Avalanche photodiode

Abstract

Avalanche photodiode The present description relates to an avalanche photodiode (200) comprising:- a main PN junction (230) adapted to be reverse biased; and- a plurality of semiconductor regions including at least: - a first epitaxially grown semiconductor region (211) of a first conductivity type; and - a second semiconductor region (212) of the second conductivity type, said second region being arranged to at least partially surround the first region, and comprising surfaces in contact with surfaces of said first region. The present description also relates to a method of manufacturing such a photodiode. Figure for abstract: Fig. 2

Description

Description Title of the invention: Avalanche photodiode Technical field The present description relates generally to avalanche photodiodes, in particular single photon avalanche diodes, called SPAD photodiodes (from the English "Single Photon Avalanche Diode") and their method of manufacturing. Prior art A photodiode is a semiconductor component capable of capturing radiation from the optical domain and transforming it into an electrical signal. A SPAD photodiode is a photodiode comprising a PN junction adapted to be reverse biased to a voltage greater than the breakdown voltage, or avalanche voltage, of said junction. At this bias voltage, the electric field around the PN junction is sufficient for a photogenerated carrier reaching the space charge region, or depleted region, of the junction to trigger an avalanche phenomenon, resulting in an increase in the current in the photodiode. In a SPAD photodiode, a single photogenerated carrier can trigger an avalanche in the depleted region by impact ionization. SPAD photodiodes can detect very low light intensity radiation and are used in particular for single photon detection and photon counting. Electronic devices may include multiple photodiodes, or multiple pixels each including at least one photodiode. For example, an image sensor is an electronic device that may include a plurality of photodiodes, the photodiodes enabling the image sensor to obtain an image of a scene at a given time. The image generally consists of an array of pixels, each pixel including at least one photodiode. The increasing miniaturization of electronic devices, particularly image sensors, is leading to the miniaturization of pixels and / or photodiodes. To miniaturize a SPAD photodiode, one solution is to reduce at least one of its dimensions in the plane of the electronic circuit integrating this photodiode, for example its width. However, reducing the width of the photodiode has the disadvantage of reducing the photon collection volume, and thus notably reducing the photon detection probability (PDP). To compensate for this reduction in volume, one solution is to increase the height of the photodiode, i.e. the dimension in the direction perpendicular to the plane of the electronic circuit. However, increasing the height of the photodiode may have the disadvantage of decreasing the probability that a photogenerated carrier will reach the multiplication zone, i.e. the zone where the electric field is strong enough for impact ionization effects to take place and therefore an avalanche to occur, or of increasing the time during which the carrier can reach this multiplication zone, especially since the photogenerated carrier is at a further distance from this multiplication zone of the photodiode. Indeed, beyond a certain distance from the PN junction, the electric field resulting from the reverse bias of the PN junction is canceled or greatly attenuated, and allows less, or even no longer allows, the photogenerated carriers to be driven towards the multiplication zone.Beyond a certain distance from the PN junction, only random diffusion in the photodiode can then be likely to drive the photogenerated carriers towards the multiplication zone, with a non-negligible probability that the photogenerated carriers never reach the multiplication zone or reach it with a significant delay. As a corollary, this can increase time jitter, which is the statistical fluctuation of the detection delay. Summary of the invention There is a need for an avalanche photodiode, for example of the SPAD type, which overcomes all or part of the disadvantages of known avalanche photodiodes. In particular, there is a need for a SPAD photodiode whose size can be reduced without degrading the photon detection probability (PDP) and / or increasing timing jitter. It would be advantageous to have such a photodiode without complicating the manufacturing process of this photodiode, and increasing the manufacturing cost. One embodiment provides an avalanche photodiode comprising: - a main PN junction adapted to be reverse biased; and - a plurality of semiconductor regions including at least: - a first epitaxially grown semiconductor region of a first conductivity type; and - a second semiconductor region of the second conductivity type, said second region being arranged to at least partially surround the first region, and comprising surfaces in contact with surfaces of said first region. One embodiment provides a method of fabricating an avalanche photodiode, comprising a main PN junction adapted to be reverse biased, the method comprising forming a plurality of semiconductor regions including at least : - a first epitaxially grown semiconductor region of a first conductivity type; and - a second semiconductor region of the second conductivity type, said second region being arranged to at least partially surround the first region, and comprising surfaces in contact with surfaces of said first region. According to one embodiment, the main PN junction is formed by fourth and fifth regions of the plurality of semiconductor regions; the fourth region being heavily doped with the first conductivity type, for example more heavily doped than the first region, and being formed from a first face of the photodiode; and the fifth region being heavily doped with the second conductivity type, for example more heavily doped than the second region, and being formed under, and in contact with, the fourth region; the fourth and fifth regions being for example located at the center of the photodiode in a plane parallel to the first face. According to one embodiment, a sixth region of the plurality of semiconductor regions is formed laterally around, and at a distance from, the fourth region from the first face of the photodiode, the sixth region being heavily doped with the second conductivity type, for example more heavily doped than the second region. According to one embodiment, a third semiconductor region of the plurality of semiconductor regions is formed between the first face of the photodiode and the first region, the third region being an epitaxial region of the first conductivity type, and the fourth, fifth and sixth regions being formed by doping, for example by ion implantation, in said third region. According to a particular embodiment, the first and third regions are epitaxially grown and doped under similar conditions, for example corresponding to regions of the same epitaxially grown layer. In one embodiment, a guard ring region of the plurality of semiconductor regions is formed around the fourth region, the guard ring region being of the first conductivity type and being more lightly doped than the fourth region. According to one embodiment, the semiconductor regions are formed in a box delimited by a deep insulating trench extending in a direction perpendicular to the first face of the photodiode. According to one embodiment, the second region comprises a lateral portion located between the first region and the deep insulating trench. According to one embodiment, the second region comprises a buried portion located between the first region and a second face of the photodiode opposite the first face. According to one embodiment, the first region is doped substantially constantly in a direction perpendicular to the first face of the photodiode. According to one embodiment, the first region is doped gradually in a direction perpendicular to the first face, for example in a decreasing manner between said first face and a second face of the photodiode opposite the first face. According to one embodiment, the first region comprises a first portion and a second portion which is less wide, in a plane parallel to the first face, than the first portion, and located between the first portion and the main PN junction. According to one embodiment, the first region surrounds the main PN junction, for example comprises a first portion located under the main PN junction, and a second portion extending to the first face of the photodiode. According to one embodiment, the first region is inside the second region. According to one embodiment, each dimension of the photodiode, in a plane parallel to a first face of the photodiode, is less than 6 um and / or the height of the photodiode is between 3 and 15 um, for example between 6 and 10 um. According to one embodiment, the formation of the plurality of semiconductor regions comprises: - the formation by epitaxial growth of an epitaxial layer of the first conductivity type intended to form the first semiconductor region; then - the formation, by partial doping of the epitaxial layer, for example by ion implantation, of at least a first portion of the second region; and the fourth region and the fifth region are formed by partial doping of the epitaxial layer, for example by ion implantation, from the first face of the photodiode. According to one embodiment, the lateral portion of the second region is formed by partial doping of the epitaxial layer, for example by ion implantation, by partial change in conductivity of the epitaxial layer, by diffusion from the deep insulating trench, or by a charge inversion material in the deep insulating trench. According to one embodiment, the buried portion of the second region is formed by epitaxial growth, or by partial doping of the epitaxial layer, for example by ion implantation from the second face of the photodiode. Brief description of the drawings These and other features and advantages will be set forth in detail in the following description of particular embodiments given without limitation in relation to the attached figures, among which: [Fig. 1] is a partial, schematic sectional view showing an example of a photodiode; [Fig.2] is a partial and schematic sectional view showing a photodiode according to one embodiment: [Fig.3] illustrates in a very simplified manner doping profiles in arbitrary units of a photodiode similar to the photodiode of [Fig.2]; [Fig.4] is a partial, schematic sectional view showing another example of a photodiode; [Fig. 5] is a partial and schematic sectional view showing a photodiode according to another embodiment: and [Fig.6A], [Fig.6B], [Fig.6C], [Fig.6D], [Fig.6E], [Fig.6F] and [Fig.6G] are sectional views showing steps of a method of manufacturing a photodiode according to one embodiment. Description of the embodiments The same elements have been designated by the same references in the different figures. In particular, the structural and / or functional elements common to the different embodiments may have the same references and may have identical structural, dimensional and material properties. For the sake of clarity, only the steps and elements useful for understanding the described embodiments have been shown and are detailed. In particular, a SPAD photodiode generally comprises ancillary circuits, in particular a circuit for biasing the PN junction to a voltage greater than its avalanche threshold, a reading circuit adapted to detect a triggering of an avalanche of the photodiode, as well as a quenching circuit having the function of interrupting the avalanche of the photodiode once it has been triggered. These ancillary circuits have not been shown in the figures and will not be detailed, the described embodiments being compatible with the ancillary circuits equipping known SPAD photodiodes. Unless otherwise specified, when referring to two elements connected together, this means directly connected without intermediate elements other than conductors, and when referring to two elements connected (in English "coupled") together, this means that these two elements can be connected or be connected by means of one or more other elements. In the following description, when referring to position qualifiers absolute, such as the terms 'front', 'back', 'top', 'bottom', 'left', 'right', etc., or relative, such as the terms 'above', 'below', 'upper', 'lower', etc., or orientation qualifiers, such as the terms 'horizontal', 'vertical', etc., unless otherwise specified, are referred to the orientation of the figures or to a photodiode in a normal position of use. In the following description, a height or depth corresponds to a dimension taken in the Z direction, which may correspond to a vertical direction, and a width corresponds to a dimension taken in the X direction, which may correspond to a horizontal direction. Unless otherwise specified, the expressions 'about', 'approximately', 'substantially', and 'of the order of' mean to within 10%, preferably to within 5%. In the following description, when referring to a region, reference is made to a semiconductor region. In the following description, when referring to an epitaxial or epitaxially grown layer, or region, reference is made to a layer, or region, formed by epitaxial growth. In the examples given below, it is considered that the first conductivity type is N-type and the second conductivity type is P-type, but the examples can also apply if the conductivity types are reversed, i.e. the first conductivity type is P-type and the second conductivity type is N-type with, for example, appropriate reversals of anode and cathode voltages. For example, P-type regions are doped with boron (B) atoms, and / or N-type regions are doped with arsenic (As) atoms. [Fig. 1] is a partial, schematic sectional view showing an example of a SPAD-type photodiode 100. The photodiode 100 comprises a first epitaxial P-type semiconductor region 111 and a second P-type semiconductor region 112, more doped than the first region 111. The second region 112 surrounds the first region 111. For example, the first region 111 forms a collection volume within which carriers have a significant probability of being generated upon absorption of a photon (photogenerated). The photodiode 100 further comprises a third epitaxially grown P-type semiconductor region 113, also less doped than the second region 112, flush with the upper face 100A of the photodiode 100. For example, the third region 113 is epitaxially grown and doped under conditions similar to the first region 111. The first region 111 may comprise a first portion 111 A and a second portion 111B less wide than the first portion 111A and connecting the first portion 111A to the third region 113. The photodiode 100 further comprises a fourth heavily doped N-type semiconductor region 114 flush with the upper face 100A of the photodiode 100, and a fifth heavily doped P-type semiconductor region 115, more heavily doped than the second region 112, under, and in contact with, the fourth region 114. For example, the fourth region 114 has a cylindrical shape. In other words, the fourth region 114 may have, when viewed from above, a round or oval shape. Similarly, the fifth region 115 may have a cylindrical shape. Alternatively, the fourth region and / or the fifth region may have other shapes, for example, when viewed from above, a quadrilateral shape or a quadrilateral with rounded corners, or a substantially hemispherical shape, or may include a portion of a substantially hemispherical shape, as shown in [Fig. 4] described later. The connection between the fourth region and the fifth region provides an upper PN junction 130, or main PN junction. For SPAD photodiode operation, the PN junction may be reverse biased beyond the breakdown voltage, so as to form an electric field around this PN junction sufficient for a photogenerated carrier reaching the space charge region, or depleted region, of the PN junction to trigger an avalanche phenomenon. The area surrounded by dotted lines corresponds approximately to the multiplication region 131, that is to say a region where the electric field is sufficiently strong so that the avalanche phenomenon has a high probability of occurring. In the example shown, a guard ring region 117 surrounds the fourth region 114. The guard ring region is an N-type semiconductor region, more lightly doped than the fourth region 114. For example, the guard ring region 117 has a round or oval ring shape, a square frame shape, or a square frame shape with rounded corners. The photodiode 100 also comprises a sixth heavily doped P-type semiconductor region 116, more heavily doped than the second region 112, flush with the upper face 100A of the photodiode, and located around and laterally at a distance from the guard ring region 117. For example, the sixth region 116 has, in top view, the shape of a round or oval ring, the shape of a square frame or a square frame whose corners are rounded. The fourth region 114, the fifth region 115, the sixth region 116 and the guard ring region 117 are generally formed by ion implantation, for example at least partially in the third region 113. The semiconductor regions 111, 112, 113, 114, 115, 116, 117 are for example in silicon. These semiconductor regions are located in a box 110 delimited by a deep trench isolation (DTI) 120. For example, the trench 120 surrounds the box 110. For example, the trench 120 has, in top view, the shape of a round or oval ring, the shape of a square frame or a square frame with rounded corners. Alternatively, the trench 120 has any closed shape. The photodiode 100 also includes a cathode 121 and an anode 122. The cathode 121 is located opposite the fourth region 114. The cathode 121 is, for example, located on and in contact with the fourth region 114, or with a contact recovery zone located in or on the fourth region 114. For example, the cathode 121 has substantially the shape of a parallelepiped or cylinder. The anode 122 is located opposite the sixth region 116. The anode 122 is, for example, located on and in contact with the sixth region 116, or with a contact recovery zone located in or on the sixth region 116. Preferably, the anode 122 has, in top view, a shape similar to the shape of the sixth region 116. For example, for the photodiode shown, in operation the anode is connected to ground and the bias voltage is applied to the cathode. For example, the photodiode 100 is included, in top view, in a square whose side is, for example, less than about 6 μm. For example, the height of the photodiode 100 is between about 3 and 15 μm, for example equal to about 6 μm or about 10 μm. In the objective of miniaturization of SPAD photodiodes, for example in the objective of reducing the width of a photodiode and compensating for this reduction by increasing the height of the photodiode, a solution is sought so that the increase in the height of the photodiode does not result in a degradation of the timing jitter, and / or a reduction in the photon detection probability (PDP). In the example of the photodiode 100, the first epitaxial region 111 included in the second region 112, and less doped with the same conductivity type as the second region 112, generates a depletion zone 132, schematically represented by the dot-dash curve. This depletion zone defines a volume within which the transport of carriers is substantially rapid. Having the largest possible depletion volume allows an improvement in the timing jitter compared to a photodiode with a smaller depletion volume. However, it is observed that this depletion zone 132 extends to the center of the epitaxial region 111, but that it does not reach the bottom of the epitaxial region 111, that is to say the bottom of the collection volume of the photogenerated carriers. Thus, a carrier photogenerated at the bottom of the volume may not reach the multiplication region 131, or with a random and significant travel time. A fortiori, when the height of the photodiode increases, the depletion zone 132 of the photodiode 100 may not be sufficient to maintain an acceptable timing jitter. Indeed, by increasing the height of the photodiode, the depletion zone remaining constant, critical zones are added to the bottom of the epitaxial region, not depleted and far from the multiplication zone. If carriers are photogenerated in these critical zones, they could take an even longer time to trigger an avalanche as the height increases, leading to a degradation of the timing jitter. The inventors propose a photodiode and a method for manufacturing such a photodiode making it possible to meet the improvement needs described above, and to overcome all or part of the drawbacks of the photodiodes described above. In particular, the inventors propose a photodiode and a method for manufacturing such a photodiode making it possible to increase its height without increasing the temporal jitter, and without reducing the photon detection probability (PDP), or even by increasing it. Embodiments of photodiodes will be described below. The embodiments described are non-limiting and various variants will become apparent to those skilled in the art from the indications in this description. [Fig.2] is a partial, schematic sectional view showing an avalanche photodiode 200 according to one embodiment. According to a preferred embodiment, the photodiode 200 is a SPAD type photodiode. The photodiode 200 has an upper face 200A (first face) and a lower face 200B (second face) opposite the upper face, preferably substantially parallel to the upper face. The photodiode 200 comprises a first epitaxially grown semiconductor region 211 of a first conductivity type (in the example, N-type) and a second semiconductor region 212 of a second conductivity type (in the example, P-type). The second region 212 surrounds, at least partially, the first region 211, and comprises surfaces in contact with surfaces of the first region 211. For example, the first region 211 is included in the second region 212. Thus, photodiode 200 differs from photodiode 100 primarily in that the first epitaxially grown region 211 is N-type, rather than P-type. The photodiode 200 has an upper semiconductor region 213 (third region) flush with the upper face 200A of the photodiode 200. The third region 213 is an N-type epitaxial region. For example, the first and third regions are epitaxially grown and doped under similar conditions, for example they correspond to regions of the same layer formed by epitaxial growth. For example, the third region 213 may correspond to an upper portion (or third portion) of the layer formed by epitaxial growth. The photodiode 200 further comprises a fourth heavily doped N-type semiconductor region 214, preferably more heavily doped than the first region 211, and flush with the upper face 200A of the photodiode 200, and a fifth heavily doped P-type semiconductor region 215, preferably more heavily doped than the second region 212, under, and in contact with, the fourth region 214. For example, the fourth region 214 has a cylindrical shape. In other words, the fourth region 114 may have, in top view, a round or oval shape. Similarly, the fifth region 215 may have a cylindrical shape. Alternatively, the fourth region and / or the fifth region may have other shapes, for example, in top view, a quadrilateral shape or a quadrilateral with rounded corners, or a substantially hemispherical shape, or may include a portion of substantially hemispherical shape, as shown for example in [Fig. 5] described later. The connection between the fourth region and the fifth region provides an upper PN junction 230, or main PN junction. For SPAD photodiode operation, the PN junction can be reverse biased beyond the breakdown voltage, so as to form an electric field around the PN junction sufficient for a photogenerated carrier reaching the multiplication region to trigger an avalanche phenomenon. The area surrounded by dotted lines corresponds approximately to a region of multiplication 231, that is to say a region where the electric field is sufficiently strong so that the avalanche phenomenon has a high probability of occurring. As shown in [Fig. 2], a guard ring region 217 may surround the fourth region 214. The guard ring region 217 is an N-type semiconductor region, more lightly doped than the fourth region 214. The guard ring region 217 makes it possible in particular to avoid premature triggering of the photodiode at the periphery of the main PN junction, and to reduce the probability of minority carriers coming from the periphery of the junction. For example, the guard ring region 217 has a round or oval ring shape, a square frame shape or a square frame with rounded corners. The photodiode 200 also comprises a sixth heavily doped P-type semiconductor region 216, preferably more heavily doped than the second region 212, flush with the upper face 200A of the photodiode 200, and located laterally around and at a distance from the guard ring region 217. This sixth region may be a contact recovery region. For example, the sixth region 216 has, in top view, the shape of a round or oval ring, the shape of a square frame or a square frame with rounded corners, or more broadly the shape of a quadrilateral or a quadrilateral with rounded corners. The fourth region 214, the fifth region 215, the sixth region 216 and the guard ring region 217 are preferably formed by ion implantation, for example at least partially in the third region 213. The semiconductor regions 211, 212, 213, 214, 215, 216, 217 are for example made of silicon. The semiconductor regions are located in a box 210 delimited by a deep insulating trench 220 (DTI, from the English "Deep Trench Insolation”). For example, the trench 220 surrounds the box 210. For example, the trench 220 has, in top view, the shape of a round or oval ring, the shape of a square frame or a square frame with rounded corners, or more broadly the shape of a quadrilateral or a quadrilateral with rounded corners. Alternatively, the trench 220 has any closed shape. For example, the trench may be filled with a conductive or semiconductive element, such as metal or polysilicon, isolated from the rest of the photodiode by a layer of insulator. Trench 220 may thus form capacitive deep trench isolation (CDTI). The contacting surfaces of the first 211 and second 212 regions form PN junctions complementary to the main PN junction 230. These complementary junctions make it possible to form depletion zones 232, schematically represented by the curve in dashed lines, and to create electric field lines deep in the photodiode, with the corollary of accelerating the transport of the photogenerated carriers in the most critical zones of the photodiode, at the bottom of the photodiode for example, towards the multiplication zone. Indeed, the transport of the photogenerated carriers in a depleted zone is faster than in a non-depleted zone. These complementary junctions can therefore generate holes, which can also participate in the avalanche phenomenon, in addition to the contribution of electrons from the main PN junction. Indeed, the holes are collected via the anode 222 (described later) and the anode is directly connected, by P-type doping continuity, to the P-type regions adjacent to these complementary junctions. Compared to the 100 photodiode of [Fig.1], this allows the depletion zone to be stretched towards the bottom of the photodiode, and thus to accelerate the transport of photogenerated carriers in the most critical areas of the photodiode, at the bottom of the photodiode for example. Indeed, in the photodiode 200, the photogenerated carriers already being in the depletion zone 232, or at least in its close vicinity, they have a higher probability of reaching the multiplication zone in a shorter time than in the case of the photodiode 100 of [Fig.1] where the depletion zone was further from the bottom of the photodiode. As a corollary, the depletion zone 232 of the photodiode 200 can make it possible to reduce the timing jitter, even when the height of the photodiode increases. For example, photodiode 200 forms a fully depleted SPAD photodiode. For example, the first region 211 forms a collection volume within which carriers have a significant probability of being generated upon absorption of a photon. For example, carriers that are photogenerated in the first region 211 can be driven to a multiplication region 231, and this is all the more efficient as the electric field lines are optimized. Advantageously, the doping of the first epitaxial region 211 can be adjusted to optimize the field lines. As described later, the first epitaxial region 211 can have a substantially constant doping in the Z direction, or a gradual doping in the Z direction, for example a doping decreasing between the upper face 200A of the photodiode and the lower face 200B of the photodiode. As shown in [Fig.2], the first region 211 may comprise a first portion 211A and a second portion 211B less wide than the first portion 211A and connecting the first portion 211A to the third region 213. Such a shape may correspond to a funnel shape promoting the conduction of the photogenerated carriers towards the multiplication region 231. Additionally, the second region 212 may include: - a first portion 212A included laterally between the second portion 211B of the first region 211 and the deep insulating trench 220; - a second portion 212B (lateral portion) comprised laterally between the first portion 211A of the first region 211 and the deep insulating trench 220; and - a third portion 212C (buried portion) located under the first region 211, between the first region 211 and the lower face 200B of the photodiode 200. The second region 212 may extend vertically towards the upper region 200A of the photodiode 200 to also be included laterally between the third region 213 and the deep insulating trench 220. The presence of the third portion 212C of the second P-type region 212 under the first N-type region 211 makes it possible to form a passivation zone for the photodiode 200, for example so that it is less impacted by the current dark current (in English "dark current"), as well as the presence of the second portion 212B of the second P-type region 212 between the first N-type region 211 and the deep insulating trench 220. For example, this makes it possible to reduce the dark count rate (in English "DCR" for "Dark Count Rate"). Preferably, the width and / or the height of the first epitaxial region 211 is optimized so that the volume of this first region is as large as possible, and thus to be able to collect as many photons as possible, for example, to be able to increase the probability of detection of PDP photons, while maintaining a distance on the one hand with the lower face 200B of the photodiode 200 and on the other hand with the deep insulating trench 220 sufficient to limit the dark counting rate DCR. For example, the second portion 212B of the second region 212 has a width of the order of a hundred nanometers, the third portion 212C of the second region 212 has a height of the order of a few hundred nanometers up to a micrometer, the dimensions of the first epitaxial region 211 being deduced from the dimensions of the second and third portions of the second region 212, and from the dimensions of the diode.As a non-limiting example, the first epitaxial region 211 may have a width of approximately 5 μm and a height of approximately 7 μm. Photodiode 200 also includes a cathode 221 and an anode 222. The cathode 221 is located opposite the fourth region 214. The cathode 221 is, for example, located on and in contact with the fourth region 214, or with a contact recovery zone located in, or on, the fourth region 214. For example, the cathode 221 has substantially the shape of a parallelepiped or cylinder. The anode 222 is located opposite the sixth region 216. The anode 222 is, for example, located on and in contact with the sixth region 216, or with a contact recovery zone located in, or on, the sixth region 216. Preferably, the anode 222 has, in top view, a shape similar to the shape of the sixth region 216. For example, for the photodiode shown, for which the first conductivity type is N-type and the second conductivity type is P-type, in operation, the anode is connected to ground and the bias voltage is applied to the cathode. Alternatively, if the first conductivity type is P-type and the second conductivity type is N-type with an inversion of the anode and cathode, in operation the cathode is connected to ground and a bias voltage is applied to the anode. For example, the photodiode 200 is included, in top view, in a square whose side is less than about 6 µm. For example, the height of the photodiode 200 is between about 3 and 15 um, for example equal to about 6 um or equal to about 10 um. [Fig. 3] illustrates in a very simplified manner doping profiles, in arbitrary units, of the first epitaxial region and the second semiconductor region of a photodiode similar to photodiode 200 of [Fig. 2]. The doping profile is determined in the Z direction, along a vertical line substantially centered laterally (marked in dotted lines in [Fig. 2]). The first epitaxial region 211 of type N can have a substantially constant N doping, in particular in the Z direction. This is illustrated in [Fig.3] by the dotted curve 301. Alternatively, the doping of the first epitaxial region 211 may be gradual in the Z direction, for example decreasing between the upper face 200A of the photodiode and the lower face 200B of the photodiode, for example between 10!5 cm* and 5.10! cm. This is illustrated in [Fig.3] by the continuous line curve 302. Curve 303 represents a gradual P doping of the second region 212, more precisely of the third portion 212C of the second region located under the first region 211, which increases between the first region 211 and the lower face 200B of the photodiode 200, for example between 2.10"? cm” and 9.10'$ cm3, The inventors were able to demonstrate by simulations that a photodiode according to one embodiment, for example of the type of photodiode 200 of [Fig. 2], makes it possible to increase the population of carriers that trigger an avalanche in a very short time. They further demonstrated that the constant N doping of the epitaxial region can be acted upon to act on the operation of the photodiode, and that a gradual N doping profile of the epitaxial region makes it possible to further increase the population of carriers that trigger an avalanche in a very short time, to reduce the avalanche triggering time, and thus to reduce the time jitter. The inventors also demonstrated that a photodiode without a portion of P-type region under the N-type epitaxial region makes it possible to further increase the population of carriers that trigger an avalanche in a very short time and to reduce the time jitter.However, as described further, it may have the disadvantage of increasing the dark count rate DCR. [Fig. 4] is a partial, schematic sectional view showing another example of photodiode 400. Only half of the photodiode in the X direction is shown. Photodiode 400 differs from photodiode 100 mainly by the shapes of the first epitaxial region 411 and the second region 412, both of the same conductivity type, P-type in the example. Furthermore, photodiode 400 does not have a guard ring region. The second region 412 may be doped gradually with doping decreasing from the bottom face 400B to the first region 411. The second region 412 surrounds the first region 411, and includes surfaces in contact with surfaces of the first region 411. The photodiode 400 further comprises a fourth heavily doped N-type semiconductor region 414 flush with the upper face 400A of the photodiode 400, and a fifth heavily doped P-type semiconductor region 415, in contact with, and surrounding, the fourth region 414, the connection between the fourth region 414 and the fifth region 415 providing an upper PN junction 430, or main PN junction. The photodiode 400 also comprises a sixth heavily doped P-type semiconductor region 416 flush with the upper face 400A of the photodiode 400, also corresponding to an upper face of the second region 412, and located around the first region 411. The first region 411 comprises a first portion 411A located under the fifth region 415 and a second portion 411B included laterally between the fifth region 415 and the second region 412 and extending to the upper face 400A of the photodiode 400. The semiconductor regions are located in a box 410 delimited by a deep insulating trench 420. [Fig. 5] is a partial, schematic, sectional view showing a photodiode 500 according to another embodiment. Only one half of the photodiode in the X direction is shown. Photodiode 500 differs from photodiode 400 of |Fig. 4] primarily in that the first epitaxial region 511 is N-type, instead of P-type. The contacting surfaces of the first 51 | and second 512 regions form PN junctions complementary to the main PN junction 530, formed by the connection between the fourth region 514 and the fifth region 515 of the photodiode 500. These complementary PN junctions make it possible to form a depletion zone 532 schematically represented by the dot-dash curve, which extends further towards the bottom of the photodiode 500 than the photodiode 400 of [Fig. 4]. For example, as shown, the first region 511 surrounds the main PN junction 530. The semiconductor regions are located in a box 510 delimited by a deep insulating trench 520. The first region 511 comprises a first portion 511A located under the fifth region 515 and a second portion 511B included laterally between the fifth region 515 and the second region 512 and extending to the upper face 500A of the photodiode 500. The second region 512 comprises a lateral portion 512B between the first region 511 and the deep insulating trench 520, and a buried portion 512C between the first region 511 and the lower face 500B of the photodiode 500. The inventors were able to demonstrate by simulations that a photodiode according to one embodiment, with complementary PN junctions, for example of the type of photodiode 500 of [Fig. 5] makes it possible to create electric field lines in the deeper regions of the photodiode compared to a photodiode of the type of photodiode 400 of [Fig. 4], without complementary PN junction. The inventors were also able to demonstrate by simulations that a photodiode according to one embodiment, for example of the type of photodiode 500 of [Fig. 5], makes it possible to increase the height of the photodiode without increasing the avalanche triggering time. Conversely, the inventors were able to observe that for a photodiode of the type of photodiode 400 of [Fig. 4], increasing its height significantly increases the avalanche triggering time. Thus, a photodiode according to one embodiment makes it possible to increase the height of the photodiode, and thus have a larger collection volume at equal width, for example to increase the photon detection probability (PDP), without increasing the time jitter. [Fig.6A], [Fig.6B], [Fig.6C], [Fig.6D], [Fig.6E], [Fig.6F] and [Fig.6G] are sectional views showing steps of a method of manufacturing a photodiode according to one embodiment. [Fig.6A] represents a structure obtained at the end of an epitaxial growth step on a substrate 608, for example a silicon (Si) substrate, forming an epitaxial layer 601 of a first conductivity type, N-type in the example method represented. The epitaxial growth can be carried out at a constant doping concentration, or at a gradual doping concentration in the Z direction, for example a doping concentration decreasing between the upper face 600A of the photodiode and the substrate 608. The epitaxial layer 601 makes it possible to form the future first N-type region. [Fig.6B] represents a structure obtained at the end of a doping step by ion implantation, so as to form, in the epitaxial layer 601, a first portion 602A of the second conductivity type (first portion of the second region), of type P in the example of the method represented. The first portion 602A is, for example, formed from a distance dl to a distance d2 from the upper face 600A of the photodiode, for example by configuring the parameters of the ion implantation to allow this implantation between dl and d2. The first P-type portion 602A delimits a first portion 601A of the N-type epitaxial layer (first portion of the first region) located under the first P-type portion 602A and a third portion 601C of the epitaxial layer type N located above the first portion 602A. The ion implantation in the epitaxial layer 601 to form the first portion 602A can be carried out through a mask 604, for example a mask obtained by photolithography. For example, the mask 604 is adapted so that the first portion 602A has a crown shape, that is to say by retaining a second portion 601B of the epitaxial layer 601 of type N (second portion of the first region) at the center of the first portion 602A of type P. This second N-type portion 601B allows the first region to have a funnel shape, which makes it possible to promote the conduction of photogenerated carriers from the first portion 601A of the first region towards the upper PN junction 630 described later. [Fig.6C] represents a structure obtained at the end of several doping steps by ion implantation, so as to form, in the third portion 601C of the epitaxial layer 601: - a heavily doped N-type region 604 (fourth region) and a heavily doped P-type region 605 (fifth region), said fourth and fifth regions forming an upper PN junction 630, the N-type region 604 being flush with the upper face 600A of the photodiode 600 and surmounting the P-type region 605; - a guard ring region 607 located around the N-type region 604, and less heavily N-doped than the 604 region; - a heavily P-doped region 606 (sixth region) flush with the upper face 600A, formed laterally around and at a distance from the guard ring region 607. The N-type region 604 and the P-type region 605 may be formed above the second portion 601B of the epitaxial layer 601, or even partially overflow above the first P-type portion 602A. Each ion implantation can be carried out through a mask (not shown), for example a mask obtained by photolithography, then removed after ion implantation. [Fig.6D] represents a structure obtained at the end of an etching step over a width L1 of a lateral edge 601D over the entire circumference of the epitaxial layer 601 from the upper face 600A of the photodiode, the etching extending for example to a depth d3 in the silicon substrate 608. This forms a lateral trench 622. Alternatively, the etching can be carried out up to the limit between the silicon substrate 608 and the epitaxial layer 601. [Fig.6E] represents a structure obtained after: - a step of filling the lateral trench 622 with an insulating material, for example silicon dioxide (SiOz), in order to form a deep insulating trench 620; and - a step of forming a second portion 602B of type P (second portion of the second region) laterally between the deep insulating trench 620 and the etched epitaxial layer 601. For example, the deep insulating trench 620 surrounds the previously formed regions. For example, the second P-type portion 602B forms a passivation zone for the photodiode, for example so that it is less impacted by the dark current. For example, the second portion 602B can be done, depending on the type of passivation desired: - by doping, for example by ion implantation; - by adding a doping material or a charge inversion material into the insulating material when filling the side trench 622. Thus, the conductivity along the deep insulating trench 620 can be reversed without ion implantation by the use of a specific material. Figures 6A-6E show that the deep insulating trench 620 is formed after the doping steps. This is not limiting, and, alternatively, the deep insulating trench may be formed before the doping steps. [Fig.6F] represents a structure obtained at the end of an etching step of the substrate 608. Figures 6A-6F show that the deep insulating trench 620 is formed before the substrate etching step 608. This is not limiting, and, alternatively, the deep insulating trench may be formed after this substrate etching step. Furthermore, the deep insulating trench may be formed from the bottom face 600B of the photodiode. [Fig.6G] represents a structure obtained at the end of an optional step of forming a third 602C portion of type P (third portion of the second region). The third P-type portion 602C may be formed by doping, for example by ion implantation, for example from the lower face 600B of the photodiode. Alternatively, the third P-type 602C portion may be formed by epitaxial growth, in this case prior to the epitaxial growth of the N-type 601 epitaxial layer. For example, the third P-type portion 602C forms a passivation zone for the photodiode, for example so that it is less impacted by the dark current. The resulting photodiode 600 shown in [Fig.6G] is similar to the photodiode 200 of [Fig.2]. From the exemplary manufacturing method described above, the skilled person can adapt the method to manufacture other photodiodes according to the embodiments, for example the photodiode 500 of [Fig. 5]. Thus, an advantage of the described embodiments is that the photodiode can be produced without complicating the manufacturing process. Another advantage of the described embodiments is that they are suitable for small photodiodes, for example, having sides less than 6 μm, and / or for photodiodes with a height for example between 4 and 10 μm, or even more. Another advantage of the described embodiments is that they allow the size, e.g., width, of the photodiode to be reduced without degrading the photon detection probability and / or increasing timing jitter. Various embodiments and variations have been described. Those skilled in the art will understand that certain features of these various embodiments and variations could be combined, and other variations will occur to those skilled in the art. Finally, the practical implementation of the embodiments and variants described is within the reach of those skilled in the art from the functional indications given above.

Claims

Claims

1. Avalanche photodiode (200; 500) comprising: - a main PN junction (230; 530) adapted to be polarized in 1Inverse; and - a plurality of semiconductor regions including at least: - a first epitaxially grown semiconductor region (211; 511) of a first type of conductivity; and - a second semiconductor region (212; 512) of the second type of conductivity, said second region being arranged to surround at less partially the first region, and comprising surfaces in contact with surfaces of said first region.

2. Photodiode (200; 500) according to claim 1, wherein the main PN junction (230; 530) is formed by fourth (214; 514) and fifth (215; 515) regions of the plurality of semi-regions drivers: the fourth region being heavily doped with the first type of conductivity, for example more heavily doped than the first region, and being formed from a first face (200A; 500A) of the photodiode; and the fifth region being heavily doped with the second type of conductivity, for example more heavily doped than the second region, and being formed under, and in contact with, the fourth region; the fourth and fifth regions being for example located at center of the photodiode in a plane parallel to the first face.

3. Photodiode (200; 500) according to claim 2, in which a sixth region (216; 516) of the plurality of semiconductor regions is formed laterally around, and at a distance from, the fourth region (214; 514) from the first face (200A; 500A) of the photodiode, the sixth region being heavily doped with the second type of conductivity, for example more heavily doped than the second region.

4. The photodiode (200) of claim 3, wherein a third semiconductor region (213) of the plurality of semiconductor regions is formed between the first face (200A) of the photodiode and the first region (211), the third region being an epitaxial region of the first conductivity type, and the fourth, fifth and sixth regions (214, 215, 216) being formed by doping, for example by implantation ionic, in the said third region.

5. The photodiode (200) of claim 4, wherein the first and third regions are epitaxially grown and doped under si- miliaries, for example corresponding to regions of the same layer epitaxial.

6. A photodiode (200) according to any one of claims 2 to 5, in which a guard ring region (217) of the plurality of regions semiconductors is formed around the fourth region (214), the guard ring region being of the first conductivity type and being more lightly doped than the fourth region (214).

7. Photodiode (200; 500) according to any one of claims 2 to 6, in which the semiconductor regions are formed in a box (210; 510) delimited by a deep insulating trench (220; 520) extending in a direction {Z) perpendicular to the first face (200A; 500A) of the photodiode.

8. Photodiode (200; 500) according to claim 7, wherein the second region (212; 512) comprises a lateral portion (212B; 512B) located between the first region (211; 511) and the trench deep insulating {220; 520).

9. Photodiode (200; 500) according to any one of claims 2 to 8, wherein the second region (212; 512) comprises a portion buried (212C; 512C) located between the first region (211; 511) and a second face (200B; 500B) of the photodiode opposite the first side.

10. A photodiode (200) according to any one of claims 2 to 9, in which the first region (211) is substantially doped constant in a direction (Z) perpendicular to the first face (200A) of the photodiode.

11. A photodiode (500) according to any one of claims 2 to 9, in which the first region (511) is gradually doped in a direction (Z) perpendicular to the first face (5SO0A), for example decreasing manner between said first face and a second face (500B) of the photodiode opposite the first face.

12. A photodiode (200) according to any one of claims 2 to 11, wherein the first region (211) comprises a first portion (211 A) and a second portion (211B) less wide, in a plane parallel to the first face (200A), than the first portion, and located between the first portion (211 A) and the main PN junction (230).

13. Photodiode (500) according to any one of claims 2 to 11, wherein the first region (511) surrounds the main PN junction (530), for example, comprises a first portion (511 A) located under the main PN junction (530), and a second portion (511B) extending to the first face (500A) of the photodiode (500).

14. A photodiode according to any one of claims 1 to 13, in wherein the first region is within the second region.

15. A photodiode according to any one of claims 1 to 14, in which each dimension of the photodiode, in a plane parallel to a first face (200A) of the photodiode, is less than 6 um and / or the height of the photodiode is between 3 and 15 um, for example between 6 and 10 um.

16. A method of manufacturing an avalanche photodiode (600), comprising a main PN junction (630) adapted to be reverse biased, the method comprising forming a plurality of semicon- regions drivers including at least: - a first epitaxially grown semiconductor region of a first type of conductivity: and - a second semiconductor region of the second type of conductivity, said second region being arranged to surround at less partially the first region, and comprising surfaces in contact with surfaces of said first region.

17. A method according to claim 16, wherein the formation of the plurality of semiconductor regions comprises: - the formation by epitaxial growth of an epitaxial layer (601) of the first type of conductivity intended to form the first semi-region driver; then - the formation, by partial doping of the epitaxial layer, for example by ion implantation, of at least a first portion (602A) of the second region; and the fourth region and the fifth region are formed by doping partial of the epitaxial layer, for example by ion implantation, to from the first face (600A) of the photodiode.

18. The method of claim 17, wherein the second region has a lateral portion located between the first region and the deep insulating trench, and the lateral portion (602B) of the second region is formed by partial doping of the epitaxial layer, for example by ion implantation, by partial change of conductivity of the epitaxial layer, by diffusion from the deep insulating trench, or by a charge-reversing material in the insulating trench deep.

19. A method according to claim 17 or 18, wherein the second region has a buried portion located between the first region and a second face of the photodiode opposite the first face, and the buried portion (602C) of the second region is formed by growth epitaxial, or by partial doping of the epitaxial layer, for example by ion implantation, from the second face (600B) of the photodiode.