor device, comprising: a first electrode in ohmic contact with a semiconductor layer; and a second electrode in contact with the first electrode and having a different shape from the first electrode,
The first electrode and the second electrode have a bonding layer region including an upper layer of the first electrode forming a first electrode surface and a lower layer of the second electrode deposited on the heat-treated first electrode;
The bonding layer region is made of a platinum group element. A nitride semiconductor device, comprising: a first electrode in ohmic contact with a semiconductor layer; and a second electrode in contact with the first electrode and having a different shape from the first electrode,
The first electrode and the second electrode have a bonding layer region including an upper layer of the first electrode forming a first electrode surface and a lower layer of the second electrode deposited on the heat-treated first electrode;
The upper layer of the first electrode and the lower layer of the second electrode are made of the same element or the same material. The nitride semiconductor device according to claim 1, wherein the first electrode has a lower layer made of a heat-treatable alloying material. 4. The nitride semiconductor according to claim 1, wherein the upper layer of the first electrode is a platinum group single layer made of a single platinum group element or an alloyed layer made of a homologous element in the platinum group element. 5. element. 5. The nitride semiconductor device according to claim 1, wherein the lower layer of the second electrode comprises a platinum group single layer made of a single platinum group element or a platinum group element alloy layer. 6. 6. The nitride semiconductor device according to claim 1, wherein an upper layer of said first electrode is made of Pt. 7. The nitride semiconductor device according to claim 1, wherein a lower layer of said second electrode is made of Pt. The semiconductor layer surface provided with the first electrode has an electrode forming region and an insulating film forming region, and the second electrode covers the insulating film forming region from the electrode forming region. Item 7. The nitride semiconductor device according to Item 7. 9. The nitride semiconductor device according to claim 8, wherein the insulating film formation region is a plurality of regions sandwiching the stripe-shaped electrode formation region, or a plurality of regions separated by the electrode formation region. 10. The nitride semiconductor device according to claim 1, wherein the nitride semiconductor device is a laser device having a ridge formed of a convex portion, and wherein the first electrode is formed in contact with an upper surface of the ridge. A first insulating film on a side surface of the semiconductor layer continuous from both side surfaces of the ridge and a side surface thereof, and a second insulating film continuous from the first insulating film to the side surface of the semiconductor layer; 11. The nitride semiconductor device according to claim 10, wherein the first electrode is formed so as to be separated from the second insulating film. The nitride semiconductor device according to claim 10, wherein the first insulating film and / or the second insulating film is covered with a single-layer or multilayer adhesive layer. 13. The nitride semiconductor device according to claim 12, wherein an uppermost layer of the adhesion layer is a layer containing a platinum group element. 14. The nitride semiconductor device according to claim 12, wherein an uppermost layer of the adhesion layer is made of the same element or the same material as an upper layer of the first electrode. The nitride semiconductor device according to claim 12, wherein an upper layer of the adhesion layer is Pt. The nitride semiconductor device according to claim 12, wherein the adhesion layer is formed in contact with above or below the first electrode. A nitride semiconductor device, comprising: a first electrode in ohmic contact with a semiconductor layer; and a second electrode in contact with the first electrode and having a different shape from the first electrode,
The surface of the semiconductor layer provided with the first electrode has an electrode forming region and an insulating film forming region,
The nitride semiconductor element, wherein the second electrode covers the electrode formation region and the insulating film formation region. 18. The nitride semiconductor device according to claim 17, wherein the insulating film formation region is a plurality of regions sandwiching the stripe-shaped electrode formation region or a plurality of regions separated by the electrode formation region. 19. The nitride semiconductor device according to claim 17, wherein the nitride semiconductor device is a laser device provided with a ridge having a convex portion, and wherein the first electrode is formed in contact with an upper surface of the ridge. A first insulating film is provided on both sides of the ridge and on a plane of the semiconductor layer continuous from the side surface thereof, and a second insulating film is provided from the first insulating film to the side surface of the semiconductor layer. 20. The nitride semiconductor device according to claim 19, wherein said first electrode is formed so as to be separated from said second insulating film. 21. The nitride semiconductor device according to claim 20, wherein the first insulating film and / or the second insulating film is covered with a single-layer or multilayer adhesive layer. 22. The nitride semiconductor device according to claim 21, wherein an uppermost layer of the adhesion layer is a layer containing a platinum group element. 23. The nitride semiconductor device according to claim 22, wherein the uppermost layer of the adhesion layer is Pt.

Images