![]() IMPROVED BI-SPECTRAL DETECTOR
专利摘要:
The invention relates to an optical detector (10) sensitive in at least two infrared wavelength ranges called first spectral band (SB1) and second spectral band (SB2), and having a set of pixels, comprising: - an absorbent structure (Sabs) disposed on a lower face (11) of a substrate (Sub) and comprising a stack of at least one absorbent layer (AL, AL1, AL2) of semiconductor material, the detector further comprising a plurality dielectric resonators (Res) on the upper face (12) of said substrate forming an upper surface called meta-surface (13), the meta-surface being configured to diffuse, deflect and focus in the pixels of the detector in a resonant manner, when illuminated by incident light (IL), a first beam (FL1) having at least a first wavelength (λ1) included in the first spectral band (SB1) and a second beam (FL2) having at least a second wavelength (λ2) included in the second band, the meta-surface being further configured so that said first (FL1) and second (FL2) beams are focused on pixels different from the detector. 公开号:FR3083645A1 申请号:FR1800717 申请日:2018-07-05 公开日:2020-01-10 发明作者:Alexandre Delga;Jean-Luc Reverchon 申请人:Commissariat a lEnergie Atomique CEA;Thales SA;Commissariat a lEnergie Atomique et aux Energies Alternatives CEA; IPC主号:
专利说明:
Improved Bi-spectral detector FIELD OF THE INVENTION The invention relates to an improved bi-spectral detector, for the fields of infrared IR imaging. More particularly, the invention relates to a bi-spectral detector, that is to say sensitive in two IR spectral bands, for example in the MWIR band (Medium Wavelength Infra Red in English) typically between 3 pm and 5 pm and in the LWIR band (Long Wavelength Infra Red in English) typically between 8 pm and 12 pm. STATE OF THE ART In thermal infrared imaging, there are mainly two wavelength ranges benefiting from the transmission of the atmosphere: -the infrared or MWIR medium from 3 to 5pm, itself subdivided into two windows on either side of the CO2 absorption band. The emission of the black body varies greatly with temperature and the transmission is less disturbed by humidity. -the far infrared or LWIR from 8 to 12pm. It corresponds to the emission peak of a black body at room temperature. It is not very sensitive to light scattering phenomena. Each of its two windows therefore has advantages depending on operational conditions (maritime, coastal, desert area, etc.) beyond the intrinsic qualities of the imagers making it possible to address these two wavelength ranges (uniformity, detectivity). ..). Also, an offer based on a MWIR / LWIR bi-spectral camera was proposed in the 2000s. According to a first technology called Bayer matrix, the detector comprises a semiconductor layer absorbing simultaneously in the two IR bands, and interference filters are placed on the pixels to detect an LWIR image lmi_ and an MWIR image Imw- The realization of these filters is complex, sensitivity and resolution are low, and the two images lm L and lm M recovered by the detector do not exhibit spatial coherence. By spatial coherence between the two images is meant the fact that the light incident on the same pixel of each image comes from the same area of the scene. The pixels LWIR and MWIR being here adjacent, there can be no spatial coherence between the two images. The detection of two coherent images has the advantage of being able to precisely compare the MWIR and LWIR signature of an element of the scene which sometimes extends only over a few pixels. A second technology for maintaining spatial coherence is based on a detector comprising two superposed absorbent layers, one sensitive in the LWIR and one in the MWIR, as illustrated in FIG. 1. In this example, each stage consists of a QWIP (for “Quantum Well Infrared Photodetector” in English) based on GaAs / AIGaAs. The pixel has a mesa structure corresponding to an etching of the active layers to delimit the pixel, these then being in contact with the air. It is appropriate in this type of structure to take a contact by layer (Clwir and Cmwir) and a common contact CC. Preferably, these three contacts are carried on the same plane, to facilitate connection to the reading circuit. QWIP technology is an inter-band (ISB) technology in which the photo-carriers are generated from one sub-band of the conduction band to another sub-band of the conduction band. The main problems encountered are poor optical coupling of the MWIR stage, compromising the critical quantum efficiency in the MWIR range. Problems of crosstalk through the common contact complicated the realization. In general, the implementation of two contacts is difficult at the matrix level for these two-stage structures. For a mesa pixel, the contact between the active layer and the air creates a dark current, this being particularly critical for detectors based on inter-band detection, that is to say by generation of photocarriers of the valence band to the conduction band. An example is a type II super-array sensor on GaSb having a mesa structure on at least one level. For detectors based on inter-band detection, it is therefore necessary to passivate the mesa structure, which poses technological problems. Thus, none of the approaches for bi-spectral IR imaging according to the state of the art has really succeeded in imposing itself for reasons of complexity of the optics and the detector. An object of the present invention is to overcome certain aforementioned drawbacks by proposing an improved bi-spectral detector having a simplified technology, better resolution and / or FTM (Modulation Transfer Function), and / or better gain and / or better sensitivity. DESCRIPTION OF THE INVENTION The subject of the present invention is an optical detector sensitive in at least two infrared wavelength ranges called first spectral band and second spectral band, and having a set of pixels, comprising: an absorbent structure arranged on a lower face of a substrate and comprising a stack of at least one absorbent layer of semiconductor material, sensitive in the two spectral bands and able to photogenerate carriers by absorption of an incident beam on an upper face of said substrate, the absorbent structure being further connected to a reading circuit by at least one set of contacts associated with said pixels, the detector further comprising a plurality of dielectric resonators on the upper face of said substrate forming an upper surface called meta-surface, the meta-surface being configured to diffuse, deflect and focus in the pixels of the detector in a resonant manner, when illuminated by the incident light, a first beam having at least a first wavelength included in the first spectral band and a second beam having at least a second wavelength included in the second band, the meta-surface being further configured so that said first and second beams are focused on pixels different from the detector. Advantageously, the first spectral band is included in the 3-5 µm band and the second spectral band is included in the 8-12 µm band. According to a first variant, the stack comprises a single sensitive absorbent layer in both the first and the second spectral band, the first beam being focused on a first subset of pixels called first pixels, and the second beam being focused on a second subset of pixels called second pixels, said first and second pixels being arranged in the same plane. According to a second variant, the stack comprises a first sensitive absorbent layer in the first spectral band and arranged in a first plane, and a second sensitive absorbent layer in the second spectral band and arranged in a second plane different from the first plane, the first layer absorbent being closest to the substrate, the first beam being focused substantially in the foreground on a first subset of pixels called first pixels, and the second beam being focused substantially in the second plane on a second subset of pixels called second pixels . According to one embodiment, the second absorbent layer is discontinuous and etched so that a second pixel has a mesa structure. Preferably, the second absorbent layer performs inter-band detection. According to one embodiment, a second pixel comprises a diffraction grating configured to diffract in reflection the second beam. According to a sub-variant, the first absorbent layer is continuous. Advantageously, the first absorbent layer (AL1) performs inter-band detection. According to one embodiment, each first pixel (Pix1) is connected to a first contact and each second pixel is connected to a second contact, and a mesa structure of a second pixel out of two is used as a pillar so as to report the first contacts on substantially the same horizontal plane as the second contacts. According to one embodiment, each first pixel is connected to a first contact and each second pixel is connected to a second contact, and the first absorbent layer is discontinuous and etched so that each first pixel has a mesa structure serving as a basis for the second pixel mesa structure. According to one embodiment, a mesa structure of a second pixel out of two is used as a pillar (15) so as to relate the first contact (CL1) to substantially the same horizontal plane as the second contact (CL2). According to one embodiment, a first meta-lens is defined as the fraction of the meta-surface focusing the first beam in a first associated pixel, and a second meta-lens as the fraction of the metasurface focusing the second beam in a second pixel. associated, and wherein the meta-surface is configured so that the first and second meta-lenses are offset from each other. According to another embodiment, a first meta-lens is defined as the fraction of the metasurface focusing the first beam in a first associated pixel, and a second meta-lens as the fraction of the metasurface focusing the second beam in a second associated pixel. , and wherein the meta-surface is configured so that the first and second meta-lenses are superimposed. Preferably the sum of the respective areas of a first and a second pixel is less than or equal to the area of a meta-lens. Other characteristics, objects and advantages of the present invention will appear on reading the detailed description which follows and with reference to the appended drawings given by way of nonlimiting examples and in which: FIG. 1, already cited, illustrates a bi-spectral detector with a mesa pixel according to the state of the art. FIG. 2 illustrates a first variant of a detector according to the invention at a single level combined with an arrangement of meta-lenses in offset mode. FIG. 3 illustrates examples of forms of resonators according to the invention, FIG. 3a illustrates a resonator of round shape and FIG. 3b illustrates a resonator of square shape. FIG. 4 illustrates an arrangement of an interlaced type meta-lens applied to the first variant of the detector according to the invention, seen from above. FIG. 5 illustrates a first variant of detector according to the invention at a single level combined with an arrangement of meta-lenses in superimposed mode. FIG. 6 illustrates an arrangement of meta-lens of superimposed type of the detector according to the invention, combined with an arrangement of “diluted” type of pixels, in top view. FIG. 7 illustrates the sub-variant with a first continuous absorbent layer of a second variant of detector according to the invention on two levels, combined with an arrangement of the offset type of the meta-lenses and having a simplified contact geometry, all the contacts being reported on the same plane. FIG. 8 illustrates the sub-variant with first continuous absorbent layer of the second variant of detector according to the invention at two levels, combined with an arrangement of superimposed type of meta-lenses and having a simplified contact geometry, all the contacts being added on the same plane. FIG. 9 illustrates the sub-variant with a first continuous absorbent layer of the second variant of detector according to the invention on two levels, combined with an arrangement of superimposed type of meta-lenses and having an arrangement of “dense” type of pixels. FIG. 10 illustrates an arrangement of meta-lens of the superimposed type of the detector according to the invention, combined with an arrangement of “dense” type of pixels, in top view. FIG. 11 illustrates the sub-double mesa variant of the second variant of detector according to the invention on two levels, combined with an arrangement of superimposed type of meta-lenses and having an arrangement of “diluted” type of pixels. DETAILED DESCRIPTION OF THE INVENTION The invention relates to an optical detector 10 sensitive in at least two infrared wavelength ranges known as the first spectral band SB1 and the second spectral band SB2, commonly known as a bi-spectral detector. According to a preferred embodiment, the first spectral band SB1 is included in the 3-5 pm MWIR band and the second spectral band SB2 is included in the 8-12 pm LWIR band. And other bi-spectral configurations are possible, such as: -MWIR "blue" around 4.2 pm and MWIR "red" around 4.8 pm, that is to say on either side of the CO 2 absorption band (to facilitate the detection of hot spots for example) , -LWIR [8-1 Opm] and LWIR [10-12pm], - MWIR and near infrared (known as SWIR). The bi-spectral detector according to the invention comprises a set of pixels, preferably arranged in the form of a matrix, subdivided into a first subset of pixels called first pixels Pix1 configured to detect a light signal in SB1, and a second subset of pixels referred to as second Pix2 pixels configured to detect a light signal in SB2. Different variants of the detector 10 according to the invention are illustrated in Figures 2, 5, 7, 8, 9 and 11 which will be described separately below. It comprises an absorbing structure Sabs disposed on a lower face 11 of a Sub substrate comprising a stack of at least one absorbing layer of semiconductor material, also called active layer, the stack being sensitive in the two spectral bands SB1, SB2 and able to photo-generate carriers by absorption of an incident light beam IL on an upper face 12 of the substrate Sub. Typically the thickness of the substrate is between a few tens to a few hundred pm and the material of the substrate is chosen from: InP, GaAs, GaSb, Si, SiC, Diamond. The absorbent structure typically comprises, in addition to the active layer or layers, several other layers (not shown) such as carrier injection layers 5. An active layer can consist of a large number of sublayers, such as quantum multi-then or super-networks. Examples of semiconductor material for the absorbent layer are: InGaAs; InAsSb; InAsSb / lnAs super network; InSb; GaSb / lnAs super network; GaN; AIGaN, AIGaAsSb; GaAs / AIGaAs or InGaAs / AllnAs quantum multi-wells, InAs / AISb superlattice. In a first variant illustrated in Figures 2 and 5, the stack comprises a single absorbent layer AL sensitive both in the first spectral band SB1 and the second spectral band SB2. In a second variant illustrated in FIGS. 7, 8, 9 and 11, the stack comprises a first absorbent layer AL1 sensitive in the first spectral band SB1 disposed in a first plane PL1, and a second absorbent layer AL2 sensitive in the second spectral band SB2 arranged in a second plane PL2 different from the first plane. The two planes PL1 and PL2 are superimposed. By convention, the first absorbent layer AL1 is closest to the substrate. The absorbent structure is also connected to a reading circuit (not shown in the figures) via at least one set of contacts associated with the pixels, and there is also a contact common to all the pixels CC, for example consisting of a heavily doped area at the end of the substrate located on the side of the absorbent layer AL or AL1. The detector according to the invention further comprises a set of dielectric resonators Res on the upper face 12 of the substrate Sub forming an upper surface called meta-surface 13. When illuminated by the incident beam IL, the meta-surface 13 is configured to diffuse in a resonant manner and to focus in the pixels of the detector, a first light beam FL1 having at least a first wavelength A1 included in the first spectral band SB1 and a second light beam FL2 having at least a second wavelength Λ2 included in the second spectral band SB2. The meta-surface is further configured so that the first beam FL1 and the second beam FL2 are focused on pixels different from the detector, it spatially separates the two wavelengths λ1, A2. The first beam FL1 is focused on a first subset of pixels made up of first pixels Pix1, and the second beam FL2 is focused on a second subset of pixels made up of second pixels Pix2. The meta-surface 13 therefore makes it possible to focus the incident flux on the desired pixels. For the first variant of the detector according to the invention illustrated in FIGS. 2 and 5, the first and second pixels are arranged in the same plane, the detector has only one level. Spectral selectivity operates by "photon sorting" using the meta-surface. Contacts CL1 associated with pixels Pix1 and contacts CL2 associated with pixels Pix2 are defined. The read circuit reconstructs, from the incident beam IL coming from a scene, an image Im1 to SB1 coming from pixels Pix1, and an image Im2 to SB2 coming from pixels Pix2, from electrical signals coming from the arriving photo-carriers respectively on contacts CL1 and CL2. For the first variant, the layer AL being sensitive in the two spectral bands, it is the meta-surface 13 which determines the pixels Pix1 which detect the image Im1 and the pixels Pix2 which detect the image Im2. The resonators Res of the meta-surface 13 are either etched directly into the Sub substrate (in this case they have no difference in index with the substrate) or produced on another surface attached to the substrate. Generally, this set of resonators forms a Huygens meta-surface which acts on the incident light so as to perform one or more optical functions. These elements constitute a monolayer of dielectric nano-antennas which function in a resonant manner, enhancing the response in the vicinity of a resonance frequency A R (spectral width of the resonance centered on Ar) while cutting the light flux around this spectral width of resonance. Each resonator is a radiating dipole or Huygens dipole and a large number of antenna modes are accessible by varying the shape, size and arrangement of these dipoles. They make up an assembly of nano-antennas whose dimensions and spacing are of the order of magnitude of the resonant wavelength, or sub-wavelength. Each resonator prints a phase specific to the incident signal, phase controlled in both SB1 and SB2. More generally, each nano-antenna can be seen as a local value of surface impedance at the two wavelengths A1 and A2. The resonators act as converging microlenses allowing the incident light to be focused in a spectral band around the resonant wavelength, as close as possible to the contact of the pixel. The effective profile of the lens is controlled by the arrangement of the antennas. Compared to optical structures of the effective index type (digital Fresnel lens type) the physical principle differs: the wavefront is modeled 20 due to a phase imparted abruptly, over a very short distance before the wavelength , and not accumulated along the optical path. An incident photon on a point of the microlens is captured by a precise nano antenna, interacts with it and is re-emitted with a given phase. The accessible lens profiles are more varied than for the effective index structures and the focal length of the lens, which can be very short, is very precisely controlled. The meta-surface of the invention is configured to resonate simultaneously at A1 and A2 (with a spectral width around these two wavelengths respectively). The ability of a single meta-surface to focus, that is to say to resonate simultaneously on several wavelengths has been demonstrated for telecom wavelengths, with co-focusing of 3 wavelengths in the publication Aieta et al : "Multiwavelength achromatic metasurfaces by dispersive phase compensation" Science, 347 35 (6228), 1342-1345 (2015). The focusing is carried out by means of an assembly of hybrid resonators formed by several coupled sub resonators. Thus, the coupling engineering makes it possible to obtain the desired impedance profile at several resonant wavelengths. The dimensions of the resonators are smaller than the dimensions of the pixels of the detector. The first meta-lens ML1 denotes the fraction of the metasurface 13 focusing the first beam FL1 in a first associated pixel Pix1, and the second meta-lens ML2 the fraction of the meta-surface focusing the second beam FL2 in a second associated pixel Pix2. ίο For example in Figures 2 and 5 the arrows 21 and 22 respectively illustrate the dimension of the pupil of ML1 and ML2. Each metal lens is made up of a plurality of resonators, typically at least 5 × 5 resonators. Typically at least one dimension of a resonator Res, chosen from width, length, height, is in the range [AT R / 2n - 50%; At R / 2n + 50%] or in the interval [At R / n - 50%; At R / n + 50%], with: At R resonance wavelength = Λ1 or Λ2, n index of the material in which the resonators are engraved. FIG. 3 illustrates two examples of basic patterns of resonators Res according to the invention. Typically they have a circular (Figure 3a) or rectangular (Figure 3b) shape. They can be in the form of walls, the center being hollow, or solid. According to the desired optical function, sub-patterns can be engraved in the basic pattern constituting the resonator, like the holes in the center 25 of the structures of FIGS. 3a and 3b, constituting sub-patterns. Thanks to the meta-surface, the phenomena of optical and electrical crosstalk (“cross talk”) are drastically limited, the sensitivity of the sensor is increased or a reduction in the size of the pixels is allowed while maintaining an acceptable sensitivity. In addition, the use of resonators according to the invention makes it possible to produce the microlenses with a simple technology in a single step, and the patterns of which are wider than those necessary for digitized Fresnel lenses. For the first variant illustrated in FIGS. 2 and 5, thanks to the meta-surface according to the invention the dielectric filters are eliminated, which greatly simplifies the method of manufacturing the detector. In addition, the detector thus produced has only one level of contact, hence a greatly simplified manufacturing 5. For the sensitive AL absorbent layer in SB1 and SB2, with SB1 included in MWIR and SB2 included in LWIR, depending on one option, a semiconductor material is used with a "gap" in LWIR, while taking care that the substrate does not absorb in LWIR or in MWIR. It is thinned if necessary. The sensitivity of the detector layer AL covers SB1 - SB2 and the range between the two (so-called broadband layer AL). For example, it is performed in MCT or with a super-network. According to another option, the sensitivity of the layer is bi-spectral, obtained for example with a quantum cascade detector called QCD (for “Quantum Cascade Detector” in English) bi-spectral, or with a hybrid QW / QCD structure. Here, the non-resonant wavelengths, not focused, pass through the layer without generating a background signal because they are not "seen" by the layer AL. The meta-lenses ML1 form a first subset of meta-lenses constituting a first two-dimensional network, and the meta-lenses ML2 form a second subset constituting a second two-dimensional network of meta-lenses. The two networks can work together in different ways. According to an embodiment illustrated in FIG. 2, the meta-lenses ML1 and ML2 are arranged in an offset mode. A top view of an example of an offset (interlaced) arrangement adapted to the first variant at one level is illustrated in FIG. 4. In this example the pixels are square, the meta-lenses have pupils which are substantially square and offset by 45 ° with respect to the squares forming the pixels, and the first and second arrays of meta-lenses are shifted along a dimension of the square of the pixels by a distance d corresponding to the distance 35 between two pixels in this same dimension. No incident flow is lost, and the loss of spatial resolution is half that of the “diluted” configurations described below. According to an embodiment illustrated in FIG. 5, the meta-surface is configured 5 so that the first and second meta-lenses are superimposed. The ML1 and ML2 meta-lenses then work off-axis and are nested one inside the other. In this case the light reaching two pixels Pix1 and Pix2 adjacent and respectively associated with ML1 and ML2 superimposed comes from the same point of the scene. We then say that there is spatial coherence between the two images Im1 and Im2 respectively detected by the first subset of first pixels and the second subset of second pixels. The spatial coherence between the two images im1 and Im2 makes it possible to detect the bi-spectral signature of very small areas in the scene. FIG. 6 illustrates an arrangement between pixels Pix1, Pix2 and meta-lenses ML1, ML2 compatible with an overlapping arrangement of meta-lenses. This arrangement is called "diluted" because the area of the pixels is smaller than the area of a meta-lens. Points 60 and 61 schematically represent the focus of ML1 and ML2 respectively. Typically the sum of the respective areas of a first and a second pixel is less than or equal to the area of a meta-lens. The fact of producing a pixel with an "electrical" surface (detection surface) smaller than the "optical" surface (pupil) of the meta-lens has the advantage of increasing the signal-to-noise ratio when the component is limited. by dark noise: the signal corresponding to the optical surface, and the noise to the electrical surface. According to the second variant of the detector according to the invention, the layers AL1 and AL2 are separated, superimposed, and located in two different planes PL1 30 and PL2 (vertical stacking). The sensitivity of AL1 and AL2 is then optimized for SB1 and SB2 respectively. The beam FL1 is focused substantially in PL1 on the first subset of pixels Pix1 and FL2 is focused substantially in PL2 on the second subset of pixels Pix2. The focal lengths of the meta35 lenses ML1 and ML2 are adjusted to focus in the associated layer and are therefore different. Each layer AL1, AL2 has an optimized sensitivity independently of the other and is therefore more efficient than a broadband or bi-spectral AL layer. In this second variant it is necessary to make two levels of contact Preferably, the second absorbent layer AL2 is discontinuous and etched so that a second pixel Pix2 has a mesa structure, as illustrated in FIGS. 7, 8, 9 and 11 to allow intermediate contact to be made. For this type of structure, it is preferable for the second absorbent layer AL2 to carry out an inter-band detection known as ISB which is easy to passivate, for example of the QWIP or QCD type. According to one embodiment, each second pixel Pix2 comprises a diffraction grating DG2 configured to diffract in reflection the second beam FL2. This network is located on the top of the mesa on the side opposite to that by which the light flux arrives, that is to say on the side of the reading circuit (not shown in the figures). According to known operation, these networks are configured to diffract an evanescent wave which is absorbed by the layer AL2. For SB2 corresponding to LWIR, and for a detector on GaSb substrate, typically an AL2 layer in QCD is preferred to QWIP because of a strong offset of the conduction band. According to a first sub-variant of the second variant, the first absorbent layer AL1 is continuous, as illustrated in FIGS. 7 to 9. The technology is then simplified and it is not necessary to choose a layer AL1 compatible with a mesa structure . Preferably, the first absorbent layer AL1 performs inter-band detection. For SB1 corresponding to MWIR, and a detector on a GaSb substrate, typically a layer AL1 of the planar nBn InAsSb type is preferred. Note that this configuration can work without meta surface 13. This type of continuous AL1 / AL2 mesa architecture makes it possible to make a simplified contact, as illustrated in FIGS. 7 and 8. According to one embodiment, a mesa structure of a second pixel Pix2 out of two is used as pillar 15 on which is deposited an insulating layer 17 then a contact CL1. The insulating layer 17 isolates the mesa from the contact CL1. This mesa 15 structure is therefore not used as a Pix2 detection pixel, it serves as a support for a CL1 contact. An intermediate contact Cint is made between AL1 and AL2, and the contact CL1 is made with the layer AL1 through the intermediate contact Cint via a hole 19. The contact CL1 therefore extends from the top of the mesa 15 to the hole 19 to layer AL1. The pixel diode detection pixel Pix1 is therefore made between the common contact CC and CL1, while the pixel pixel detection diode is made between the intermediate contact Cint and CL2, the contact Cint acting as common contact for Pix2 pixels. Recall that each pixel of the matrix must have a common contact and an individual contact. The first contacts CL1 are thus reported on substantially the same horizontal plane as the second contacts CL2, which greatly simplifies the connection of the detection part of the detector to the reading circuit (standard hybridization). This configuration of the contacts is made possible by the selective focusing carried out by the meta-surface 13, without it the electrical signal would come from an area that is too large, generating unacceptable crosstalk. The first sub-variant is compatible with an arrangement of the metal lenses offset as illustrated in FIG. 7 and an arrangement of the metal lenses superimposed as illustrated in FIG. 8. In this last configuration of FIG. 8, there is therefore spatial coherence between the two images. detected, and an arrangement of the “diluted” type of the pixels as illustrated in FIG. 6 is recommended. According to another embodiment of the first sub variant AL1 continuous / AL2 in mesa all the mesa of Pix2 are used as detection pixel and covered with a contact CL2, as illustrated in FIG. 9. In this case the contact CL1 remains located at the bottom of the inter-mesa space 23, and it will be necessary to come and seek this contact in the same way as what is done for the detector of the state of the art illustrated in FIG. 1 (this ascent of the contact does not is not shown). However, this architecture has the advantage of presenting the highest spatial resolution of the various variants of the invention, while guaranteeing an improved FTM compared to the state of the art. This architecture combined with the superimposed mode of meta-lenses is compatible with a so-called dense pixel architecture illustrated in FIG. 10 in which the dimension of Pix1 and Pix2 is substantially equal to that of meta-lenses, lenses and pixels all being superimposed. More precisely, the places where the carriers are collected are offset between Pix1 and Pix2, but the electrical contacts defining the electrical pixels are superimposed. According to a second sub-variant of the second variant, the first absorbent layer AL1 is discontinuous and etched so that each first pixel Pix1 has a mesa structure serving as a basis for the mesa structure of the second pixel Pix2. This double mesa structure is illustrated in FIG. 11. Here the layer AL1 is preferably also of the ISB type like the layer AL2. This double mesa structure is compatible with a superimposed arrangement of the meta-lenses of the meta-surface 13 guaranteeing spatial coherence (illustrated in FIG. 11) and with an offset arrangement not shown. According to one embodiment, a mesa structure of a second pixel out of two is used as pillar 15 so as to bring the first contact CL1 onto substantially the same horizontal plane as the second contact CL2. The layer AL1 is divided into a subset of first mesas, but only one out of two first mesas 24 is used to make an operational pixel Pix1. The second mesa 15 structuring the layer AL2 and placed above this first mesa 24 then serves as a pillar for the contact CL1, and is therefore not a pixel for detection Pix2. According to an option, an operational pixel Pix1 comprises a diffraction grating DG1 configured to diffract in reflection the first beam FL1. The first mesa positioned below an operational Pix2 pixel is also not used for detection. The intermediate contact Oint between AL2 and AL1 is, for this double mesa, connected to the common contact via layer 18. The CL1 contact of an operational Pix1 pixel extends from the top of mesa 15 to the remaining free top of the first mesa serving as its base. The detection diode of a pixel Pix1 is therefore produced between the common contact CC and CL1, while the detection diode of a pixel Pix2 is produced between the intermediate contact Cint connected to the common contact and CL2. This double mesa structure is compatible with a “diluted” arrangement of the 5 pixels, and with a “dense” arrangement, the latter being however difficult to achieve. Indeed in this case it is necessary to lose a part of the surface of Pix2 to bring the contact Cint on the plane of CL2 at the level of each pixel. We come back to the hybridization difficulty mentioned in the state of the art.
权利要求:
Claims (14) [1] 1. Optical detector (10) sensitive in at least two infrared wavelength ranges called first spectral band (SB1) and second spectral band (SB2), and having a set of pixels, comprising: an absorbent structure (Sabs) arranged on a lower face (11) of a substrate (Sub) and comprising a stack of at least one absorbent layer (AL, AL1, AL2) of semiconductor material, sensitive in the two spectral bands (SB1, SB2) and able to photo-generate carriers by absorption of a beam (IL) incident on an upper face (12) of said substrate, the absorbent structure being further connected to a reading circuit by at least one set of contacts (CL1, CL2) associated with said pixels, the detector further comprising a plurality of dielectric resonators (Res) on the upper face (12) of said substrate forming an upper surface called meta-surface (13), the meta-surface being configured to diffuse, deflect and focus in the pixels of the detector resonantly, when illuminated by the incident light (IL), a first beam (FL1) having at least a first wavelength (A1) included in the first spectral band (SB1) and a second beam (FL2 ) having at least a second wavelength (A2) included in the second band, the meta-surface being further configured so that said first (FL1) and second (FL2) beams are focused on pixels different from the detector. [2] 2. Detectors according to claim 1, in which the first spectral band (SB1) is included in the 3-5 pm band (MWIR) and the second spectral band is included in the 8-12 pm band (LWIR). [3] 3. Detector according to one of the preceding claims in which the stack comprises a single absorbent layer (AL) sensitive both in the first and the second spectral band, the first beam (FL1) being focused on a first subset of pixels called first pixels (Pix1), and the second beam (FL2) being focused on a second subset of pixels called second pixels (Pix2), said first and second pixels being arranged in the same plane. [4] 5 4. Detector according to one of claims 1 or 2 wherein said stack comprises a first absorbent layer (AL1) sensitive in the first spectral band (SB1) and arranged in a first plane (PL1), and a second absorbent layer ( AL2) sensitive in the second spectral band (SB2) and arranged in a second plane (PL2) different from the foreground, the first absorbent layer being closest to the substrate, the first beam (FL1) being focused substantially in the first plane (PL1) on a first subset of pixels called first pixels (Pix1), and the second beam (FL2) being focused substantially in the second plane (PL2) on a second subset of pixels called second pixels (Pix2). 5. Detector according to the preceding claim wherein the second absorbent layer (AL2) is discontinuous and etched so that a second 20 pixel (Pix2) has a mesa structure. [5] 6. Detector according to the preceding claim wherein the second absorbent layer (AL2) performs inter-band detection (ISB). 25 [6] 7. Detector according to one of claims 5 or 6 in which a second pixel (Pix2) comprises a diffraction grating (DG2) configured to diffract in reflection the second beam (FL2). [7] 8. Detector according to one of claims 4 to 7 wherein the first absorbent layer (AL1) is continuous. [8] 9. Detector according to one of claims 4 to 8 wherein the first absorbent layer (AL1) performs inter-band detection. [9] 10. Detector according to one of claims 5 to 9 wherein each first pixel (Pix1) is connected to a first contact (CL1) and each second pixel (Pix2) is connected to a second contact (CL2), and in which a mesa structure of a second pixel out of two is used as a pillar (15) 5 so as to bring the first contacts (CL1) onto substantially the same horizontal plane as the second contacts (CL2). [10] 11. Detector according to one of claims 5 to 7 in which each first pixel (Pix1) is connected to a first contact (CL1) and each second ίο pixel (Pix2) is connected to a second contact (CL2), and in which the first absorbent layer (AL1) is discontinuous and etched so that each first pixel (Pix1) has a mesa structure serving as a basis for the mesa structure of the second pixel (Pix2). 15 [11] 12. Detector according to claim 11 in which a mesa structure of a second pixel out of two is used as a pillar (15) so as to relate the first contact (CL1) to substantially the same horizontal plane as the second contact (CL2) . 20 [12] 13. Detector according to one of the preceding claims, in which a first meta-lens (ML1) is defined as the fraction of the meta-surface focusing the first beam in a first associated pixel, and a second meta-lens (ML2) as the fraction of the meta-surface focusing the second beam in a second associated pixel, and in which the 25 meta-surface is configured so that the first and second meta-lenses are offset from each other. [13] 14. Detector according to one of claims 1 to 12 in which a first meta-lens (ML1) is defined as the fraction of the metasurface 30 focusing the first beam in a first associated pixel, and a second meta-lens (ML2) as the fraction of the metasurface focusing the second beam in a second associated pixel, and in which the meta-surface is configured so that the first and second metal lenses are superimposed. [14] 15. Detector according to one of claims 13 or 14 wherein the sum of the respective areas of a first and a second pixel is less than or equal to the area of a meta-lens.
类似技术:
公开号 | 公开日 | 专利标题 EP2180512B1|2018-12-12|Polarimetric imaging device optimised in relation to polarisation contrast EP2368273B1|2016-07-13|Photodetector having a very thin semiconducting region EP2324377B1|2013-06-05|Nanostructured spectral filter, image sensor and manufacturing method WO2020007622A1|2020-01-09|Improved bi-spectral detector EP3098858A1|2016-11-30|Photodetector with high quantum efficiency EP2180511B1|2018-08-29|Multi-spectrum imaging device based on quantum multi-wells FR2900279A1|2007-10-26|MICRO-STRUCTURE SPECTRAL FILTER AND IMAGE SENSOR FR2983297A1|2013-05-31|INFRARED DETECTOR BASED ON SUSPENDED BOLOMETRIC MICRO-PLANKS EP3153831A1|2017-04-12|Bolometer with high spectral sensitivity EP2477231A2|2012-07-18|Photodetector optimised by metal texturing on the rear surface EP2276072B1|2011-11-16|Photodetector element FR2917846A1|2008-12-26|NETWORK TYPE LIGHT RECEPTION DEVICE AND LIGHT COLLECTION METHOD. WO2017187029A1|2017-11-02|Multispectral imaging device EP2432033B1|2019-11-27|Multi-layer bispectral detector with photodiodes EP2865017B1|2019-11-13|Semiconductor structure with absorbing region in a focusing cavity FR2964795A1|2012-03-16|PHOTODETECTEUR AND CORRESPONDING DETECTION MATRIX EP3471151B1|2020-02-19|Helmholtz resonator photodetector EP3281217B1|2018-12-12|Multi-band photocathode FR3069956B1|2019-09-06|IMPROVED INFRARED DETECTOR EP2865015B1|2016-07-06|Photodetector element for infrared light radiation, and photodetector including such a photodetector element EP2801115B1|2017-02-15|Photodetection device EP0923141A1|1999-06-16|Photodetecting device, process of manufacturing and use for multispectral detection CA3105594A1|2020-02-06|Multispectral imaging sensor provided with means for limiting crosstalk
同族专利:
公开号 | 公开日 US11171168B2|2021-11-09| FR3083645B1|2020-07-31| IL279876D0|2021-03-01| WO2020007622A1|2020-01-09| EP3818566A1|2021-05-12| US20210288095A1|2021-09-16|
引用文献:
公开号 | 申请日 | 公开日 | 申请人 | 专利标题 EP2180511A2|2008-10-24|2010-04-28|Thales|Multi-spectrum imaging device based on quantum multi-wells| US20160322516A1|2013-12-17|2016-11-03|Centre National De La Recherche Scientifique - CNR S|Quantum detection element with low noise and method for manufacturing such a photodetection element| US6420728B1|2000-03-23|2002-07-16|Manijeh Razeghi|Multi-spectral quantum well infrared photodetectors| FR2868602B1|2004-04-05|2006-05-26|Commissariat Energie Atomique|PHOTON DETECTION CIRCUIT WITH MESA STRUCTURE| FR3083645B1|2018-07-05|2020-07-31|Thales Sa|IMPROVED BI-SPECTRAL DETECTOR|FR3083645B1|2018-07-05|2020-07-31|Thales Sa|IMPROVED BI-SPECTRAL DETECTOR| US11201993B1|2020-06-15|2021-12-14|Samsung Electronics Co., Ltd.|Multi-camera on a chip and camera module design|
法律状态:
2019-06-27| PLFP| Fee payment|Year of fee payment: 2 | 2020-01-10| PLSC| Publication of the preliminary search report|Effective date: 20200110 | 2020-06-25| PLFP| Fee payment|Year of fee payment: 3 | 2021-06-24| PLFP| Fee payment|Year of fee payment: 4 |
优先权:
[返回顶部]
申请号 | 申请日 | 专利标题 FR1800717A|FR3083645B1|2018-07-05|2018-07-05|IMPROVED BI-SPECTRAL DETECTOR| FR1800717|2018-07-05|FR1800717A| FR3083645B1|2018-07-05|2018-07-05|IMPROVED BI-SPECTRAL DETECTOR| PCT/EP2019/066468| WO2020007622A1|2018-07-05|2019-06-21|Improved bi-spectral detector| US17/258,089| US11171168B2|2018-07-05|2019-06-21|Bi-spectral detector| EP19732639.0A| EP3818566A1|2018-07-05|2019-06-21|Improved bi-spectral detector| IL279876A| IL279876D0|2018-07-05|2020-12-30|Improved bi-spectral detector| 相关专利
Sulfonates, polymers, resist compositions and patterning process
Washing machine
Washing machine
Device for fixture finishing and tension adjusting of membrane
Structure for Equipping Band in a Plane Cathode Ray Tube
Process for preparation of 7 alpha-carboxyl 9, 11-epoxy steroids and intermediates useful therein an
国家/地区
|