PLANAR PHOTODIODINE BASED ON GERMANIUM WITH A COMPRESSIVE PERIPHERAL SIDE ZONE

Description

TECHNICAL FIELD

[0001] The field of the invention is that of planar photodiodes made from germanium. The invention finds application particularly in the field of detecting light radiation in the near-infrared. STATE OF THE PRIOR ART

[0002] The optoelectronic photodetection devices may comprise a matrix of planar photodiodes. The photodiodes then extend along the same main plane, between first and second opposite and parallel faces. They then each comprise a detection portion originating from the same semiconductor layer, formed for example from a first region (base, seed layerin English) in p+ doped germanium and flush with the first face, a second region (box) in n+ doped germanium and flush with the second face, and an intermediate region in intrinsic or very lightly p-doped germanium, and located between the first and second doped regions. This intermediate region surrounds the second n+ doped region in the principal plane of the photodiode, and is flush with the second face. A passivation layer made of a dielectric material usually covers the second face, and makes it possible to limit the contribution of the dark current to the electric current measured by each photodiode.

[0003] This detection portion may be surrounded by a peripheral lateral portion, made of an electrically conductive material, which surrounds the detection portion in the main plane of the photodiode and comes into contact with the first p+ doped region. This peripheral lateral portion then provides an optical isolation function with the neighboring planar photodiodes and makes it possible to apply an electrical potential to the first p+ doped region from the second face. It may also make it possible to generate mechanical stresses in tension in the detection portion.

[0004] Indeed, when the material of the absorption portion is based on germanium, it can be tensioned so as to make its energy band structure direct and no longer indirect. In addition, the photodiode can then be adapted to absorb light radiation up to a cut-off wavelength greater than 1550nm, which is the absorption cut-off wavelength of relaxed germanium, for an application in particular in the near infrared (SWIR, for Short Wavelength IR in English).

[0005] In this respect, document WO 2019 / 202250 A1 describes an example of such a planar photodiode, where the peripheral lateral portion is made of a piezoelectric material. The deformation of the peripheral lateral portion in the main plane, by inverse piezoelectric effect, makes it possible to tension the detection portion.

[0006] Document EP 3 806 167 A1 describes another example of a planar photodiode. Here, the peripheral lateral portion is made of a p+ doped silicon-based material. The voltage of the detection portion is obtained during the epitaxial production of the peripheral lateral portion, due to the thermal expansion differential between the germanium and the silicon.

[0007] Furthermore, it appears that the presence of the dielectric passivation layer can still contribute to generating a significant dark current. Thus, the article by Sood et al. entitled Characterization of SiGe-Detector Arrays for Visible-NIR Imaging Sensor Applications, Proc. of SPIE VOL. 8012, 801240, 2011, describes a method for manufacturing a passivated photodiode to limit the dark current. The dark current is linked to the presence of a depleted zone located in the semiconductor material of the photodiode, at the interface with the dielectric passivation layer. The manufacturing process then includes a step of annealing the photodiode under N 2 H 2 , making it possible to transform this depleted zone into a hole accumulation zone. This step then makes it possible to reduce the intensity of the dark current.

[0008] However, it appears that this annealing step, intended to modify the depleted zone into an accumulation zone, can induce a degradation of the performance of the photodiode, in particular due to an undesired modification of the dimensions of the n+ doped well, in particular when the lateral diffusion of the n-type doping elements is significant. Furthermore, the presence and characteristics of the depleted zone can be linked to the technique used for depositing the dielectric passivation layer as well as to the operating conditions. As a result, the annealing in question may then not make it possible to reproducibly obtain the desired accumulation zone and therefore the desired reduction in the dark current.

[0009] Also, document EP 3 657 556 A1 describes an example of a passivated photodiode made from germanium comprising a p-doped peripheral region surrounding the n+ doped well and flush with the germanium face covered by the passivation dielectric layer. This p-doped peripheral region is located between the n+ doped well and the peripheral lateral portion, and makes it possible to reduce the dark current by limiting the surface component of the dark current.

[0010] We also know of document EP 3 660 930 A1, which describes another example of a germanium-based passivated photodiode. The passivation layer is not made of a dielectric material but is silicon-based. Annealing is performed to induce interdiffusion between the silicon of the passivation layer and the germanium of the detection layer. Thus, the n+-doped chamber is surrounded by a peripheral SiGe-based zone that forms a "bandgap opening," limiting the surface component of the dark current.

[0011] Finally, document EP 3 971 995 A1 is known, which describes another example of a passivated photodiode made from germanium. It comprises a peripheral portion made from a ferroelectric material, located between and in contact with the intermediate region and the passivation dielectric layer, and which surrounds the n+ doped well in the main plane. This ferroelectric peripheral portion, by the desired orientation of the ferroelectric dipoles during the polarization of the photodiode, makes it possible to limit or avoid the formation of a depleted zone or an inversion zone in the detection portion, and makes it possible to avoid the circulation of charge carriers in this same ferroelectric peripheral portion, between the n+ doped well and the peripheral lateral region.

[0012] However, there is a need for another solution to reduce the dark current in a planar photodiode and in particular its surface component located between the n+ doped box and the peripheral lateral portion. STATEMENT OF THE INVENTION

[0013] The invention aims to remedy at least in part the drawbacks of the prior art, and more particularly to propose a planar photodiode making it possible to obtain a low dark current, and in particular of its surface component.

[0014] For this, the object of the invention is a planar photodiode comprising: a so-called detection portion of a main semiconductor layer, made of at least one crystalline germanium-based semiconductor material M0, having first and second faces parallel to a principal plane, and comprising: a first p-type doped region, flush with the first face; a second n-type doped region, flush with the second face; and an intermediate region, extending between the first and second doped regions, and surrounding the second doped region in the principal plane by flush with the second face; a peripheral lateral portion, intended to be electrically polarized, made of at least one first electrically conductive material M1, extending, along a vertical axis orthogonal to the principal plane, until it comes into contact with the first doped region, and extending into contact with the detection portion by surrounding the intermediate region in the principal plane;the first material M1 having a coefficient of thermal expansion CTE1 lower than that of the material M0, extending from the first face, the intermediate region comprising a first part P1, surrounded by the first material M1 in the main plane, having mechanical stresses in tension.;

[0015] According to the invention, the peripheral lateral portion comprises several distinct electrically conductive materials, stacked on top of each other along the vertical axis, including: at least one second material M2, extending between the first material M1 and the second face, having a thermal expansion coefficient CTE2 greater than or equal to that of the material M0.

[0016] The intermediate region then comprises a second part P2 flush with the second face F2, surrounded by the second material M2 in the principal plane, formed by a peripheral lateral zone Z2p exhibiting compressive mechanical stresses. This peripheral lateral zone Z2p surrounds a central zone Z2c in the principal plane, exhibiting tensile mechanical stresses and where the second doped region is located.

[0017] Some preferred but non-limiting aspects of this photodiode are as follows.

[0018] The first material M1 can extend, along the thickness axis of the photodiode, from the first face F1 over a distance of between 50% and 80% of the thickness of the detection portion, and the second material M2 extends, along the thickness axis of the photodiode, from the first material M1 to the second face F2 over a distance of between 20% and 50% of the thickness of the detection portion.

[0019] The material M1 may be based on p-type doped silicon, and the material M2 may be selected from a metallic material and a p-type doped germanium-based semiconductor material.

[0020] The planar photodiode may comprise a thin interposed portion, made of at least one metallic material, and located between the p-type doped silicon-based material M1 and the metallic material M2.

[0021] The first material M1 can be made from p-type doped silicon and the second material M2 can be made from a metallic material, the detection portion comprising a p-type doped lateral region extending over the entire height of the peripheral lateral portion and in contact with the latter.

[0022] The first material M1 can be made from p-type doped silicon, the detection portion comprising a lateral zone made from silicon germanium, extending over the entire height of the peripheral lateral portion and in contact with the latter.

[0023] The planar photodiode may comprise an upper insulating layer, made of a dielectric material, extending over the second face in contact with the intermediate region and at the peripheral lateral zone Z2p.

[0024] The invention also relates to a method of manufacturing at least one planar photodiode according to any one of the preceding characteristics, comprising the following steps: fabrication of the main semiconductor layer; fabrication of at least one trench to form the detection portion; deposition of the first material M1 at a deposition temperature T1 higher than the ambient temperature, so as to fill the trench, then return to ambient temperature; fabrication of at least one notch by partial etching of the first material M1; deposition of the second material M2 at a deposition temperature T2 higher than the ambient temperature, so as to fill the notch, then return to ambient temperature, thus obtaining the central zone Z2c and the peripheral lateral zone Z2p.

[0025] The method may include a step of producing the second doped region located in the central zone Z2c.

[0026] The main semiconductor layer can be made from a support layer having a thermal expansion coefficient lower than the CTE0 coefficient of the material M0. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Other aspects, aims, advantages and characteristics of the invention will appear better on reading the following detailed description of preferred embodiments thereof, given by way of non-limiting example, and made with reference to the appended drawings in which: there Figure 1A is a schematic cross-sectional view of a planar photodiode according to one embodiment; Figure 1B is a schematic cross-sectional view of a planar photodiode according to an alternative embodiment; figures 2A to 2L illustrate, schematically and partially, different stages of a method of manufacturing a matrix of planar photodiodes identical or similar to that illustrated in the fig.1B . DETAILED DESCRIPTION OF SPECIFIC EMBODIMENTS

[0028] In the figures and in the remainder of the description, the same references represent identical or similar elements. In addition, the different elements are not shown to scale so as to enhance the clarity of the figures. Furthermore, the different embodiments and variants are not mutually exclusive and may be combined with each other. Unless otherwise indicated, the terms "substantially", "approximately", "of the order of" mean to within 10%, and preferably to within 5%. Furthermore, the terms "between ... and ..." and equivalent mean that the limits are included, unless otherwise indicated.

[0029] The invention relates generally to a planar photodiode, and preferably to a matrix of photodiodes, as well as to a method for manufacturing such a photodiode. Each photodiode comprises a detection portion made from germanium, and is adapted to detect light radiation in the near infrared (SWIR) corresponding to the spectral range from 0.8µm to approximately 1.7µm, or even to approximately 2.5µm. In other words, it is adapted to detect light radiation at a cut-off wavelength greater than 1550nm.

[0030] The photodiode(s) are made from the same main semiconductor layer. The latter extends in a main plane, and has a first face and a second face opposite each other and parallel to the main plane. The two faces therefore extend along identical planes for each of the photodiodes, and vertically delimit (along the thickness axis) the detection portion.

[0031] The photodiodes do not have a mesa structure since they are made from the same main semiconductor layer and are optically isolated from each other by a peripheral lateral portion. They thus have a particularly high fill factor. Furthermore, they are preferably passivated here since the face where the n+ doped well is flush is at least partly covered by a dielectric passivation layer. This helps to reduce the surface component of the dark current.

[0032] The dark current of a photodiode is the electric current present within the photodiode in operation, when it is not subjected to light radiation. It can be formed by currents generated thermally inside the volume of the detection portion (diffusion currents, depletion currents, tunnel currents, etc.) and by surface currents. Surface currents can be linked to the presence of electric charges in the passivation dielectric layer. Indeed, these electric charges can induce a modification of the curvature of the energy bands near the surface, leading to the formation of a depleted zone or an inversion zone. The depleted zone, when located in the space charge zone of the photodiode, can give rise to parasitic currents of the generation-recombination type.Furthermore, the inversion zone, which is then electrically conductive, can allow the movement of electric charges between electrically polarized regions located at the interface with the passivation layer.

[0033] In the context of the invention, the planar photodiode comprises: ∘ a so-called detection portion of the main semiconductor layer, made of a germanium-based crystalline semiconductor material M0, and comprising: a first p-type doped region (here a p+ doped sole), a second n-type doped region (here an n+ doped well), and an intermediate region located between the two doped regions and surrounding the n+ doped well in the main plane of the photodiode; ∘ a peripheral lateral portion, intended to be electrically polarized, extending along a thickness axis of the photodiode until it comes into contact with the first p-type doped region, and extending into contact with the detection portion by surrounding the intermediate region in the main plane, and made of several electrically conductive materials (doped semiconductor and / or metal) stacked on top of each other along the thickness axis.

[0034] As will be described in detail later, the peripheral lateral portion comprises: at least one first electrically conductive material M1, having a coefficient of thermal expansion CTE1 lower than the coefficient CTE0 of the germanium-based semiconductor material M0 of the detection portion; and at least one second electrically conductive material M2, having a coefficient of thermal expansion CTE2 greater than or equal to the coefficient CTE0 of the semiconductor material M0.

[0035] The first material M1 and the second material M2 are stacked on top of each other along the thickness axis of the photodiode, such that the first material M1 extends vertically from the first face, and the second material M2 extends vertically between the first material M1 and the second face. This arrangement results in the formation, in the intermediate region: of a first part P1, surrounded by the first material M1 in the main plane, and having mechanical stresses in tension; and of a second part P2 flush with the second face F2, surrounded by the second material M2 in the main plane, and formed of a peripheral lateral zone Z2p having mechanical stresses in compression. This peripheral lateral zone Z2p surrounds a central zone Z2c in the main plane, which has mechanical stresses in tension and where the n+ doped well is located.

[0036] Note that the mechanical stresses in question here are oriented in the principal plane of the photodiode. Mechanical stresses cause deformation of the meshes of the crystal lattice of the stressed material. When the material is in compression, respectively in tension, it has a lower, respectively higher, mesh parameter than its natural value when the material is relaxed (unstressed).

[0037] Also, the presence of this peripheral lateral zone Z2p of the intermediate region where the mechanical constraints are in compression, which is flush with the second face and surrounds the n+ doped box in the main plane, results in a local increase in the forbidden band energy (gap opening) of the germanium-based M0 material, and therefore in a reduction in the surface component of the dark current, which improves the performance of the photodiode.

[0038] THE Figures 1A and 1Bare schematic and partial cross-sectional views of a planar photodiode 1 according to different embodiments, belonging to a matrix of preferably identical photodiodes.

[0039] The planar photodiode 1 of the fig.1B differs essentially from that of the fig.1A in that the peripheral lateral portion 3 comprises a thin intercalated portion 26 situated between and in contact with materials M1 and M2, allowing for improved ohmic contact between these materials. This is particularly the case when material M1 is p+-doped polysilicon and material M2 is titanium nitride. The planar photodiode 1 of the fig.1B also differs from that of the fig.1A by the presence of a p-doped lateral region 14 and a SiGe lateral region 15. Note that either of these features can also be present in the planar photodiode 1 of the fig.1A .

[0040] Here and for the remainder of the description, a three-dimensional direct reference XYZ is defined, where the X and Y axes form a plane parallel to the main plane of the photodiodes 1, and where the Z axis is oriented along the thickness of the detection portion 10 of the photodiode 1, from the first face F1 towards the second face F2. The terms “lower” and “upper” refer to an increasing positioning along the +Z direction defined on the fig.1A .

[0041] Each photodiode 1 is made from germanium, and is here reverse-polarized from the second face F2, while being optically isolated from each other by peripheral lateral portions 3. Each photodiode 1 comprises a detection portion 10 made from germanium, and is adapted to detect light radiation in the near infrared (SWIR).

[0042] The photodiode 1 comprises a so-called detection portion 10 of a main semiconductor layer 2, which extends along the Z axis between a first face F1 and a second face F2. The faces F1 and F2 are parallel to each other and opposite each other. The first face F1 and the second face F2 are common to the photodiodes 1 of the matrix. In this example, the faces F1 and F2 are substantially planar.

[0043] The thickness of the detection portion 10, defined along the Z axis between the first and second faces F1, F2, is here substantially constant from one photodiode to another, for example is between a few hundred nanometers and a few microns, for example between approximately 1 µm and 5 µm. The thickness is chosen so as 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 which can be between a few hundred nanometers and a few tens of microns, for example between approximately 1 µm and 20 µm.

[0044] The detection portion 10 is made of a crystalline semiconductor material based on germanium, denoted M0, preferably monocrystalline. By germanium-based, it is meant that the material M0 is germanium or is a compound (binary or ternary, etc.) formed 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 thus 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. It can alternatively be a stack of sub-layers of different semiconductor materials (heterojunction), which are then formed based on germanium.

[0045] The detection portion 10 is thus formed of a first p-doped region 11, which is flush with the first face F1 and forms a p-doped “base”, and a second n-doped region 12, which is flush with the second face F2 and forms an n-doped “well”. By flush, we mean “reaching the level of”, or “extending from”. An intermediate region 13, not intentionally doped (in the case of a pin junction), or p-doped (in the case of a pn junction), is located between and in contact with the two doped regions 11, 12, and surrounds the n+ doped well 12 in the main plane. In this example, the semiconductor junction is of the pin type, the base 11 being p+ doped, the well 12 being n+ doped and the intermediate region 13 being unintentionally doped.

[0046] The p+ doped sole 11 extends in the XY plane from the lateral edge 10b, flush with the first face F1. As illustrated in fig.1A, it may have a substantially uniform thickness along the Z axis and thus only be flush with a lower zone of the lateral edge 10b. Alternatively, as illustrated in fig.1B , the p+ doped sole may also have a p+ doped lateral region 14 which is continuously flush with the lateral edge 10b along the Z axis and which extends over the entire periphery of the detection portion 10. The p+ doped sole 11 may have a doping which may be between 10 18< and 10 20< at / cm 3< approximately.

[0047] The n+ doped well 12 extends here from the second face F2 and is surrounded by the intermediate region 13 in the main plane. It is distant from the lateral edge 10b of the detection portion 10 in the XY plane insofar as it is surrounded laterally by the intermediate region 13. It may have a doping which may be between 10 19< and 10 21< at / cm 3< approximately.

[0048] The intermediate region 13 is located vertically between the p+ doped sole and the n+ doped well. It laterally surrounds the n+ doped well in the main plane. Here it is made of a non-intentionally doped semiconductor material, so as to form a pin junction but can be doped according to the second type of conductivity, here p-type, to form a pn junction.

[0049] The photodiode 1 preferably comprises a lower insulating layer 21, made of a dielectric material, covering the first face F1 of the detection portion 10, as well as, in this example, the lower face of the peripheral lateral portion 3. The lower insulating layer 21 may further be adapted to form an anti-reflection function with respect to the incident light radiation. It in fact forms the receiving face of the light radiation intended to be detected.

[0050] The detection portion 1 is delimited laterally, in the main plane, by a trench, preferably continuous, filled with several electrically conductive materials stacked along the vertical axis Z, forming a peripheral lateral portion 3. The peripheral lateral portion 3 participates in electrically polarizing the photodiode 1, here from the second face F2, and in pixelating the photodiode matrix (optical isolation). It extends here over the entire thickness of the detection portion 10 to open onto the lower insulating layer 21, but alternatively, it may not open onto the lower insulating layer 21 and may end in the p+ doped sole 11. The internal face of this peripheral lateral portion 3 is in contact with the lateral edge 10b of the detection portion 10.As detailed below, the peripheral lateral portion 3 is formed of at least one first electrically conductive material M1 having a thermal expansion coefficient CTE1 lower than the coefficient CTE0 of the germanium-based material M0 of the detection portion 10, and of at least one second electrically conductive material M2 having a thermal expansion coefficient CTE2 greater than or equal to the coefficient CTE0 of the material M0.

[0051] Furthermore, as illustrated by the fig.1B, and in the case where the material M1 is silicon-based, the detection portion 10 advantageously comprises a p-type doped lateral region 14 located at the level of the lateral edge 10b. This lateral region 14 has a doping level higher than that of the intermediate region 13 when it is doped. The p+ doped lateral region 14 is flush with the lateral edge 10b and is in contact here with the peripheral lateral portion 3. In this example, the lateral region 14 extends over the entire height of the peripheral lateral portion 3, even at the level of the second material M2. This comes from the fact that the material M1 was deposited over the entire height of the trench 23 (cf. fig.2D). Thus, the polarization of the p+ doped sole 11 is improved to the extent that the contact surface with the peripheral lateral portion 3 is increased. In addition, this p+ doped lateral region 14 makes it possible to prevent the space charge zone of the photodiode 1 from extending to the lateral edge 10b. Thus, the contribution of this zone (potentially not free from defects linked to the production of the trenches) to the dark current is limited. This improves the performance of the photodiode 1.

[0052] Furthermore, as also illustrated by the fig.1B, in the case as here where the detection portion 10 is made of germanium, and where the material M1 of the peripheral lateral portion 3 is based on silicon, the detection portion 10 advantageously comprises a lateral zone 15 made from silicon germanium. The lateral zone 15 is flush with the lateral edge 10b and is in contact with the peripheral lateral portion 3. In this example, the lateral zone 15 extends over the entire height of the peripheral lateral portion 3, even at the level of the second material M2. This comes from the fact that the material M1 has been deposited over the entire height of the trench 23 (cf. fig.2D Thus, the lateral zone 15 has a higher band gap energy than the germanium detection portion 10. This lateral "band gap opening" reduces the sensitivity of photodiode 1 to defects near trenches 24. This also improves the performance of photodiode 1.

[0053] An upper insulating layer 22 covers the second face F2 of the photodiode 1, and allows the material M0 at face F2 to be passivated, and provides electrical insulation for the metallic contacts 32.1 and 32.2. It is thus in contact with the n+-doped housing 12 and the intermediate region 13. It is preferably made of a dielectric material, such as silicon oxide, silicon nitride, or silicon oxynitride. This is then a passivation dielectric layer. Other dielectric materials can be used, such as hafnium or aluminum oxide, or even aluminum nitride, among others. Its thickness, for example, is between 50 nm and 500 nm.

[0054] The photodiode 1 also includes metallic contacts 27.1 and 27.2 that allow it to be reverse-biased from the side of face F2. A metallic contact 27.2 is positioned on and in contact with the n+-doped cell 12, ensuring its electrical bias. A metallic contact 27.1 is positioned on and in contact with the peripheral lateral portion 3, allowing the p+-doped base 11 to be electrically biased. The metallic contacts 27.1 and 27.2 are electrically isolated from each other in the XY plane by the upper insulating layer 22. The photodiode 1 is intended to be reverse-biased, for example, by applying a negative electrical potential to the peripheral lateral portion 3 and grounding the n+-doped cell 12.

[0055] Generally speaking, for illustration purposes, the photodiode 1 may have dimensions in the XY plane of between approximately 1 µm and 100 µm. The thickness of the p+ doped base 11 may be between approximately 20 nm and 500 nm. The thickness of the intermediate region 13 may be between approximately 0.7 µm and 2.5 µm when the photodiode 1 is intended to detect light radiation in the SWIR or near infrared (NIR) range. The n+ doped well 12 may have a depth of between approximately 10 nm and 600 nm, for example approximately 300 nm. The upper insulating layer 22 can have a thickness that allows the F2 face of the photodiode 1 to be completely covered, for example between approximately 10nm and 600nm, and the thickness of the lower insulating layer 21 can be between approximately 50nm and 1µm.

[0056] To reduce the surface component of the dark current, the peripheral lateral portion 3 is formed of several electrically conductive materials (doped semiconductors, metals, etc.), stacked vertically along the Z axis and arranged in electrical contact with each other, and in contact with the detection portion. They have well-chosen thermal expansion coefficients.

[0057] The objective here is to generate different mechanical stresses within the intermediate region 13: on the one hand, mechanical stresses in tension in the light radiation absorption zone, i.e. between the p+ doped base 11 and the n+ doped box 12, so as to increase the absorption cut-off wavelength and thus obtain a photodiode capable of operating in the SWIR; and on the other hand, mechanical stresses in compression in a zone flush with the second face F2 and surrounding the n+ doped box 12 in the main plane, so as to locally open the gap and thus reduce the surface component of the dark current.

[0058] The first material M1 and the second material M2 are electrically conductive, insofar as they participate in ensuring the electrical polarization of the p+ doped sole 11 from the metal contact 27.2 located at the second face F2. Also, the first material M1 and the second material M2 may be chosen from p-type doped semiconductors and metals. For example, the first material M1 may be a p-type doped silicon-based material, for example p+ doped polysilicon. The second material M2 may be a metallic material or a p-type doped germanium-based semiconductor material, for example p+ doped germanium or a titanium nitride, or even a Ti / TiN, Ni / TiN, Ti / Cu or Ti / TiN / Cu combination.

[0059] The first material M1 and the second material M2 are stacked on top of each other along the Z axis. They can be in contact with each other (cf. fig.1A). This is the case in particular when the second material M2 is p+ doped germanium which rests on the p+ doped polysilicon material M1. Alternatively, they can be separated vertically from each other by a thin interposed portion 26 making it possible to improve the ohmic contact (cf. fig.1B ). This is the case in particular when the second material M2 is titanium nitride which rests on the material M1 in p+ doped polysilicon. They are also in contact with the lateral edge 10b of the detection portion 10, so as to be able to generate the desired mechanical stresses within the detection portion 10.

[0060] The first material M1 extends from the first face F1 over a distance along the +Z direction preferably at least equal to 50% of the thickness of the detection portion 10, and preferably over a distance between 50% and 80% of this thickness. The second material M2 extends from the first material M1 to the second face F2, over a distance along the +Z direction preferably at most equal to 50% of the thickness of the detection portion 10, and preferably over a distance between 20% and 50% of this thickness, depending on the height of the first material M1.

[0061] The first material M1 has a coefficient of thermal expansion CTE1 lower than the coefficient CTE0 of the germanium-based semiconductor material M0 of the detection portion 10. The second material M2 has a coefficient of thermal expansion CTE2 greater than or equal to the coefficient CTE0 of the semiconductor material M0.

[0062] For example, in the case where the detection portion 10 is made of germanium, the first material M1 can be made of p+ doped polysilicon, and the second material M2 can be made of titanium nitride or p+ doped germanium. The CTE0 coefficient of germanium is equal to 5.9 µm / (m×K), that of silicon is equal to 2.6 µm / (m×K), and that of titanium nitride is equal to 9.0 µm / (m×K).

[0063] The fact that the peripheral lateral portion is made from this first material M1 and this second material M2 generates different mechanical stresses within the intermediate region 13.

[0064] More specifically, the intermediate region 13 has a first part P1, which is surrounded in the XY plane by the first material M1. Due to the thermal expansion differential between the first material M1 and that of the intermediate region 13, the first part P1 has mechanical stresses in tension, while the first material M1 has mechanical stresses in compression.

[0065] The mechanical stresses in tension undergone locally in the part P1 of the intermediate region 13 then result in a local modification of the optical and / or electrical properties of the photodiode 1, including in particular a broadening of the spectral range of absorption of light radiation: the absorption cut-off wavelength can then be greater than 1550 nm, so that the photodiode 1 can then operate in a larger spectral range, or even in the entire spectral range of the SWIR. In addition, the tensile stress can also be sufficient to make the energy band structure of the germanium in the part P1 of the intermediate region 13 substantially direct, whereas the band structure of the relaxed germanium is indirect.

[0066] In addition, the intermediate region 13 has a second part P2, which is surrounded in the XY plane by the second material M2. This second part P2 is therefore flush with the second face F2 and is located above the first part P1 along the Z axis. Due to the thermal expansion differential between the second material M2 and that of the intermediate region on the one hand, and due to the thermal expansion differential between the second material M2 and the first material M1 on the other hand, the second part P2 is formed of a peripheral lateral zone Z2p, which is flush with the second face F2 and is in contact with the second material M2, and which has mechanical stresses in compression. It appears that this peripheral lateral zone Z2p surrounds a central zone Z2c (of the detection portion 10, not of the intermediate region 13), which is also flush with the second face F2 and has mechanical stresses in tension.Note also that the second material M2 presents mechanical stresses in tension.

[0067] The n+ doped well 12 is located mainly in the central zone Z2c, that is to say that it can be entirely located in the central zone Z2c or can have a lateral border which is located in the peripheral lateral zone Z2p. Furthermore, the mechanical stresses remain continuously in tension when going from the central zone Z2c towards the part P1 of the intermediate region 13, which makes it possible to maintain good performances of the photodiode 1.

[0068] The peripheral lateral zone Z2p is also flush with the second face F2, and comes laterally into contact with the second material M2. It surrounds the central zone Z2c in the main plane, and therefore also the n+ doped well 12. It is located above the first part P1. Insofar as the intermediate region 13 locally presents, i.e. in the peripheral lateral zone Z2p, mechanical compressive stresses, the intermediate region 13 presents a gap opening there which reduces the surface component of the dark current.

[0069] Also, by this configuration of the peripheral lateral portion 3 which induces zones of different mechanical constraints within the intermediate region 13, and in particular the part P1 in tension and the zone Z2p in compression, the photodiode 1 is then able to detect light radiation up to a cutoff wavelength greater than 1550nm, and also exhibits good performance due to the reduction of the surface component of the dark current at the level of the face F2.

[0070] THE figures 2A to 2L illustrate, schematically and partially, different stages of a manufacturing process for a planar photodiode array 1 identical or similar to that illustrated on the fig.1B .

[0071] Photodiodes 1 are made of germanium and have a pin junction. They are suitable for detecting infrared radiation in the SWIR range. Photodiodes 1 are planar and passivated, and are reverse biased from the second face F2 by a control chip 30 hybridized to the photodiode matrix.

[0072] In a first step ( fig.2A ), a first semiconductor sub-layer 2.1 of monocrystalline germanium is produced. The first semiconductor sub-layer 2.1 is secured to a support layer 20, preferably a thick substrate ( waferin English) made of a material having a coefficient of thermal expansion CTEw lower than that of the material M0 of what will be the detection portion 10. In this example, the support layer 20 is a silicon substrate. The first semiconductor sub-layer 2.1 is assembled to the support layer 20 via the lower insulating layer 21, here made of a silicon oxide. This stack takes the form of a GeOI substrate (for Germanium On Insulator, This stack is preferably made using the method described in the publication by Reboud et al. entitled Structural and optical properties of 200mm germanium-on-insulotor (GeOI) substrates for silicon photonics applications, Proc. SPIE 9367, Silicon Photonics X, 936714 (February 27, 2015).

[0073] The process of Reboud et al. 2015 has the advantage of producing a germanium 2.1 semiconductor sub-layer with a total absence or a low rate of structural defects such as dislocations. The germanium may be unintentionally doped or be doped, for example p-type. The 2.1 semiconductor sub-layer may have a thickness of between approximately 100nm and 500nm, for example approximately 300nm, and may be covered with a protective layer (not shown) made of silicon oxide. The lower insulating layer 21 (BOX, for Buried Oxide, in English) can have a thickness between 50nm and 1µm, for example between 100nm and 500nm, and advantageously provides an anti-reflective function.

[0074] A p-type doping of the first sub-layer 2.1 in germanium is then carried out, by ion implantation of a dopant such as boron or gallium, when the first sub-layer 2.1 was initially in intrinsic germanium. The protective layer, if any, has been previously removed by surface cleaning, and the first sub-layer 2.1 of germanium can be coated with a pre-implantation oxide layer (not shown) with a thickness of a few tens of nanometers, for example equal to 20 nm. The sub-layer 2.1 of germanium then has a doping level of between 10 18< and 10 20< at / cm 3< approximately. A diffusion annealing of the dopant can then be carried out under nitrogen, for a few minutes to a few hours, for example 1 hour, at a temperature which can be between 600°C and 800°C, for example equal to 800°C. This annealing may not be performed when sublayer 2.1 was growth-doped.Another way to fabricate this p+ layer is by epitaxy of a boron-doped germanium layer in situ at approximately 10¹⁸ to 10²⁰ at / cm³ onto an intrinsic germanium sublayer. This epitaxy can be carried out between 400 and 800°C, but preferentially at 400°C.

[0075] A second germanium semiconductor sublayer 2.2 is then created by epitaxy from the first sublayer 2.1. The two sublayers 2.1 and 2.2 are intended to form the main semiconductor layer 2 as well as the detection portions 10. The second sublayer 2.2 is formed by epitaxy, for example by chemical vapor deposition (CVD). Chemical Vapor Deposition (in English) and reduced pressure precedence (RPCVD, for Reduced Pressure Chemical Vapor Deposition) or by any other epitaxy technique. Annealing can be carried out to reduce the dislocation rate in the sub-layer 2.2. The pre-implementation oxide layer, if any, has been previously removed by surface cleaning. The second sub-layer 2.2 of germanium is here intrinsic, i.e. not intentionally doped since it is desired to produce a pin junction. It is intended to form the light absorption zone of the photodiodes 1. Its thickness depends on the wavelength range of the light radiation to be detected in the case of a photodiode 1. In the context of SWIR photodiodes, the sub-layer 2.2 of intrinsic germanium has a thickness for example between 0.5 µm and 3 µm, preferably equal to 1.5 µm.

[0076] An upper insulating layer 22 is then deposited so as to continuously cover the upper face of the second sub-layer 2.2, that is to say so as to cover what will be the detection portions 10 of the photodiodes 1. The upper insulating layer 3 is made of a dielectric material, for example a silicon oxide, nitride or oxynitride. The upper face of the second sub-layer 2.2 may have been cleaned beforehand. The upper insulating layer 22 may have a thickness of between 10nm and 600nm.

[0077] In reference to the fig.2B, the trenches 23 intended to pixelate the photodiodes 1 and to participate in their electrically reverse polarization by the peripheral lateral portions 3 are produced by photolithography and etching. A localized etching of the upper insulating layer 22, of the intrinsic germanium sub-layer 2.2, and of the p+ doped germanium sub-layer 2.1 is thus carried out, until it opens here onto the upper face of the lower insulating layer 21 (but the trenches 23 can open onto the sub-layer 2.1 without crossing it). Each trench 23 thus preferably extends continuously around a photodiode 1. A plurality of detection portions 10 are thus obtained, separated from each other by a continuous trench 23. They are preferably obtained by an anisotropic etching technique, so as to obtain a lateral border 10b substantially vertical along the Z axis.The trenches 23 have a transverse dimension (width) in the XY plane which can be between 300nm and 2µm, for example equal to 1µm. The detection portions 10 can thus have a shape in the XY plane, for example circular, oval, polygonal, for example square, or any other shape.

[0078] In reference to the fig. 2C to 2E , we begin the production of the peripheral lateral portions 3 by depositing the first material M1 at high temperature.

[0079] For this, as illustrated by the fig.2C, the structure is brought to a deposition temperature T1 higher than ambient temperature, for example to a few hundred degrees Celsius, such as for example at least equal to approximately 500°C, for example to 650°C in the case here of silicon deposition. Due to the presence of the trenches 23, the germanium of the detection portions 10 expands in the XY plane. This expansion of the germanium is accompanied here by a curvature of the main layer 2 due to the thermal expansion differential between the silicon of the substrate 20 and the germanium of the main layer 2.

[0080] Note that the fig.2Cillustrates the deformation field in the main plane of the structure when heated to 650°C. These results are from a numerical simulation carried out using COMSOL software. In this example, the trenches 23 are 2.3µm deep and 1µm wide, and the germanium detection portion 10 has a width of 4µm. The germanium rests on a silicon layer 20 with a thickness of 10µm (here to reduce calculation times). But the Si layer preferably has a thickness of 725µm ( Bulk Silicon ). It appears that germanium here exhibits a maximum tensile deformation of the order of 0.65%.

[0081] As illustrated by the fig.2D, the epitaxial growth of the first material M1, here p+ doped silicon, is carried out in the trench 23, and here from the free surface of the lateral edge 10b of the detection portions 10. Insofar as the first material M1 is here silicon, the epitaxial growth is accompanied by a formation of the lateral region 14 based on SiGe and of the lateral zone 15. The epitaxially grown silicon grows from the free germanium surface and fills the trenches 23. It is therefore in contact with the lateral edge 10b of the detection portions 10. It is deposited so as to protrude from the trenches 23 and thus partially covers the upper insulating layer 22.

[0082] Following the epitaxy step, the temperature is brought back to a final temperature lower than the epitaxy temperature, for example to room temperature, for example to 25°C, which causes the germanium of the detection portions 10 to be put under tension, as well as the compression of the first material M1.

[0083] The material M1 is preferably a silicon-based material, for example amorphous silicon, polycrystalline silicon, silicon germanium. The material M1 is doped according to the second type of conductivity, here p-type with boron or gallium, with a dopant concentration of the order of 10 19< to 10 20< at / cm 3< approximately.

[0084] As illustrated by the fig.2E, a chemical mechano-polishing (CMP) step is then carried out, stopping on the upper face of the upper insulating layer 23, to remove the excess material M1 and planarize the upper face formed by the upper insulating layer 23 and the first material M1. This gives a peripheral lateral portion which is currently entirely filled by the first p+ doped silicon material M1.

[0085] In reference to the fig. 2F to 2I , we continue the production of the peripheral lateral portions 3 by depositing the second material M2 at high temperature.

[0086] For this reason, as illustrated by the fig.2F, a partial etching of the first material M1 is carried out to form a notch 25. A notch 25 is here a non-through trench, opening onto a non-etched part of the material M1. A resin etching mask 24 is deposited on the upper insulating layer 22 so as to leave the first material M1 free. Then, the first material M1 is thus etched so as to leave free a surface of the lateral edge 10b of the detection portion 10. The first material M1 is partially etched along the Z axis, up to a predefined depth so that the first material M1 extends from the first face F1 over a distance at least equal to 50% of the thickness of the detection portion 10.

[0087] Following this partial etching step of the material M1, the intermediate region 13 presents a part P1, surrounded by the material M1, which presents mechanical stresses in tension, surmounted by a part P2, surrounded by the notch 25, where the germanium is then substantially relaxed.

[0088] If the second material M2 is TiN, it can be advantageous to then form a thin interlayer 26, a few nanometers thick, of Ni or Ti covering the upper surface of the first material M1. The second material M2 is then in electrical contact with the material M1 via the thin interlayer 26. This improves the ohmic contact between the materials M1 and M2. If the material M2 is p+-doped germanium, such a thin interlayer 26 can be omitted.

[0089] As illustrated by the fig.2G, the second material M2 is then grown by epitaxy in the notch 25, from the first material M1 and the free surface of the lateral edge 10b of the detection portion 10. The material M2 is deposited so as to completely fill the notches 25, and here it covers at least in part the upper insulating layer 22. The deposition temperature T2 is higher than ambient temperature, for example a few hundred degrees Celsius, for example at least equal to 300°C or 400°C in the case of TiN, or at least equal to 500°C in the case of p+ doped germanium.

[0090] Following the epitaxial step, the temperature is lowered to a final temperature below the epitaxial temperature, for example, to room temperature, such as 25°C. This causes the formation, in the P2 portion of the intermediate region 13, of the central tension zone Z2c and the peripheral compression zone Z2p. The material M2 is then under tension. Note that if the material M2 is a doped semiconductor, such as p+-doped germanium, the doping can be carried out during growth or after deposition by ion implantation.

[0091] As illustrated by the fig.2H, a chemical mechanical polishing (CMP) step is then carried out, stopping on the upper face of the upper insulating layer 22, to remove the excess material M2 and planarize the upper face formed by the upper insulating layer 22. In this example, a peripheral lateral portion 3 is thus obtained which is formed by the material M1 in p+ doped silicon topped with the material M2 here in TiN.

[0092] There fig.2I illustrates the deformation field in the main plane of the structure then brought to 650°C, which is similar to that of the fig.2CIn this example, the p+ doped polysilicon material M1 extends over a height of 1.7µm and the TiN material M2 extends over a height of 0.8µm. The germanium is then in tension in the P1 part of the intermediate region 13 (and the M1 material is in compression), and in tension in the central zone Z2c while being in compression in the peripheral lateral zone Z2p (and the M2 material is in tension).

[0093] In a next step ( fig.2J), the n+-doped well 12 is produced. To do this, a protective layer of a dielectric material, for example silicon oxide, is deposited, with a thickness of, for example, between approximately 10 nm and 30 nm (this protective layer and the upper insulating layer together form the layer 22). Then, by photolithography, an opening corresponding to an implantation zone defining the n+-doped well 12 is defined in a photosensitive resin (not shown). An ion implantation of a dopant such as phosphorus or arsenic is then carried out through the openings. The well 12 is preferably n+-overdoped, and can thus have a doping level of between approximately 5.10 18< and 3.10 20< at / cm 3<. The n+ doped box 12 is delimited in the XY plane and along the Z axis by the intermediate region 13. After removal of the photosensitive resin, a diffusion annealing of the dopants can be carried out.Note here that the n+ doped well 12 is mainly located in the central zone Z2c. Here it is entirely located in the central zone Z2c, but it is possible that its lateral border is located in the peripheral lateral zone Z2p (cf. . fig.1A and 1B ).

[0094] Note that the n+ doped well 12 can also be made, not by ion implantation, but by diffusion of dopants from a p-doped germanium pad made by localized epitaxy on the F2 face. This improves the crystalline quality of the germanium in the detection portion 10.

[0095] In a next step ( fig.2K), openings are then made by localized etching of the upper insulating layer 22, which open onto the n+ doped well 12 and onto the peripheral lateral portion 3. Contact metallizations 27 are then made in the openings, which can be formed of a Ti / TiN cladding and a tungsten W or copper core. For this, a thin metallic layer is deposited in the openings in a conformal manner, for example formed of a sub-layer of titanium Ti with a thickness of approximately 5 nm then a sub-layer of titanium nitride TiN with a thickness of approximately 10 nm. Annealing is carried out so as to improve the electrical contact between the titanium and the germanium of the n+ doped well 12, and between the titanium and the peripheral lateral portion 3 when it is made of germanium. During this annealing, the dopants of the n+ doped well 12 can diffuse so that it occupies a larger volume in the detection portion 10.A metallic layer, for example made of tungsten W or copper Cu, is then deposited to fill the openings. A mechano-chemical polishing step is then carried out, stopping on the upper insulating layer 22. This results in contact metallizations 27.1 and 27.2 consisting of a Ti / TiN sheath and a W or Cu core.

[0096] In a next step ( fig.2L), the hybridization of the photodiode matrix 1 thus obtained is carried out on a control chip 30. The connection face of the control chip 30 can thus be coated with an insulating layer 31, made of a dielectric material, crossed by contact metallizations 32.1, 32.2. The photodiode matrix 1 and the control chip 30 are thus assembled by hybrid molecular adhesion. A bonding annealing can be carried out so as to increase the surface bonding energy between the two contacting faces. The support layer 20 is then removed, for example by grinding ( grinding ), so as to expose one face of the lower insulating layer 21. This thus forms the face receiving the light radiation to be detected.

[0097] The manufacturing method thus makes it possible to obtain one or more passivated planar photodiodes 1, which have improved performance, due to the reduction of the surface component of the dark current at the level of the face F2 between the n+ doped well 12 and the peripheral lateral portion 3. The detection portion 10 is tensioned in the part P1 and locally compressed in the zone Z2p, following the steps of high-temperature deposition of the materials M and M2 of the peripheral lateral portion 3, due to the thermal expansion differentials.

[0098] Specific embodiments have just been described. Different variations and modifications will be apparent to those skilled in the art.

Claims

1. Planar photodiode (1), comprising: ∘ a detection portion (10) of a main semiconductor layer (2), produced from at least one germanium-based crystalline semiconductor material M0, having first and second faces (F1, F2) parallel to a main plane, and comprising: a p-doped first region (11) flush with the first face (F1); an n-doped second region (12) flush with the second face (F2); and an intermediate region (13) extending between the first and second doped regions (11, 12) and surrounding the doped second region (12) in the main plane while fitting flush with the second face (F2); ∘ a peripheral lateral portion (3) intended to be electrically biased, produced from at least one first electrically conductive material M1 extending, along a vertical axis orthogonal to the main plane, until it comes into contact with the first doped region (11), and extending in contact with the detection portion (10) while surrounding the intermediate region (13) in the main plane; • the first material M1 having a coefficient of thermal expansion CTE1 lower than that of the material M0, extending from the first face (F1), ▪ the intermediate region (13) comprising a first part P1 surrounded by the first material M1 in the main plane, having mechanical tensile stresses; ∘ characterised in that: the peripheral lateral portion (3) comprises a plurality of distinct electrically conductive materials, stacked one on the other along the vertical axis, among them: • at least one second material M2, extending between the first material M1 and the second face (F2), having a coefficient of thermal expansion CTE2 higher than or equal to that of the material M0, ▪ the intermediate region (13) comprising a second part P2 fitting flush with the second face F2, surrounded by the second material M2 in the main plane, formed by a peripheral lateral zone Z2p having mechanical compressive stresses, which surrounds in the main plane a central zone Z2c having mechanical tensile stresses and where the second doped region (12) is located.

2. Planar photodiode (1) according to claim 1, wherein the first material M1 extends, along the thickness axis of the photodiode, from the first face F1 over a distance of between 50% and 80% of the thickness of the detection portion (10), and the second material M2 extends, along the thickness axis of the photodiode, from the first material M1 as far as the second face F2 over a distance of between 20% and 50% of the thickness of the detection portion (10).

3. Planar photodiode (1) according to claim 1 or 2, wherein the material M1 is based on p-doped silicon, and the material M2 is selected from a metal material and a semiconductor material based on p-doped germanium.

4. Planar photodiode (1) according to any one of claims 1 to 3, comprising an intermediate end portion produced from at least one metal material and located between the material M1 based on p-doped silicon and the metal material M2.

5. Planar photodiode (1) according to any one of claims 1 to 4, wherein the first material M1 is produced from p-doped silicon and the second material M2 is produced from a metal material, the detection portion (10) comprising a p-doped lateral region (14) extending over the entire height of the peripheral lateral portion (3) and in contact with the latter.

6. Planar photodiode (1) according to any one of claims 1 to 5, wherein the first material M1 is produced from p-doped silicon, the detection portion (10) including a lateral zone (15) produced from silicon germanium, extending over the entire height of the peripheral lateral portion (3) and in contact with the latter.

7. Planar photodiode (1) according to any one of claims 1 to 6, including an upper insulating layer (22) produced from a dielectric material, extending over the second face (F2) in contact with the intermediate region (13) and at the peripheral lateral zone Z2p.

8. Method for manufacturing at least one planar photodiode (1) according to any one of claims 1 to 7, comprising the following steps: ∘ producing the main semiconductor layer (2); ∘ producing at least one trench (23) to form the detection portion (10); ∘ depositing the first material M1 at a deposition temperature T1 higher than ambient temperature, so as to fill the trench (23), and then returning to ambient temperature; ∘ producing at least one recess (25) by partial etching of the first material M1; ∘ depositing the second material M2 at a deposition temperature T2 higher than ambient temperature, so as to fill the recess (25), and then returning to ambient temperature, thus obtaining the central zone Z2c and the peripheral lateral zone Z2p.

9. Manufacturing method according to claim 8, including a step of producing the second doped region (12) located in the central zone Z2c.

10. Manufacturing method according to claim 8 or 9, wherein the main semiconductor layer (2) is produced from a support layer (20) having a coefficient of thermal expansion lower than the coefficient CTE0 of the material M0.