• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    The light-emitting diode (1) comprises: - a stack (2) of semiconductor layers comprising a first face (3) and a second face (4) opposite in a thickness of the stack (2), - a first electrode ( 5) having a face (5a) in contact with the first face (3) of the stack (2), and - a second electrode (6) in contact with the stack (2). Furthermore, the light-emitting diode (1) is such that: - a recess (11) is formed in the second face (4) of the stack (2) from which it follows that the stack (2) comprises a part thinned (12), - the face (5a) of the first electrode (5) in contact with the first face (3) is in contact only with the thinned portion (12) of the stack (2), - the second electrode (6) is in contact with a zone of the stack (2) distinct from said thinned portion (12) of the stack (2).
    公开号:FR3069106A1
    申请号:FR1756778
    申请日:2017-07-17
    公开日:2019-01-18
    发明作者:David Vaufrey
    申请人:Commissariat a lEnergie Atomique CEA；Commissariat a lEnergie Atomique et aux Energies Alternatives CEA；
    IPC主号:
    专利说明:

    The field of the invention relates to light emitting diodes, in particular based on inorganic semiconductors.
    STATE OF THE ART [002] The efficiency of a light-emitting diode can be increased by varying the various characteristics of the latter.
    For example, it is known to implement a texturing forming a roughness on an emission surface of a layer of GaN (Gallium Nitride) doped with N type of a stack of a light emitting diode like l 'teaches the document "Increase in the extraction efficiency of GaN-based light-emitting diodes via surface roughening" by T. Fujii et al. published in Applied Physics Letters volume 84, number 6, February 9, 2004 on pages 855 to 857. Such texturing makes it possible to limit the light reflections internal to the stack, and to diffuse this light towards the outside of the stack .
    [004] Although the texturing of the light-emitting diode makes it possible to improve the extraction of photons, it is limited to improving the extraction of photons which reach the emission surface. In this sense, it is understood that there is a need to improve the quantity of photons which reach the emitting surface of the light-emitting diode.
    The document “Evaluation of InGaN / GaN light-emitting diodes of circular geometry” by X. H. Wang et al. published in Optics Express Vol. 17, issue 25 pp. 22311-22319 (2009) describes that the shape of a light emitting diode can influence the extraction of light generated by the light emitting diode. Indeed, in a stack of a light emitting diode, photons can be generated, preferably by quantum wells, and then propagate in different directions. According to this document, the extraction from above can be done via photon emissions carried out at an associated solid angle, and when the photons are emitted outside this solid angle, they can escape via the sides of the light-emitting diode. , or can be reflected within the light emitting diode. In short, without any particular provision aiming to increase the extraction of the light-emitting diode, only the photons emitted at a particular solid angle are taken into account for the calculation of the efficiency of the light-emitting diode if this efficiency of the light emitting diode for one side, usually from the top or "Top surface" in English. It is understood that there is also a need to improve the efficiency of the light-emitting diode, in particular by increasing the extraction of photons emitted by the light-emitting diode, in particular from the same side of the light-emitting diode.
    In this sense, it is clearly understood that there is a need to develop a new light-emitting diode structure responding to the general problem of improving the efficiency of the light-emitting diode, in particular by optimizing the optical extraction of the light-emitting diode.
    OBJECT OF THE INVENTION The aim of the invention is to meet all or part of the needs mentioned above.
    To tend towards this aim, the invention relates to a light-emitting diode comprising:
    a stack of semiconductor layers comprising a first face and a second face opposite according to a thickness of the stack,
    a first electrode comprising a face in contact with the first face of the stack, and
    - a second electrode in contact with the stack, the light-emitting diode is characterized in that:
    - a recess is formed in the second face of the stack, from which it results that the stack comprises a thinned part,
    - the face of the first electrode in contact with the first face is in contact only with the thinned part of the stack,
    - The second electrode is in contact with a zone of the stack distinct from said thinned part of the stack.
    This in particular makes it possible to improve the efficiency of the extraction of the light-emitting diode while retaining a suitable spatial distribution of the current within the light-emitting diode in operation.
    The light emitting diode may include one or more of the following characteristics:
    • the recess defines said thinned part of the stack with a part of the first face, and the face of the first electrode in contact with the thinned part of the stack has an area less than or equal to the area of said part from the first side;
    • the stack comprises an active region capable of generating photons, a layer of first type doped semiconductor material and a layer of second type doped semiconductor material, the recess being formed in the layer of semi-material first type doped conductor which has an electrical conductivity strictly greater than the electrical conductivity of the layer of second type doped semiconductor material;
    The stack has a first thickness between its first and second faces and a second thickness d between a bottom of the recess and the first face, the second thickness d being strictly less than the first thickness, the second thickness d satisfying the following equation:
    d = z + L with z = k z . -, d = k d .—, and L being such that -. (K d -kd) = L with k d >
    Z 2n a 2n 2n y az ' a k z , k d and k z being positive integers different from 0, λ being a main emission wavelength in operation of the light-emitting diode, and n being an optical index of the stack determined at the main emission wavelength;
    • k z is strictly greater than 1 and strictly less than 6, k d is strictly less than 10, and k z is such that the layer of second type doped semiconductor material has, within the thinned part, a thickness greater than or equal to 50nm, and k d is such that the layer of first type doped semiconductor material present, between the bottom of the recess and one face of said layer of first type doped semiconductor material oriented towards the layer second type doped semiconductor material, a thickness greater than or equal to 100 nm;
    • the thinned part of the stack comprises:
    a first part connecting the first electrode to a zone located in the active region of the stack, the first part having a thickness equal to z, a second part extending from the first part to the bottom of the recess, the second part having a thickness equal to L;
    • the active region comprises several active layers each allowing the recombination of charge carriers, and the area located in the active region is located in the active layer closest to the first electrode;
    • the recess is delimited by an opening of the recess formed in the second face, a bottom of the recess and a wall of the recess connecting the bottom to the opening;
    The opening has dimensions smaller than the dimensions of the recess measured in a plane situated between the opening and the bottom of the recess, and the opening has an area greater than or equal to that of said face of the first electrode ;
    • the wall comprises, according to a cutting plane perpendicular to the bottom, a concave contour;
    The light-emitting diode has a length of dispersion of the current, and, orthogonally to a direction of stacking of the semiconductor layers of the stack, the recess has a dimension comprised between once and five times the length of dispersion of the current of the light emitting diode.
    The invention also relates to a method for producing a light-emitting diode, said light-emitting diode to be produced comprising:
    a stack of semiconductor layers comprising a first face and a second face opposite according to a thickness of the stack,
    a first electrode comprising a face in contact with the first face of the stack, and
    - a second electrode, and the production method comprises a step of designing the stack, the second face of which has a recess to form a thinned part of the stack, the design step taking into account the following characteristics:
    - said face of the first electrode of the light-emitting diode is intended to be in contact only with the thinned part of the stack,
    - The second electrode of the light-emitting diode is intended to be in contact with a zone of the stack distinct from said thinned part of the stack.
    The production process may include one or more of the following characteristics:
    • the stacking design stage includes the following stages:
    a step of determining a main emission wavelength λ of the light-emitting diode during its operation, a step of determining, at the main emission wavelength, an optical index n of the stacking, a step of determining a thickness d of the thinned part between the first face and a bottom of the recess, the thickness d satisfying d = z + L with z = k 7 .—, d = k d . -, and L being such that -. (K d -k 7 ) = L with k d > k z , k d and k z being positive integers other than 0;
    The step of determining the thickness d comprises a step of choosing the value of k z , k z having to be strictly greater than 1 and strictly less than 6, and a step of choosing the value k d , k d must be strictly less than 10;
    The step of designing the stack comprises a step of determining a length of dispersion of the current of the light-emitting diode, and a step of dimensioning the recess so that, orthogonally to a direction of stacking of the semi layers -conductive of the stack, said recess has a dimension between once and five times the length of dispersion of the current of the light-emitting diode.
    Brief description of the drawings The invention will be better understood on reading the description which follows, given solely by way of nonlimiting example and made with reference to the figures for which:
    FIG. 1 illustrates a sectional view of a light-emitting diode according to a particular embodiment of the invention,
    FIG. 2 is a perspective view of the light-emitting diode of FIG. 1,
    FIG. 3 is a sectional view of the light-emitting diode of the type of FIG. 1 for which the hatching has been removed in order to reveal current lines,
    FIG. 4 illustrates an alternative embodiment of the light-emitting diode seen from above,
    FIG. 5 illustrates, in section view, part of the layers of the stack of the light-emitting diode,
    FIGS. 6 to 10 are sectional views of different embodiments of the light-emitting diode,
    - Figure 11 schematically illustrates steps of a process for developing the light emitting diode.
    In these figures, the same references are used to designate the same elements.
    Furthermore, the different elements shown in the figures are not necessarily shown on a uniform scale to make the figures more readable.
    Description of particular embodiments Within the framework of the present invention, it is sought preferably to produce a light-emitting diode (LED for “Light Emitting Diode”) whose structure tends to concentrate photons emitted from an active region of a stacking of semiconductor layers of the light-emitting diode, in particular by quantum wells, in a solid angle for which the emitted photons escape through an upper emitting surface, or more generally an emitting surface of the light-emitting diode. This emitting surface can be located on one side of the light emitting diode opposite to an anode of the light emitting diode for an LED based on GaN (gallium nitride) or AIGaN (aluminum nitride-gallium). To optimize the operation of the LED, a first electrode can be in contact with a charge carrier transport layer of the stack having a first electrical conductivity, while a second electrode can be in contact with another transport layer charge carriers of the stack having a second electrical conductivity strictly greater than the first electrical conductivity. In particular, the contact surface of the first electrode with the stack is strictly greater than the contact surface of the second electrode with the stack. Such a structure can be obtained by limiting the thickness of a part of the stack which then comprises the emission surface.
    To respond to the problem of improving the efficiency of the extraction of the light-emitting diode, it is proposed to limit the thickness of part of the stack, while retaining a suitable spatial distribution of the current. within the operating light emitting diode. A light emitting diode is a device for generating electromagnetic waves, for example centered on a particular wavelength, or light in the visible spectrum or not. In other words, a light emitting diode is configured to emit photons. Preferably, the proposed thinning makes it possible judiciously to take account of the interference between the electromagnetic waves emitted direct and the electromagnetic waves reflected on the first electrode, as well as the interferences between the electromagnetic waves emitted within the stack induced by the optical cavity. of the light-emitting diode in order to improve the extraction of photons, and therefore the efficiency of the light-emitting diode. The extraction efficiency of a light emitting diode can be seen as the number of photons coming out of the light emitting diode compared to the number of photons generated in the light emitting diode. Furthermore, the conservation of the suitable spatial distribution of the current is implemented by a particular arrangement of the electrical contacts (anode and cathode) adapted to the constraints imposed by the conditions of constructive interference, therefore by the presence of this part. thinner stacking. In the present description, when reference is made to a main emission wavelength of the light emitting diode (also called main emission wavelength in operation of the light emitting diode or wavelength the main emission of the light emitting diode during operation), this is the wavelength of the maximum of the light emitting spectrum of the light emitting diode. In Figures 1 to 10, there is shown a three-dimensional orthogonal coordinate system of axes X, Y, Z, the axes X, Y and Z being orthogonal to each other and extending from the same origin. The axes X and Y define a plane considered to be parallel to the horizontal of the reference frame of the light-emitting diode 1. The axis Z is directed upwards, and makes it possible to represent a vertical direction of the light-emitting diode 1. In particular, the Z axis is orthogonal to the emission plane of the light-emitting diode 1, that is to say to the plane through which the photons emitted by the light-emitting diode 1 escape. It is notably along this vertical direction that we can interpret the terms "above" and "above". Figures 1, 3, 5, 6, 7, 8, 9 and 10 are sectional views along a section plane parallel to the plane defined by the X and Z axes.
    In the present description, the stack of semiconductor layers of the light-emitting diode 1 is associated with a stacking direction which corresponds to the axis along which the semiconductor layers of the stack are stacked. The concept of thickness used in the present description is linked to the Z axis, and is measured along an axis parallel to the Z axis, that is to say along an axis parallel to the stacking direction of the layers. semiconductors of the stack of layers.
    In the present description, by the formulation "A based on B" it is meant that B mainly enters the composition of A.
    In the present description, an emitting surface corresponds to a surface of the stack of semiconductor layers of the light-emitting diode through which the photons generated within the stack escape. This emitting surface is located on one side of the light emitting diode, also called the emitting face of the light emitting diode, and makes it possible to define the emission plane mentioned above.
    The concept of interference between direct emitted electromagnetic waves and indirect emitted electromagnetic waves (those reflected) is a concept known to those skilled in the art under the terms "Lloyd's mirror" which describes an interference phenomenon between a source point emitting photons and a mirror. By considering the mirror as the first electrode, and the source point as the photon emitting place closest to the mirror, adjusting the separation distance between the mirror and the source point can generate constructive interference with a view to optimizing the extraction of these waves from the stack of the light-emitting diode via the emission surface, in particular situated directly above the first electrode.
    The notion of interference induced by the cavity is known to those skilled in the art, in particular according to the principle of the Fabry-Perrot interferometer. The optical cavity in the context of the stack of the light-emitting diode can be constituted by all the parts of the stack which lie directly above the first electrode.
    By studying these two concepts of interference and the expressions of the differences in optical paths which result therefrom, it can be concluded that these two types of interference can be optimized independently of one another, but that their combination acts in synergy to improve the extraction efficiency of the light-emitting diode. However, it can be seen that the optimization results in reducing the usual thickness of the stack of semiconductor layers of the light-emitting diode, in particular by reducing the thickness of a hole transport layer and a transport layer. of stack electrons. The reduction in thickness mentioned above makes it possible to improve the extraction efficiency of the photons generated, but is done to the detriment of the spatial distribution of the current in the stack when the light-emitting diode is in operation: it results therefrom that the charge carriers are concentrated in the areas closest to the electrodes. Consequently, a significant part of the active region of the stack capable of generating photons is not or only slightly crossed by a current capable of generating photons. In fact, many light-emitting diodes suffer from a drop in efficiency at high injection: consequently it is always desirable to standardize the current passing through the LED.
    In this sense, the present invention provides a new light-emitting diode structure making it possible in particular to improve the efficiency of photon extraction at the same side of the light-emitting diode. Such a structure can be implemented by the stack of semiconductor layers with a thinned portion, and by judicious cooperation of the electrodes of the light-emitting diode with this stack of semiconductor layers. In particular, it is proposed that the stack comprises a thinned part in contact with an anode, for example in the case of LEDs based on GaN, or in contact with a cathode, for example in the case of LEDs based on InP , and to position the cathode (LED based on GaN), or if necessary the anode (for example LED based on InP) outside the thinned part in order to improve the distribution of the current during the operation of the light-emitting diode .
    As illustrated in Figure 1, the light emitting diode 1 comprises the stack 2 of semiconductor layers. This stack 2 of semiconductor layers comprises a first face 3 and a second face 4. In particular, these semiconductor layers are inorganic. The semiconductor layers of the stack 2 may each comprise a material chosen from the materials III-V and II-VI. The stack 2 of semiconductor layers has a thickness, in particular given by the sum of the thicknesses of the stacked semiconductor layers of said stack 2. The first and second faces 3, 4 are opposite according to the thickness of the stack 2, c ' that is, they are oriented in opposite directions, and in particular formed at least in part in planes parallel to each other and orthogonal to the direction of stacking of the semiconductor layers of stack 2 parallel to the Z axis. The stacking direction of the semiconductor layers of stack 2 is orthogonal to the planes of the semiconductor layers of stack 2. In other words, the semiconductor layers of stack 2 are stacked between the first and second faces 3, 4. The light-emitting diode 1 also comprises the first electrode 5 comprising a face 5a in contact with the first face 3 of the stack 2, and the second element trode 6 in contact with stack 2.
    The first electrode 5 can be an anode, or a cathode, and the second electrode 6 of the type opposite to the first electrode 5 can be a cathode, or an anode. The first and second electrodes 5, 6 can be metallic, and each have a thickness of between a few nanometers and 5 μm. For example, the first and second electrodes 5, 6 can comprise at least one material chosen from nickel, silver, platinum, titanium, aluminum and gold. The first and second electrodes 5, 6 can be formed by multilayers of electrically conductive materials whose resistivity is preferably less than 10 ' 3 Ohms-cm. In particular, each of the first and second electrodes 5, 6 has an electrical conductivity greater than that of the semiconductor layer of the stack 2 with which it is in contact, for example by a factor of 10, 100 or 1000.
    Conventionally, the anode makes it possible to inject holes into the stack 2 and the cathode makes it possible to inject electrons into the stack 2. In this sense, the face 5a of the first electrode 5 in contact with the first face 3 is also called, where appropriate, hole injection face or electron injection face. The face 5a is oriented towards the active region 8 of the stack 2. In particular, the injection face 5a of the first electrode 5 is parallel to the first face 3, and is orthogonal to the direction of stacking.
    Preferably (FIG. 1), the stack 2 comprises a layer 9 of first type doped semiconductor material (in particular in contact with the second electrode 6) and a layer 7 of second doped semiconductor material type (in particular in contact with the first electrode 5). The first type and the second type are opposite, and are chosen from N and P. When the doping is of type P, the doped layer P is in contact with the anode which injects holes in said layer of doped semiconductor material P type when the light-emitting diode 1 is in operation. When the doping is of type N, the doped layer N is in contact with the cathode which injects electrons into said layer of doped semiconductor material of type N when the light-emitting diode 1 is in operation. The layer 9 of first type doped semiconductor material and the layer 7 of second type doped semiconductor material are charge carrier transport layers, possibly holes if the doping is P type, or electrons if the doping is of type N. The stack 2 also comprises the active region 8 within which the recombination of charge carriers in order to generate photons is possible. A charge carrier can be an electron or a hole. The active region 8 is capable of generating photons. In particular, the active region 8 is arranged / located between the layer 7 of second type doped semiconductor material and the layer 9 of first type doped semiconductor material. The active region 8 can be an intermediate region located between the layer of N doped material and the layer of P doped material. The active region 8 can also be the simple junction of the layers of first type doped semiconductor material and semi material -doped conductor of the second type (PN junction), or may be a layer of the stack 2 which may comprise a single layer (PIN junction for N-doped Intrinsic P doped) semiconductor or a plurality of semiconductor layers as it will be seen later. The active region 8 can be in contact with the layer 9 of first type doped semiconductor material, and in contact with the layer 7 of second type doped semiconductor material. The thickness of the active region 8 can be adapted by the skilled person as required. Alternatively, the stack 2 may include a blocking layer 10 of charge carriers of semiconductor material arranged between the active region 8 and the layer 7 of doped semiconductor material of the second type. In this case, the blocking layer 10 is in contact with the layer 7 of doped semiconductor material of the second type and the active region 8, and can block the electrons if the second type is of type P, or block the holes if the second type is type N.
    According to a first example, the material of the layers 7, 9 of doped semiconductor material of the first type and of doped semiconductor material of the second type may comprise - or be based on - a semiconductor material such as for example gallium nitride (GaN). In this first example, the second type is the P type, and the layer 7 of doped semiconductor material of the second type can comprise galium nitride doped with magnesium. In this first example, the first type is N, and the layer 9 of doped semiconductor material of the first type can comprise gallium nitride doped with silicon. According to this first example, the first electrode 5 is the anode.
    According to a second example, the material of the layers 7, 9 in doped semiconductor material of the first type and in doped semiconductor material of the second type may comprise - or be based on - a semiconductor material such as for example indium phosphide (InP). In this second example, the second type is the N type, and the layer 7 of doped semiconductor material of the second type may include indium phosphide doped with sulfur or tin. In this second example, the first type is P, and the layer 9 of doped semiconductor material of the first type may include indium phosphide doped with zinc. According to this second example, the second electrode 6 is the anode.
    In the present description, when we speak of doping, it is an electrical doping: impurities have been added to a pure or desired pure substance (in the example GaN or InP) in order to intentionally modify its electrical conductivity properties.
    In general, still illustrated in Figure 1, a recess 11 is formed in the second face 4 of the stack 2 where it follows that the stack 2 has a thinned portion 12. The recess 11 is in particular as it is formed in one of the semiconductor layers of the stack 2 so as to have a depth strictly less than the maximum thickness of said layer in which it is formed, the recess 11 can then be formed by a localized thinning of one of the semiconductor layers of the stack 2. The recess 11 is in particular formed in the layer 9 of doped semiconductor material of the first type which has an electrical conductivity strictly greater than the electrical conductivity of the layer 7 made of doped semiconductor material of the second type: this makes it possible to locally reduce the thickness of the stack in order to ensure proper distribution of the current during f light-emitting diode operation. In particular, the recess 11 is formed so that a portion of the layer 9 made of first type doped semiconductor material remains between a bottom of the recess 11 and the active region 8. Thus, the thinned portion 12 of the stack 2 can be such that it comprises a portion of each of the semiconductor layers of the stack 2. This thinned part 12 of the stack 2 therefore extends from the recess 11 to the first face 3. It is then said that the stack 2 comprises the recess 11. Thus, the recess 11 can be delimited by an opening 11a of the recess formed / formed in the second face 4, the bottom 11b of the recess 11 and a wall 11 c of the recess 11 connecting the bottom 11b to the opening 11a. The bottom 11b and the wall 11c may correspond to surfaces of the layer 9 of first type doped semiconductor material. The bottom 11b can be flat. The wall 11c extends in particular from the bottom 11b to, for example, form a straight hollow cylinder. It is said in particular that the recess 11 is a non-through hole, that is to say a blind hole, formed in the second face 4 in particular in the layer 9 of doped semiconductor material of the first type (FIG. 2) . It is in this sense that FIG. 2 schematically illustrates the light-emitting diode 1 comprising the recess 11, the wall 11c of which connecting the bottom 11b to the opening 11a of the recess has four sides perpendicular to the bottom 11b and perpendicular to each other two by two so as to form a frame. The recess 11 within the meaning of the present description is not a simple surface roughness.
    In one embodiment, for example illustrated in Figure 1, the recess 11 delimits said thinned portion 12 with a portion 13 of the first face 3. In this case, the portion 13 of the first face 3 has a strictly lower area to the area of the first face 3. In particular, it is said that the recess 11 makes it possible to delimit one side of the thinned part of the stack 2, and that the part 13 of the first face 3 delimits another side of the thinned part of the stack 2. According to another formulation, the thinned part extends from the part 13 of the first face 3 to the recess 11. The thinned part 12 of the stack 2 makes it possible to form a optical cavity of the stack 2 adapted to a main emission wavelength of the light-emitting diode 1. In particular, it is said that the thinned part 12 of the stack 2 comprises the first part 13 of the first face 3 and a stack area 2 by ticipating in delimiting the recess 11. In particular, the first face 3, or at least the part 13 of the first face 3 is planar, and for any point of the part 13 of the first face 3, there is a straight line passing through said point and perpendicular to part 13 of the first face 3 such that said straight line passes through the recess 11. Preferably, any point of the first electrode 5 can be projected orthogonally in a plane orthogonal to the direction of stacking of the layers semiconductors of the stack 2 and passing through the recess 11, in particular passing through the opening 11 a of the recess 11 so that all of the projected points of the first electrode 5 are in an area of the plane included in the recess 11 to promote the recombination of charge carriers within the thinned part 12.
    The thinning of the stack 2 makes it possible to exacerbate the constructive interference described above with a view to improving the emission of photons by the light-emitting diode 1, in particular from the top of the thinned part 12 of the stack opposite to the first face 3. Thus, the emitting surface of the light-emitting diode 1 is in particular formed by the surface of the stack 2 (or at least part of this surface of the stack 2) participating in delimiting the recess 11. The emission face of the light-emitting diode 2 is therefore on the side of the second face 4.
    The face 5a of the first electrode 5 in contact with the first face 3 of the stack 2 is in contact only with the thinned part 12 of the stack
    2. In FIG. 1, the face 5a of the first electrode 5 is in contact only with the part 13 of the first face 3. This makes it possible to locate the recombination of the charge carriers (holes and electrons) in the thinned part 12 of the stack 2, that is to say in at least part of the active region 8 located in the thinned part 12, when the light-emitting diode is in operation.
    Preferably, the face 5a of the first electrode 5 in contact only with the thinned part 12 of the stack 2, and therefore where appropriate only with the part 13 of the first face 3, has a smaller or equal area to the area of said part 13 of the first face 3. This allows a better efficiency of the light-emitting diode 1 in the sense that the layer 7 of doped semiconductor material of second type has a conductivity such as the current lines within it are generally vertical, that is to say parallel to the direction of stacking of the semiconductor layers of stack 2. We will thus concentrate the probabilities of recombination of the charge carriers in the thinned part 12 of the stack 2, in particular within a part 8a (FIG. 1) of the active region 8 located in the thinned part 12 of the stack 2. In fact, it is within the part 8a of the active region 8 that the maj ority of the charge carriers recombine. It is then understood that only a negligible part of the charge carriers can recombine in the active region 8 outside of the thinned part 12 of the stack 2. The fact that the area of the part 13 of the first face 3 is less than the area of the first face 3 makes it possible, for example, to avoid a problem, in particular related to “quenching”, in the event of etching of the layer 7 of second type doped semiconductor material.
    Figure 3 is a simplified version of Figure 1 for which the hatching has been removed to show the current lines (arrows F1) in the layer 7 of second type doped semiconductor material between the first electrode 5 and the active region 8. In the field, it is generally said that the shadow of the first electrode 5 in the active region 8 characterizes the majority region of charge carrier recombination. The shadow of the first electrode 5 in the active region 8 is in particular such that each point of the active region 8 shaded by the first electrode 5 can be projected orthogonally onto the face 5a of the first electrode 5.
    Furthermore, the second electrode 6 is in contact with a zone of the stack 2 separate from said thinned part 12 of the stack 2 (Figure 1). Thus, the second electrode 6 is in particular not in contact with the thinned part 12, the second electrode 6 can then be spaced from the thinned part
    12. This zone of the stack 2 corresponds in particular to an outer surface of the stack 2. This allows, on the one hand, to authorize the emission of photons from the thinned part 12 of the stack 2 on the side of the second face 4 without the second electrode 6 hindering this emission from the top of the thinned part 12, and, on the other hand, improving the distribution of the current in the light-emitting diode 1 by increasing the length of dispersion of the current (“Spreading length” in English) of the light-emitting diode 1. FIG. 3 precisely illustrates the current lines (arrows F2), coming from the second electrode 6 placed on the second face 4 of the light-emitting diode 1, which pass through layer 9 of first type doped semiconductor material. These lines have a greater elongation than if the second electrode 6 were in contact with the thinned part 12. Thus, the positioning of the first and second electrodes 5, 6 relative to the thinned part 12 makes it possible to increase the efficiency of extraction of the LED (more photons will be able to escape from the light-emitting diode by the combination of the thinned part with the adapted positioning of the first and second electrodes).
    Preferably, as illustrated in Figures 1 to 4, the second electrode 6 is arranged, that is to say formed, on the second face 4, in particular at the periphery of the opening 11a of the recess 11 The second electrode 6 can, for example, take the form of a closed electrically conductive track. With this arrangement, the second electrode 6 is in particular situated at a distance from the emission surface of the stack, or the second electrode 6 and the emission surface are arranged so that there is a contact, or an interface, between the second electrode 6 and a peripheral contour of the emission surface. In FIG. 1, the second electrode 6 is positioned on an area of the stack 2 associated with a part of the stack 2 which is not thinned so that the current emanating from the second electrode 6 can be distributed uniformly in the part 8a of active region 8 located in the thinned part 12 of the stack 2. The second electrode 6 extends in particular on the second face 4 from a contour of the opening 11a of the recess 11.
    Preferably, the light-emitting diode 1 has a length of dispersion of the current and, orthogonally to the direction of stacking of the semiconductor layers of the stack 2, the recess 11 has a dimension d ev (visible in FIG. 1), for example a diameter (FIG. 4) or a width, comprised between once and five times the length of dispersion of the current of the light-emitting diode 1. This dimension is in particular that separating two opposite sides one on the other side of the wall 11c of the recess 11 (FIG. 1), in particular when the wall 11c of the recess has four flanks connected in pairs in the manner of what is illustrated in FIG. 2. The choice of 'such a dimension makes it possible to limit the current losses at the center of the light-emitting diode 1 because the current decreases according to an exponential parameter divided by the length of dispersion of the current of the light-emitting diode 1. In this sense, the di mension referred to here is in particular a maximum dimension of between once and five times the length of dispersion of the current of the light-emitting diode 1. This is particularly valid when the second electrode 6 is placed on the second face 4 at the periphery of the opening 11a of the recess 11. In general, the dimensions of the recess 11 depending on the length of dispersion of the current given above are particularly valid when the bottom 11a of the recess 11 is located between two planes, in particular parallel , passing respectively through the first electrode 5 (in particular through the face 5a of the first electrode 5) and the second electrode 6 (in particular through a face of the second electrode 6 in contact with the stack 2), or according to another formulation when the bottom 11a is located at a level located between the first electrode 5 and the second electrode 6. Thus, a plane, passing through the bottom 11a of the recess 11 and ort hogonal to the direction of stacking of the semiconductor layers of the stack 2, can be located between the first and second electrodes 5, 6, this plane being located at a distance from the first and second electrodes 5, 6.
    The current dispersion length of the light emitting diode 1 is a concept known to those skilled in the art. By way of example, the dispersion length L s of the current of the light-emitting diode 1 can be calculated as follows:
    Ls - (r c +
    Pti tti
    Pe
    -1/2 with p tl the resistivity of layer 9 in first type doped semiconductor material, p e the resistivity of first electrode 5, t tl the maximum thickness of layer 9 in doped semiconductor material first type, t e the thickness of the first electrode 5, r c corresponding to the sum of the contact resistances between the different materials of the light-emitting diode 1, p t2 corresponding to the resistivity of the layer 7 of doped semiconductor material of second type, t t2 corresponding to the thickness of layer 7 of second type doped semiconductor material. In the formula giving L s , "x" represents the multiplication. It is understood from what has been described above that the thickness of the thinned part 12 of the stack 2 can be chosen so that, when the light-emitting diode 1 is in operation, favor constructive interference between the direct electromagnetic waves and the reflected electromagnetic waves emitted from the active region 8 of the stack 2, and favoring the interference linked to the optical cavity of the stack 2 included in the thinned part 12. For this, the stack 2 preferably has a first thickness d1 (Figure 1) between its first and second faces 3, 4 called maximum thickness of the stack, and a second thickness d (Figure 1) between the bottom 11a of the recess 11 and the first face 3. Of course, the second thickness d is strictly less than the first thickness. The second thickness d satisfies the following equation:
    With z = k z . ^, D = k d . ^, And L being such that ^ (k d - k z ) = L, with k d > k z , k d and k z being non-harmful positive integers ( that is to say different from 0), λ being the main emission wavelength in operation of the light-emitting diode, and n an optical index of the stack 2 determined at the emission wavelength main. Here represents the multiplication operator in the equations. In particular, the optical index n can be that of the material which is mainly used in the composition of the transport layers, that is to say layers 7, 9 of first type doped semiconductor material and of semiconductor material of the second type. Thus, if necessary, n can be equal to the optical index of GaN or of InP. Alternatively, the optical index n may be an average optical index, determined at the main emission wavelength, of semiconductors used in stack 2. The optical index is also known by the name "refractive index" , and corresponds to the square root of the dielectric constant of the material considered at the optical frequency (the wavelength) considered.
    Preferably, k z is strictly greater than 1 and strictly less than 6. Indeed, for a value of k z below 2, the first electrode 5 would be too close to the active region 8, which would increase the probability of “quenching” in English (extinction of luminescence in French), that is to say in particular that an exciton can recombine non-radiatively in the level continuum of the first electrode 5. With a value of k z greater than or equal to 6, the layer 7 of doped semiconductor material of the second type would become too thick, which would have the consequence of inducing, on the one hand, ohmic losses linked to the transport of charge carriers by the layer 7 in second type doped semiconductor material, and, on the other hand, an increase in optical absorption because the semiconductor layers do not have zero absorption (in particular P-doped GaN whose mesh is strongly altered by the introducti there is massive dopant Mg), and therefore to alter the operation of the light-emitting diode 1. Preferably, k z is such that the layer 7 of doped semiconductor material of second type present, within the thinned part 12 of the stack 2, a thickness greater than or equal to 50 nm and preferably strictly less than 200 nm. Furthermore, preferably k d is strictly less than 10, this is explained by the fact that it is desired to minimize the optical absorption within the light-emitting diode. In addition, k d is such that the layer 9 of doped semiconductor material of the first type is present, between the bottom 11b of the recess 11 and a face of the layer 9 of doped semiconductor material of the first type oriented towards the layer 7 of second type doped semiconductor material, a thickness greater than or equal to 100 nm. In case of ambiguity on the choice of k d , it is preferable to opt for the lowest value among those preselected according to the criteria explained above to limit the optical absorption. In other words, we set k z to the smallest value allowing to obtain at least 50nm of thickness for the layer of doped semiconductor material of second type, and we set k d to the smallest value allowing obtain a minimum thickness of 100 nm of the layer 9 of doped semiconductor material of the first type in the thinned part of the stack.
    It is therefore understood that the value choices for k z and k d have an influence on the thickness of the layer 7 of second type doped semiconductor material, and on the thickness of the layer 9 of semi material -doped conductor of the first type between the active region 8 and the bottom 11b of the recess. To visualize the influence of the thickness of the layers of second type doped semiconductor material and first type doped semiconductor material, it is possible to simulate a map of the radiance (or energy luminance) of which the unit is the watt per square meter per steradian (Wm ' 2 .sr' 1 ) given according to the thickness of the layer 7 in doped semiconductor material of second type and according to the thickness of the layer 9 of first type doped semiconductor material. Such a simulation is associated with an active region 8 whose structure is fixed, and if necessary with a charge carrier blocking layer interposed between layer 7 of second type doped semiconductor material and active region 8 whose characteristics are also fixed. With such a mapping, it is possible to notice thickness value pairs for which the radiance reaches optimized values. We also note that the choice of a pair of values k z and k d , according to the choice constraints given above, corresponds to a pair of thickness values adapted from layers 7, 9 of second type doped semiconductor material and of first type doped semiconductor material.
    In particular (Figure 1), the thinned part 12 of the stack 2 has a first part connecting the first electrode 5 to an area (also called internal area of the stack 2) located in the active region 8 of the stack 2. The first part has a thickness equal to z obtained according to the calculations mentioned above. The first part comprises in particular a portion of the layer 7 of doped semiconductor material of the second type. The thinned part 12 also includes a second part extending from the first part to the bottom 11b of the recess 11, the second part has a thickness equal to L obtained according to the calculations mentioned above.
    In particular, as illustrated in Figure 1, the first part of the thinned part 12 of the stack 2 comprises at least a portion 70 of the layer 7 of doped semiconductor material of second type whose thickness corresponds to the thickness of the layer 7 of doped semiconductor material of the second type within the thinned part 12. The first part of the thinned part 12 of the stack 2 can also, if necessary, include a portion 100 of the charge carrier blocking layer 10 whose thickness corresponds to the thickness of the charge carrier blocking layer 10 within the thinned part 12. The first part of the thinned part 12 of the stack 2 can also comprise a portion 80a of the active region 8 whose thickness is strictly less than the thickness of the active region 8 within the thinned part 12.
    In particular, as illustrated in Figure 1, the second part of the thinned part 12 has, depending on its thickness, at least a portion 90 of the layer 9 of first type doped semiconductor material, the thickness of the portion 90 of the layer 9 of doped semiconductor material of the first type corresponding to the distance separating the face of the layer 9 of doped semiconductor material of the first type oriented towards the active region 8 (or towards the layer 7 of semiconductor material doped with second type) of the bottom 11b of the recess 11. Furthermore, the second part of the thinned part 12 of the stack 2 may also include a portion 80b of the emissive region 8 distinct from the portion 80a of the emissive region 8 that includes the first part.
    The active region 8 can be formed by a PN junction, or in a single layer (PIN junction), or can comprise a plurality of active layers each forming a quantum well and separated in pairs by a barrier layer ("cladding" layer "in English). The structure of the active region 8 depends on the main emission wavelength desired for the light-emitting diode 1, the production of an active region 8 is well known to those skilled in the art. According to one embodiment, the active region 8 comprises several active layers 8b (FIG. 5) - which therefore belong to the stack of semiconductor layers each allowing the recombination of charge carriers at least in the part of the active region 8 located in the thinned part of the stack. Each active layer 8b can form a quantum well. The adjacent active layers 8b can be separated by a barrier layer 8c adapted to the active layers that it separates. The active layer 8b closest to the layer 9 of first type doped semiconductor material, can be connected to said layer 9 of first type doped semiconductor material by an additional barrier layer 8d. The active layer 8b closest to the first electrode can be in contact either with the blocking layer 10 if it is present, or with a corresponding barrier layer. Each active layer 8b can have a thickness between 1 nm and 8 nm, each barrier layer can have a thickness between 3 nm and 15 nm. An active layer can be made of gallium nitride, aluminum gallium nitride, or gallium nitride indium, the atomic percentages of the compounds of which can be adapted as a function of the materials used in the transport layers or the barrier layers. The barrier layers can be: in ln x Gai- X N if the quantum well is in ln y Gai- y N with x <y and x can be equal to 0, or in Al x Gai_ x N if the quantum well is in Al y Gai- y N with x> y and y can be equal to 0.
    In the example where the active region 8 has several active layers each allowing the recombination of charge carriers, the area located in the active region 8 is located in the active layer 8b allowing the recombination of closest charge carriers of the first electrode 5. This makes it possible, in correspondence with the values of z and L, to optimize the thickness of the thinned part. In this sense, the area located in the active region 8 is preferably located in a median plane, orthogonal to the direction in which the thickness of the active layer is measured, of said active layer 8b closest to the first electrode 5.
    According to a particular implementation in which the active region 8 comprises a plurality of active layers each forming a quantum well, the source of emission of the photons is not located at a point or in a plane but on the contrary in several wells, that is to say several planes arranged in the active region 8 of the stack 2. To optimize such a light-emitting diode structure, it would for example be possible to take into account the different layers forming the quantum wells as independent point sources. Then, the results should be summed to optimize the thickness of the optical cavity, that is to say of the stack of layers between the first electrode and the bottom of the recess, by varying the thickness of the portion of the layer of second type doped semiconductor material present in the thinned part of the stack, and over the thickness of the portion of the layer of first type doped semiconductor material present in the thinned part of the stacking.
    However, in particular in the case of an LED based on GaN, InGaN or AIGaN, the first electrode 5 is an anode, and the reasoning relating to the independent point sources can be simplified in the following way: the mobility of electrons being greater than that of the holes, the majority of the recombinations within the active region 8 will take place in the layer forming the quantum well closest to the first electrode 5. It is in this sense that the first part of the part thinned 12 of the stack 2 preferably extends from the part 13 of the first face 3 to the area located in the active region 8 mentioned above. Said zone located in the active region 8 is then preferably situated at the median of the layer, depending on its thickness, forming the quantum well closest to the first electrode 5. It is then said that the first part of the thinned part of the stack 2 comprises the portion of the layer 7 of doped semiconductor material of the second type and a wafer / portion of the layer forming the quantum well closest to the anode 5. In the particular case of an LED based on InP, these are the holes which are more mobile than the electrons, this is why the above reasoning is reversed considering that the first electrode is a cathode.
    According to one embodiment, the shape of the recess 11 can also contribute to improving the efficiency of the light-emitting diode 1 by taking into account the length of current dispersion of the light-emitting diode 1.
    In Figure 1, the flanks of an etching defining the thinning of the layer 9 of first type doped semiconductor material are shown with right angles to the bottom 11b. However, this is not an essential characteristic of the invention since the wall 11c of the recess 11 can be formed by flanks preferably at an angle, then forming a trapezium (FIG. 6), the smallest base of which is proximal to the active region 8 and so that the area of the bottom 11b of the recess 11 is strictly less than the area of the opening 11a of the recess 11: with such an arrangement, the distribution of the current is facilitated, especially when the second electrode 6 is formed on the second face 4. However, it can also be imagined a wall 11c comprising one or more concave (or FIG. 7) or convex flanks, or a wall 11c whose flanks are biased (Figure 8) to form a narrowing of the recess 11 at the opening 11a of the recess 11 at the top of the light-emitting diode 1 so as to form a diaphragm limiting the emission cone of the light-emitting diode 1. [0057 ] The FIG. 8 represents an implementation according to which the opening 11a of the recess has dimensions smaller than the dimensions of the recess 11 measured in a plane situated between the opening 11 a of the recess and the bottom 11 b of l recess 11, this plane being in particular parallel to the planes of the layers of the stack 2. According to another formulation, the area of the bottom 11b is here strictly greater than the area of the opening 11a of the recess. Here the photons should preferably pass through the opening 11a of the recess 11. In this sense, the opening 11a of the recess 11 has an area greater than or equal to the area of said face 5a of the first electrode 5 in contact with the first face 3 of the stack. Here preferably, the face 5a the first electrode 5 is oriented towards the opening 11a of the recess 11, it is said in particular that the opening 11a of the recess is located above, or plumb, of the first electrode 5 then located in the lower part of the light-emitting diode 1. In this sense, the face 5a of the first electrode 5 being planar, any point on the face 5a of the first electrode 5 oriented towards the active region 8 can be associated with a straight line passing through this point and orthogonal to said face 5a of the first electrode 5, said straight line crossing the opening 11 a of the recess 11. This improves the distribution of the current within the light-emitting diode 1 for the 'desired application. Furthermore, such an implementation makes it possible to control the beam emitted by the light-emitting diode 1 on the side where the second electrode 6 is located.
    To optimize the spatial distribution of the current during the operation of the light-emitting diode 1, preferably when the second electrode 6 is arranged / formed on the second face 4 (FIG. 7), the wall 11c of the recess 11 connecting the bottom 11b of the recess at the opening 11a of the recess 11 is such that it has a contour connecting a point of the opening 11a of the recess 11 to a point of the bottom 11b of the recess 11. Said contour is in particular taken along a section plane orthogonal to the layers of the stack 2 and in particular at the bottom of the recess 11. The shape of this contour is preferably an exponential dependent on the length of dispersion of the current of the light-emitting diode 1 In particular, this shape taken in a plane parallel to the plane ZX is of the type / (%) = 1. e ^) for the left part of the wall of the recess in FIG. 7 and of the type / (%) = 1. had) for the right part of the par oi of the recess in Figure 7. In other words, the wall 11c of the recess 11 may be concave, that is to say that it may comprise, according to a cutting plane perpendicular to the bottom 11b of the recess 11, a concave contour, preferably representative of an exponential function dependent on the length of dispersion of the current Ls of the light-emitting diode 1. This contour connects, in the section plane, a point of the opening 11a of the recess 11 at a point at the bottom 11b of the recess 11. According to an embodiment where the opening of the recess is circular, the wall of the recess can be delimited by a surface of revolution obtained by the rotation of a plane curve whose equation is representative of an exponential dependent on the length of dispersion of the current of the light-emitting diode. The rotation of the plane curve takes place around an axis parallel to the Z axis and passing through the bottom of the recess.
    The recess 11 may have a depth corresponding to a value between 2% and 98% of the maximum thickness of the layer 9 of first type doped semiconductor material. In particular, according to the example illustrated in FIGS. 1 and 6 to 8, the thickness of the layer 9 of first type doped semiconductor material between the second electrode 6 and the face of the layer 9 of doped semiconductor material. first type oriented towards layer 7 of doped semiconductor material of second type can be between 1pm and 10pm. Below 1pm the current dispersion length of the light-emitting diode will impose a first electrode 5 too small for the emitting surface of the light-emitting diode to be correct, and above 10pm the price of the light-emitting diode will be too high . For the rest of stack 2, the thicknesses will be chosen as a function of the optical index, the main emission wavelength as well as the values k d and k z . In general, we will try to keep a maximum thickness for the stack 2 of between 1pm and 800pm.
    It is understood from what has been described above that the invention also relates to a method of production (Figure 11) of a light emitting diode 1 as described, that is to say that said light-emitting diode 1 to be produced comprises the stack of semiconductor layers comprising the first face 3 and the second face 4 opposite according to the thickness of the stack 2. The process of preparation is in particular implemented with a view to subsequently fabricating said light-emitting diode as prepared. The advantages linked to the production process follow from those described in the context of the light-emitting diode 1 with a view to manufacturing it so that it has the desired advantages. The light-emitting diode 1 to be produced also includes the first electrode 5 comprising the face 5a in contact with the first face 3 of the stack 2, and the second electrode 6. In this case, the production method comprises a design step E1 of the stack 2, the second face 4 of which has the recess (then formed in the second face 4) to form the thinned part 12 of the stack 2. This design step E1 takes into account the following characteristics:
    said face 5a of the first electrode 5 of the light-emitting diode 1 is intended to be in contact only with the thinned part 12 of the stack 2,
    - the second electrode 6 of the light-emitting diode 1 is intended to be in contact with a zone of the stack distinct from said thinned part 12 of the stack 2.
    It is understood that the design step E1 makes it possible to implement the development of the light-emitting diode 1.
    In particular, the design step E1 of the stack 2 comprises the following steps:
    a step of determining E1-1 of the main emission wavelength λ of the light-emitting diode during its operation,
    a step of determining E1-2, at the main emission wavelength, of the optical index n of the stack 2, in particular as described above,
    a step E1-3 of determining the thickness d of the thinned part 12 between the first face 3 and the bottom 11b of the recess 11, the thickness d
    Λ Λ satisfying d = z + L with z = k 7 .—, d = k d .— and L being such that z 2n a 2n
    Λ
    -. (k d - k z ) = L with k d > k z , k d and k z being positive integers other than 0.
    In particular, the step E1-3 of determining the thickness d comprises a step of choosing the value of k z , k z having to be strictly greater than 1 and strictly less than 6, and a step of choosing the value k d , k d having to be strictly less than 10. All the constraints linked to the values of k z and k d described in combination with the light-emitting diode can be applied to the production process.
    The steps E1-1 to E1-3 make it possible in particular to determine the maximum thickness of the stack 2, and to determine the different semiconductor layers which will have to compose it.
    Furthermore, the design step E1 of the stack 2 may include a step of determining E1-4 of the length of dispersion of the current of the light-emitting diode 1. The design step E1 also includes a step of dimensioning E1-5 of the recess 11 so that, orthogonally to a direction of stacking of the semiconductor layers of the stack 2, said recess 11 has a dimension (for example a diameter, a width), in particular maximum, between once and five times the current dispersion length of the light-emitting diode 1.
    In particular, everything that has been described in the context of light-emitting diode 1 can also be applied in the context of the production process. In this case, the design step E1 can also take into account the fact that the stack 2 must include, for its thinned part 12, the first and second parts described above.
    The invention may also relate to a method for manufacturing a light-emitting diode 1 as described, comprising a step for producing the light-emitting diode 1 implementing the method for producing the light-emitting diode 1, and a step of manufacturing the light-emitting diode 1 according to the results of the step of making the light-emitting diode 1.
    There is given hereinafter a particular example of a light-emitting diode, the stack of which is based on GaN, that is to say the material of the layers of doped material of first type (N doping) and second type (P doping) is GaN. In this case, the optical index n is equal to 2.49 for a main emission wavelength of 450 nm in operation of the light-emitting diode. In this stack, the following successive layers are arranged between the layer of doped semiconductor material of second type and the layer of doped semiconductor material of first type: an AIGaN electron blocking layer and at least one layer forming a quantum well of thickness equal to 10nm in InGaN. According to this
    For example, we therefore have - ~ 90nm. In this example, z represents the thickness of the first part of the thinned part of the stack, and corresponds to the sum of the thicknesses of the layer of second type doped semiconductor material, of the electron blocking layer , and half the thickness of the quantum well closest to the blocking layer. Thus, for an emission at wavelength of 450nm, one will opt in priority at k z = 2 whence z = 90 * 2 = 180nm, that is to say a thickness of the layer of P doped material of around 150nm if the thickness of the blocking layer is 25nm and the thickness of a quantum well is around 10nm. The value k d is then chosen so as to make it possible to form the rest of the layers making up the active region and the layer of doped semiconductor material of the first type whose thickness is at least 100 nm.
    In the present description, by "of the order of" is meant to within 10%. The light-emitting diode architecture illustrated in FIG. 1 corresponds in particular to a diode with a thin film configuration better known in the field under the acronym VTF for "Vertical Thin Film" in English. However, the invention can also be applied to other structures such as for example that illustrated in FIG. 9 of a TFFC structure (acronym for "Thin Film Flip Chip") for which the first electrode 5 and the second electrode 6 are accessible from the same side of the light-emitting diode 1, the thinned part 12 of the stack 2 being represented between the dotted lines 11 and
    I2. In FIG. 9, a substrate 14 is shown on which the light-emitting diode 1 is transferred after its formation on a sacrificial substrate which is not visible in FIG. 9 because it is removed so as to allow the formation of the recess 11. The substrate 14 therefore comprises connection terminals 14a, 14b each coming into contact with one of the corresponding first and second electrodes 5, 6. As mentioned above, the level of the bottom 11b is lower than the level of the second electrode 6 in FIG. 9 (that is to say that the level of the bottom 11b is located between the level of the first electrode 5 and the level of the second electrode 6) to satisfy the conditions related to the length of dispersion of the current with respect to the lateral dimensioning of the recess 11.
    In general, the lines 11 and I2 shown in FIGS. 1 and 6 to 9 virtually represent the limits of the thinned part 12 of the stack 2 in contact with the first electrode 5 and making it possible to view the recess 11, the bottom 11b, the opening 11 a and the wall 11 c. In FIGS. 1 to 3 and 6 to 10, the stack 2 illustrated comprises the layer 7 of doped material of second type, the layer 9 of doped material of the first type, the active region 8, the blocking layer 10 of the diode light emitting 1.
    The light-emitting diode described above is intended to emit electromagnetic radiation, that is to say photons, when it is in operation. The operating wavelength of the diode is not limited to the visible wavelengths. Indeed, a person skilled in the art can adapt the stack to center the emission of the light-emitting diode on a wavelength other than the visible.
    The light emitting diode may also include at least one Bragg mirror. The structure of a Bragg mirror is known and will not be detailed here. For example, the light-emitting diode 1 may comprise (FIG. 10) a Bragg mirror 15 arranged in the recess 11, for example at least on the bottom 11b of the recess 11 so that the Bragg mirror is in contact with the layer 9 of first type doped semiconductor material. The positioning of the Bragg mirror 15 in the recess 11 makes it possible to favor a directive emission of the light-emitting diode 1. Alternatively, or in combination with, the Bragg mirror arranged in the recess 11, the light-emitting diode 1 may also include a Bragg mirror 16 arranged on the first electrode 5, in particular on a face of the first electrode opposite to the face 5a of the first electrode 5 in contact with the stack 2 and oriented towards the active region 8. The arrangement of the mirror Bragg 16 in contact with the first electrode 5 makes it possible to return photons reaching the first electrode 5 towards the recess 11. The Bragg mirror 16 is used in particular if the first electrode 5 is transparent or semi-transparent. Bragg mirrors increase the reflectivity of the light emitting diode at the interface where they are located, and consequently increase the intensity of constructive interference.
    The light-emitting diode described according to the present invention also makes it possible to promote a directive emission of photons, even if the Bragg mirror (s) mentioned above are not present.
    Furthermore, the light-emitting diode can also include any device making it possible to improve the reflectivity of the interface of the stack with the first electrode or of the air / semiconductor interface by forming, for example, a mirror Bragg as mentioned above or photonic crystals of semiconductor material in the recess, especially on the bottom of the recess. According to another example, at least the bottom of the recess can be covered with a layer (forming a "capping layer" in English) in a material having a refractive index greater than the optical index of the material used to form layer 9 of first type doped semiconductor material.
    The present invention may have industrial applications such as in lighting, or display, where a light emitting diode as described can be used.
    In particular, the present invention is particularly suitable for forming a matrix of light emitting diodes dedicated to the display, or to the display, because the contrast is increased. In particular, when, for each light-emitting diode, the cathode is formed on the second face of the stack and completely or partially surrounds the opening of the recess, this makes it possible to avoid crosstalk (or "crosstalk").
    In general, to increase the contrast of a light emitting diode when the cathode is located on the second face of the stack, a layer of chromium can be arranged on the cathode so that the cathode is caught between the layer of chromium and the layer of first type doped semiconductor material.
    Another advantage of the present invention is that such a diode can be manufactured while limiting the number of masks necessary due to the possible alignment of the anode, more generally the first electrode 5, with the recess. Furthermore, the same mask can be used to make the cathode, more generally the second electrode, and the recess.
    权利要求:
    Claims (12)
    [1" id="c-fr-0001]
    1. Light-emitting diode (1) comprising:
    - a stack (2) of semiconductor layers comprising a first face (3) and a second face (4) opposite according to a thickness of the stack (2),
    - a first electrode (5) comprising a face (5a) in contact with the first face (3) of the stack (2), and
    - a second electrode (6) in contact with the stack (2), characterized in that:
    - a recess (11) is formed in the second face (4) of the stack (2) from which it follows that the stack (2) has a thinned part (12),
    - the face (5a) of the first electrode (5) in contact with the first face (3) is in contact only with the thinned part (12) of the stack (2),
    - the second electrode (6) is in contact with a zone of the stack (2) distinct from said thinned part (12) of the stack (2).
    [2" id="c-fr-0002]
    2. Light-emitting diode (1) according to claim 1, characterized in that the recess (11) delimits said thinned part (12) of the stack (2) with a part (13) of the first face (3), and in that the face (5a) of the first electrode (5) in contact with the thinned part (12) of the stack (2) has an area less than or equal to the area of said part (13) of the first side (3).
    [3" id="c-fr-0003]
    3. Light emitting diode (1) according to any one of claims 1 to 2, characterized in that the stack (2) comprises an active region (8) capable of generating photons, a layer (9) of semi-material first type doped conductor and a layer (7) of second type doped semiconductor material, the recess (11) being formed in the layer (9) of first type doped semiconductor material which has strictly electrical conductivity higher than the electrical conductivity of the layer (7) of second type doped semiconductor material.
    [4" id="c-fr-0004]
    4. Light-emitting diode (1) according to any one of the preceding claims, characterized in that the stack (2) has a first thickness (d1) between its first and second faces (3, 4) and a second thickness d between a bottom (11a) of the recess (11) and the first face (3), the second thickness d being strictly less than the first thickness, the second thickness d satisfying the following equation:
    d = z + L withz = k z . ^, d = k d . ^, and L being such that (k d - k z ) = L with k d > k z , k d and k z being different positive integers 0, λ being a main emission wavelength in operation of the light-emitting diode, and n being an optical index of the stack (2) determined at the main emission wavelength.
    [5" id="c-fr-0005]
    5. Light-emitting diode (1) according to claim 4 and claim 3, characterized in that k z is strictly greater than 1 and strictly less than 6, k d is strictly less than 10, and in that k z is such that the layer (7) of doped semiconductor material of the second type has, within the thinned part (12), a thickness greater than or equal to 50 nm, and in that k d is such that the layer (9) of material first type doped semiconductor present, between the bottom (11 b) of the recess (11) and one face of said layer (9) of first type doped semiconductor material oriented towards the layer (7) of material second type doped semiconductor, a thickness greater than or equal to 100 nm.
    [6" id="c-fr-0006]
    6. Light-emitting diode (1) according to claim 5, or claim 4 and claim 3, characterized in that the thinned part (12) of the stack (2) comprises:
    a first part connecting the first electrode (5) to a zone located in the active region (8) of the stack (2), the first part having a thickness equal to z,
    - a second part extending from the first part to the bottom (11b) of the recess (11), the second part having a thickness equal to L.
    [7" id="c-fr-0007]
    7. Light-emitting diode (1) according to claim 6, characterized in that the active region (8) comprises several active layers (8b) each allowing the recombination of charge carriers, and in that the area located in the active region ( 8) is located in the active layer (8b) closest to the first electrode (5).
    [8" id="c-fr-0008]
    8. Light-emitting diode (1) according to any one of the preceding claims, characterized in that the recess (11) is delimited by an opening (11a) of the recess formed in the second face (4), a bottom ( 11b) of the recess (11) and a wall (11c) of the recess (11) connecting the bottom (11b) to the opening (11a).
    [9" id="c-fr-0009]
    9. Light-emitting diode (1) according to the preceding claim, characterized in that the opening (11a) has dimensions smaller than the dimensions of the recess (11) measured in a plane located between the opening (11a) and the bottom (11b) of the recess (11), and in that the opening (11a) has an area greater than or equal to that of said face (5a) of the first electrode (5).
    [10" id="c-fr-0010]
    10. Light-emitting diode (1) according to claim 8, characterized in that the wall (11c) comprises, according to a cutting plane perpendicular to the bottom (11b), a concave contour.
    [11" id="c-fr-0011]
    11. Light-emitting diode (1) according to any one of the preceding claims, characterized in that it has a length of dispersion of the current, and in that, orthogonally to a stacking direction of the semiconductor layers of the stack (2), the recess (11) has a dimension between once and five times the length of current dispersion of the light emitting diode (1).
    [12" id="c-fr-0012]
    12. Method for producing a light-emitting diode (1), said light-emitting diode (1) to be produced comprising:
    - a stack (2) of semiconductor layers comprising a first face (3) and a second face (4) opposite according to a thickness of the stack (2),
    - a first electrode (5) comprising a face (5a) in contact with the first face (3) of the stack (2), and
    - a second electrode (6),
    characterized in that the production method comprises a design step (E1) of the stack (2), the second face (4) of which has a recess to form a thinned part (12) of the stack (2), the design stage (E1) taking into account the following characteristics:
    - said face (5a) of the first electrode (5) of the light-emitting diode (1) is intended to be in contact only with the thinned part (12) of the stack (2),the second electrode (6) of the light-emitting diode is intended to be in contact with a zone of the stack (2) distinct from said thinned part (12) of the stack (2). 13. Preparation process according to the preceding claim, characterized in that
    that the design step (E1) of the stack (2) comprises the following steps:
    - a step of determining (E1-1) a main emission wavelength λ of the light-emitting diode (1) during its operation, a step of determining (E1-2), at the wavelength d main emission, from an optical index n of the stack (2), a step of determining (E1-3) a thickness d of the thinned part (12) between the first face (3) and a bottom ( 11b) of the recess (11), Λ Λ the thickness d satisfying d = z + L with z = k z . -, d = k d .—, and L being such that —- (k d - k z ) = L with k d > k z , k d and k z being positive integers other than 0. 14. Production method according to claim 13, characterized in that
    the step of determining (E1-3) the thickness d comprises a step of choosing the value of k z , k z having to be strictly greater than 1 and strictly less than 6, and a step of choosing the value k d , k d must be strictly less than 10.
    15. Production method according to any one of Claims 12 to 14,
    characterized in that the design step (E1) of the stack (2) comprises a step of determining (E1-4) a length of dispersion of the current of the light-emitting diode (1), and a dimensioning step ( E1 -5) of the recess (11) so that, orthogonally to a direction of stacking of the semiconductor layers of the stack (2), said recess (11) has a dimension of between once and five times the length for dispersing the current of the light-emitting diode (1).
    类似技术:
    公开号 | 公开日 | 专利标题
    FR3069106B1|2019-10-18|LIGHT-EMITTING DIODE HAVING AN AMINO-SECTIONAL STACK AND METHOD OF PRODUCING THE LIGHT EMITTING DIODE
    EP3061137A1|2016-08-31|Light-emitting device
    EP3667728B1|2021-03-31|Method for manufacturing a device with light-emitting and/or light-receiving diodes and with a self-aligned collimation grid
    EP3347917B1|2019-10-30|Electroluminescent device with integrated light sensor
    FR3030995A1|2016-06-24|ELECTROLUMINESCENT LIGHT SOURCE WITH ADJUSTABLE OR ADJUSTABLE LUMINANCE LUMINANCE PARAMETER AND METHOD FOR ADJUSTING A LUMINANCE PARAMETER OF THE LIGHT EMITTING LIGHT SOURCE
    FR3000608A1|2014-07-04|SEMICONDUCTOR STRUCTURE OF THE AVALANCHE PHOTODIODE TYPE AND PROCESS FOR PRODUCING SUCH A STRUCTURE
    FR3015113A1|2015-06-19|LOW NOISE QUANTUM DETECTION ELEMENT AND METHOD FOR MANUFACTURING SUCH PHOTODETECTION ELEMENT
    FR2896944A1|2007-08-03|LIGHT EMITTING DEVICE WITH CHROMATIC CONTROL
    EP3347916B1|2019-08-14|Electroluminescent device with integrated light sensor
    EP3109900B1|2017-10-11|Method for manufacturing a plurality of dipoles in the form of islands with self-aligned electrodes
    WO2018100205A1|2018-06-07|Optoelectronic component with improved absorption
    FR3087580A1|2020-04-24|METHOD FOR PRODUCING AN OPTOELECTRONIC DEVICE COMPRISING HOMOGENEOUS LIGHT-EMITTING LIGHT-EMITTING DIODES
    EP3549177B1|2020-10-28|Optoelectronic device with light-emitting diode with extraction enhancement
    TWI594455B|2017-08-01|Photoelectric semiconductor chip and its manufacturing method
    EP3843162A1|2021-06-30|Method for manufacturing a light-emitting diode with extraction layer comprising a step of sizing of a semiconductor layer
    EP3798695A1|2021-03-31|Diffusive structure for light source
    EP3664141B1|2021-08-25|Peripheral absorption photosensor arraywith focusing structures
    EP2840620A1|2015-02-25|Method for elaborating an electrode for light-emitting diode using the current spreading length in said diode
    FR3059787A1|2018-06-08|OPTICAL ELECTROLUMINESCENT DIODE DEVICE WITH INCREASED EXTRACTION
    EP3884523A1|2021-09-29|Light-emitting diode and process for producing a light-emitting diode
    FR3103322A1|2021-05-21|Manufacturing process of a set of light emitters
    FR3105748A1|2021-07-02|Device for laser treatment and method of laser treatment
    WO2020240140A1|2020-12-03|Optoelectronic device comprising two wire-shaped light-emitting diodes each having a layer that limits the leakage currents
    FR3058571A1|2018-05-11|OPTICAL DEVICE FOR A MOTOR VEHICLE HAVING A SEMICONDUCTOR LIGHT SOURCE
    同族专利:
    公开号 | 公开日
    EP3432368B1|2021-12-29|
    EP3432368A1|2019-01-23|
    US10490700B2|2019-11-26|
    FR3069106B1|2019-10-18|
    US20190019918A1|2019-01-17|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US4700210A|1984-11-21|1987-10-13|American Telephone And Telegraph Company, At&T Bell Laboratories|Asymmetric chip design for LEDS|
    US4990972A|1988-09-14|1991-02-05|Ricoh Company, Ltd.|Semiconductor light emitting device|
    JPH0786635A|1993-09-16|1995-03-31|Daido Steel Co Ltd|Light emitting diode|
    US20020047131A1|1999-12-22|2002-04-25|Ludowise Michael J.|Selective placement of quantum wells in flipchip light emitting diodes for improved light extraction|
    US20020113235A1|2001-02-20|2002-08-22|Showa Denko K.K.|Group-III nitride semiconductor light-emitting diode|
    US20120037885A1|2009-04-20|2012-02-16|3M Innovative Properties Company|Non-radiatively pumped wavelength converter|
    EP2846366A1|2013-09-09|2015-03-11|Commissariat à l'Energie Atomique et aux Energies Alternatives|Light-emitting diode with localised photonic crystals|
    GB1478152A|1974-10-03|1977-06-29|Standard Telephones Cables Ltd|Light emissive diode|
    US4706101A|1984-10-27|1987-11-10|Kabushiki Kaisha Toshiba|Light emitting diode formed of a compound semiconductor material|DE102016100317A1|2016-01-11|2017-07-13|Osram Opto Semiconductors Gmbh|Optoelectronic component and method for producing an optoelectronic component|
    FR3068514B1|2017-06-30|2019-08-09|Aledia|OPTOELECTRONIC DEVICE|
    DE102019133214A1|2019-12-05|2021-06-10|OSRAM Opto Semiconductors Gesellschaft mit beschränkter Haftung|Radiation-emitting semiconductor chip|
    WO2021188920A1|2020-03-20|2021-09-23|Sensor Electronic Technology, Inc.|Optoelectronic device with reduced optical loss|
    法律状态:
    2019-01-18| PLSC| Publication of the preliminary search report|Effective date: 20190118 |
    2019-07-31| PLFP| Fee payment|Year of fee payment: 3 |
    2020-07-31| PLFP| Fee payment|Year of fee payment: 4 |
    2021-07-29| PLFP| Fee payment|Year of fee payment: 5 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1756778|2017-07-17|
    FR1756778A|FR3069106B1|2017-07-17|2017-07-17|LIGHT-EMITTING DIODE HAVING AN AMINO-SECTIONAL STACK AND METHOD OF PRODUCING THE LIGHT EMITTING DIODE|FR1756778A| FR3069106B1|2017-07-17|2017-07-17|LIGHT-EMITTING DIODE HAVING AN AMINO-SECTIONAL STACK AND METHOD OF PRODUCING THE LIGHT EMITTING DIODE|
    EP18183183.5A| EP3432368B1|2017-07-17|2018-07-12|Light emitting diode comprising a stack with a thinned portion and method of manufacturing the same|
    US16/037,430| US10490700B2|2017-07-17|2018-07-17|Light-emitting diode comprising a stack with a thinned part, and method for developing the light-emitting diode|
    [返回顶部]