Low-power current-assisted photonic demodulator

Abstract

The invention relates to a current-assisted photonic demodulator (1) adapted for detecting light radiation, comprising: a detection portion (10), made of germanium, extending vertically between a rear face (Far) and a front face (Fav) parallel to a principal plane, the front face (Fav) being intended to receive the light radiation, comprising modulation regions (11) and collection regions (12); a dielectric layer (22), made of at least one electrically insulating material, extending over and in contact with the rear face (Far); at least one central electrode (EC), intended to be positively polarized, partially traversing the dielectric layer (22) and spaced from the rear face (Far) by a non-zero distance, located opposite the central region (Zc) and arranged, in projection onto the principal plane, between the modulation regions (11). Figure for the abstract: Fig. 1A

Description

Title of the invention: Low-power current-assisted photonic demodulator. Technical field

[0001] The field of the invention is that of current-assisted photonic demodulators (CAPDs) adapted to detect light radiation in the near-infrared. The invention finds application particularly in telemetry, biological analysis, and industrial inspection (non-contact detection of surface defects). PREVIOUS STATE OF THE ART

[0002] Current-assisted photonic demodulators are photodetectors in which the distribution of a drift electric field is modulated. They were initially described, notably in the scientific article by Van Nieuwenhove et al. entitled "Novel Standard CMOS Detector using Majority Current for giding Photo-Generated Electrons towards Detecting Junctions," Proc. Symp. IEEE / LEOS Benelux Chapter, pp. 229-232, 2005. This type of optoelectronic device is particularly used in time-of-flight (TOF) telemetry.

[0003] Such a demodulator typically comprises a detection portion made of an unintentionally doped or lightly doped p-type crystalline semiconductor material, which has, on one of its faces, two p+ doped regions for generating and modulating a drift current, and two n+ doped regions located near the p+ doped regions for collecting the photocurrent. An electrical potential difference is applied between the p+ doped regions, while a positive electronic potential (positive voltage) is applied to the n+ doped regions so as to form two reverse-biased (p+n+) diodes on each side of the CAPD, which generates a drift electric field in the detection portion.Also, when light is absorbed in the detection region, an electron-hole pair is generated. The photogenerated hole then propagates under the influence of the drift field towards the p+ doped region with the lowest electrical potential, while the photogenerated electron is directed towards the opposite p+ doped region and then collected by the adjacent n+ doped region. Thus, the photocurrent (minority electrons) can be efficiently measured by the demodulator.

[0004] Due to the separation between the majority hole current and the photocurrent (minority electrons), the contribution of the majority hole current to Schottky noise (shot noise in English) as well as thermal noise (thermal noise in English).

[0005] However, it is also important to reduce the power consumption associated with the majority hole current (modulation current). Several possibilities exist, such as reducing the doping level of the detection portion to increase the material's resistivity. However, this solution can affect the demodulator's bandwidth and thus reduce the operating frequency. Another solution is to decrease the potential difference applied between the modulation electrodes. However, this can result in a reduction of the demodulation contrast. Description of the invention

[0006] The invention aims to remedy at least in part the drawbacks of the prior art, and more particularly to propose a current-assisted photonic demodulator, suitable for detection in the near-infrared, and exhibiting good performance (in terms of demodulation contrast and bandwidth) while having reduced energy consumption.

[0007] To this end, the object of the invention is a current-assisted photonic demodulator, adapted to detect light radiation, comprising: • a detection portion, made of germanium, extending vertically between a rear face and a front face parallel to a principal plane, the front face being intended to receive the light radiation, comprising: • at least two p-type doped modulation regions, adapted to generate and modulate a drift current in the detection portion, flush with the rear face and located on either side of a central zone of the detection portion; • at least two n-type doped collection regions, adapted to collect minority charge carriers photogenerated during the absorption of light radiation in the detection portion, flush with the rear face and located on either side of the central zone and modulation regions; • a dielectric layer, made of at least one electrically insulating material, extending over and in contact with the rear face.

[0008] According to the invention, the demodulator comprises at least one central electrode, intended to be positively polarized, partially traversing the dielectric layer and spaced from the rear face by a non-zero distance, located opposite the central area and arranged, in projection in the principal plane, between the modulation regions.

[0009] Some preferred but not limiting aspects of this current-assisted photonic demodulator are as follows.

[0010] The central electrode can be spaced from the rear face of the detection portion by a distance of between 20 and 50nm.

[0011] The dielectric layer, in an intercalated part located between the central electrode and the rear face of the sensing portion, can be made of at least one material chosen from a silicon, aluminum, hafnium, zirconium or germanium oxide.

[0012] The central electrode can extend laterally, in projection in the principal plane, to the edges of the modulation regions.

[0013] The demodulator may include a peripheral lateral portion surrounding the detection portion in the main plane, made of a silicon-based semiconductor material.

[0014] The detection portion may include a lateral zone made of SiGe, located at the interface with the peripheral lateral portion.

[0015] The central electrode may include a thin conductive layer extending into the dielectric layer and a conductive pad coming into contact with the thin conductive layer.

[0016] The demodulator may include collection electrodes, passing through the dielectric layer and coming into electrical contact with the collection regions; and modulation electrodes, passing through the dielectric layer and coming into electrical contact with the modulation regions.

[0017] The invention also relates to a method for manufacturing a photonic demodulator according to any one of the preceding characteristics, comprising the following steps: • realization of the detection portion in a p-type unintentionally doped or lightly doped germanium-based compound; • realization of p-type doped modulation regions and n-type doped collection regions in the detection portion; • deposition of a dielectric layer on the back face of the detection portion; • fabrication of the central electrode through a part of the dielectric layer. Brief description of the drawings

[0018] Other aspects, objectives, advantages and features of the invention will become clearer upon reading the following detailed description of preferred embodiments of this, given by way of non-limiting example, and made with reference to the attached drawings on which:

[0019] Figures IA and IB are schematic and partial views, in cross-section ([Fig.1A]) and in top view ([Fig.1B]), of a current-assisted photonic demodulator according to one embodiment;

[0020] Figures 2A and 2B illustrate the evolution, as a function of the difference of electrical potential applied between the modulation electrodes, of the intensity of the modulation current ([Fig.2A]) and of the electrical power dissipated ([Fig.2B]), for different values ​​of electrical potential applied to the central electrode;

[0021] Figures 3A to 3H illustrate different stages of a manufacturing process for a demodulator similar to that of [Fig.1A].

[0022] DETAILED DESCRIPTION OF SPECIFIC EMBODIMENTS

[0023] In the figures and throughout the description, the same reference numerals represent identical or similar elements. Furthermore, the various elements are not drawn to scale in order to enhance the clarity of the figures. Moreover, the different embodiments and variants are not mutually exclusive and may be combined. Unless otherwise indicated, the terms "approximately," "about," and "in the order of" mean within 10%, and preferably within 5%. Furthermore, the terms "between ... and ..." and equivalents mean that the limits are inclusive, unless otherwise stated.

[0024] Figures IA and IB are schematic and partial views, in cross-section ([Fig.1A]) and in top view ([Fig.1B]), of a current-assisted photonic demodulator 1 according to an embodiment, belonging here to a matrix of identical planar demodulators.

[0025] A three-dimensional orthogonal XYZ direct coordinate system is defined herein and for the remainder of this description, where the X and Y axes form a plane parallel to the principal plane of the demodulator 1, and where the +Z direction is oriented from the front face Fav to the rear face Far (following the layer growth direction). In the remainder of this description, the terms "lower" and "upper" are understood to refer to an increasing position along the +Z direction. Furthermore, the term "horizontal" refers to an orientation parallel to the XY plane and the term "vertical" refers to an orientation parallel to the Z axis.

[0026] The current-assisted photonic demodulator 1 is suitable for detecting light radiation in the near-infrared (SWIR) spectral band corresponding to the spectral range from approximately 0.8 pm to 1.7 pm, or even approximately 2.5 pm. It is therefore suitable for detecting light radiation with wavelengths ranging from 800 nm up to a cutoff wavelength greater than 1550 nm.

[0027] In general, the demodulator 1 comprises: • a detection portion 10 made of germanium, extending vertically between a rear face Far and a front face Fav through which the light radiation is received. It comprises: • at least two p-type doped modulation regions 11, adapted to generate and modulate the drift current, flush with the rear face Far and located on either side of a central zone Zc of the detection portion 10; and • at least two n-type doped collection regions 12, adapted to collect photogenerated minority carriers, outcropping the rear face Far and located on either side of the central zone Zc and the modulating doped regions 11; • a dielectric layer 22, made of at least one electrically insulating material, which extends over and in contact with the rear face Far of the detection portion 10; • electrodes located at the rear face Far, including modulation electrodes M1, M2 in electrical contact with the doped modulation regions 11; and collection electrodes Cl, C2 in electrical contact with the doped collection regions 12; • at least one central electrode EC, intended to be positively polarized, partially traversing the dielectric layer 22 and spaced from the rear face Far by a non-zero distance, located opposite the central zone Zc and arranged, in projection in the principal plane, between the two doped modulation regions 11.

[0028] The demodulator 1 preferably belongs to a matrix photodetector comprising an array of identical detection pixels, where each detection pixel includes a planar demodulator. The configuration is said to be planar in the sense that each demodulator 1 is made from the same main semiconductor layer 21 (see [Fig. 3A]). This layer extends in a principal plane XY between the opposite rear face Far and front face Fav, which are parallel to the principal plane. The two faces Far and Fav thus extend along identical planes for each of the demodulators 1, and vertically delimit (along the thickness axis Z) the detection portions 10 of the demodulators 1. Therefore, the demodulators 1 do not have a mesa structure since they are made from the same main semiconductor layer. The front face Fav is the one that receives the light radiation to be detected.

[0029] Each demodulator 1 comprises a detection portion 10 made of germanium and therefore adapted to detect light radiation in the near-infrared (SWIR). The detection portion 10 is part of the layer main semiconductor 21. Insofar as the detection portion 10 is made of germanium and can undergo mechanical stresses in tension in the XY plane (as described later), the demodulator 1 can be adapted to detect light radiation at a cutoff wavelength greater than 1.55pm.

[0030] The thickness of the detection portion 10, defined along the vertical axis Z between the rear face Far and the front face Fav, is substantially constant from one demodulator 1 to another; for example, it is between a few hundred nanometers and a few microns, for example, between approximately 1pm and 5pm, and preferably 1.5pm. The thickness is chosen to obtain good absorption in the wavelength range of the light radiation to be detected. The detection portion 10 has a transverse dimension in the XY plane that can be between a few hundred nanometers and a few tens of microns, for example, between approximately 1pm and 20pm, for example, equal to 10pm.

[0031] The detection portion 10 is made of at least one crystalline semiconductor material based on germanium, that is to say, the semiconductor material(s) are germanium or a compound (binary or ternary...) composed of at least germanium. Thus, the detection portion 10 can be made, for example, of germanium Ge, silicon germanium SiGe, germanium tin GeSn, or even silicon germanium tin SiGeSn. It can therefore be made of the same semiconductor material and have regions of different types of conductivity (homojunction) so as to form a pn or pin junction. Alternatively, it can be a stack of sublayers of different semiconductor materials (heterojunction), which are then formed from germanium. Preferably, the detection portion 10 is made of germanium.

[0032] A central zone Zc of the detection region 10 is defined as the principal absorption site of the light radiation to be detected. This central zone Zc is advantageously delimited in the XY plane by an optical cover 4 located on the side of the front face Fav that receives the light radiation to be detected (see [Fig. 3H]).

[0033] The detection portion 10 comprises an intermediate region 13, unintentionally doped (with possible residual p-type doping) or slightly p-doped. The intermediate region 13 extends between the Far and Fav faces, as well as in the XY plane, and forms the main absorption region of the light radiation to be detected. It surrounds in the XY plane the modulating doped regions 11 and the collection doped regions 12.

[0034] The detection portion 10 comprises at least two modulation regions 11, doped with p-type, here p+ doped, adapted to generate and modulate the drift current via the electrical potential imposed on them by the modulation electrodes M1 and M2. They are here p+ doped, and have, for example, a doping level between 1 QAA 'J 1 Û 7 Approximately 10 and 10 at / cm, preferably 10 at / cm. They are flush with the rear face Far and extend towards the front face Fav along the Z-axis to a predefined depth. The depth can be defined here as the distance along the Z-axis between the rear face Far and a zone where the doping level is locally equal to half the maximum doping level. Furthermore, the two doped modulation regions 11 are located in the XY plane on either side of the central zone Zc of the detection portion 10.

[0035] The detection portion 10 also includes at least two n-type doped collection regions 12, here n+ doped, adapted to collect the photogenerated minority carriers (photocurrent) resulting from the absorption of the light radiation to be detected in the intermediate region, via the electrical potential imposed on them by the collection electrodes Cl and C2. They are n+ doped and may have a doping level ranging from approximately 5 x 10⁻¹⁰ to 10⁻¹⁰ at / cm². They are flush with the rear face Far and extend towards the front face Fav along the Z-axis to a predefined depth. Furthermore, the two doped collection regions 12 are located adjacent to and very close to the modulating doped regions 11.

[0036] Note that the modulating doped regions 11 and the collection doped regions 12 can be made, as in [Fig.1A], by localized ion implantation in the detection portion 10 from the rear face FL. Alternatively, as mentioned in patent application FR2212468 filed on November 29, 2022, it can be made by growth doping during a resumption of epitaxy in notches formed from the rear face Far of the detection portion 10.

[0037] By "outcropping," we mean "reaching the level of" or "extending from." The doped collection regions 12 and the doped modulation regions 11 are arranged in the XY plane on either side of the central zone Zc. In a configuration illustrated in [Fig. 1A], the doped modulation regions 11 are arranged near the central zone Zc, while the doped collection regions 12 are far from it.

[0038] The detection portion 10 is advantageously delimited laterally, in the XY plane, by a peripheral lateral portion 24, filled with a semiconductor material preferably based on silicon optionally doped of type p. The peripheral lateral portion 24 provides lateral optical isolation of the demodulators 1 in the XY plane, and advantageously provides a tension (in terms of mechanical stresses) in the XY plane of the material of the detection portion 10, thus increasing the absorption cutoff wavelength of the incident light radiation. It preferably extends here over the entire thickness of the detection portion 10 to reach the support layer 20. The inner face of this peripheral lateral portion 24 then defines the lateral border of the detection portion 10.

[0039] The semiconductor material is preferably silicon-based, for example amorphous silicon, monocrystalline or polycrystalline silicon, or germanium silicon, so as to advantageously form a lateral region 14 made of germanium silicon. The lateral region 14 is flush with the lateral edge and in contact with the peripheral lateral portion 24. Thus, the lateral region 14 has a higher band gap energy than the detection portion 10 made of germanium. This lateral "band gap opening" reduces the sensitivity of the demodulator 1 to defects present near the trenches. This also improves the performance of the demodulator 1.

[0040] The demodulator 1 comprises a dielectric layer 22, made of at least one electrically non-conductive material, such as an insulating material and / or an intrinsic semiconductor material, which covers the rear face Far and allows the detection portion 10 to be passivated and the electrodes M1, M2, Cl, C2, EC to be electrically isolated from each other. It is thus in contact with the modulating doped regions 11 and the collection doped regions 12, as well as the intermediate region 13. It is preferably made of an oxide, such as silicon, aluminum, germanium, hafnium, zirconium oxide, etc., or, for example, of unintentionally doped (intrinsic) silicon. It has a thickness, for example, between 20 nm and 500 nm.

[0041] The dielectric layer 22 includes an interlayer of non-zero thickness, preferably between 20 and 50 nm, which vertically separates the central electrode EC from the rear face Far of the detection portion 10. The thickness of this interlayer depends on the dielectric constant of the material from which it is composed. This interlayer is made of a material selected from, for example, a silicon oxide such as SiO2, an aluminum oxide such as Al2O3, a hafnium oxide such as HfO2, a zirconium oxide such as ZrO2, or even a germanium oxide such as GeO2, among others. Preferably, such a material has a high dielectric constant (high-k). This interlayer may also include a thin layer (a few nanometers, e.g., 2 nm) of intrinsic silicon.

[0042] Furthermore, the detection portion 10 rests on a support layer 20, made here of a semiconductor crystalline material adapted to the epitaxy of the germanium of the detection portion 10. The support layer 20 is made of a material optically transparent to the light radiation to be detected. It may include a thin layer 20.1 made of silicon adapted to the epitaxy of the main layer 21 (cf. [Fig.3A]), and possibly an oxide layer 22.2. This support layer 20 can be derived from an SOI substrate, a SiGeOi substrate or a GeOi substrate.

[0043] The demodulator 1 comprises modulation electrodes M1, M2, which generate and modulate the drift current. These electrodes pass through the dielectric layer 2 to contact the doped modulation regions 11 and apply a positive or zero electrical potential to them. They are connected to a control chip 2 (see [Fig. 3H]). It also comprises collection electrodes C1, C2, which collect the photogenerated electrons (photocurrent). These electrodes pass through the dielectric layer 22 to contact the doped collection regions 12 and apply a positive electrical potential to them. They are also connected to the control chip 2.

[0044] According to the invention, the demodulator 1 also comprises at least one central electrode EC, partially traversing the dielectric layer 22 and separated from the rear face Far by a non-zero distance. It is located opposite the central zone Zc and is positioned, in projection onto the principal plane, between the doped modulation regions 11. This spacing distance, defined along the Z-axis, is preferably between 5 nm and 50 nm, and preferably approximately 40 nm in the case of SiO2 (depending on the dielectric constant of the material of the intercalated portion of the dielectric layer 22). This central electrode EC is therefore not in contact with the rear face Far of the detection portion 10.

[0045] The central electrode EC is intended to be positively biased, and the modulation electrodes M1, M2 and collection electrodes Cl, C2 are intended to be positively biased. The electrical potential applied to the central electrode EC is obviously lower than a threshold value at which breakdown of the intercalated portion of the dielectric layer 22 can occur.

[0046] In this example, a single central electrode EC is located opposite the central zone Zc. Preferably, the central electrode EC extends laterally, projecting in the XY plane, to the edges opposite the doped modulation regions 11. The edges are defined as an area where the doping level is locally equal to half the maximum doping level.

[0047] The inventors have observed that the presence of this central electrode EC, located opposite the central zone ZC and positioned between the modulating doped regions 11 (projected onto the XY plane), makes it possible to reduce the intensity of the modulation current by electric field effect via the intercalated portion of the dielectric layer 22. Indeed, this field effect results in an accumulation of minority carriers (electrons) under the dielectric layer 22, at the level of the rear face Far, under the central electrode EC, and therefore between the modulating doped regions 11. This amounts to pinching the majority carrier conduction channel between the regions Doped modulation elements (11) reduce the modulation current intensity. This lowers energy consumption without affecting demodulation contrast or bandwidth, as the modulation voltage and resistivity of the detection material remain unchanged. This leads to improved demodulator performance (1).

[0048] Figures 2A and 2B illustrate the evolution, as a function of the value of the electrical potential applied to the central electrode EC, of ​​the intensity IM of the modulation current ([Fig. 2A]) and the energy consumption PM ([Fig. 2B]). The energy consumption corresponds to the electrical power PM = AVMiM2xIM

[0049] In this example, the demodulator 1 comprises a 2 µm thick, unintentionally undoped germanium detection portion 10 (residual p-type doping on the order of 10¹⁶ cm³), within which are located doped modulation regions 11 and collection regions 12. The central electrode EC is separated from the rear face Far of the detection portion 10 by an intercalated portion of the 40 nm thick dielectric layer 22 of SiO₂. The modulation electrodes M1, M2 and the collection electrodes Cl, C2, as well as the central electrode EC, extend into the dielectric layer 22. Furthermore, the support layer 20 is a thin film of single-crystal silicon, and the peripheral lateral portion 24 is made of polysilicon. Furthermore, a potential difference AVMim2 is applied between the modulation electrodes M1, M2, and an electrical potential Vec is applied to the central electrode EC.

[0050] The potential difference AVMiM2 is varied between 0 and IV for different values ​​of the electrical potential Vec: -IV, 0V, +1V, and +2V. Figures 2A and 2B show a clear reduction in the intensity IM of the modulation current and in the energy consumption PM when the central electrode EC is biased at +1V. This reduction becomes more significant if the electrical potential applied to the central electrode EC is increased to +2V. Conversely, the energy consumption PM increases if the central electrode EC is negatively biased.

[0051] It is also possible to simulate (e.g., using the ATLAS-SILVACO software) the hole current density in the detection region. When the central electrode EC is positively biased, a reduction in the majority hole current is observed between the modulating doped regions 11 due to an accumulation of minority electrons under the intercalated portion of the dielectric layer 22, beneath the central electrode EC. The demodulator 1 therefore exhibits reduced power consumption with a bandwidth and demodulation contrast that are not affected by the presence of the positively biased central electrode EC.

[0052] Figures 3A to 3H illustrate different stages of a manufacturing process for a demodulator 1 belonging to an array of identical planar demodulators.

[0053] With reference to [Fig. 3A], a main semiconductor layer 21 is formed by epitaxy from the support layer 20. In this example, the support layer 20 comprises a thin layer 20.1 of monocrystalline silicon of a Silicon On Insulator (SOI) substrate. This SOI substrate comprises a thick layer 20.3 of silicon (which will be removed at the end of the process), a buried oxide layer 20.2, and the thin layer 20.1 of monocrystalline silicon with a thickness between 10 and 100 nm.

[0054] The main semiconductor layer 21 is made of unintentionally doped germanium and has a thickness of between 700 nm and approximately 3 pm, for example 1.5 pm. It can be made as described in particular in the publication by Hartmann & Aubin entitled "Assessment of the growth / etch back technique for the production of Ge strain-relaxed buffers on Si", Journal of Crystal Growth, 488 (2018), 43. The main semiconductor layer 21 then has a very low density of emergent dislocations (for example, on the order of 107 dislocations / cm2), which helps to reduce the dark current in the detection portion 10 of the demodulator 1.

[0055] Alternatively, the main semiconductor layer 21 can be formed from a GeOI (Germanium-on-insulator) substrate. Thus, the support layer 20 can be a germanium nucleation layer of a few tens to a few hundred nanometers thick resting on a lower layer of approximately 2 nm of silicon, which in turn rests on an insulating layer of a few tens of nm to a few microns thick, and then on a silicon substrate. Such a GeOI substrate can be produced using the process described in the publication by Reboud et al. entitled "Structural and optical properties of 200 mm germanium-on-insulator (GeOI) substrates for Silicon photonics applications," Proc. SPIE 9367, Silicon Photonics X, 936714 (February 27, 2015).

[0056] Next, a dielectric layer 22 is deposited on the upper surface of the main semiconductor layer 21. This dielectric layer 22 can be formed of a first protective sublayer 22.1, deposited by epitaxy on the germanium of the detection portion 10, for example a thin layer of undoped silicon with a thickness ranging from 1 to 4 nm, for example 2 nm, or even an aluminum oxide deposited by atomic layer deposition (ALD) with a thickness on the order of 10 to 50 nm. Then, a second sublayer 22.2, made for example of a silicon oxide such as SiO2 TEOS (tetraethyl orthosilicate) with a thickness on the order of 20 to 500 nm, is deposited on the first sublayer 22.1.

[0057] With reference to [Fig. 3B], trenches 23 are created by photolithography and etching to pixelate the demodulators 1 by means of the peripheral lateral portions 24. This creates a localized etch of the main germanium semiconductor layer 21 to reach the silicon support layer 20. Each trench 23 preferably extends continuously in the XY plane around a demodulator 1. This results in a plurality of detection portions 10 separated from each other by a continuous trench 23. These are preferably obtained by an anisotropic etching technique, so as to obtain a substantially vertical lateral edge along the Z-axis. The trenches 23 have a transverse dimension (width) in the XY plane that can be between 0.5 pm and 2 pm, for example, equal to 1 pm.The detection portions 10 can thus present a shape in the XY plane, for example circular, oval, polygonal, for example square, or any other shape.

[0058] With reference to [Fig. 3C], the peripheral lateral portion 24 is formed by epitaxy in the trenches 23 of a silicon-based crystalline semiconductor material. This may be silicon or polysilicon. It may be p-type doped, for example with boron, with a doping level on the order of 4 x 10¹⁹ cm³. This material has a coefficient of thermal expansion lower than that of the germanium-based detection portion 10, so that upon returning to room temperature (after the silicon epitaxy in the trenches), the detection portion 10 exhibits tensile mechanical stresses in the XY plane. Subsequently, an interdiffusion annealing is performed to form the SiGe-based lateral region 14. Finally, a chemical mechano-polishing (CMP) step is then carried out, stopping on the top face of the dielectric layer 22, to remove excess silicon-based material and planarize the top face of the stack.

[0059] With reference to [Fig. 3D], the modulation doped regions 11 and the collection doped regions 12 are produced, here by ion implantation. To this end, notches 25 are formed in the dielectric layer 22, opposite areas of the detection portion 10 intended to form the doped regions 11 and 12. In this example, the notches 25 are formed in the oxide sublayer 22.2 to open onto the silicon sublayer 22.1. Alternatively, they could open onto the material of the detection portion 10. The width of the notches 25 can be on the order of 0.5 to 1pm. A thin pre-implantation layer 26 is then deposited, for example an oxide with a thickness of 10 to 30nm, so as to cover the upper face of the sublayer 22.2 and to extend conformally into the notches 25.

[0060] The doped regions 11, 12 are then produced. The modulation doped regions 11 can be produced first by implanting boron in the detection portion 10 through the dedicated notches (through an implantation mask, not (shown). The p-type doping level can be on the order of 10¹⁹ cm³. The doped collection regions 12 are then created by implanting phosphorus or arsenic into the detection portion 10 through dedicated notches. The n-type doping level can be on the order of 5 x 10⁻¹⁰ cm³. The implantation mask and the pre-implantation oxide layer 26 are thus removed.

[0061] With reference to [Fig.3E], a new dielectric sublayer is deposited on the underlying dielectric layer 22 (the whole being noted with the same reference 22), so as to cover the modulating doped regions 11, the collecting doped regions 12 and the peripheral lateral portion 24. It can be an oxide such as TEOS, with a thickness between 100 and 500nm, followed by a mechano-chemical planarization step.

[0062] With reference to [Fig. 3F], a central notch 27 is made in the dielectric layer 22 in order to subsequently form the central electrode EC. The notch 27 extends opposite the central region Zc, between the modulating doped regions 11. It opens here onto the sublayer 22.1 of the dielectric layer 22. A thin dielectric layer 22.3 of a material (preferably with a high dielectric constant) selected from SiO2, Al2O3, HfO2, ZrO2, GeO2, among others, is then deposited in conformal fashion, with a thickness, for example, between 20 and 50 nm. At least two sublayers can also be used for this layer 22.3, e.g., of the type SiO2 / HfO2 or HfO2 / Al2O3, etc. The thickness depends on the dielectric constant of the dielectric material, and can take into account the presence of the 22.1 sublayer (here silicon 2nm thick).

[0063] A thin conductive film EC.1, made of at least one electrically conductive material, is then deposited so as to minimally cover the bottom surface of the central notch 27. Here, the thin conductive film EC.1 also covers the sides of the notch 27 and part of the upper face of the dielectric layer 22. The thin conductive film EC.1 can be made of a Ti / TiN stack (among others), for example 10 and 40 nm.

[0064] With reference to [Fig. 3G], the various electrodes are fabricated. First, an additional dielectric layer, here TEOS, is deposited so as to cover the underlying dielectric thin layer 22.3 and to fill the central notch 27, followed by a chemical-mechanical planarization step. The various electrodes are then fabricated. The central electrode EC has a conductive pad EC.2 that extends through part of the dielectric layer 22 to come into contact with the conductive thin layer EC.L. It is therefore separated from the rear face Far of the detection portion 10 by the sublayer 22.1 and by the dielectric thin layer 22.3. The collection electrodes Cl, C2 and the modulation electrodes M1, M2 extend to through the dielectric layer 22 to make electrical contact with the corresponding doped regions 12, 11. They may also include a lower part made of Ti / TiN (among other materials) to optimize the metal / germanium electrical contact, followed by a copper filling section. A mechano-chemical planarization step is then performed.

[0065] With reference to [Fig. 3H], the demodulator 1 is assembled and connected to a control chip 2. The demodulator is thus inverted so that the face where the electrodes M1, M2, Cl, C2, EC open is brought into contact with an interconnection face of the control chip 2. The contact pads 3 make contact with the electrodes and ensure hybrid Cu / Cu bonding. Then the thick layer 20.3 of the SOI substrate is removed. Furthermore, an optical cover 4 can be positioned opposite the front face Fav of the demodulator, allowing light radiation to pass to the central area Zc of the detection portion 10. Note that the central electrode EC also acts as a reflector of the incident light radiation, thus optimizing absorption and therefore the performance of the demodulator.

[0066] This gives us a matrix of current-assisted photonic demodulators 1, here in planar configuration, which exhibits improved performance, and more specifically reduced energy consumption for demodulation contrast and bandwidth which remain optimal.

[0067] Specific embodiments have just been described. Various variants and modifications will be apparent to those skilled in the art.

Claims

Demands

1. Current-assisted photonic demodulator (1) adapted to detect light radiation, comprising: • a detection portion (10), made of germanium, extending vertically between a rear face (Far) and a front face (Fav) parallel to a principal plane, the front face (Fav) being intended to receive the light radiation, carrying: • at least two p-type doped modulation regions (11), adapted to generate and modulate a drift current in the detection portion (10), flush with the rear face (Far) and located on either side of a central zone (Zc) of the detection portion (10);• at least two n-type doped collection regions (12), adapted to collect the minority charge carriers photogenerated during the absorption of light radiation in the detection portion (10), flush with the rear face (Far) and located on either side of the central zone (Zc) and the modulation regions (11); • a dielectric layer (22), made of at least one electrically insulating material, and extending over and in contact with the rear face (Far); • characterized in that it comprises at least one central electrode (EC), intended to be positively polarized, partially traversing the dielectric layer (22) and spaced from the rear face (Far) by a non-zero distance, located opposite the central zone (Zc) and disposed, in projection in the principal plane, between the modulation regions (11).

2. Photonic demodulator (1) according to claim 1, wherein the central electrode (EC) is spaced from the rear face (Far) of the detection portion (10) by a distance of between 20 and 50nm.

3. Photonic demodulator (1) according to claim 1 or 2, wherein the dielectric layer (22), in an intercalated portion located between the central electrode (EC) and the rear face (Far) of the detection portion (10), is made of at least one material selected from an oxide of silicon, aluminium, hafnium, zirconium or germanium.

4. Photonic demodulator (1) according to any one of claims 1 to 3, wherein the central electrode (EC) extends laterally, in projection in the principal plane, to the edges of the modulation regions (11).

5. Photonic demodulator (1) according to any one of claims 1 to 4, comprising a peripheral lateral portion (24) surrounding the detection portion (10) in the principal plane, made of a silicon-based semiconductor material.

6. Photonic demodulator (1) according to claim 5, wherein the detection portion (10) comprises a lateral area (14) made of SiGe, located at the interface with the peripheral lateral portion (24).

7. Photonic demodulator (1) according to any one of claims 1 to 6, wherein the central electrode (EC) comprises a conductive thin layer (EC.1) which extends into the dielectric layer (22) and a conductive pad (EC.2) coming into contact with the conductive thin layer (EC.1).

8. Photonic demodulator (1) according to any one of claims 1 to 7, comprising collection electrodes (Cl, C2), passing through the dielectric layer (22) and coming into electrical contact with the collection regions (12); and modulation electrodes (M1, M2), passing through the dielectric layer (22) and coming into electrical contact with the modulation regions (11).

9. A method for manufacturing a photonic demodulator (1) according to any one of the preceding claims, comprising the following steps: • making the detection portion (10) out of an unintentionally doped or lightly doped p-type germanium compound; • making the p-type doped modulation regions (11) and the n-type doped collection regions (12) in the detection portion (10); • deposition of a dielectric layer (22) on the rear face (Far) of the detection portion (10); realization of the central electrode (EC) through a part ie of the dielectric layer (22).

Images