LIGHT EMITTING DEVICE WITH TRANSPARENT CONDUCTIVE GROUP-III NITRIDE LAYER

Abstract

A group III-nitride semiconductor device comprises a light emitting semiconductor structure comprising a p-type layer and an n-type layer operable as a light emitting diode or laser. On top of the p-type layer there is arranged an n+ or n++-type layer of a group III-nitride, which is transparent to the light emitted from the underlying semiconductor structure and of sufficiently high electrical conductivity to provide lateral spreading of injection current for the light-emitting semiconductor structure.

Claims

1. A device comprising: a light-emitting semiconductor structure operable to emit light and comprising a first layer, which is p-type and composed of a nitride of at least one group-III element, and a second layer, which is n-type and composed of a nitride of at least one group-III element; and a transparent, current spreading layer of a nitride composed of at least one group-III element, which is transparent to light emitted from the light-emitting semiconductor structure and of sufficiently high electrical conductivity to provide lateral spreading of injection current for the light-emitting semiconductor structure within the transparent, current spreading layer.

2. The device according to claim 1, wherein the electrical conductivity of the transparent, current spreading layer exceeds the electrical conductivity of the first layer by at least a factor of 10.

3. The device according to claim 1, wherein the transparent, current spreading layer is n-type.

4. The device according to claim 1, further comprising a current aperture stop defining a current aperture between a portion of the first layer and a portion of the transparent, current spreading layer.

5. The device according to claim 4, wherein the current aperture stop has a resistivity of at least 10 times a resistivity of the first layer.

6. The device according to claim 4, wherein the current aperture stop is composed of a nitride of at least one group-III element.

7. The device according to claim 1, wherein the first layer is doped with Mg as a p-type dopant.

8. The device according to claim 1, wherein a point defect density in the transparent, current spreading layer is at least one of: above 5.Math.10.sup.19/cm.sup.3; between 5-10.sup.19/cm.sup.3 and 1.Math.10.sup.21/cm.sup.33; and between 5.Math.10.sup.19/cm.sup.3 and 5.Math.10.sup.20/cm.sup.3.

9. A method for manufacturing a device, the method comprising: producing a light-emitting semiconductor structure operable to emit light by depositing a first layer, which is p-type and composed of a nitride of at least one group-III element, and depositing a second layer, which is n-type and composed of a nitride of at least one group-III element; and depositing a transparent, current spreading layer, which is n-type and composed of a nitride of at least one group-Ill element, the transparent, current spreading layer being configured to be transparent to light emitted from the light-emitting semiconductor structure and of sufficiently high electrical conductivity to provide lateral spreading of injection current for the light-emitting semiconductor structure within the transparent, current spreading layer.

10. The method according to claim 9, wherein at least the first layer of the light-emitting semiconductor structure is deposited using metal organic vapor phase epitaxy, and wherein the transparent, current spreading layer is deposited using molecular beam epitaxy.

11. The method according to claim 9, wherein as deposited the first layer has passivated dopants, the method further comprising activating said passivated dopants in at least a portion of the first layer.

12. The method according to claim 11, wherein the passivated dopants are activated in said portion of the first layer by applying a local heat treatment.

13. The method according to claim 9, further comprising producing a current aperture stop between the transparent, current spreading layer and die first layer so as to define a current aperture therebetween.

14. The method according to claim 13, wherein the current aperture stop has a resistivity amounting to at least 10 times a resistivity of the first layer.

15. The method according to claim 13, wherein the current aperture stop is produced from a part of the first layer by locally increasing the resistivity in said part of the first layer, wherein locally increasing the resistivity in said part of the first layer comprises effecting a diffusion of foreign atoms locally into the first layer.

16. The method according to claim 15, wherein effecting the diffusion of the foreign atoms locally into the first layer comprises: depositing on the first layer, in a lateral area where the current aperture stop is to be produced, a diffusion-promoting mask containing the foreign atoms; and effecting the diffusion of the foreign atoms into the first layer by applying a heat treatment.

17. The method according to claim 15, wherein effecting the diffusion of the foreign atoms locally into the first layer comprises: depositing on the first layer, in a lateral area where the current aperture is to be produced, a diffusion-inhibiting mask; and applying a plasma containing the foreign atoms to diffuse into the first layer.

18. The method according to claim 9, wherein the first layer is doped with Mg as a p-type dopant.

19. The method according to claim 9, wherein the electrical conductivity of the transparent, current spreading layer exceeds the electrical conductivity of the first layer by at least a factor of 10.

20. The method according to claim 9, wherein a point defect density in the transparent, current spreading layer is at least one of: above 5.Math.10.sup.19/cm.sup.3: between 5.Math.10.sup.19/cm.sup.3 and 1.Math.10.sup.21/cm.sup.33; and between 5.Math.10.sup.19/cm.sup.3 and 5.Math.10.sup.20/cm.sup.3.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0115] Below, the invention is described in more detail by means of examples and the included drawings. The figures show schematically in a cross-sectional view:

[0116] FIG. 1 a device comprising a TCN layer and a current aperture stop;

[0117] FIG. 2A a prior art GaAs-based VCSEL;

[0118] FIG. 2B a prior art GaAs-based VCSEL with TCO current spreading layer;

[0119] FIG. 3 a device without current aperture;

[0120] FIG. 4 a flip chip LED device comprising a TCN layer;

[0121] FIG. 5A an LED device comprising a TCN layer and current aperture stop, with non-conductive substrate;

[0122] FIG. 5B an LED device comprising a TCN layer and current aperture stop, with conductive substrate;

[0123] FIG. 6 a ridge waveguide laser diode device;

[0124] FIG. 7 a stripe injection gain-guided laser device;

[0125] FIG. 8 an air-cladding laser device;

[0126] FIG. 9 a VCSEL device;

[0127] FIG. 10a device in which the current aperture stop extends from the TCN layer to the active region of the semiconductor structure;

[0128] FIG. 11A a semiconductor structure on a substrate;

[0129] FIG. 11B the semiconductor structure of FIG. 11A with a current aperture in the first layer;

[0130] FIG. 11C a device obtained from the semiconductor structure of FIGS. 11A, 11B by applying the TCN layer across the current aperture and a portion of the first layer.

DETAILED DESCRIPTION OF THE INVENTION

[0131] The described embodiments are meant as examples or for clarifying the invention and shall not limit the invention.

[0132] We describe in several examples a transparent and conductive III-nitride layer (herein also referred to as TCN layer or transparent layer) that can be present, e.g., on an uppermost Mg-doped layer of a III-nitride p-i-n junction grown by, e.g., conventional MOVPE.

[0133] FIG. 1 illustrates schematically such a device 10 in a cross-sectional view.

[0134] Even though p-i-n type semiconductor structures are illustrated through most of the present description, corresponding embodiments with p-n type semiconductor structures are also possible embodiments.

[0135] The TCN layer 5, grown as a monocrystalline layer (e.g. by MBE or MOVPE) or deposited as a polycrystalline layer (e.g., by sputtering, evaporation or atomic layer deposition), provides very low resistivity, in particular- to metallic contacts, and offers excellent transparency to UV and/or visible light. The TCN layer 5 has a high concentration of impurities (like for example silicon, oxygen, carbon, magnesium, beryllium, germanium, zinc atoms) or other types of point defects, like for example Nitrogen vacancies.

[0136] The semiconductor structure 4 is a p-i-n structure, e.g., grown by MOVPE. The p-doped layer 1 is also referred to as first layer 1, the n-doped layer 2 is also referred to as second layer 2, and the layer 3 in between can be an i-layer, i.e. a non-intentionally doped layer.

[0137] Semiconductor structure 4 is post-growth thermally annealed to activate the p-dopant in the first layer 1, i.e. to remove hydrogen bonded to the Mg dopant, and it can be also processed by conventional techniques (photolithography, etching, plasma passivation, ion implantation, etc.), e.g., to optionally create a current aperture 7 defined by a current aperture stop 6 before the growth/deposition of the TCN layer 5.

[0138] III nitride based p-i-n structures 4 grown by MOVPE are current technology required for producing highly efficient and long-lifetime light-emitting diodes and laser diodes with emission in the UV or visible part of the spectrum. Despite the strong degree of development that such structures have experienced over the last 20 years, some aspects related to the quality of the contact layers, in particular to the p-side of the junction, remain unsolved. The present description relates i.a. to the following aspects:

[0139] Low values for the specific resistance of metallic contacts provided for current injection to the p-doped layers, even though highly desirable, are difficult to achieve. Most values reported in literature lie in the high 10.sup.−4 Ω.Math.cm.sup.−2 or even 10.sup.−3 Ω.Math.cm.sup.−2 range for metallic contacts deposited over Mg-doped GaN in light-emitting diodes with visible emission. UV-emitters with aluminium-containing uppermost layers suffer from even worse resistances. This resistance has a direct impact on the current-voltage (I-V) characteristics of the final device, and also impacts the electric-to-optic power conversion efficiency.

[0140] The known Mg-doped p-layers themselves have a considerable resistivity, which apparently renders impossible in known technology the realization of device geometries with lateral injection (current flowing along a lateral direction, i.e. in-plane), as it can be for example desirable for intracavity Vertical Cavity Surface Emitting Laser (VCSEL) contacts in the GaAs material system. FIG. 2A illustrates a prior art GaAs-based VCSEL. Metal contacts are referenced 8 and 9, reflectors are referenced 11 and 12. The open arrow indicates the direction of light emission. This limitation has been overcome in prior art by the use of transparent conductive oxide layers (TCO) that can provide a decent specific resistance to the uppermost p-layer 1 in the epitaxial structure (e.g., 4 to 9 times 10.sup.−4 Ω.Math.cm.sup.−2). And at the same time, it can offer a certain degree of transparency. FIG. 2B illustrates such a device with a TCO layer 15. Despite strong development efforts and use of such semi-transparent layers, it is a challenge in prior art to provide contact layers that are better matched to the uppermost p-layer 1 of the epitaxial structure and that at the same time offer a high degree of transparency.

[0141] We herein describe the use of a III-nitride layer (in particular AlInGaN with any possible composition of Al, In and Ga atoms) that is deposited, e.g., regrown or epitaxially grown on, e.g., a conventional p-i-n structure produced by MOVPE.

[0142] The as grown MOVPE structure comes with p-layers that are not active due to the magnesium being bond to hydrogen atoms that are widely present in the epitaxial growth environment. In order to achieve current injection and to properly operate the device, the p-dopant is activated by post-growth thermal annealing and hydrogen segregation from the structure resulting therefrom. The activation of the p dopant can be performed, e.g., either in-situ inside the MOVPE-growth equipment or ex-situ in a thermal annealing oven. After the activation of the Mg dopant, the semiconductor structure 4 can be processed to create one or more current apertures and then produce the TCN layer 5 (cf. FIG. 1), or the TCN layer 5 can be produced directly on first layer 1. The latter case with no current aperture is illustrated in FIG. 3.

[0143] In FIG. 3 is also shown that the semiconductor structure 4 can be (in any embodiment) present on a substrate 18.

[0144] The TCN layer, e.g., a highly Si-doped GaN layer, can be grown by MBE, e.g., over a conventional p-i-n structure 4. Nevertheless, other growth or deposition techniques such as MOVPE, sputtering, evaporation can, in instances be used, too. In contrast to MOVPE as usually applied, MBE can be carried out in a hydrogen-free environment. Therefore, using MBE or other hydrogen-free techniques for depositing the TCN layer 5 can prevent that Mg which is already present (in the first layer 1) is deactivated (by hydrogen present during deposition of the TCN layer 5). Moreover, the TCN layer 5 can be epitaxially grown over the existing structure (which includes the first layer 1 and, if present, also the current aperture stop 6) and thus offers an excellent crystalline purity and can therefore be designed for very high transparency to the light emitted from the active region of the underlying semiconductor structure 4, i.e. from the i-layer in case of a p-i-n structure 4.

[0145] Light emitting diodes (LEDs) and laser diodes (LDs) can be designed with the TCN layer 5, e.g., regrown by means of MBE. On such devices, the inventors already have achieved contact resistances as low as 1.Math.10.sup.−6 Ω.Math.cm.sup.−2 (which is two orders of magnitude lower than in conventional devices without TCN layer 5), while no adverse effect on the optical performance or quality of the devices has been found.

[0146] On the TCN layer 5 of the device 10, one or more electrical contacts (such as prior art contact 8, cf. FIGS. 2A, 2B) can be applied.

[0147] Examples of Possible Device Designs Including a TCN Layer 5

[0148] The device 10 can be, e.g., one of the devices sketched in the following. The meaning of the reference symbols can be inferred from the previous Figures, where not explicitly explained. The open arrows indicate directions of light emission. Electrical contacts 8, 9 can be metallic contacts.

[0149] Light Emitting Diode [0150] a. FIG. 4 illustrates flip chip device 10: Herein, the TCN layer 5 can contribute to minimizing the resistance to the electrical contact 8 which is used as reflector 13. [0151] b. FIGS. 5A and 5B illustrate examples with current apertures 7, in which a transparent contact geometry is realized such that light is extracted through the TCN layer 5. This can contribute to achieving very high transparency and homogeneous current spreading over the p-layer (first layer 1) grown by MOVPE. FIG. 5A illustrates a case with a non-conductive substrate 18a, FIG. 5B with a conductive substrate 18b. Current aperture stop 6 can be, but need not be present.

[0152] Edge-Emitting Laser Diode or Optical Amplifier or Superluminescent Diode [0153] a. FIG. 6 illustrates a ridge waveguide laser diode device 10. The TCN layer 5 can contribute to minimizing the resistance of the p-side metallic contact 8 used to inject current into the active region 3. A dielectric 17 is provided. The direction of light emission is perpendicular to the drawing plane. [0154] b. FIG. 7 illustrates a stripe injection gain-guided laser device 10. The TCN layer 5 can contribute to minimizing the resistance to the metallic contact 8. And it can contribute to making the current injection uniform over a stripe-shaped region. The stripe region can be defined by standard fabrication techniques in the p-i-n structure grown by MOVPE before the deposition of the TCN layer 5. [0155] c. FIG. 8 illustrates an air-cladding laser device 10. The TCN layer 5 is used in this case as a current spreading layer in a p-i-n laser structure 4 designed for a large penetration of the optical mode towards the p-side. The inset illustrates the modal intensity I as a function of the coordinate −z. The interface between air and the TCN layer 5 squeezes the laser mode back into the active region 3 and can contribute to maximizing the optical confinement. A stripe region can be defined by standard fabrication techniques in the p-i-n structure grown by MOVPE before the deposition of the TCN layer 5.

[0156] Vertical Cavity Surface Emitting Laser [0157] a. FIG. 9 illustrates a VCSEL device 10 in which the TCN layer 5 is used as a current spreading layer to achieve efficient intracavity injection of the VCSEL active region 3. It can be of advantage for the operation of the VCSEL when the TCN layer has a high transparency (such as more than 99%). A current aperture (not illustrated in FIG. 9) can be defined by standard fabrication techniques in the p-i-n structure 4 grown by MOVPE before the deposition of the TCN layer 5. Reflectors 11, 12 are provided.

Manufacturing Details and Further Embodiments

[0158] In several examples above with a current aperture stop 6, the first layer 1 extends below the current aperture stop 6. However, this needs not be the case, neither in the embodiments above, nor in those below. In FIG. 10, an embodiment is illustrated in which a thickness of the current aperture stop 6 is identical to a thickness of the first layer 1. While in illustrated embodiments above, only a portion of the first layer 1 is present in the lateral area defined by the current aperture 7, in FIG. 10, the whole first layer is present in the lateral area defined by the current aperture 7.

[0159] FIGS. 11A, 11B, 11C illustrate method steps for manufacturing a device 10.

[0160] Initially provided/manufactured is the semiconductor structure 4 (which also can be considered the body of the light emitting device 10). An example is illustrated in FIG. 11A. It includes one or more light emitting layers that are sandwiched between doped layers 1, 2 of different type. P-type layers can be above, towards the surface of the device. N-type layers can be below, in between the light emitting layers and an optional substrate 18. Substrate 18 can be, e.g., free-standing III-nitride (e.g., GaN), sapphire, Silicon, SiC. It may include reflectors, in particular optical reflectors having a high reflectivity such as Bragg reflectors In FIG. 11A, i-layer 3 represents the active region of the semiconductor structure 4.

[0161] In a next step, one or more current aperture stops 6 can be produced (cf. FIG. 11B), in particular in the semiconductor structure 4 and even more specifically in the first layer 1. As illustrated in FIG. 11B, a current aperture 7 can be located within the top p-layer 1. Current aperture 7 can have lower resistivity than the surrounding areas, i.e. than the current aperture stop 6. In the current aperture stop, conductivity due to the p-type dopant can be, e.g., compensated by the presence of other impurities of different type, i.e. of n-type, like for example Si, Oxygen, Titanium, Carbon. Or the p-type dopant (Mg) can be passivated by Hydrogen, in order to form the current aperture stop in the first layer 1.

[0162] A current aperture stop 6 can also be considered a current confining area or current confining structure or current blocking structure.

[0163] In a next step (cf. FIG. 11C), the TCN layer 5 is applied on top. The p-i-n structure 4 (more specifically a portion of the first layer 1) and the current aperture stop 6 are overgrown by the TCN layer, and thus by a transparent highly n-type or p-type doped III-layer (the group-III elements being one or more of Al, In, Ga). The TCN layer 5 minimizes the resistance to metallic layers that can be deposited thereabove. And it can also provide a tunnel junction at the interface with the uppermost layer (first layer 1) for efficient current injection through the non-resistive current apertures, in case the TCN layer 5 is n-type doped and the first layer is p-type doped or vice versa.

[0164] The body of the light emitting device (semiconductor structure 4, cf. FIG. 11A) can be made by, e.g., MOCVD or MBE.

[0165] The current aperture 7 can be formed, e.g., from a first layer 1 which is a resistive p-layer, by decreasing the resistivity in local areas (where the current aperture shall be located), e.g., by removal of a type of impurities (like for example Hydrogen or Oxygen impurities that are present in the layer together with the Mg).

[0166] Or, the current aperture can be formed, e.g., from a low-resistivity first layer which is a p-layer, by increasing the resistivity in local areas (where the current aperture stop shall be located), by the introduction of doping species of different type (i.e. n-type, like Si, O.sub.2, Ti, C), or by locally passivating the p-type Mg with Hydrogen.

[0167] The highly doped TCN layer 5 (e.g., n-type doped; or p-type doped) can be epitaxially deposited by MBE. In case of an n-doped TCN-layer 5, this can result in an efficient tunnel junction at the interface with the uppermost p-layer.

[0168] Some details of exemplary current aperture stops:

[0169] The current aperture seen from the top of the wafer (i.e. in vertical direction) can have, e.g., a circular or rectangular shape. [0170] a. Circular: diameter can be, e.g., between 2 μm and 100 μm; [0171] b. Rectangular: Dimension in x: can be, e.g., between 2 μm and 100 μm; and dimension in y: can be, e.g., between 2 μm and 2000 μm

[0172] A device can include several current apertures. E.g., a current aperture can be repeated several times within a single device, e.g., to produce linear or 2D current aperture arrays.

[0173] The thickness t (along the vertical axis) of the current aperture stop 6 (resistive region) can be, e.g., between 10 nm and 1 μm or between 10 nm and the full thickness of the first layer. Its thickness can be dependent on the method used for creating the current aperture stop: [0174] a. Selective Mg activation: Thickness t can be identical with the full thickness of the first layer 1 [0175] b. Local passivation by mask and annealing: e.g., 50 nm<t<500 nm [0176] c. Local passivation by plasma treatment: e.g., 5 nm<t<100 nm [0177] d. Resistive region created by deposition (e.g., regrowth), such as after etching: e.g., 10 nm<t<full thickness of the first layer 1.

[0178] In the following, some ways for producing a current aperture stop will be briefly described (first layer p-doped with Mg assumed). [0179] 1. Selective activation by mask deposition and annealing: [0180] Mask selectively deposited only over regions outside current aperture 7. Mask can be made of, e.g., SiN, or SiO.sub.2. [0181] Mg activation by thermal annealing: T>450° C. [0182] Removal of mask [0183] 2. Selective passivation by mask deposition and annealing: [0184] Mg activation under thermal annealing, e.g., at above >450° C. [0185] Mask selectively deposited only over regions where resistivity will be increased, i.e., where current aperture stop 6 shall be created. Mask can be made of, e.g., SiN, or SiO.sub.2 or Ti. [0186] Annealing, e.g., at temperature −100° C.<T<450° C. [0187] Removal of mask [0188] 3. Local passivation by plasma treatment: [0189] Mg activation under thermal annealing, e.g., at above >450° C. [0190] Mask selectively deposited (e.g., a photoresist material) to protect current aperture region. [0191] Plasma treatment: plasma power supply typically in the 50 V to 500 V range. Pressure: 1<p<100 mTorr. Possible atomic species in the plasma: one of more of H, O, C, Si, Cl, Ar [0192] Removal of mask [0193] 4. Resistive region created by etching and regrowth: [0194] Mg activation under thermal annealing, e.g., at above >450° C. [0195] Mask selectively deposited to protect current aperture region. [0196] Etching down exposed surface to a desired thickness, e.g., by ion etching techniques including Cl atoms [0197] Regrowth of, e.g., codoped (Mg and Si dopants), resistive region(s), e.g., by MBE [0198] Removal of mask

[0199] In all above cases, deposition of the TCN layer can take place after removal of the mask.

[0200] Some further exemplary details of devices 10 (cf., e.g., FIG. 3; but the details apply also if a current aperture stop 6 is present): [0201] The substrate 18 is suited for epitaxial growth of an AlInGa—N compound, and is, e.g., bulk GaN (thickness between 0.2 mm and 1 mm) or sapphire (thickness between 0.2 mm and 1 mm). In the case of bulk GaN, the substrate is, e.g., n-doped with Si or O impurities as dopant. [0202] The second layer 2 can be made of a III-nitride crystal and e.g., can contain any possible combination of the group-III elements Al, Ga, In. The composition can change along the vertical axis to create desired effects, e.g., for improving light confinement and/or to optimize current injection towards the active region 3 and/or for improved structural strain management. [0203] The second layer 2 can be, e.g., n-doped with Si. The doping level can vary over the vertical axis, and is typically lower closer to the active region 3 (e.g., between 1.Math.10.sup.17/cm.sup.3 and 1.Math.10.sup.18/cm.sup.3) and higher towards the substrate (e.g., more than 1.Math.10.sup.18/cm.sup.3). The thickness of the second layer 2 can depend on the device geometry. It can be, e.g., between 0.2 μm and 5 μm. [0204] The active region 3 is typically non intentionally doped (n.i.d.) and can be made of or at least comprise quantum wells, such as lower band-gap III-nitride crystal material (such as one or multiple nitride layers containing In and Ga and Al) embedded in a larger bandgap group-III-nitride crystal material acting as a barrier (such as one or multiple III-nitride layers containing In and Ga and Al). The thickness of the active region 3 can depend on the device geometry. It can be, e.g., between 2 nm and 500 nm. [0205] The first layer 1 can be made of a III-nitride crystal and e.g., can contain any possible combination of the group-III elements Al, Ga, In. The composition can change along the vertical axis to create desired effects, e.g., for improving light confinement and/or to optimize current injection towards the active region 3 and/or for improved structural strain management. [0206] The first layer 1 can be, e.g., p-doped with Mg. The doping level can vary over the vertical axis, and can be lower closer to the active region 3 (e.g., between 1.Math.10.sup.18/cm.sup.3 and 5.Math.10.sup.18/cm.sup.3) and higher towards the substrate (e.g., more than 5.Math.10.sup.18/cm.sup.3 and can be lower than 2.Math.10.sup.20/cm.sup.3). The thickness of the first layer 1 can depend on device geometry. It can be, e.g., between 0.05 μm and 1 μm. [0207] The semiconductor structure 4 (body of the light emitting device), being the total of first layer 1 plus active region 3 plus second layer 2 can be epitaxially grown by Metal Organic Vapour Phase Epitaxy (MOVPE) or MOCVD (Metal Organic Chemical Vapor Deposition). [0208] The transparent layer 5 (Transparent Conductive Nitride layer, TCN layer) is in contact with the first layer 1 and can be grown by MBE. It can be highly n-doped, e.g., with Si. Typical n-doping levels can be between 1.Math.10.sup.18/cm.sup.3 and 5.Math.10.sup.20/cm.sup.3. The dopant concentration can vary along the vertial axis. Typical thickness t of the TCN layer 5 can depend on the device geometry and amount to, e.g., between 30 nm and 300 nm.

[0209] Summarizing some points of the described devices and methods, we note that MOVPE can make possible to produce III-nitride based semiconductor structures of excellent crystal quality capable of high light emission efficiency, but Mg atoms (as p-dopant) are passivated by hydrogen such that as-grown structures have a high resistivity (in the p-doped layer). The hydrogen can be removed by annealing above about 400° C., resulting in a good conductivity. But due to relatively low hole mobility in an (activated) Mg-doped first layer, lateral current injection is problematic (because of still relatively high resistivity), and producing ohmic contacts of low resistance on the first layer is problematic, too. The deposition of the transparent layer 5 (TCN layer) can contribute to achieving a good lateral current spreading and low-resistance ohmic contacts.

[0210] Provision of current aperture stops, e.g., as described, can provide the effects as described. It can be produced in a way that it is epitaxial with the first layer, cf. above.

[0211] With or without current aperture stop 6, the transparent (TCN) layer can be grown without exposing the first layer 1 to hydrogen, e.g. when using MBE. Epitaxial deposition of the transparent layer 5 (on the first layer 1) can result in low resistances at the interface to the first layer and to low resistivity in the transparent layer 5. And in case a current aperture stop 6 is present, epitaxial deposition of the transparent layer (on the current aperture stop) can result in low resistances at the interface to the current aperture stop 6.

[0212] And the interfaces of the transparent layer to the first layer and to the current aperture stop can both be planar and lie both in one and the same (lateral) plane.

[0213] High doping levels of the transparent layer, e.g., with Si as dopant, can result in excellent electrical properties.

[0214] The foregoing description and accompanying drawings illustrate the principles, preferred embodiments and modes of operation of the invention. However, the invention should not be construed as being limited to the particular embodiments discussed above. Additional variations of the embodiments discussed above will be appreciated by those skilled in the art.

[0215] Therefore, the above-described embodiments should be regarded as illustrative rather than restrictive. Accordingly, it should be appreciated that variations to those embodiments can be made by those skilled in the art without departing from the scope of the invention as defined by the following claims.