Photodetector device

Abstract

The photodetector device (PDD), comprising at least one group of pixels (P1, P2, P3), includes: - a reference electrode (Eref) common to all pixels, - a continuous thin layer of colloidal quantum dots (CQD) common to all pixels, comprising a photosensitive region (n-CQD) capable of photogenerating electrical charges (h+, e-), and - a charge-collecting electrode (EC1, EC2, EC3) for the photogenerated charges specific to each pixel (P1, P2, P3), the charge-collecting electrodes (EC1, EC2, EC3) being configured to collect electrically positive photogenerated charges (h+). Figure for the abbreviation: Fig 4A

Description

Title of the invention: Photodetector device

[0001] Embodiments relate to integrated circuits incorporating photodetector devices, in particular photodetector devices using thin-film technology of colloidal nanoparticles of quantum dots.

[0002] Fig. 1 A illustrates a classic example of a photodetector pixel using a thin layer of colloidal nanoparticles of CQD quantum dots.

[0003] Thin films or films containing colloidal (nano-)conducting semiconductor particles (called "colloidal quantum dots" (CQDs), an expression derived from the English term "colloidal quantum dot" and indicating zero-dimensional quantum confinement) enable a significant and exacerbated photoelectric effect at the wavelength corresponding to the excitonic peak characteristic of quantum confinement. These quantum dot layers can be integrated into conventional microelectronic fabrication processes to obtain photodiodes.

[0004] Classically, the photodiode (n-CQD, p-CQD) made in a thin layer of colloidal CQD quantum dots is polarized between two electrodes El, E2, at potentials allowing modulation of the external quantum efficiency of the photoelectric generation, in particular between an "active" state (in English "on") where the quantum efficiency is positive and constant, and an "inactive" state (in English "off") where the quantum efficiency is substantially zero (Cf. [Fig.2]).

[0005] The photogenerated charges, i.e. the electrical charges generated by the photoelectric effect, are collected by the electrodes El, E2, via respective charge transfer layers ETL, HTL.

[0006] One of the electrodes El is called "collecting", typically intended to collect negative charges (electrons), to which is associated an electron transfer layer ETL located between the electrode El and the thin layer of colloidal quantum dots CQD (n-CQD).

[0007] The other electrode E2 is called "reference" E2, typically intended to collect positive charges (holes) and for example coupled to a common ground terminal for all pixels, to which is associated an HTL hole transfer layer located between the electrode E2 and the thin layer of colloidal quantum dots CQD (p-CQD).

[0008] Dielectric regions DL1 (for example, silicon oxide or nitride, or resin, or other) allow for the approximate lateral delimitation of the collection electrodes El and the respective electron transfer layers ETL. Dielectric regions DL2 may optionally be provided to delimit laterally the reference electrodes E2 and / or the respective hole transfer layers HTL.

[0009] This electron collection configuration is classically used because readout circuits usually operate with electrons.

[0010] However, this configuration presents a problem of leakage between neighboring collection electrodes El when one is polarized to control an active state and the other an inactive state.

[0011] Fig. IB illustrates in this respect the electrostatic potential in the CQD thin film subjected to such polarization on a central electrode Elc at the potential of the active state and two lateral electrodes Elg, Eld at the potential of the inactive state on either side of the central electrode Elc.

[0012] Fig.1C illustrates the energy band levels in eV (electronvolts) of electrons and holes in the CQD thin film of Fig.1B along the VCUT path between the reference electrode E2 and the central electrode Elc, and along the HCUT path between a lateral electrode Eld (or Elg by symmetry) and the central electrode Elc.

[0013] On the one hand, we can see on the VCUT path the "normal" migration in the presence of an electric field of electrons e- towards the central electrode Elc and the migration of holes h+ towards the reference electrode E2.

[0014] On the other hand, we can see on the HCUT path the possibility of "leakage" of electrons between the non-polarized lateral electrodes Eld (or Elg by symmetry) and the polarized central electrode Elc by crossing a very narrow potential barrier, according to the phenomenon usually called "tunnel effect".

[0015] Fig. 1D illustrates the total current density that results from the energy bands of electrons e- on the path HCUT, between a lateral electrode Elg (or Eld by symmetry) and the central electrode Elc.

[0016] Thus in particular we see the flux FL of electrons e- from the unpolarized lateral electrode Elg (or Eld by symmetry) towards the polarized central collecting electrode Elc.

[0017] Figure [1E] illustrates the intensity of the currents Ilg and Hd flowing across the lateral electrodes Elg and Eld, and of the current Ile across the central collecting electrode Elc, for example, in the absence of incident light and the photoelectric effect, as a function of the bias voltage of the central electrode of the photodiode. It can be seen that under these conditions, the current Ile across the central electrode Elc is equal to the sum of the currents Ilg and Ild leaking from the lateral electrodes Elg and Eld.

[0018] Thus, in classical structures as described above, there is a problem of charge leakage from the non-polarized collecting electrodes to the polarized collecting electrode.

[0019] This problem causes difficulties, particularly in applications that selectively operate certain pixels whose photosensitive regions are located in the same thin layer of common CQD colloidal quantum dots, for example in two interlaced pixel arrays, and alternately collecting photogenerated charges.

[0020] This is particularly the case for the following applications: a binning mode of operation where a smaller number of pixels is selectively used, in order to benefit from a larger volume of photogenerated charge to increase sensitivity; a plural (e.g. dual) mode of operation, where at least two pixel arrays of different functions are interlaced and share the same thin layer of colloidal quantum dots.

[0021] Conventional techniques for limiting current leakage problems rely on pixel control and readout circuits, notably using switching transistors to limit current leakage.

[0022] Conventional solutions consequently generate additional bulk in the control circuits of each pixel, which can also be highly complex. This hinders the reduction of pixel size (and consequently limits the sensor resolution) and / or leads to complex and expensive designs, such as a stack of several silicon wafers glued together.

[0023] Embodiments according to the aspect defined below make it possible to remedy the classical problems by means of a configuration made on the energy bands of the photogenerated charges, based on the collection of holes instead of the collection of electrons, so as to take advantage of a potential barrier preventing current leakage.

[0024] In addition to the absence of charge leakage, the architecture defined in the embodiments below eliminates the need for switching circuits for each collecting electrode, which improves integration.

[0025] Also, the embodiments defined below allow for binning operation and plural operation, for example including interlacing of different types of pixels, for example a static acquisition network of the global shutter type, an event-driven acquisition network, and / or a time-of-flight measurement acquisition network.

[0026] According to one aspect, a photodetector device is proposed in this regard comprising at least one group of pixels including: - a common reference electrode between the pixels, - a thin layer of continuous colloidal quantum dots common to all pixels, comprising a photosensitive region capable of photogenerating electric charges, and - a photogenerated charge collection electrode respective to each pixel configured to collect electrically positive photogenerated charges.

[0027] Configuring the collection electrodes, i.e. the means enabling the reading of the light signal detected by the device, to collect electrically positive photogenerated charges, called "holes", makes it possible to benefit from a potential barrier effect preventing current leakage directly at the level of the pixel architecture and therefore without requiring additional circuits in the control means.

[0028] According to one embodiment, the device comprises a hole transfer layer between the thin layer of colloidal quantum dots and the respective collection electrodes, configured to allow transfers of positive electrical charges in an active state of the pixel, and to generate an electrostatic potential barrier large enough to block transfers of positive electrical charges in an inactive state of the pixel.

[0029] According to one embodiment, the hole transfer layer may comprise a transparent metal oxide. The metal oxide is advantageously heavily doped so as to be used as a hole-extracting layer. For example, the hole transfer layer comprises molybdenum oxide, nickel oxide, copper oxide, tungsten oxide, or vanadium oxide.

[0030] According to one embodiment, the thin layer of colloidal quantum dots may further comprise a P-type doped region at the interface with the hole transfer layer.

[0031] According to one embodiment, the device further comprises control means configured to selectively control the pixels in the active state or in the inactive state, by polarizing the common reference electrode to a reference voltage, and by polarizing the collection electrodes respectively.

[0032] According to one embodiment, the collection electrodes are configured to be polarized on polarization channels in direct contact with the respective collection electrodes in a manner devoid of switching means.

[0033] Indeed, the structure of the photosensitive region capable of photogenerating electrical charges and the collection of the photogenerated positive charges according to the aspect defined above, allow an absence of current leakage with a control carried out by a polarization and consequently eliminate the need for palliative switching circuits for each collection electrode.

[0034] According to one embodiment, in a shared operating mode, the photodetector device is configured to collect the photogenerated charges in the thin layer of colloidal quantum dots, on the collection electrode of only one of the pixels of said at least one group of pixels.

[0035] In other words, in a binning application, a reduction in resolution can be configured to increase sensitivity. Conversely, depending on the need, an increase in resolution can be configured at the expense of sensitivity.

[0036] According to one embodiment, the pixels of said at least one group of pixels are of the same type dedicated to a photodetection of the same nature.

[0037] For example, the type of pixel dedicated to a photodetection of the same nature are dedicated to a photodetection of the same spectral component of light, such as a red, green, blue, or infrared component.

[0038] According to one embodiment, said at least one pixel group comprises a first type of pixel and a second type of pixel, the photodetector device being configured to collect the photogenerated charges in the thin layer of colloidal quantum dots, either on the collection electrodes of the pixels of the first type, or on the collection electrodes of the pixels of the second type.

[0039] According to one embodiment, the pixels of the first type and the pixels of the second type are each and distinctly configured for one of the following acquisitions: - a static acquisition by global shutter; - dynamic acquisition through event detection; - a measurement of flight time.

[0040] Other advantages and features of the invention will become apparent upon examination of the detailed description of embodiment and implementation, which is by no means limiting, and the accompanying drawings in which:

[0041] [Fig.lA] ;

[0042] [Fig.lB] ;

[0043] [Fig.lC] ;

[0044] [Fig.lD] ;

[0045] [Fig.1E] previously described, illustrate an example of a pixel;

[0046] [Fig.2] ;

[0047] [Fig.3] ;

[0048] [Fig.4A] ;

[0049] [Fig.4B] ;

[0050] [Fig.4C] ;

[0051] [Fig.4D] ;

[0052] [Fig.4E];

[0053] [Fig.5] ;

[0054] [Fig.6] illustrate embodiments of the invention.

[0055] Fig. 2 schematically illustrates the operating principle of a photodiode made in a thin layer of colloidal quantum dots (CQD).

[0056] The photodiode is polarized between two electrodes at potentials V allowing modulation of the external quantum efficiency of the photoelectric generation, which is reflected in the intensity of the current I of the photodiode.

[0057] The photogenerated charges, that is to say the electrical charges generated by the photoelectric effect, comprise electrons (represented by a solid circle) and holes (represented by an empty circle), and are respectively collected by an electron transfer layer ETL and by a hole transfer interface HTL, which are also connected to the electrodes controlling the photodiode.

[0058] The flow of the current I from the photogenerated charges depends on the modulation of the energy bands by the control potentials V.

[0059] Thus, the photodiode can be controlled in particular in:

[0060] - a blocked BLCK mode, that is to say an "inactive" (in English "OFF") state, in in which the charge carriers are in partial desertion and recombine in an electrically neutral region of the CQD thin film

[0061] - an RVRS inversion mode, that is to say an "active" (in English "ON") state, in in which the charge carriers are in complete desertion in the CQD thin film, and benefiting from an absence of electrostatic barrier in both charge transfer interfaces ETL, HTL (in both electrodes)

[0062] - a threshold state (in English, "reach through") RCHT, at a threshold voltage Vrt Between the BLCK blocked mode and the RVRS inversion mode, charge carriers are completely depleted of the CQD thin film, and an electrostatic barrier blocks the ETL electron transfer interface.

[0063] - a direct conduction state DRCT, at positive control voltages V, in into which a strong current is injected from the electrodes.

[0064] Fig. 3 illustrates an example of realization of a pixel based on a photodiode in a thin layer of colloidal quantum dots (CQD).

[0065] The pixel includes, in particular: - a reference electrode Eref - a thin layer of colloidal quantum dots (CQDs) comprising a photosensitive region, called a photodiode, capable of photogenerating electric charges, i.e., generating electric charges by the photoelectric effect, - an EC1 collection electrode.

[0066] In addition, the pixel comprises: - an HTL hole transfer layer located between the EC1 collection electrode and the p-CQD colloidal quantum dot thin layer having P-type doping, designed to transfer the photogenerated positive charges, called holes, so as to be collected on the EC1 collection electrode, and - an electron transfer layer ETL located between the reference electrode Eref and the thin layer of n-CQD colloidal quantum dots having N-type doping, intended to transfer the photogenerated negative charges, called electrons, to a reference voltage node (ground).

[0067] Dielectric regions DL1 (for example silicon oxide or nitride, or resin, or other) allow roughly to laterally delimit the collection electrode EC1 as well as the hole transfer layer HTL.

[0068] Dielectric regions DL2 may optionally be provided to laterally delimit the reference electrode Eref and / or the electron transfer layer ETL.

[0069] Fig. 4A illustrates an example of an embodiment of a DSP photodetector device, typically belonging to an integrated circuit IC, comprising at least one group of pixels P1, P2, P3 using a thin layer of colloidal quantum dots CQD.

[0070] The DSP device includes, in particular: - a common reference electrode Eref between pixels PI, P2, P3, - a thin layer of continuous and common colloidal CQD quantum dots between pixels PI, P2, P3 comprising a photosensitive region capable of photogenerating electric charges h+, e-, and - an EC1, EC2, EC3 collection electrode of the photogenerated charges respective to each pixel PI, P2, P3.

[0071] The charge collection electrodes EC1, EC2, EC3 are configured to collect electrically positive photogenerated charges, called h+ holes.

[0072] The reference electrode Eref, common to all pixels PI, P2, P3, is configured to carry the negative electrical photogenerated charges, called electrons e-, for example to a ground potential terminal.

[0073] The reference electrode Eref covers, for example, the outer side of the thin layer of colloidal quantum dots CQD, i.e., the side that is exposed to the incident light flux IL. Thus, the reference electrode Eref is made of an electrically conductive material capable of being transparent to the incident light IL (depending, for example, on the material and / or its thickness).

[0074] An electron transfer layer ETL, usually a metal oxide, suitable for enabling the transfer of negative electrical charges e- can advantageously be provided to interface between the colloidal film and the metal of the reference electrode Eref.

[0075] The collection electrodes EC1, EC2, EC3, for example made of copper, are located on the opposite side of the colloidal film CQD, and are connected to metallic vias VM of a BEOL interconnect circuit of the integrated circuit IC.

[0076] The VM metallic vias electrically connect the EC1-EC3 collection electrodes to electronic circuits made in a FEOL semiconductor part. The electronic circuits are configured to implement the device's function, particularly with regard to pixel control and readout (see below).

[0077] Advantageously, an HTL hole transfer layer is located between the colloidal film of CQD quantum dots and the respective collection electrodes EC1, EC2, EC3. The HTL hole transfer layer is configured to allow the transfer of positive electrical charges h+ in an active pixel state, and to generate an electrostatic potential barrier large enough to block the transfer of positive electrical charges h+ in an inactive pixel state.

[0078] For example, the HTL hole transfer layer may in this respect comprise a transparent metal oxide whose work function and band gap are chosen so as to form an electrostatic potential barrier blocking the photogenerated h+ holes when the collecting electrode EC2 is not polarized and to allow all to pass when a control voltage is applied (EC1, EC3), for example less than or equal to 5V for the applications concerned.

[0079] The metal oxide of the HTL layer is advantageously heavily doped so as to modulate the work function and the band gap in order to be used as a hole-extracting layer. For example, the HTL hole transfer layer comprises molybdenum oxide, nickel oxide, copper oxide, tungsten oxide, or vanadium oxide.

[0080] Optionally, the HTL hole transfer layer may include in this respect a pCQD region comprising quantum dots passivated by ligands behaving like a P-type doped semiconductor. This pCQD layer allows the nCQD layer to be passivated. By "passivate," we mean "generate a passivation."

[0081] Thus the pixels PI, P2, P3 comprise on the one hand a photosensitive region (thin layer of colloidal quantum dots CQD) between two electrodes Eref, EC1-EC3, and on the other hand control and reading means in the circuit made in the semiconductor part FEOL.

[0082] CQD colloidal quantum dot thin films are technologies suitable for photodetection which allow the development of low cost, high resolution image sensors with very good performance in the short wave infrared (or usually "SWIR" for "Short Wave Infrared" in English), particularly in terms of external quantum efficiency, crosstalk and dark current.

[0083] In summary, "quantum dots" usually refer to formations of nanometric dimensions (for example less than 100 nm), comprising a core of a semiconductor material, enveloped by molecules called ligands which extend radially outwards from the core.

[0084] These formations derive their name from "quantum box" because they form a confinement zone by quantum effect in all directions of space. The dimensions of quantum dots are, for example, less than 100 nm, preferably between 2 nm and 15 nm.

[0085] The materials composing the quantum dots and the dimensions of each quantum dot, in particular the dimensions of the core, determine the absorption wavelengths of the quantum dots, that is to say the wavelengths generating the photoelectric effect in the CQD photosensitive region.

[0086] The absorption wavelengths correspond for example to the short infrared, i.e. wavelengths between 700 nm and 1.6 pm, and / or to the mid-infrared, i.e. wavelengths between 1.6 pm and 4 pm, and / or to the visible, i.e. wavelengths between 300 nm and 700 nm.

[0087] The semiconductor core is, for example, a compound of lead sulfide, indium arsenide, or mercury telluride. The ligands are preferably organic molecules or organometallic and inorganic molecules.

[0088] The doping and fermi level of the thin film of colloidal CQD n-CQD, p-CQD quantum dots is related to the stoichiometry of the quantum dots, the net charge and the dipole moment of the ligands

[0089] Indeed, the ligands can be molecules acting as N-type dopants, for example organic molecules such as thiolates; or molecules acting as P-type dopants, for example organic molecules of the carbon chain type.

[0090] Finally, a polarization of the thin layer of colloidal quantum dots (CQDs) makes it possible to modulate the external quantum efficiency of the photoelectric generation and charge recombination, in particular between an "active" state where the intensity of the current flowing over the collection electrodes is the representation of the quantity of photogenerated charge, and an "inactive" state where no current flows regardless of the quantity of photogenerated charge, the latter being recombined within the thin layer due to the low diffusivity and mobility of the photogenerated charges in the absence of an electric field, (see [Fig. 2]).

[0091] In this regard, the DSP photodetector device advantageously further comprises CMD control means configured to selectively control the pixels PI, P2, P3 in the active state, capable of collecting electrically positive photogenerated charges h+, or in the inactive state respectively unsuitable for collecting electrically positive photogenerated charges h+, by biasing the common reference electrode Eref to a reference voltage, for example 0V, and by biasing respectively the collection electrodes EC1, EC2, EC3.

[0092] It should be noted in particular that the collection electrodes EC1, EC2, EC3 are configured to be polarized in order to control the active and inactive states of pixels PI, P2, P3, and can therefore be accessed by polarization channels in direct contact on the respective collection electrodes EC1, EC2, EC3 in a manner without any switching means in particular.

[0093] Indeed, given the configuration made to collect h+ holes, and in particular the electrostatic potential barrier blocking h+ holes at the collection electrodes EC1, EC3 in the inactive state, the DSP device benefits from an absence of current leakage between electrodes in the active and inactive states.

[0094] In other words, the design of the CQD photosensitive region blocking the collected h+ charges by the potential barrier of the PI, P3 pixels in the inactive state, intrinsically eliminates the need to block leakage of collected charges with switching circuits for all pixels in the inactive state, when neighboring EC1, EC2, EC3 collection electrodes are distinctly polarized to control an active and an inactive state.

[0095] Reference is made in this regard to figures 4B, 4C, 4D and 4E.

[0096] Fig. 4B illustrates the electrostatic potential in the thin layer of colloidal quantum dots CQD subjected to a polarization in which a centrally positioned collecting electrode EC2 is brought to the potential of the active state, for example -3V, and two adjacent collecting electrodes in lateral positions EC1, EC3, are brought to the potential of the inactive state, for example 0V.

[0097] The central collecting electrode EC2 will be referred to hereafter as "central electrode EC2", while the two collecting electrodes in lateral positions EC1, EC3, located on either side of the central electrode EC2, will be referred to hereafter as lateral electrodes EC1, EC3.

[0098] Fig. 4C illustrates the energy band levels in eV (electronvolts) of the charge carriers e-, h+ in the CQD colloidal quantum dot thin layer of Fig. 4B along the vertical path VCUT between the reference electrode Eref and the central electrode EC2, and along the "lateral" path HCUT between a lateral electrode EC3 (or EC1 by symmetry) and the central electrode EC2.

[0099] On the one hand, we can see on the vertical path VCUT the "normal" migration of the holes h+ collected by the central electrode EC2 and the migration of the electrons e- towards the ground supported by the reference electrode Eref.

[0100] On the other hand, we can see on the HCUT path the electrostatic potential barrier EPB preventing the migration of the h+ holes between the central electrode EC2 and the lateral electrodes EC3 (or EC1 by symmetry) by crossing a very narrow potential barrier, according to the phenomenon usually called "tunneling effect".

[0101] The electrostatic potential barrier EPB thus prevents the migration of h+ holes, i.e. the photogenerated positive electrical charges in the photosensitive region CQD, between two neighboring collection electrodes EC2, EC3 and supporting distinct potentials (i.e. potentials controlling an active state, for example -3V, and an inactive state, for example 0V).

[0102] Fig. 4D illustrates the total current density that results from the configuration of the energy bands of the h+ holes on the path HCUT, between a lateral electrode EC1 (or EC3 by symmetry) and the central electrode EC2.

[0103] Thus we see that there is no flow of positive charges h+ from the lateral electrode EC1 (or EC3 by symmetry) towards the central collecting electrode EC2, and that there are flows FL1, FL2 directing globally the positive charges h+ from the volume of the photosensitive region CQD towards the central electrode EC2.

[0104] Fig. 4E illustrates the intensity of the currents II, 13 flowing over the lateral electrodes EC1 and EC3 and of the current 12 over the central electrode EC2, for example in the presence of incident light, as a function of the bias voltage of the photodiode.

[0105] It can be seen that under these conditions, the currents II, 13 are strictly nuisanced on the side electrodes EC1 and EC3, while the current 12 on the central electrode EC2 flows from active state control voltages below -2.5V, for example -3V.

[0106] Thus, in the DSP device described above in relation to figures 4A to 4E, there is no problem of charge leakage between neighboring collection electrodes polarized at different potentials.

[0107] Consequently, the DSP device enables advantageous and efficient applications, particularly for shared operation and / or plural operation.

[0108] Fig. 5 illustrates an example of shared operation of the DSP photodetector device.

[0109] In binning mode, a smaller number of pixels are selectively used for charge collection within a given volume of the photosensitive region. This increases detection sensitivity, thereby improving operating speed and event detection capacity, while simultaneously reducing associated noise.

[0110] Indeed, the DSP photodetector device is for example configured to collect the photogenerated charges in the photosensitive region (thin layer of colloidal quantum dots CQD), on the collection electrode of a single PXon pixel of a group of GRP pixels.

[0111] The PXon pixel collecting the charges of the group is commanded in the active state, and is thus designated as the "active pixel" of the group, while all other PXoff pixels of the GRP group are commanded in the inactive state, and are thus designated as the "inactive pixels" of the group.

[0112] It is recalled that the inactive PXoff pixels allow the photogeneration of h+ charges in the associated photosensitive region, but prevent the flow of h+ hole current on their collecting electrode.

[0113] Thus the h+ charges generated in the photosensitive region of all the pixels of the GRP group are collected by the only active pixel PXon of the GRP group.

[0114] In this respect, the active PXon pixel is chosen at a position within the GRP group that allows it to be as close as possible to the charge generation location. For example, the active PXon pixel of the group is chosen at a central position, for example, in the middle of a group of 9 pixels in a 3x3 pixel square. Other pixel arrangements can be used to form GRP groups suitable for sharing.

[0115] In this example the pixels of the GRP pixel groups are dedicated to the same type of photodetection, that is to say for example that all the pixels of the GRP group are dedicated to a static detection by global shuttering of the components of the visible, red R, green G and blue B, respectively in each group GRPG, GRPR, GRPB.

[0116] Conversely, depending on the need, each pixel of the GRP group can also be individually controlled, for example in order to increase the resolution at the expense of the sensitivity, possibly by means of a remosaicing technique (expression taken from the common English term "remosaicing"), that is to say a rearrangement by extrapolation into an elementary pattern (or mosaic), for example of the Bayer type.

[0117] Alternatively, the pixels of the GRP pixel groups are dedicated to different types of photodetection.

[0118] In this regard, reference is made to [Fig.6].

[0119] Fig. 6 illustrates examples of plural operation, for example involving an interlacing of several types of pixels dedicated to different kinds of photodetection, for example: a static acquisition network of the global shutter or rolling shutter type, an event-driven (IR) acquisition network, and / or a time-of-flight (IR) measurement acquisition network.

[0120] Thus, the DSP device comprises at least two pixel arrays of different functions, are interlaced and share the same thin layer of colloidal quantum dots CQD.

[0121] In practice, to avoid excessive complexity, it is advantageous to implement a dual operation, involving an interlacing of two types of pixels dedicated to different types of photodetection.

[0122] Thus, in a first example, part of the pixels of the pixel groups GRP33, GRP22, GRP5, are dedicated to a static acquisition by global shuttering of the visible components R, G, B, and another part to a dynamic acquisition by event detection of an infrared component IR.

[0123] In a second example, part of the R, G, B pixels of the GRP33, GRP22, GRP5 pixel groups are dedicated to a static acquisition by global shuttering of the visible RGB components, and another part to an avalanche effect acquisition for a time-of-flight measurement of an infrared IR component.

[0124] In these two examples, each GRP33 group includes R, G, B pixels dedicated to acquisitions made in the visible, red, green and blue, and at least one IR pixel dedicated to acquisitions made in the infrared.

[0125] For example, the GRP33 group comprises nine pixels in a 3x3 pixel square, including two red R pixels, four green G pixels, two blue B pixels, and one infrared IR pixel at the middle position of the square.

[0126] For example, the GRP22 group comprises four pixels in a 2x2 pixel square, including a red R pixel, a green pixel, a blue pixel and an infrared IR pixel.

[0127] For example, the GRP5 group comprises five pixels in a 2x2 pixel square, including one red R pixel, two green pixels, and one blue pixel, with an additional infrared IR pixel positioned in the middle of the square.

[0128] By providing red, green and blue filters transparent to the infrared component, then the pooling technique (“binning”) can be applied by ordering the infrared IR pixel as the active PXon pixel of the GRP group, in order to benefit from the increase in sensitivity and speed, while reducing the associated noise.

[0129] For example, in this context of multiple detections, it will be possible to implement said global shutter, event-driven, or time-of-flight measurement acquisitions during acquisition phases distinct from each other.

[0130] During the separate acquisition phases, we will thus have either the first type of pixel controlled in the active state and the second type of pixel controlled in the inactive state, or the first type of pixel controlled in the inactive state and the second type of pixel controlled in the active state.

[0131] The photodetector device is in this case configured to collect the photogenerated charges in the photosensitive region common to the group GRP33, GRP22, GRP5, either with the first type of pixel (for example the visible R, G, B pixels), or with the second type of pixel (for example the infrared IR pixel).

[0132] In both cases, the collected currents will not be tainted by leakage between neighboring collection electrodes controlling the active and inactive states respectively.

Claims

Demands

1. Photodetector device (DSP) comprising at least one group of pixels (PI, P2, P3) having: - a reference electrode (Eref) common between the pixels, - a continuous thin layer of colloidal quantum dots (CQD) common between the pixels having a photosensitive region (n-CQD) capable of photogenerating electric charges (h+, e-), and - a collection electrode (EC1, EC2, EC3) of the photogenerated charges respective to each pixel (PI, P2, P3), the charge collection electrodes (EC1, EC2, EC3) being configured to collect electrically positive photogenerated charges (h+).

2. Device according to claim 1, comprising a hole transfer layer (HTL) between the colloidal quantum dot thin layer (CQD) and the respective collection electrodes (EC1, EC2, EC3), configured to permit positive electrical charge transfers (h+) in an active pixel state, and to generate an electrostatic potential barrier (EPB) large enough to block positive electrical charge transfers (h+) in an inactive pixel state.

3. Device according to claim 2, wherein the hole transfer layer (HTL) comprises a transparent metal oxide.

4. Device according to any one of claims 2 or 3, wherein the colloidal quantum dot thin layer has a P-type doped region (p-CQD) at the interface with the hole transfer layer (HTL).

5. Device according to any one of claims 2 to 4, further comprising control means (CMD) configured to selectively control the pixels (PI, P2, P3) in the active state or in the inactive state, by biasing the common reference electrode (Eref) to a reference voltage, and by biasing the collection electrodes (EC1, EC2, EC3) respectively.

6. Device according to claim 3, wherein the collecting electrodes (EC1, EC2, EC3) are configured to be polarized on polarization channels (VM) in direct contact with the respective collecting electrodes in a manner devoid of switching means.

7. Device according to any one of claims 1 to 4, wherein, in a shared operating mode, the photodetector device (DSP) is configured to collect the photogenerated charges in the thin layer of colloidal quantum dots (CQD), on the collecting electrode of a single pixel (PXon) of said at least one pixel group (GRP).

8. Device according to claim 5, wherein the pixels of said at least one group of pixels (GRP) are dedicated to the same type of photodetection (G).

9. Device according to any one of claims 1 to 6, wherein said at least one pixel group (GRP33, GRP22, GRP5) comprises a first type of pixel and a second type of pixel, the photodetector device (DSP) being configured to collect the photogenerated charges in the colloidal quantum dot (CQD) thin layer, either on the collecting electrodes of the first type of pixels or on the collecting electrodes of the second type of pixels.

10. Device according to claim 9, wherein the pixels of the first type and the pixels of the second type are each and distinctly configured for one of the following acquisitions: - a static acquisition by global shutter; - a dynamic acquisition by event detection; - a time-of-flight measurement.

Images