TW202445884A - Optoelectronic device and method for processing the same - Google Patents

Abstract

The invention concerns an optoelectronic device, comprising an epitaxially grown layer stack with a first charge carrier transportation layer of a first doping type, a second charge carrier transportation layer of a second doping type and an active layer stacked between the first charge carrier transportation layer and the second charge carrier transportation layer. An emission surface is arranged on top of the second charge carrier transportation layer. The active layer comprises a first region centrally located beneath the emission surface and a second region surrounding the first region. The first region of the active layer comprises at least one first portion that is configured to emit light and at least one second portion that is configured not to emit light.

Description

Optoelectronic device and method for preparing the same

[Related applications]

This application claims priority from international patent application PCT/EP2023/052084, filed on January 27, 2023, the entire contents of which are hereby incorporated by reference.

The present invention relates to an optoelectronic device and a method for preparing the optoelectronic device.

For several display applications, the size of the individual optoelectronic devices in each pixel is shrinking. A decrease in the fill factor of a pixel (the ratio between the area of the optoelectronic device and the area of the pixel) from 1:10 to 1:100 is common in some display applications. Similarly, in some applications, the size of the display itself is getting smaller, while the optical system is used to guide the light. Typical applications for the former are various larger displays or panels, while applications for the latter include AR or VR displays.

These displays use different optoelectronic devices to represent RGB colors, each with various implementations. However, certain requirements are placed on the emission behavior for implementation purposes (such as when using a converter), or to provide a certain level of viewing experience.

This time, a constant color impression (color variation with angle) should be obtained in certain applications, where a user can view the display from slightly different angles, or multiple users can view the screen from different angles at the same time. Another aspect is the overall brightness, which should be constant within a given and expected angle range.

Optoelectronic devices, also called µ-LEDs for their size less than 50µm and as small as 2µm to 4µm, can contain their own radiation profile, which depends on their respective material system, polarization, size of the active area, size and geometry of the emitting surface and geometry of the device body. Changes in the latter two usually result in a loss of light extraction efficiency.

This problem can be solved by structuring the emitting surface; this is still often done because it has been found that without surface structuring the extraction efficiency of µ-LEDs at low current levels is significantly reduced.

Conventional techniques for adjusting the radiation characteristics of optoelectronic devices are usually based on modifications of the device body and package design. Thus, examples can be found, for example, in US 10.367.122 B2 and DE 10 2020 113716, the latter using additional optical elements.

While these techniques generally work well, they are not applicable in all cases and do not improve the low light efficiency drop of µ-LEDs. Therefore, it remains desirable to improve the control of the radiation characteristics of optoelectronic devices, and in particular µ-LEDs, without changing the geometry of the device body.

This and other objects are met by the subject matter of the independent claims. The features and other aspects of the proposed concept are outlined in the dependent claims.

The inventors propose to modify the radiation pattern of a photoelectric device by selectively deactivating specific areas of the active layer to obtain adjustment of the far-field radiation and light distribution of the entire emitting surface. The expression "deactivation" in this application includes any modification of a specific area or part of the active layer so that in normal operation the light emitted by this part is substantially less than another part of the active layer that has not been modified. In the sense of this application, reduced light emission is considered to be a reduction of 10% or more, and particularly more than 50%, and more particularly more than 70% compared to the unmodified part.

This modification can be achieved by various means, but may often require a structured mask layer to selectively modify the region within the active layer. In this regard, the inventors note that Zn diffusion has been used primarily to prevent surface recombination at the outer edges of μ-LEDs based on ternary or quaternary phosphide material systems (e.g., InGaAlP). The purpose of this dopant diffusion is to avoid non-radiative efficiency losses, a process known as quantum well intermixing.

Among other aspects, the inventors now propose to reuse the diffusion of dopants into the parts of the active layer that should be selectively deactivated, so that light emission occurs mainly in the non-doped regions. Although such partial modification can also be achieved by other means, the diffusion of dopants provides a local quantum well intermixing part of the active layer, which produces a local change in the energy gap, which to a certain extent causes the electron carriers injected into the active layer to fall into a lower energy valley. Therefore, the regions in those parts of the active layer where the energy gap increases will show reduced light emission, hence the term deactivation.

With appropriate simulation and process control, one can tune the radiation pattern to reflect the desired far-field distribution of the optoelectronic device. Other process approaches (e.g., conventional quantum well intermixing or terrace trimming with subsequent regrowth) can also be maintained and implemented. Thus, the resulting optoelectronic device combines the advantages of increased quantum efficiency due to the reduction of available non-radiating recombination centers at the edges of the sloped sidewalls (which are blocked by electrostatic fields from the regrowth layer or QWI edge region) with the advantages of optimized radiation pattern due to selective tuning of radiating and non-radiating regions within the active layer.

The proposed principle is not limited to specific material systems such as the above-mentioned phosphide systems, but can also be implemented in nitride material systems such as InGaN, InAlGaN or GaN. Similarly, other techniques including annealing steps, the use of Mo:SiO2 coatings and "repeated rapid thermal annealing" to produce regions with higher band gaps are also possible. In alternative embodiments, one can induce defects in the areas that should be deactivated, thereby leading to a higher defect density in those parts of the active layer. However, defects generally cause non-radiative recombination, which may reduce the overall quantum efficiency because charge carriers still recombine in those deactivated parts. Therefore, combination with other techniques may be appropriate. Furthermore, one can imagine performing a selective etching process in those parts of the active layer or its sublayers to define the non-radiative parts.

In some embodiments, the present inventors provide an optoelectronic device comprising an epitaxially grown layer stack having a first charge carrier transport layer of a first doping type, a second charge carrier transport layer of a second doping type, and an active layer stacked between the first charge carrier transport layer and the second charge carrier transport layer. The layer stack can be epitaxially grown on some carrier substrates.

An emission surface of the optoelectronic device can be defined above the second charge carrier transport layer. The active layer according to the proposed principle comprises a first region centrally located below the emission surface and a second region surrounding the first region. Depending on the respective implementation, the emission surface can be smaller than the first region, but can also be larger than the respective first region. In the latter case, parts of the second region can be located below the emission surface.

Herein, an "emitting surface" is generally considered to be one of the major surfaces of an optoelectronic device through which light is emitted during operation of the device. In some cases, the light emitting surface is one of the top surfaces of the device, and portions of the top surface are covered by output coupling structures, transparent contact layers, etc.

According to the proposed principle, the first region of the active layer includes at least one first portion configured to emit light and at least one second portion configured to not emit light. The configuration of the emitting portion and the not emitting or reduced emitting portion of the first region produces a radiation pattern of the optoelectronic device with a specific far-field distribution. The radiation pattern and the associated far-field distribution can be adjusted by appropriately selecting and forming the respective first portion and the second portion.

Thus, different light distribution properties can be obtained and thus the light emission of the optoelectronic device can be adjusted to meet individual requirements. Especially for µ-LEDs, where mechanical components like optical elements are difficult to realize, the possibility to adjust the light emission by modifying the active layer offers several advantages. Producing optoelectronic devices using optical elements and the device itself becomes cheaper and easier to manufacture.

It has been found that, in order to tune the far field properties, a central portion configured for light emission may be provided in the active layer, and one or more portions around the central portion may include different emission patterns. In some cases, for example, at least one second portion surrounds the at least one first portion. The at least one first portion may be centered in the center of the active layer (or first region) and surrounded by a second portion. The first and second portions may alternate, and the thickness (i.e., the area of the respective regions when viewed from above) may be different.

In some cases, for example, when viewed from above, the at least one first portion comprises a circular or elliptical shape. In some other cases, when viewed from above, the at least one first portion comprises more than one ring, and the more than one ring is surrounded by a respective annular second portion. In some cases, the annular first and second portions can also alternate, thereby forming alternating circular radiating regions and non-radiating regions in the active layer.

Some aspects relate to the preparation of the second portions, i.e., the modification of the active layer material in the second portion. In some cases, the at least one second portion comprises a higher defect density than the at least one first portion. The higher defect density can be achieved by introducing dopants or other atoms in the second portion of the active layer. These defects act as non-radiative recombination centers or scattering centers, so that significantly reduced emission occurs in the second portion. In some cases, the defect density of the at least one second portion is one order of magnitude higher than the defect density of the at least one first portion, in particular at least two orders of magnitude or between 1.5 and 3 orders of magnitude. In some cases, the defect density of the at least one first portion can be within the measurement limit or slightly greater than the measurement limit. Therefore, the defect density of the at least a second portion may be in a range between 1 order of magnitude and 2.5 orders of magnitude higher than the measurement limit.

Alternatively or additionally, dopants may be introduced into the respective second portions in the active layer, the dopants causing local quantum well intermixing or at least local energy gap distortion, resulting in a higher energy gap at the location of the second portion. In some cases, the dopants trigger defect movement, resulting in the exchange of lattice sites of Al and Ga in the high/low aluminum-rich layers. The higher energy gap generates a local electrostatic field to repel charge carriers from those regions and to capture or retain them to a certain extent in the first portions. In some cases, the dopant concentration may be in the range between 5e17 1/cm ³ and 7e18 1/cm ^3. There may be a gradient in the growth direction. For example, in some cases, the dopant concentration may vary between several values with a maximum value in the at least one second portion and a corresponding minimum value in the at least one first region. The dopant concentration may be different and may be controlled by various process steps during the fabrication of the device. Various different techniques may be combined to modify the active layer and/or internal bandgap structure.

The dopant itself may include one of Mg or Zn. Diffusion of the dopant into the second portion of the active layer may be achieved by a variety of means including but not limited to vapor phase epitaxy, CVD, MBE, ALD with subsequent annealing, and the like.

In some cases, the second region of the active layer includes a quantum well intermixing region. As described above, the second region surrounds the first region. By creating a quantum well intermixing region in the second region, an electrical barrier can be created that primarily traps charge carriers in the first region. This is particularly useful if the optoelectronic device element is terraced with an overall length or diameter that is smaller than the diffusion length of the charge carriers. Material systems with large diffusion lengths are based on phosphides, such as InGaP, InGaAlP, GaP, etc. However, due to the possibility of far-field tuning and light shaping, nitride-based semiconductor materials (e.g., GaN, InGaN, AlGaN, InGaAlN, etc.) can also use QWI in the second region.

In some cases, the epitaxial growth layer stack includes a terrace etch structure with tilted side surfaces. Terrace etching generally separates active layers of adjacent optoelectronic devices from each other so that those devices can be separated or individually positioned and supplied during preparation. In some cases, the outer edge of the second region constitutes a portion of the tilted side surfaces. Thus, the second region can be quantum well intermixed so that the outer edge of the second region has a higher bandgap than the region of the active layer near the center.

The inclined sidewalls of the terrace etched optoelectronic device can be treated to further reduce undesirable non-radiative recombination centers. In some cases, a dielectric layer is disposed on the inclined side of the surface. The dielectric layer may include Al2O3, SiO2, etc. Alternatively, the surface of the inclined side is covered by a regrown semiconductor layer. The material of the regrown layer has a larger energy gap than the material of the active layer. In some cases, the material of the regrown layer contains a higher Al content than the material of the active layer.

The top surface of the second charge carrier transport layer typically includes an emitting region. In some cases, for example, as a volume emitter, the side surfaces of the optoelectronic device are not covered with a reflective material. For a top emitter, the side surfaces may be covered with a metal or a reflective layer. In some cases, the emitting surface (i.e., the region of the top surface that is typically configured to emit light) has approximately the same size as the top surface of the second charge carrier transport layer or a contact layer disposed on the second charge carrier transport layer. In some other cases, the emitting surface may be smaller than the actual top surface.

In some embodiments, the optoelectronic device includes a top contact layer having a transparent conductive layer disposed on or adjacent to the emitting surface and electrically connected to the second charge transport layer. In this regard, the optoelectronic device may include a contact layer comprising a highly doped semiconductor material.

In some other cases, the active layer of the layer stack includes a single quantum well or a multiple quantum well structure having a plurality of alternating barrier layers and quantum well layers, wherein the barrier layers optionally have a higher Al content than the quantum well layers. To modify the layers according to the proposed principles, some or all of the quantum well layers and the barrier layers or a combination thereof may be modified individually.

A plurality of optoelectronic devices according to the proposed principles can be arranged in rows and columns to form part of a display. Individual optoelectronic devices can have their radiation pattern adjusted individually, but specific optoelectronic devices can be combined together and adjusted in a similar or at least specific manner to provide a predetermined and desired radiation pattern.

In some embodiments, three optoelectronic devices configured to emit light of different wavelengths are combined together as sub-pixels, each sub-pixel providing a specific radiation pattern. Therefore, their active layers are modified accordingly. However, the modifications of the optoelectronic devices associated with the same pixel may vary due to different material compositions, different light emission, etc. In some cases, the display thus includes a plurality of optoelectronic devices according to the proposed principles and one or more optical structures arranged above the emission surface of the pixel and/or the plurality of optoelectronic devices.

Other aspects relate to a method for preparing an optoelectronic device with predetermined light distribution characteristics. In order to prepare an optoelectronic device according to the proposed principle, it is necessary to obtain the desired shape and configuration of the activated and deactivated parts of the active layer according to the desired radiation field pattern.

Therefore, in order to find the mask layers for preparing the first part and the second part, simulations of the desired results may be appropriate, if not trials and tests. Various simulation possibilities are suitable. For example, a uniform light distribution over the entire emitting surface and the active layer can be assumed. In order to correctly simulate the incoherent light source of the optoelectronic device, a plurality of dipoles are virtually densely distributed on the active surface and simulated individually. Such a simulation provides information about the light extraction and the far field for each dipole, which corresponds to a small area of the active layer.

By selectively switching virtual dipoles on or off, the radiation pattern of the emitting surface and the active layer can be generated. Post-processing of the data will therefore provide the dipoles that must be excited or activated in the simulation to achieve the desired radiation pattern. A binary mask is thus obtained, which identifies a first part and a second part of the first region of the epitaxially grown active layer, where the first part should emit light and the second part should not.

In some cases, the inventors propose to reuse Zn diffusion into the active layer, which is used in phosphide-based semiconductor materials to induce potential barriers through quantum well intermixing.

A similar process is applied to construct the mask layer of the active and deactivated parts of the first region. The same mask layer can also be used to construct the quantum well intermixing in the outer edge of the active layer. In this regard, the proposed principle can be used to guide light around a known absorber located above the emitting surface to improve the overall light extraction efficiency.

In some aspects, the inventors now propose a method for preparing an optoelectronic device, comprising the steps of: epitaxially growing a first charge carrier transport layer of a first doping type on a carrier substrate; and epitaxially growing an active layer on the first charge carrier transport layer. The active layer includes a first region and a second region surrounding the first region. The active layer is also configured to emit light in response to a current passing through the active layer.

According to the proposed principle, a first structured mask layer is provided. The structured mask layer covers at least the first region of the active layer. More specifically, the first mask layer is structured to cover a first portion of the active layer and include at least one groove exposing a top surface of a second portion of the first region.

In a subsequent step, the second portion of the first region is modified or altered such that the second portion is configured to emit no light or to emit significantly reduced light compared to the first portion.

Furthermore, the method also includes the step of epitaxially growing a second charge carrier transport layer on the active layer. At this point, the step of epitaxially growing the second charge carrier transport layer on the active layer can be performed before the step of providing the structured mask layer, but can also be performed after.

When a current is injected into the active layer, the modification of the active layer or parts thereof also changes the radiation pattern. Thus, appropriate modification will result in the desired radiation pattern. The fabrication of optoelectronic devices according to the proposed principle is not limited to phosphide material systems, but can be generally applied to nitride, arsenide and phosphide systems as well as to more exotic direct semiconductor materials.

The step of changing or modifying the second portion can be performed in different ways, one of which includes the step of adding a dopant to the second portion, wherein the dopant concentration is in a range between 5e17 1/cm ³ and 7e18 1/cm ^3. In some cases, the dopant can be first deposited on the exposed surface above the second portion and then diffused therein. The steps of deposition and diffusion can be performed with different process parameters, such as but not limited to different temperatures. Furthermore, an annealing step can be performed after the dopant is diffused in the active layer.

In some cases, the first structured mask layer exposes a top surface portion of the second region of the active layer. Dopants, in particular Zn or Mg, then diffuse into the second portion of the first region and into the second region of the active layer. As a result, local quantum well intermixing is induced in the second portion of the first region and in the second region of the active layer. In this respect, the expression local quantum well intermixing means more or less well-defined boundaries of dopant material in the active layer, which result in electrical potential barriers.

Dopant diffusion creates a local potential barrier within the active layer that prevents charge carriers from recombination in the portion where the energy gap is increased. The potential barrier acts like a trap for charge carriers, concentrating them in the potential "valleys" in between. In some other cases, the defect density within the active layer is locally increased so that charge carriers are scattered at those local high defect densities. Some charge carriers may recombine in areas of increased defect density, but they are non-radiative. The defect density can be in a range of at least one order of magnitude greater than the measurement limit of the defect density. In some other cases, the exposed surfaces of the second portions can be selectively etched to create perturbations in the energy gap and potential structure.

Various approaches can be combined to achieve the desired bandgap perturbation and thus a pattern of regions in the active layer that emit light and regions that do not emit light during device operation.

In some embodiments, the material of the first structured mask layer covers a first portion centrally located in the first region of the active layer. At least one groove surrounds the material. In some embodiments, the shape of the first portion covered by the material of the structured mask layer can include a substantially circular or elliptical form when viewed from above. Alternatively or additionally, the shape of the first portion covered by the material of the structured mask layer includes more than one ring when viewed from above. The more than one ring of material of the structured mask layer is surrounded by a respective annular groove. Therefore, in some cases, the structured mask layer includes alternating circular portions of material and grooves, and the width of the material and/or grooves can be different.

In some other cases, the first mask for modifying the first and second portions of the first region is separate from the mask for quantum well intermixing in the second region of the active layer. Therefore, in some cases a second structured mask layer is provided to cover the first region and expose the top surface portion of the second region of the active layer. Then, a dopant (particularly one of Zn and Mg) is diffused into the second region, thereby causing quantum well intermixing in the second region of the active layer. In some cases, the first mask layer and the second mask layer are deposited simultaneously.

Some aspects relate to the preparation of the optoelectronic device and the preparation of the layer stack. In some cases, a structured etch mask is set on the second charge carrier transport layer. Then, a terrace etching is performed to etch at least the second charge carrier transport layer and the active layer. The resulting structure includes a tilted surface. The outer edge of the second region constitutes a portion of the tilted side surface. The above-mentioned terrace etching can be performed before setting the first structured mask layer.

The inclined surfaces may be further treated to reduce hanging bonds, surface defects, etc. For example, a dielectric layer may be deposited on the inclined side surfaces. As an alternative, in some cases, a layer of semiconductor material is regrown on the inclined side surfaces, wherein the energy gap of the semiconductor material layer is greater than the energy gap of the active layer. The use of a regrown layer can reduce the overall size of the second region in favor of the first region until the second region of the active layer is essentially the interface of the regrown layer. Thus, the total area or size of the second region can be quite small and limited to a few layers adjacent to the edge of the active layer.

In some embodiments, a top contact layer is formed having a transparent conductive layer disposed on or adjacent to an emitting surface and electrically connected to the second charge carrier transporting layer.

Some aspects are related to the preparation of the layer stack and the active layer. In some cases, the step of epitaxially growing the active layer includes the step of epitaxially growing a plurality of alternating quantum well layers and barrier layers. The barrier layers may contain Al. The Al content in the barrier layers is higher than the Al content in the quantum well layers.

As described above, the first structured mask can be simulated according to the desired radiation pattern. In this case, the step of providing the first structured mask layer includes structuring a mask layer material in response to simulation results of a remote field evaluating the desired distribution characteristics of a plurality of virtual dipoles, each virtual dipole representing a small area of the active layer.

The following embodiments and examples disclose various aspects and combinations thereof according to the proposed principles. The embodiments and examples are not always drawn to scale. Similarly, different elements may be shown enlarged or reduced in size to emphasize individual aspects. It goes without saying that individual aspects of the embodiments and examples shown in the drawings can be combined with each other without additional effort and without contradicting the principles according to the invention. Some aspects show regular structures or forms. It should be noted that slight differences and deviations from the ideal form may occur in practice, but this does not contradict the concept of the invention.

In addition, the various figures and aspects are not necessarily shown in the correct size, and the proportions between individual elements are not necessarily correct. Some aspects are highlighted by enlarging the display. However, terms such as "above", "directly above", "below", "directly below", "larger", "smaller", etc. are correct representations of the elements in the figures. Therefore, the relationship between the elements can be inferred based on the figures.

FIG1 illustrates a conventional optoelectronic device, but the proposed principle is also applicable to a complete array of multiple µ-LEDs arranged in rows and columns. The optoelectronic device comprises a terraced etched semiconductor body having an inclined surface 30, a top surface and a bottom surface. The bottom surface forms a bottom contact with a highly p-doped contact layer 11, on which a charge carrier transport layer 12 is deposited.

The multi-quantum well structure 20 is arranged on the first charge carrier transport layer 12 and includes a plurality of alternating barrier layers and quantum well layers (not shown in detail here). The top side of the optoelectronic device includes a second n-doped charge carrier transport layer 14, followed by a current distribution layer 15 on top of the second charge carrier transport layer 14. A transparent contact layer 16 (e.g., a conductive metal oxide) is deposited on top of the semiconductor material of the current distribution layer 15. The top surface of the contact layer 16 forms the top emitting surface 19 of the optoelectronic device, but light is also emitted from the side.

The sloping sidewalls are covered with a respective insulating material 31 extending from the bottom side of the device (i.e. the bottom contact layer 11) all the way to the top surface of the contact layer 16. The material of the layer 31 on the sloping sidewalls 30 comprises an insulating transparent material which also extends over part of the top surface of the contact layer 16. As a result, the main emitting surface 19 can be slightly smaller than the corresponding area of the top surface of the underlying charge distribution layer 15. However, with a suitable thickness the passivation layer 31 can act as an AR coating and even increase the outcoupling of light.

In operation of the optoelectronic device according to FIG1 , charge carriers are induced from the bottom side of the contact layer 11 and from the top side via the transparent conductive metal oxide of layer 16. The resulting optoelectronic device is thus similar to a so-called vertical µ-LED, which is characterized in that charge carriers are injected from two opposite sides. Alternatively, the optoelectronic device can be implemented as a horizontal µ-LED, in which the top contact area is located on the same side of the optoelectronic device.

In operation, the induced individual charge carriers will diffuse through the doped charge carrier transport layers 12 and 14, respectively, and recombine in the active layer 20 and its multiple quantum well structure. Due to the small size of the optoelectronic device, measures such as intermixing or re-growth of quantum wells can be taken to increase the quantum efficiency of the optoelectronic device by providing an electric potential barrier at the side of the active layer 20 (i.e., the interface between layer 31 and the active layer 20).

When viewed from above, the optoelectronic device according to Fig. 1 resembles a substantially circular shape, wherein the main emitting surface 19 is present in the center of the optoelectronic device with a diameter of approximately 5 μm. For example, the radiation pattern of the optoelectronic device will therefore be mainly isotropic, providing a substantially equal light distribution.

The far field produced by such a structure is shown in Figure 2, and its Lambertionality factor LF is 1.24. As shown in the figure, the light distribution of the far field provides two maxima at an angle of about 20°, and the light radiation decreases by about 10% within an angle range of about 0° around the center. Therefore, when viewing directly from the front, people may feel slightly smaller brightness compared to an angle of about 20°. When using optical components, taking this phenomenon into consideration will cause the optical components to be more expensive.

The principles presented herein now utilize various techniques to improve the radiation pattern of optoelectronic devices and, more specifically, to tune the far field of an optoelectronic device by locally modifying and changing the active area of the device to emit light with a desired radiation pattern. Figures 3A and 3B illustrate two examples of such modifications that can adjust its radiation pattern.

FIG3A illustrates an optoelectronic device having a layer stack including a first charge carrier transport layer 12, a second charge carrier transport layer 14, and an active layer 20 disposed between the two charge carrier transport layers. More specifically, the first charge carrier transport layer 12 is deposited and disposed on the contact and current distribution layer 11. The current distribution layer 11 is highly doped to inject charge carriers into the charge carrier transport layer 12. The charge carrier transport layers 12 and 14 may have different dopant concentrations along their thickness. Any distribution of dopants in the charge carrier transport layers 12 and 14 is beneficial to the distribution and injection of charge carriers into the active layer 20.

Furthermore, a small undoped region of the multi-quantum well structure may be provided directly adjacent to the active layer 20, and in particular between the charge carrier transport layer 12 and the active layer 20, to prevent undesired diffusion of n dopants from the first charge carrier transport layer 12 into the active layer 20. A superlattice structure may be deposited on the n-side of the device.

A further charge distribution layer 15 is deposited on top of the second charge carrier transport layer 14. A transparent metal conductive oxide layer 16 is deposited on the layer 15, which acts as a contact layer. The top surface of the layer 16 also represents the main emitting surface 19 of the optoelectronic device according to the proposed principle.

The optoelectronic device in this embodiment includes a terrace-type etched sidewall, which is subsequently covered by a dielectric layer 31. In a top view, the optoelectronic device has a circular or polygonal shape, for example, a hexagonal or octagonal shape. The interface between the dielectric layer 31 and the active layer 20 typically includes a plurality of dangling bonds and other non-radiative recombination centers, which typically reduce the overall quantum efficiency of the optoelectronic device. This effect becomes greater as the device size decreases and in certain material systems (primarily due to the diffusion length of charge carriers). For certain material systems, the diffusion length of charge carriers may be within the range of the corresponding size of the optoelectronic device itself, or even longer than the corresponding size of the optoelectronic device itself. As a result, charge carriers may diffuse into the outer edge of the active layer 20 during operation, be subsequently trapped at surface defects at its interface, and then recombine non-radiatively, thereby reducing the overall quantum efficiency.

In order to increase its quantum efficiency, quantum well intermixing is utilized, and additional p-doped materials diffuse from the second charge carrier transporting layer 14 to the quantum layer 20 at the side. The diffused doping material (e.g., Zn or Mg) causes quantum well intermixing in the outer and side portions of the active layer 20, thereby increasing the energy gap at the edge and interface between the active layer 20 and the dielectric layer 31 in the active layer 20. This quantum well intermixing generates an electric potential barrier to prevent charge carriers from diffusing to the side.

According to the proposed principle, the dopants used to induce quantum well intermixing at the sides are further used to locally adjust the energy gap at certain regions within the central region of the active layer 20. More specifically, the active layer 20 is actually divided into a first central region 21 surrounded by a second region 22, wherein the second region 22 also includes a quantum well intermixing region between the interface of the active layer 20 and the dielectric layer 31. However, the central region 21 also includes local portions 41 and 41' where quantum well intermixing caused by additional dopants occurs. The adjacent portion 40 of the quantum well intermixing in the second region 22 may or may not completely extend laterally to the etched platform 30.

Thus, a local potential barrier is generated at the location of the first portions 41, 41' in the central region 21 to capture charge carriers in the undoped region 42 therebetween. When viewed from above, the active layer 20 comprises a plurality of second portions 42 in the first region, which lead to light emission during operation of the device, and a plurality of first portions 41 in the first region, which lead to significantly reduced light emission or no light emission at all during operation of the device.

FIG. 3B illustrates another embodiment of an optoelectronic device according to the proposed principle. This optoelectronic device comprises a terrace-type etched layer stack having two charge carrier transport layers 12 and 14 and an active layer 20 therebetween. The bottom surface of the layer stack is covered by a contact and charge distribution layer 11. The top surface forming the main emitting surface 19 comprises the material of the transparent metal contact layer 16. The side 30 of the optoelectronic device is inclined and covered by a passivation layer 31. As an alternative to the quantum well intermixing region 40, the regrown material 32 here prevents recombination of non-radiating surfaces in a similar manner. Further required preparation steps are not shown.

The active layer 20 of the optoelectronic device comprises a central first region 21 surrounded by a second region 22, which forms the inclined side and the interface with the regrown material 32. The first region 21 of the active layer 20 comprises a first portion 42 and a second portion 41 arranged therebetween (or vice versa). The first portions 42 are configured for light emission of the optoelectronic device, while the second portions 41 exhibit significantly reduced light emission in operation. This is achieved by modifying the energy gap structure within those portions 41, in the present case by introducing defects into those portions. These defects cause the charge carriers to recombine or simply scatter back into the first portion 42 without light emission.

The second portion can be modified in different ways, some of which are presented in the present application, but are not limited thereto. In addition to the diffusion of dopants into the second portion as already described, recesses thereof can also be formed, filling these recesses with insulating materials or with higher energy gap materials. In both cases, an electric potential barrier is created, which traps charge carriers in the first portion 42 therebetween in a repulsive manner. Alternatively, or in addition, the defect density can be increased so that charge carriers scatter or recombine non-radiatively at those defects.

FIG. 4A illustrates three possible configurations of emission dipoles/emitting active regions forming a pattern of locally generated light M1, M2 and M3 in a top view of an active layer and its first region. For simplicity, the first region of the active layer has a circular shape. Each first region comprises a plurality of first portions 42 and a plurality of second portions 41, both arranged as a ring around a center. In three possible embodiments, the central portion of the first region is arranged to emit light, thus similar to the first portion 42.

The radiation pattern shown in FIG. 4A can be simulated. For this purpose, the entire area of the first region is divided into a plurality of rings, each of which has the same thickness. The thickness may depend on the simulation resolution, but should be determined by practical considerations, namely that the individual rings should be thick enough to be fabricated in the manufacturing process of such a device. It is now possible to simulate deactivated rings (rings that do not emit light) and activated rings (rings of the pattern that emit light) and obtain the far-field characteristics therefrom.

FIG4B provides the far-field characteristics of the radiation patterns M1, M2 and M3. It can be observed that all three radiation patterns modify the far field shown in FIG2 by adding an additional maximum at a small angle of θ = 0°. The overall Lambertionality factor LF drops to about 1.1, which corresponds to a reduction of about 13%. The curves for the various patterns are very similar, indicating that for a desired or given far field, a plurality of different patterns are generally possible.

FIG. 4C illustrates three further possible radiation patterns, where the Lambertionality factor LF is equal to 1.1. For values below 1.15, it has been found that in the present design with a plurality of ring segments (23 in total), approximately 50% of those ring segments have to be switched off. Of course, even for the same Lambertionality factor LF, having ring segments of different thicknesses or providing different designs will lead to different solutions. Furthermore, the radiation pattern and in particular the position and number of active ring segments strongly depend on the overall chip design. Other chip designs or different settings for simulating the radiation pattern (e.g., a hexagonal pattern instead of a circular ring segment pattern) will lead to different results when different parts of the active layer are deactivated and activated, respectively.

In the exemplary pattern M4, almost all the rings (including the center) have been deactivated, leaving only two rings. Although such a radiation pattern produces a far field similar to that shown in FIG. 4B, the overall brightness is reduced. Therefore, other solutions (e.g., patterns M5 or M6 with a larger total luminous area) are preferred.

In the configuration M5, the central region is also deactivated, while the configuration M6 demonstrates that a Lambertionality factor of 1.1 can be achieved with the central region switched on. Here, as in some of the other configurations M2 and M3, the outer ring is activated, resulting in a relatively large area of the active layer being configured for light emission. Thus, the configurations M2, M3 and M6 provide high brightness while maintaining a low Lambertionality factor LF.

5A to 5E illustrate a first embodiment of fabricating an optoelectronic device according to the proposed principle.

The optoelectronic device is essentially grown as an epitaxial layer stack on a carrier substrate 18, which has been appropriately prepared for epitaxial growth. Typical substrates on which epitaxial growth can be performed include sapphire or Al2O3 substrates, GaN substrates, GaAs substrates, etc., depending in part on the substrate system used. The carrier substrate can have one or more buffer layers grown on it, which are used to provide a smooth and defect-free surface or to adjust the lattice constant to be consistent with the substrate system subsequently used.

A first highly doped contact layer 11 is deposited on the top surface of the prepared substrate 18. This layer or subsequent layers are epitaxially deposited using various growth and deposition techniques, either individually or in combination. These techniques include, but are not limited to, chemical vapor deposition (CVD), physical vapor deposition (PVD), atomic layer deposition (ALD), etc.

The contact layer 11 is formed as an additional buffer layer, which is subsequently also used to provide a bottom contact of the optoelectronic device according to the proposed principle. On top of the contact layer 11, an n-doped charge carrier transport layer 12 is deposited.

In this regard, the term "charge carrier transport layer" may include one or more sublayers having various material compositions (e.g., ternary or quaternary compositions with different In, Ga or Al contents, different or varying dopant concentrations, etc.). One skilled in the art will recognize various possibilities for realizing such a charge carrier transport layer. The charge carrier transport layer 12 is configured to transport and inject charge carriers into the active region 20.

The active region 20 is formed as a multiple quantum well structure having a plurality of barrier layers 25 and quantum well layers 24 stacked on top of each other in an alternating order. A plurality of barrier layers and quantum well layers 24 are described herein, but more or fewer of those layers may be used. The multiple quantum well structure in the present embodiment includes a quaternary system like InGaAlN or InGaAlP. Different energy gaps between the individual barrier layers 25 and quantum well layers 24 are achieved by using different aluminum contents in the individual layers. More specifically, a higher aluminum content generally results in a higher energy gap. Adjusting the In content causes some strain in the active layer. This can change the energy gap, thereby producing different colors of light depending on the confinement changes caused by adjusting the QW thickness. Finally, a p-doped charge carrier transport layer 14 is deposited on the multi-quantum well structure 20.

After epitaxially growing the stack as shown in FIG5A, a mask layer material 60 is deposited on top of the second charge carrier transport layer 14. The mask layer 60 serves as a hard mask for subsequent diffusion processes for intermixing of quantum wells and selective deactivation or activation of portions within the active layer 20. The mask layer material 60 is then structured to form a plurality of recesses 61 and 62, respectively.

The grooves 61 and 62 expose a portion of the top surface of the second charge carrier transport layer 14. In addition, the groove 62 also defines a second region within the active layer 20 of the optoelectronic device, the second region surrounding the first region having the first portion and the second portion. The latter is correspondingly defined by the groove 61 and the masking material. The resulting structure is illustrated in FIG. 5B. In the case of a subsequent terrace etching process after quantum well intermixing has been performed, the groove 62 can also define the boundaries of the optoelectronic device.

5C illustrates the subsequent steps. First, additional p-doping, such as Zn, is deposited on the exposed surface of the second charge carrier transport layer 14. This deposition can be achieved by using a corresponding precursor for depositing Zn on the top surface of the layer 14 at a first temperature. The first temperature is set to avoid significant diffusion of the Zn deposited on the top surface into the second charge carrier transport layer 14. This step allows accurate quantification of the total amount of dopants deposited on each exposed portion. In a subsequent step of this process, its parameters are adjusted, i.e. the temperature is increased to induce diffusion of the dopants on the top surface into the second charge carrier transport layer 14 and the respective barrier layer 24 and quantum well layer 25 of the active layer 20. The diffusion process is controlled by the amount of Zn deposited on the surface, the temperature and the diffusion time. Additional surface treatment or pre-treatment steps can be performed to optimize the quantum well intermixing process.

The process parameters are selected such that quantum well intermixing occurs in the individual portions 40 and 41 of the active layer 20. More specifically, the quantum well intermixing at the portion 40 of the active layer is similar to conventional quantum well intermixing at the edges and boundaries of an optoelectronic device formed in a mesa etching process step.

In contrast, the quantum well intermixed portions 41 represent the first portion within the first region of the active layer 20 that is deactivated and no longer configured to emit light during operation of the device. Portions 42 between the quantum well intermixed portions 41 are not doped and their energy gaps remain substantially unchanged. Therefore, during operation of the device, these portions 42 represent portions of the first region of the active layer 20 that are configured to emit light.

In the design and simulation of the mask layer 60, it must be considered a priori that the dopant may diffuse into the active layer 20 in a non-linear manner under the structured mask layer 60 in order to achieve a properly defined first and second portion within the active layer 20. An optimized diffusion process can completely prevent underdiffusion and allow the dopant to diffuse into the active layer in a substantially linear manner. This will provide a properly defined first and second portion within the active layer 20. The activated portions are those portions in which no additional dopant is diffused, while in the portions containing Zn as a dopant, quantum well intermixing will occur, and are therefore referred to as deactivated portions. When viewed from above, the overall structure resembles the previously simulated and expected radiation field pattern of the optoelectronic device. In addition, the second region 14 also includes a quantum well intermixing region, thereby preventing the charge carriers from reaching the periphery and boundaries of the optoelectronic device.

FIG. 5D illustrates a top view of an optoelectronic device that has been fabricated according to the proposed principles.

In this embodiment, the optoelectronic device is formed with a rectangular shaped second region 40. The optoelectronic device includes terrace etched sidewalls to allow intermixing of quantum wells by creating potential barriers to prevent injected charge carriers from reaching the sides.

Following the edge and the outer region of the optoelectronic device, individual portions 41 and 42 in the first region of the active layer 20 alternate starting from near the outer edge 40, wherein the central portion 42 is undoped and configured for light emission. The thickness of the individual rectangular structures configured for light emission or not is different and has been calculated to obtain the desired radiation pattern.

After removing the mask layer 60 from the second charge carrier transport layer 14, a p-doped current distribution layer 15 is deposited on the top surface, followed by a transparent metal oxide contact layer 16. The resulting structure shown in Figure 5E includes a plurality of quantum well intermixing portions in a central first region 21 of the optoelectronic device. The first region 21 is surrounded by a second region 22, wherein the second region 22 includes a quantum well intermixing region of the active layer 20 adjacent to the side of the device. In operation of the device, a portion 42 in the first region of the active layer 20 is configured to emit light, while the quantum well intermixing portion 41 in the first region will not emit any light at all or at least significantly reduce the emission of light. The resulting field pattern provides a modified far-field distribution of light emitted by the optoelectronic device.

Figures 6A to 6D illustrate another embodiment of preparing an optoelectronic device according to the proposed principle. The layer stack shown in Figure 6A is mainly similar to the previously described layer stack of Figure 5A, having a contact layer 11 deposited on a carrier substrate 18 and a charge carrier transport layer 12 deposited on layer 11. The active layer 20 of Figure 6A also includes a plurality of alternating barrier layers and quantum well layers. The active layer 20 includes InGaAlP as a material system, the barrier layers and quantum well layers having different Al contents and optional In contents, wherein different optional In contents are required to achieve wavelength matching of the active area design. Compared to the previous embodiments, the material of the mask layer 60 is directly deposited on the last sublayer of the active layer 20 and then structured as shown in Figure 6A.

According to the present invention, the portion of the active layer below the surface exposed by the structured mask 60 is modified to prevent light from being emitted therefrom.

For this purpose, as shown in FIG. 6B , exposed portions of the active layer 20, i.e. exposed portions of the barrier layer and the quantum well layer, can be etched. The depth can vary, but is sufficient to significantly reduce the light emitted from those portions. The resulting structure shown in FIG. 6B forms a plurality of recesses in the active layer material; during operation of the device, no recombination of charge carriers occurs at these recesses. Surface recombination due to, for example, etching damage, hanging keys, etc. in layer 20 can also be prevented by appropriate additional means.

Similar to the previous embodiment, the lateral dimensions of the etched portion 62 within the active layer 20 are defined by the mask layer and follow the previously simulated radiation pattern. Subsequently, as shown in Figure 6C, the groove 62 is filled with an insulating layer material 70. The insulating layer material 70 completely fills the groove 62 and also provides a small amount of insulating layer on the top surface of the uppermost sublayer of the active layer 20. In a subsequent step, the insulating layer material above the active source layer 20 is removed without removing the insulating material within the groove 62. The process of filling the groove with material can adopt a re-growth process, but other techniques can also be applied.

In the present embodiment, an insulating layer material is used. However, semiconductor materials with energy gaps exceeding the energy gap of the active layer material can also be used. In both cases, recombination of charge carriers in those portions is reduced and an electrical potential barrier is established to trap charge carriers in the portions 42 between the filled grooves.

Referring now to FIG. 6D , the subsequent steps of the proposed preparation method are illustrated. The material of the second charge carrier transport layer 14 is deposited on the active layer 20 and the grooves filled with the material. In some cases, a small amount of a barrier layer of undoped material may be deposited first to prevent the dopant of the doped layer of the second charge carrier transport layer from diffusing into the active layer or the material filled in the grooves.

Similar to the previous embodiments, a contact or charge carrier distribution layer 15 is deposited thereon. The finished structure comprises an active layer 20 having an unmodified first portion 42 and a second portion 41; in operation of the device, light emission is reduced or absent only at the second portion 41. The lateral dimensions of the various portions produce a specific radiation pattern that will provide a far field optimized to the desired needs.

The optoelectronic device may be further processed, ie, beta-etched, to form a plurality of separate optoelectronic devices.

1: Photoelectric device 11: Contact layer 12,14: Charge carrier transport layer 15: Charge distribution layer 16: Contact layer 18: Carrier substrate 19: Top emitting surface 20: Active layer 21: First region 22: Second region 24: Quantum well layer 25: Barrier layer 30: Inclined sidewall 31: Passivation layer 32: Regrown material 40: Quantum well intermixing region 41,41': Second part 42: First part 60: Mask layer 61: Removed mask layer 62: Groove 70: Insulating layer material

Additional aspects and embodiments according to the proposed principles will become apparent from the various embodiments and examples described in detail in conjunction with the accompanying drawings, in which FIG. 1 shows an embodiment of a conventional optoelectronic device implemented as a µ-LED; FIG. 2 illustrates a simulated far field of the optoelectronic device of FIG. 1 ; FIG. 3A shows a first embodiment of an optoelectronic device according to some aspects of the proposed principles; FIG. 3B illustrates a second embodiment of an optoelectronic device according to some aspects of the proposed principles; FIG. 4A shows different top views of a patterned first region of an active layer of individual optoelectronic devices to outline some aspects of the proposed principles; FIG. 4B illustrates the far field and radiation pattern of the optoelectronic device of FIG. 4A according to some aspects of the proposed principles; FIG. 4C shows different top views of the patterned first region of the active layer of each optoelectronic device to outline some aspects of the proposed principle; FIGS. 5A to 5E illustrate some steps of preparing an optoelectronic device according to some aspects of the proposed principle; FIGS. 6A to 6D show some steps of preparing an optoelectronic device according to some aspects of the proposed principle.

1: Optoelectronic devices

12,14: Charge carrier transport layer

15: Charge distribution layer

16: Contact layer

19: Top emitting surface

20: Active layer

21: First area

22: Second area

24: Quantum Well Layer

25: Barrier layer

30: Leaning sidewalls

31: Passivation layer

40: Quantum well intermixing region

41,41': Part 2

42: Part 1

Claims (25)

A photoelectric device comprises: An epitaxial growth layer stack comprising a first charge transport layer of a first doping type, a second charge transport layer of a second doping type and an active layer stacked between the first charge transport layer and the second charge transport layer; An emitting surface located above the second charge transport layer, wherein the active layer comprises a first region centered below the emitting surface and a second region surrounding the first region; wherein the first region of the active layer comprises at least a first portion configured to emit light and at least a second portion configured not to emit light. The optoelectronic device of claim 1, wherein the at least one second portion surrounds the at least one first portion; and/or wherein the at least one first portion is centered in the active layer. An optoelectronic device as claimed in claim 1, wherein the at least one second portion comprises one of the following: a defect density higher than the defect density of the at least one first portion; a defect density in a range of at least one order of magnitude greater than a measurement limit, in particular obtained by one of positron annihilation spectroscopy or SIMS, and in particular between 1.5 orders of magnitude and 3 orders of magnitude greater than the measurement limit; a dopant concentration in a range between 5e17 1/cm ³ and 8e18 1/cm ³ ; a semiconductor material having a band gap higher than the band gap of the active layer. An optoelectronic device as claimed in claim 1, wherein the second region comprises a quantum well intermixing region, wherein optionally, the dopant introduced into the second region comprises one of Zn and Mg. The optoelectronic device of claim 1, wherein when viewed from the direction of the emitting surface, the at least one first portion has a circular or elliptical shape; and/or when viewed from the direction of the emitting surface, the at least one first portion includes one or more rings, each of which is surrounded by a respective annular second portion. The optoelectronic device of claim 1, wherein the epitaxial growth layer stack includes a terrace-type etching structure having inclined side surfaces, wherein an outer edge of the second region constitutes a part of the inclined side surfaces; and/or a dielectric layer is disposed on the inclined side surfaces; and/or a re-growth layer is disposed on the inclined side surfaces. The optoelectronic device of claim 1, wherein the emitting surface above the second charge carrier transport layer is larger than the top surface of the second charge carrier transport layer. The optoelectronic device of claim 1 further comprises a top contact layer having a transparent conductive layer disposed on or adjacent to the emitting surface and electrically connected to the second charge carrier transport layer. The optoelectronic device of claim 1, wherein the active layer comprises a multi-quantum well structure having a plurality of alternating barrier layers and quantum well layers, wherein optionally, the barrier layers have a higher Al content than the quantum well layers and/or have a different Indium content than the quantum well layers. A display comprising: a plurality of optoelectronic devices as claimed in one of the preceding claims; one or more optical structures disposed above the emitting surfaces of the plurality of optoelectronic devices. A method for preparing an optoelectronic device comprises the following steps: epitaxially growing a first charge carrier transport layer of a first doping type on a carrier substrate; epitaxially growing an active layer on the first charge carrier transport layer, wherein the active layer comprises a first region and a second region surrounding the first region, and the active layer is configured to emit light in response to a current passing through the active layer; providing a first structured mask layer covering at least the first region of the active layer, wherein the first structured mask layer covers a first portion of the active layer and comprises at least one groove exposing a top surface of a second portion of the first region; altering the second portion of the first region so that the second portion is configured to be non-luminescent; A second charge carrier transport layer is epitaxially grown on the active layer. The method of claim 11, wherein the step of epitaxially growing the second charge transport layer on the active layer is performed before the step of providing the first structured mask layer. The method of claim 11 or 12, wherein the step of modifying the second portion of the first region comprises at least one of: increasing the defect density to be higher than the defect density of the at least one first portion; providing a defect density in a range of at least one order of magnitude greater than a measurement limit, in particular obtained by one of X-ray diffraction positron annihilation spectroscopy or SIMS, and in particular between 1.5 and 3 orders of magnitude greater than a measurement limit; adding a dopant in a range of between 5e17 1/cm ³ and 8e18 1/cm ³ to the second portion, optionally with a gradient dopant concentration in the growth direction; etching at least partially into the active layer; A semiconductor material is filled in a groove, wherein the energy gap of the semiconductor material is higher than the energy gap of the active layer. The method of claim 11, wherein the material of the first structured mask layer covers a first portion centered in the first region of the active layer, and the at least one groove surrounds the material. A method as claimed in claim 11, wherein the first structured mask layer exposes a portion of the upper surface of the second region of the active layer; and the step of changing the second portion of the first region includes: diffusing a dopant (particularly one of Zn and Mg) into the second portion of the first region and into the second region, thereby causing quantum well intermixing in the second portion of the first region and in the second region of the active layer. The method of claim 11 further comprises: Providing a second structured mask layer covering the first region and exposing a portion of the upper surface of the second region of the active layer; Diffusing a dopant (particularly one of Zn and Mg) into the second region, thereby causing quantum well intermixing in the second region of the active layer. The method of claim 11, wherein when viewed from the top, the shape of the first portion covered by the material of the first structured mask layer is substantially circular or elliptical; and/or when viewed from the top, the shape of the first portion covered by the material of the first structured mask layer includes more than one ring, and the more than one ring of the material of the first structured mask layer is surrounded by a respective annular groove. The method of claim 11 further comprises: Providing a structured etching mask on the second charge carrier transport layer; At least platform etching the second charge carrier transport layer and the active layer, wherein the outer edge of the second region constitutes a portion of an inclined side surface. The method of claim 17, wherein the material of the structured etching mask and the material of the first structured mask layer comprise the same material. A method as claimed in claim 17 or 18, wherein the step of terrace etching is performed before providing the first structured mask layer. The method of claim 17 further comprises one of the following: Depositing a dielectric layer on the inclined side surfaces; and/or Regrowing a semiconductor material layer disposed on the inclined side surfaces, wherein the energy gap of the semiconductor material layer is greater than the band gap of the active layer. A method as claimed in claim 17, wherein an emitting surface above the second charge carrier transporting layer is larger than a top surface of the second charge carrier transporting layer. The method of claim 17 further comprises: Forming a top contact layer having a transparent conductive layer, the transparent conductive layer being disposed on or adjacent to an emitting surface and electrically connected to the second charge carrier transport layer. The method of claim 17, wherein epitaxially growing the active layer comprises the following steps: Epitaxially growing a plurality of alternating quantum well layers and barrier layers, wherein the barrier layers contain a higher Al content than the quantum well layers. A method as in claim 11, wherein the step of providing the first structured mask layer includes structuring a mask layer material in response to simulation results of a far field evaluating expected distribution characteristics of a plurality of virtual dipoles, each virtual dipole representing a small area of the active layer.