Light-emitting device with improved extraction

Abstract

The invention relates to a light-emitting device comprising: - a light concentrator (CL) made of a solid fluorescent or scintillation material comprising at least seven faces: two parallel faces referred to as large faces (FE1, FE2); and at least five faces referred to as side faces (FL1, FL2, FS, FT1, FT2), including: - a first side face (FL1) and a second side face (FL2) having a surface greater than a surface of the other side faces; - a third side face (FS), referred to as the exit face, adjacent to the first side face and to the second side face; - a side face referred to as the first facet (FT1) adjacent to the first side face and a side face referred to as the second facet (FT2) adjacent to the second side face, the first facet and the second facet forming an angle a that is smaller than 175° and greater than 0°; - a first mirror (M1) arranged so as to cover the first facet (FT1); - a second mirror (M2) arranged so as to cover the second facet (FT2).

Description

DESCRIPTION Title of the invention: Light emitting device with improved extraction Technical field:

[0001] The present invention relates to the field of concentrators pumped by external radiation and more particularly to the improvement of the extraction of light emitted within such a concentrator. Previous technique:

[0002] Light-emitting diodes (LEDs) have many applications in lighting. However, the luminance of LEDs is limited to values ​​that are not suitable for some applications.

[0003] One solution to increase the luminance of LEDs is to use LED-pumped light concentrators (see for example Barbet, Adrien, et al. "Light-emitting diode pumped luminescent concentrators: a new opportunity for low-cost solid-state lasers." Optica 3.5 (2016): 465-468.). This concentrator is, for example, a fluorescent crystal in the visible (red-orange) range such as Ce:YAG which absorbs in the blue (around 450 nm), at a wavelength where LEDs are very efficient. The crystal is cut in the form of a plane, lined with hundreds (or even thousands) of LEDs on the two large surfaces and with emission from the edges. These concentrators make it possible to obtain luminance values ​​10 to 20 times higher than that of an LED.

[0004] Figure 1A illustrates an example of a light-emitting module ME0 known from the prior art based on a concentrator crystal CL having the shape of a parallelepiped of length L , width w , and thickness e . The emission module ME0 comprises a set of LEDs intended to emit in a first spectral band and a light concentrator CL. The concentrator CL has at least one illumination face FE1 , FE2, of dimensions L xw, illuminated by the electroluminescent radiation Ld emitted by the LEDs (optical pumping). The illumination faces FE1 , FE2 are also called “large pumping faces” because their surface area is greater than that of the other faces of the concentrator in order to make the exposure to the incident pumping radiation (Ld) the largest possible. The FS output face is chosen from the two smallest faces (here exw). The isotropy of the emission implies that the same quantity of light comes out on all faces. The smallest face therefore has the greatest illumination. It is the concentrator effect which is used to increase the illumination and the luminance. The structure is therefore composed of two large pumping faces (FE1, FE2), two large lateral faces (FL1, FL2) and two small lateral faces, one of which is chosen as the FS output face.

[0005] The concentrator crystal is configured to absorb the electroluminescent radiation Ld. The luminous flux emitted by the LEDs and directed towards the illumination face is absorbed by the phosphors Lum of the fluorescent crystal which are distributed throughout the volume of the crystal and which then emit fluorescence radiation inside the crystal. Inside the concentrator, the emitted rays can be classified into two main categories: - the trapped rays noted L p : these rays are trapped in the crystal due to total internal reflection (TIR) ​​on the different faces of the crystal. These rays exist for example if the concentrator is placed in the air and if its shape is a parallelepiped with 6 faces parallel two by two and perpendicular to each other and if the index of the concentrator is greater than 2. The trapped rays never leave the crystal, apart from imperfections in the latter. - untrapped rays are the rays that eventually exit the crystal. They are divided into two subcategories: guided rays denoted L g whose characteristic is to be guided by RTI and to exit on one of the faces of the concentrator and the unguided rays noted L ou t which come out directly from the concentrator without being reflected on the faces.

[0006] Figure 1 B is a representation of the angular diagram of the rays emitted and trapped in the concentrator (parallelepiped composed of 6 parallel faces two by two). The dark caps - also called "exit cones" or "escape cones" - represent the angles corresponding to the untrapped rays (guided and unguided) for each face of the concentrator and the light areas represent the angles corresponding to the rays trapped by total internal reflection within the concentrator. In this representation given as a For example, the medium chosen as the concentrator crystal CL is a Ce:YAG crystal (index n2=1.82) with a critical angle of 33° when the ambient medium is air. This angle corresponds to the angle at the apex of the exit cone on the faces of the concentrator. The percentage of radiation trapped by RTI compared to the non-trapped radiation is fixed by the index of the crystal and that of the ambient medium by the Snell-Descartes law.

[0007] In the concentrator of Figure 1A, assuming that the exit face is in contact with air (index n=1) and that the concentrator index is n=1.82 (e.g., YAG), only 8% of the fluorescence radiation passes through the exit face FS to form the exit beam L sThe rest of the fluorescence radiation is trapped within the concentrator by total internal reflection (52%) or leaves the concentrator by exiting through the other faces (5x8%) (see figure 1 B). In the case of a parallelepiped concentrator (n=2.4) in air, the percentage of trapped rays reaches 73%.

[0008] Luminescent concentrators are isotropic light sources with light exiting from all sides and in all directions. From a usage perspective, it is important to be able to collect the light exiting from only one side. This side is the exit side FS. The extraction efficiency of the concentrator is defined by the ratio between the light power exiting from FS (the power of the exit beam L s ) and the total light power emitted by the concentrator.

[0009] Extraction efficiency in light-emitting media has been a crucial topic in the field of scintillators for more than 50 years (see for example M Ankenbrandt and EMLent "Increasing the light collection efficiency of scintillation counters", RSI Vol 34 N°6 p647 (1963)). This topic is particularly important because of the high refractive index of scintillators (Ce:YAG n=1.83, BGO n=2.15), which results in a high percentage of light trapped by total internal reflection (52% for Ce:YAG, 65% for BGO in a parallelepiped geometry) and therefore a low extraction efficiency.

[0010] The problem of extraction efficiency is also of interest in the field of quantum optics for single-photon emission using diamonds containing nitrogen-vacancy centers (NV centers). Indeed, diamonds have a high percentage of light trapped by total internal reflection (73% for a parallelepiped geometry) given their high index (n=2.4). This property limits the performance of the NV center emitter.

[0011] In order to increase the extraction efficiency, it is known to those skilled in the art to place a mirror on the side face opposite the exit face as illustrated in Figure 1 C, called standard configuration. It is then possible to double the extraction efficiency compared to the previous values ​​(2x8%=16% for Ce:YAG and 2x4.5%=9% for diamond in a parallelepiped geometry) by coupling the exit cones 1 and 2 using the mirror. However, this extraction efficiency remains very limited compared to the potential.

[0012] Another solution detailed in document EP21305827 and illustrated in Figure 2A consists of modifying the concentrator of Figure 1A by introducing a new face in the form of a beveled edge SA between two lateral faces defined as the output face and crossed by the output beam L s. By judiciously adapting the normal of the SA edge, it is possible to "recycle" a significant portion of the rays that would have been trapped in the CL concentrator without this beveled SA edge and to direct them towards the exit face. The coupling between the cones is ensured by imaging on the lateral faces considered as plane mirrors. Thus, the image of cone No. 1 by the vertical (resp. horizontal) lateral faces of Figure 2B is cone No. 4 (resp. No. 2). Figure 2B therefore shows that the exit cone has 3 antecedents by imaging on the lateral faces of the structure (cones 1, 2 and 3). It is therefore theoretically possible to obtain an extraction efficiency 4 times greater than that of Figure 1A. This configuration leads to an extraction efficiency of 4x8%=32% for Ce:YAG and 4x4.5%=18% for diamond in a parallelepiped geometry.

[0013] However, it remains desirable to further improve this extraction efficiency.

[0014] The principle of the invention is to increase the number of cones coupled towards the exit face by considering the antecedents of the exit cone by multiple imaging on the lateral faces.

[0015] For this purpose, an object of the invention is a light emitting device comprising: - a light concentrator in a solid fluorescent or scintillating material comprising at least seven faces, two large parallel faces and at least five so-called lateral faces including a first lateral face and a second lateral face having a surface greater than a surface of the other lateral faces, an exit face adjacent to the first lateral face and to the second lateral face, a first facet adjacent to the first lateral face and a facet adjacent to the second lateral face, the first facet and the second facet forming an angle a greater than 0 and less than 175° - a first mirror arranged so as to cover said first facet - a second mirror arranged so as to cover said second facet.

[0016] By a judicious choice of the angle α between the first facet and the second facet, the concentrator makes it possible to couple the output cone associated with the output face to other cones existing in the structure by imaging, so as to obtain a higher extraction efficiency than that of the solutions of the prior art. Summary of the invention:

[0017] For this purpose, an object of the invention is a light emitting device comprising: - a light concentrator in a solid fluorescent or scintillating material having an index n cand adapted to absorb incident light and then emit luminescent radiation within said concentrator, said concentrator comprising at least seven faces, two parallel faces called large faces, and at least five so-called lateral faces including: a first lateral face and a second lateral face having a surface greater than a surface of the other lateral faces, - a third lateral face, called the exit face, adjacent to the first lateral face and to the second lateral face, a portion, called the outgoing portion, of the rays of said luminescent radiation being guided by total internal reflection within the concentrator by crossing said exit face - a lateral face called the first facet adjacent to the first lateral face and a lateral face called the second facet adjacent to the second lateral face, - a first mirror arranged so as to cover said first facet - a second mirror arranged so as to cover said second facet, the first facet and the second facet forming an angle α greater than 0 and less than 175° so that a portion, called the recycled portion of the luminescent radiation, passes through said output face after at least one reflection on said first or second facet or on said first or second mirror, the outgoing portion and the recycled portion forming an output beam.

[0018] According to one embodiment, the angle α is adapted according to the dimensions and an arrangement of the facets in order to maximize an extraction efficiency of the output beam.

[0019] According to one embodiment, the angle a is between 95° and 160°.

[0020] According to one embodiment, the angle a is between 100° and 140°.

[0021] According to one embodiment, the angle a is between 100° and 120°.

[0022] According to one embodiment, the angle a is equal to 110° within ±1°.

[0023] According to one embodiment, the concentrator comprises a number of lateral faces equal to 5, the first and second facets being adjacent.

[0024] According to one embodiment, the concentrator comprises a number of lateral faces equal to 6, including an additional facet adjacent to the first facet and to the second facet.

[0025] According to one embodiment, the concentrator comprises a number of lateral faces greater than 6, including a plurality of additional facets, each additional facet being adjacent to two other additional facets, or adjacent to another additional facet and to the first facet or adjacent to another additional facet and to the second facet.

[0026] According to one embodiment, the large faces have a surface area greater than a surface area of ​​the other faces.

[0027] According to one embodiment, the device comprises a first additional mirror arranged so as to cover said first lateral face and a second additional mirror arranged so as to cover said second lateral face. Preferably, the first lateral face and the second lateral face are not parallel.

[0028] According to one embodiment, the index n c of the concentrator is greater than 1.5, the device comprising an output medium in contact with the output face and having an index n s such that n c > n s > 1. Preferably, the device comprises an optical system attached to the output medium adapted to collimate rays of the output beam.

[0029] According to one embodiment, the device comprises a third additional mirror arranged so as to cover said output face, and in which the concentrator has a beveled edge between the output face and the first lateral face so that a portion, called the additional recycled portion, of said trapped portion passes through said beveled edge forming a so-called additional output beam. Brief description of the figures:

[0030] Other characteristics, details and advantages of the invention will emerge from reading the description given with reference to the appended drawings given by way of example and which represent, respectively:

[0031] [Fig.1A] a schematic view of an example of a light emitting module known from the prior art,

[0032] [Fig.1 B], a representation of the angular diagram of the rays emitted and trapped in the concentrator of Figure 1A,

[0033] [Fig.1 C] a schematic view of an example of a light emitting module known from the prior art,

[0034] [Fig.2A] a schematic view of an example of a light emitting module known from the prior art,

[0035] [Fig.2B], a schematic representation of spatial recycling in the concentrator of Figure 2A,

[0036] [Fig.SA], [Fig.3B], a schematic perspective and top view respectively of a light emitting device according to the invention,

[0037] [Fig.4], a graphical representation of the angular direction of the exit cone antecedents, generated by imaging in the structure, for three different angles between FT1 and FT2: a = 169° (left graph), a = 135° (center graph), a = 99° (right graph)

[0038] [Fig.5], a graphical representation of the extraction efficiency as a function of the angle a of the concentrator of the invention, for three different values ​​of index n c of concentrator (n c =1 .5, n c =1.83 and n c =2.4),

[0039] [Fig.6], a graphical representation of the extraction efficiency as a function of the angle a of the concentrator of the invention, for three different values ​​of linear loss coefficient: 2 x 10 -3 cm -1 (top curve), 2 x 10 -2 cm -1 (middle curve), 2 x 10 -1 cm -1 (bottom curve),

[0040] [Fig.7], [Fig.8], a representation in the form of a planisphere of the output cones of the concentrator of the invention for a = 120° and a = 157.5° respectively,

[0041] [Fig.9], a schematic top view of a light emitting device according to the invention in which the first and second facets are adjacent and have different dimensions,

[0042] [Fig.10], a schematic top view of a light emitting device according to the invention in which the concentrator comprises a number of lateral faces equal to 6, including an additional facet adjacent to the first facet and to the second facet,

[0043] [Fig.11], a schematic view of a preferred embodiment of the invention in which the device comprises an output medium in contact with the output face having an index n s such that n c > n s > 1 and includes a collimating optical system,

[0044] [Fig.12], a schematic view of an embodiment of the invention,

[0045] [Fig.13], a schematic view of an embodiment of the invention,

[0046] In the figures, unless otherwise indicated, the elements are not to scale and identical references designate identical elements. Detailed description:

[0047] Figure 3A illustrates a schematic perspective view of a light emitting device 1 according to the invention. Figure 3B illustrates a schematic top view along an xy plane of the light emitting device 1 according to the invention.

[0048] The light emitting device according to the invention comprises a light concentrator CL made from a solid fluorescent or scintillator material suitable for absorbing incident light Ld causing the emission of luminescent radiation LL within the concentrator.

[0049] The CL concentrator of the invention has an optimized geometry in order to optimize the extraction of LL luminescent radiation.

[0050] The concentrator has an index n c preferably greater than 1.5 because the geometry of the invention is particularly advantageous for high indices (see figure 4 described later).

[0051] The concentrator CL comprises at least seven faces including two parallel faces called “large faces” or “illumination faces” FE1, FE2. By “parallel faces” is meant here and in the rest of the document that the faces are parallel to ±5°. For example, in Figures 3A and 3B, the large faces FE1, FE2 have two main dimensions: a length ! and a width w.

[0052] Preferably, the large faces have a surface area greater than the surface area of ​​the other faces in order to make the exposure to the incident pump radiation (Ld) as great as possible. This makes it possible to obtain an output beam L s exhibiting higher power.

[0053] The concentrator CL further comprises at least five so-called lateral faces, including a first lateral face FL1 and a second lateral face FL2. The lateral faces FL1, FL2 have a surface area greater than a surface area of ​​the other lateral faces in order to ensure a concentration effect on the output face, as explained previously. For example, in Figures 2A and 2B, the lateral faces FL1, FL2 are rectangular and have dimensions ex L, with e a thickness of the concentrator.

[0054] The concentrator Cl comprises a third lateral face FS, called the output face, adjacent to the first lateral face FL1 and to the second lateral face FL2. For example, in Figures 2A and 2B, the FS face is rectangular and has a dimension ex w.

[0055] Finally, the concentrator CL comprises two other lateral faces: a lateral face called the first facet FT1 adjacent to the first lateral face and a lateral face called the second facet FT2 adjacent to the second lateral face. The normals n n2 to the planes of the two facets FT1 and FT2 form an angle A between them (see figure 3B). The angle a between the facets is defined as the angle a = 180° - A. In the invention, this angle a is greater than 0° and less than 175°.

[0056] It is noted that the case a = 180° (see figure 5 for example) does not correspond to an embodiment of the invention but corresponds to the configuration of the prior art of figure 1 C (“standard” configuration).

[0057] The device 1 further comprises a first mirror M1 arranged to cover the first facet FT1 and a second mirror M2 arranged to cover the second facet FT2.

[0058] In a manner known per se, a portion, called the trapped portion L p , LL luminescent radiation is trapped by total internal reflection within the concentrator. A second portion is composed of light exiting the concentrator, with a portion guided towards the L slices g and part L ou t which comes out directly from the concentrator without reflection on one of the faces. We call the outgoing portion L s , the rays of the luminescent radiation LL guided by total internal reflection within the concentrator crossing the exit face FS.

[0059] In addition, the recycled portion is called L r , the portion of the luminescent radiation passing through the output face FS after at least one reflection on the first or second facet or on the first or second mirror. The outgoing portion L s and the recycled portion L r form the output beam L tot .

[0060] Compared with the previous configurations (Fig 1 .A and Fig.2B), the exit cone can image on 4 lateral faces (FL1 , FL2, FT 1 , FT2) instead of one (Fig 1 .A) or three (Fig .2 B). The exit cone can therefore have more antecedents by imaging on the lateral faces. Thus, more cones can be coupled to the exit, which leads to an increase in the extraction efficiency.

[0061] Preferably, the angle a is adapted according to the dimensions and arrangement of the facets in order to maximize the extraction efficiency of the output beam. Indeed, it is understood that the values ​​of the angle a allowing to maximize the extraction efficiency depend directly on the geometry of the structure. As will be detailed later, for a given geometry, it is possible to calculate by simulation the values ​​of angle a which will allow this optimization.

[0062] Figure 4 is a graphical representation of the angular direction of the exit cone antecedents, generated by imaging in the structure, for three different angles between FT1 and FT2: a = 169° (left graph), a = 135° (center graph), a = 99° (right graph). To clarify the reading of the graph, Figure 4 only gives the angular direction of the central axis of the cones considered. This is given by the straight line passing through the center of the circles and the points placed on the circle. Thus, each point in the graphs represents the axis of a cone, antecedent of the exit cone.

[0063] As a non-limiting example, in the embodiment of Figures 3A and 3B, the lateral faces each have a normal in a plane parallel to said large faces. Also, all the cones considered in Figure 4 have their central axis in the xy plane of the large faces. Indeed, their images by the large faces are the cones themselves, because the reflections by a mirror do not change the axis of the cones belonging to the plane of the mirror.

[0064] Furthermore, as a non-limiting example in the embodiment of Figures 3A and 3B, the dimensions of the facets FT1 and FT2 are identical. It is shown that in the case where: a = TI - n / p, with p integer, the number of possible antecedents for the exit cone FS is finite and is worth 2p-1. For example, for a = 168.75°, we have ap = 16 and there are therefore 2p = 32 antecedents of the exit cone. Depending on the values ​​of the angle a, the images can be very numerous (as in the case a = 99° shown in Figure 4).

[0065] More generally, in a geometry different from that of the embodiment of Figure 3A and 3B, we will say that the number of antecedents depends on the value of the angle a.

[0066] Figure 5 is a graphical representation of the extraction efficiency as a function of the angle a of the concentrator of the invention, for three different values ​​of index n c of concentrator (n c =1 .5, n c =1.83 and n c =2.4). Figure 5 presents results of ray tracing simulations using the Monte Carlo method allowing an estimation of the extraction efficiency thanks to this coupling of rays towards the exit cone.

[0067] The results of Figure 5 are obtained for a concentrator according to the geometry of the embodiment of Figures 3A and 3B modified so as to comprise a first additional mirror arranged so as to cover said first lateral face FL1 and a second additional mirror arranged so as to cover said second lateral face FL2 (“optimized” configuration). These additional mirrors make it possible to improve the extraction efficiency.

[0068] As a non-limiting example, these results are presented assuming that the mirrors are totally reflective and for a concentrator having a linear loss coefficient equal to 2 x 10 -3 cm -1 for the central wavelength of the luminescence radiation.

[0069] Figure 5 illustrates that the extraction efficiency is significantly improved compared to the standard configuration for low linear loss conditions and for perfect mirrors. These conditions are close to reality for a Ce:YAG concentrator (of index n c = 1.83) which is a crystal with very little loss.

[0070] In addition, the extraction efficiency of the configuration of Figure 5 is significantly improved compared to that of the prior art configuration of Figures 2A and 2B. As a reminder, the prior art configuration of Figures 2A and 2B makes it possible to obtain a maximum extraction efficiency of 4x8%=32% for Ce:YAG and 4x4.5%=18% for diamond. The configuration of Figure 5 makes it possible to obtain an extraction efficiency of 40% for Ce:YAG and 28% for diamond (index 2.4). It is noted that the improvement is relatively more significant when the index of the concentrator is high. This can be understood by the size of the cones, which is smaller as the index is large.

[0071] The "bell" shape of the curves can be interpreted as follows: - for angles close to 180°, it is necessary to make many back and forths in the structure to "rotate the rays" and couple a large number of rays towards the exit. Two reasons can be given. The first is linked to the angle of rotation of the ray at the exit of the reflection by the FT1, FT2 facets which is all the smaller as the angle a is close to 180°. The second is that an angle a close to 180° will not allow the FT1, FT2 facets to combine their effects because the rays will only rarely intercept the FT2 facet after a reflection on the FT1 facet (and vice versa). Thus, an angle a close to 180° implies large propagation distances of the rays before exiting through the FS facet and therefore relatively greater losses. - for angle a = 90°, the roof reflects only one direction relative to the incident ray: the opposite direction. There are therefore no multiple image effects for the exit cone.

[0072] Figure 6 is a graphical representation of the extraction efficiency as a function of the angle a of the concentrator of the invention, for three different values ​​of linear loss coefficient: 2 x 10 -3 cm -1 (top curve), 2 x 10 -2 cm -1 (middle curve), 2 x 10 -1 cm -1 (bottom curve).

[0073] Figure 6 is obtained for the same configuration as Figure 5 (“optimized configuration”) and for a diamond concentrator (index n c = 2.4) and the extraction efficiency values ​​are normalized to the standard configuration.

[0074] Figure 6 illustrates that there is a very pronounced drop in extraction efficiency for an angle of 90° and some local drops in configurations where few images are coupled on the output cone (angle of 120° or 130°). It can be seen that the best performances are obtained for angles close to 110° when the losses are low in the structure. For higher losses (bottom curves), the local drop effects diminish. Indeed, since the average propagation distance of the rays is limited by the losses, configurations requiring multiple reflections (and therefore large propagation distances) become less efficient in extraction.

[0075] Also, preferably, the angle a is between 95° and 160° to optimize the extraction efficiency. In order to further improve this extraction efficiency, the angle a is between 100° and 140°. In order to further improve this extraction efficiency, the angle a is between 100° and 120°. To obtain maximum extraction efficiency, preferably, the angle a is equal to 110° within ±1°. Through simulations, the inventors have determined that these angle a values ​​are advantageous regardless of the relative size of the facets or regardless of the number of facets.

[0076] According to the simulations, the criterion for choosing the angle a seems quite tolerant as long as the number of multiple images of the output cone associated with the FS face is sufficient. This number of multiple images depends on the index of the medium and the value of the angle a.

[0077] To illustrate this, Figures 7 and 8 are a planisphere representation of the exit cones using angular coordinates P is the angle with respect to the z axis and i is the angle in the xy plane, with respect to the y axis. The white areas between the cones correspond to the trapped rays L p . As a reminder, the angular portions L out correspond to the escape cones on the large faces FE1, FE2. The angular portion L s corresponds to the exit cone associated with the FS face. Finally, the portion L r corresponds to the cones recycled by the reflections on the facets FT1, FT2, FL1, FL2 and exiting through the face FS. Without the faces FT1 and FT2, this light L r would not reach the exit face, it would exit through the side faces or remain trapped in the concentrator. The exit beam L tot is therefore composed of the portions L s + L r .

[0078] As a non-imitative example, Figures 7 and 8 are given for the “optimized” configuration, with a diamond concentrator (index 2.4). Figure 7 corresponds to the embodiment where a = 120° and Figure 8 corresponds to the embodiment where a = 157.5°

[0079] In the embodiment of Figure 7, for angle a = 120°, the outlet cone L s has only 5 antecedents. Given the high index of the concentrator, the critical angle associated with this exit cone is small and the cone L s is small. In this case, the images of the output cone that form the recycled portion L r do not overlap and the extraction efficiency will be low. To choose the angle of In order to maximize the extraction efficiency, one criterion is the overlap of adjacent images of the output cone L s , the angle between two adjacent cone axes corresponding to the apex angle of the exit cone.

[0080] It is possible to extract the light on a "crown"; corresponding to the overlap of the exit cone associated with the FS face and its images that we see appearing in figure 8. Indeed, in the embodiment of figure 8, we combine 16 escape cones in the exit beam L tot Figure 8 shows that in this case, almost the entire crown is extracted.

[0081] It is also possible to maximize the extraction efficiency by using a lower angle a in order to ensure a rapid multiplication of the rays and thus not be limited by propagation losses. Indeed, as shown in Figures 5 and 6, the extraction efficiency is better for an angle of approximately 110°. This solution makes it possible to cover the entire crown in a few round trips. Indeed, with such an angle, the rays have 4 possibilities before returning to the exit face, depending on their position in the concentrator: they can be reflected on FT1 only, on FT2 only or on a combination FT 1 -FT2 or FT2-FT 1 .

[0082] An experimental validation of the concept was carried out with a PMMA CL concentrator (index n c= 1.5) doped with an orange dye (lumogen, Evonics) with an optimized geometry. The results obtained show an output power increased by 60% compared to a standard configuration for an angle a close to 120°. The loss coefficient estimated by simulation is 6.10 -2 cm' 1 .

[0083] The concept of the invention is therefore validated even though the experimental configuration is far from optimal because the chosen medium has a lot of losses (PMMA doped with a dye). Losses of the order of 10' 3 cm' 1 can be obtained in well-controlled crystals such as Ce:YAG. In this case, the performance will be even better and can reach the values ​​indicated in Figure 5, tripling the extraction compared to a standard configuration.

[0084] By way of non-limiting example, in the embodiment of Figures 3A and 3B, the concentrator comprises a number of lateral faces equal to 5 and the first facet FT1 and the second facet FT2 are adjacent and have identical dimensions. Alternatively, according to another embodiment illustrated in Figure 9, the first and second facets are adjacent and have different dimensions. This has the advantage of making the structure less symmetrical and therefore limiting the trapped rays that could propagate in the concentrator without meeting the exit face with an angle belonging to the exit cone.

[0085] According to another embodiment illustrated in Figure 10, the concentrator CL comprises a number of lateral faces equal to 6, including an additional facet FTA adjacent to the first facet FT1 and to the second facet FT2. To clarify the representation, the mirrors M1, M2 are not shown in Figure 10, although they are included in the device 1. The FTA facet makes it possible to promote a greater number of reflections of the rays of the portion L r and therefore an exit through the FS face on average faster rays. The FTA facet therefore potentially makes it possible to increase extraction efficiency by reducing propagation losses associated with the L beam out . The angles a and a2 (defined in Fig. 10) being linked to a by the relation a + n = a + a2, we note that it is always the angle a which conditions the extraction efficiency of the structure of the embodiment of figure 10.

[0086] According to another embodiment, the concentrator comprises a number of lateral faces greater than 6, including a plurality of additional facets. Each of these additional faces is adjacent to two other additional facets, or adjacent to another additional facet and to the first facet FT1 or adjacent to another additional facet and to the second facet FT2. Like the additional facet FTA of the embodiment of FIG. 10, these additional facets potentially make it possible to increase the extraction efficiency by reducing the propagation losses associated with the beam L out .

[0087] According to an embodiment compatible with all the embodiments recently discussed, the first lateral face FL1 and the second lateral face FL2 are not parallel. Depending on the angle a chosen, this non-parallelism makes it possible to increase the number of images of the output cone produced by reflection on the lateral faces. This therefore potentially makes it possible to increase the extraction efficiency in cases where it is not maximized by the angle a.

[0088] In structures with pumping by L radiation d optical, it is necessary to leave the large pumping faces FE1, FE2 free of access. In addition, it is difficult in practice to put dichroic mirrors on these large faces FE1, FE2 because it is necessary to reflect with an excellent coefficient (R>99.9%) multiple incidences and wide spectra while transmitting the pump radiation. Also, the exhaust cones L outon the large faces are irrecoverable. The losses through these cones are linked to the index. Table 1 below shows that for high indices (n=1.83 or n=2.4 for example), the crown extracted in the air is far from the total potential (i.e. the totality of the rays emitted less the two cones L out ). Indeed, the thickness of the crown is linked to the angle of the exit cone which is also the angle of total internal reflection (0 TIR ). Table 1

[0089] In order to increase this critical angle 0 TIR , it is preferable to use an output medium with a higher index than air, when the concentrator index is high.

[0090] Figure 11 illustrates a preferred embodiment of the invention in which the device comprises an output medium MS in contact with the output face and having an index n s such that n c > n s> 1. According to the embodiment illustrated in figure 11, the device comprises a first additional mirror MT arranged so as to cover the first lateral face FL1 and a second additional mirror M2' arranged so as to cover said second lateral face FL2. These mirrors MT, M2' are optional and make it possible to increase the extraction efficiency by preventing luminescence radiation from passing through the faces FL1, FL2 before exiting through the exit face FS (so-called optimized configuration).

[0091] The MS outlet medium increases the extraction efficiency by increasing the critical angle associated with the outlet cone of the FS face, which forms a wider outlet cone. This embodiment is of particular interest when the concentrator index is high. This effect is illustrated in the "Glue crown extraction" row of Table 1, which shows the extraction efficiency of the configuration in Figure 11. It should be compared to the "Air crown extraction" row, which shows the extraction efficiency of the configuration in Figures 3A and 3B with an outlet medium that is air. It is possible to increase the extraction efficiency from 42% to 71% in a concentrator with an index n c = 2.4 thanks to an MS output medium with an index of 1.7 for example.

[0092] Preferably, as illustrated in figure 11, the device comprises an optical system SO attached to the output medium adapted to collimate rays of the output beam L tot. The optical system SO is a non-imaging optic that will straighten the rays exiting the output medium MS and allow them to exit into the air. In this embodiment, the output medium MS is for example a glue allowing the mechanical and optical connection between the concentrator CL and the optical system SO.

[0093] Extraction will be limited by the index of the glue between the structure and the non-imaging optics. A glue with the same index as the middle of the SO optical system would be ideal, however, high index glues (n>1.7) are difficult to find today.

[0094] Figure 12 schematically illustrates an embodiment of the invention in which the device comprises: - a first additional mirror MT arranged so as to cover the first lateral face FL1, - a second additional mirror M2' arranged so as to cover the second lateral face FL2 - a third additional mirror M3 arranged so as to cover the output face FS

[0095] Furthermore, the concentrator CL has a beveled edge AB between the output face FS and the first side face FL1 (or the second side face FL2) so that a portion L r , of the trapped portion L p crosses the beveled edge AB. The rays of the portion L form an output beam called additional L totl . In the configuration of figure 12, the passage through the edge AB is the only solution for the rays to exit the structure (outside the pumping faces). It therefore forces the rays to reflect many times before reaching the right position and angle. This effect imposes a strong superposition of the rays inside, which can be described as both angular and spatial recycling. The use of an AB edge increases the propagation distance, and potentially reduces the overall extraction, but it increases the luminance.

[0096] Preferably, the normal of the beveled edge is parallel to the sum of the normals of the two lateral faces FL1, FS adjacent to the beveled edge at ±5°.

[0097] The embodiment of Figure 11 and Figure 12 are combinable. That is to say, according to one embodiment of the invention, an output medium MS is attached to the beveled edge AB to maximize the luminance of the additional output beam L totl and maximize extraction efficiency.

[0098] Figure 13 schematically illustrates an embodiment of the invention in which the device comprises: - a first additional mirror MT arranged so as to cover the first lateral face FL1, - a second additional mirror M2' arranged so as to cover the second lateral face FL2 - at least one additional mirror M3, arranged so as to partially cover the output face FS by defining a surface not covered SFS by the mirrors M3.

[0099] As explained previously, the MT, M2' mirrors make it possible to increase the extraction efficiency by preventing the luminescence radiation from passing through the FL1, FL2 faces before exiting through the SFS surface.

[0100] The M3 mirror with partial coverage of the output face allows an increase in illumination on the SFS surface. The dimensions of the emission module are adapted so that a ratio R = SFS / S SFbetween the uncovered surface SFS and the surface of the face FS is adapted so that rays of the luminescence radiation are reflected on M3 and propagate on average over an average distance L moy » L within the CL concentrator before crossing the SFS surface. The rays of luminescence radiation leaving the concentrator form an output beam L tot . By L moy » L, we hear here that L moy is greater than 7 times preferably 15 times L.

[0101] A suitable ratio R to verify condition L moy» L ensures that the effect of confining the rays in the 3 dimensions of space, that is to say that the luminescent radiation, instead of directly crossing the SFS surface after being generated, is mainly reflected within the CL concentrator and makes on average several round trips within the CL concentrator before exiting through the SFS surface. The smaller the ratio R, the greater the average distance L moy traveled by the luminescence rays within the CL concentrator before exiting increases. Thus, the illumination on the exit face is increased because the luminescence radiation has time to "fill" the CL concentrator before exiting. The inventors noticed that when the ratio R is less than or equal to 1 / 4, or preferably, less than or equal to 1 / 8 the average distance L moyis sufficient to achieve a significant concentrating effect. Also, according to a preferred embodiment of the invention, the ratio R is less than or equal to 1 / 4, or preferably, less than or equal to 1 / 8.

[0102] The structures of the devices detailed in this description and the figures are given by way of example and are intended to illustrate the invention. They should in no way be considered as limiting the scope of the invention. Various modifications and variations of the described structures, manufacturing methods, and applications and uses thereof will occur to those skilled in the art without departing from the scope and spirit of the invention.

Claims

Claims 1. Light emitting device comprising: - a light concentrator (LC) in a solid fluorescent or scintillator material having an index n c and adapted to absorb incident light (Ld) then emit luminescent radiation (LL) within said concentrator, said concentrator comprising at least seven faces, two parallel faces called large faces (FE1, FE2), and at least five so-called lateral faces (FL1, FL2, FS, FT1, FT2) including: - a first lateral face (FL1) and a second lateral face (FL2) having a surface area greater than a surface area of ​​the other lateral faces, - a third lateral face (FS), called the exit face, adjacent to the first lateral face and to the second lateral face, a portion (L s), called the outgoing portion, of the rays of said luminescent radiation (LL) being guided by total internal reflection within the concentrator by crossing said exit face (FS) - a lateral face called the first facet (FT1) adjacent to the first lateral face and a lateral face called the second facet (FT2) adjacent to the second lateral face, - a first mirror (M1) arranged so as to cover said first facet (FT1), - a second mirror (M2) arranged so as to cover said second facet (FT2), the first facet and the second facet forming an angle a greater than 0 and less than 175° so that a portion (L r ), called the recycled portion of the luminescent radiation, passes through said output face (FS) after at least one reflection on said first or second facet or on said first or second mirror, the outgoing portion (L s ) and the recycled portion (L r) forming an output beam 2. Device according to claim 1, wherein said angle a is adapted according to the dimensions and an arrangement of the facets in order to maximize an extraction efficiency of the output beam.

3. Device according to claim 1 or 2, in which the angle a is between 95° and 160°.

4. Device according to any one of the preceding claims, in which the angle a is between 100° and 140°.

5. Device according to any one of the preceding claims, in which the angle a is between 100° and 120°.

6. Device according to any one of the preceding claims, in which the angle a is equal to 110° at ±1°.

7. Device according to any one of the preceding claims, in which the concentrator comprises a number of side faces equal to 5, the first and second facets being adjacent.

8. Device according to any one of claims 1 to 6, in which the concentrator comprises a number of lateral faces equal to 6, including an additional facet (FTA) adjacent to the first facet and to the second facet.

9. Device according to any one of claims 1 to 6, in which the concentrator comprises a number of lateral faces greater than 6, including a plurality of additional facets, each additional facet being adjacent to two other additional facets, or adjacent to another additional facet and to the first facet or adjacent to another additional facet and to the second facet.

10. Device according to any one of the preceding claims, in which the large faces have a surface area greater than a surface area of ​​the other faces.

11. Device according to any one of the preceding claims, comprising a first additional mirror (MT) arranged so as to cover said first lateral face and a second additional mirror (M2') arranged so as to cover said second lateral face.

12. Device according to the preceding claim, in which the first lateral face and the second lateral face are not parallel.

13. Device according to any one of the preceding claims, in which the index n c of the concentrator is greater than 1.5, the device comprising an output medium in contact with the output face and having an index n s such that n c > n s > 1.

14. Device according to the preceding claim, comprising an optical system (SO) attached to the output medium adapted to collimate rays of the output beam.

15. Device according to any one of the preceding claims in combination with claim 11, wherein the device comprises a third additional mirror arranged so as to cover said output face, and wherein the concentrator has a beveled edge (AB) between the output face and the first lateral face so that a portion (L r ), called additional recycled portion, of said trapped portion (L p ) crosses said beveled edge (AB) forming an output beam called additional (L tot ,).

Images