METHOD FOR PRODUCING A PHOTOVOLTAIC CELL

Description

Technical field

[0001] The invention relates to the technical field of photovoltaic cells, in particular photovoltaic cells comprising a passivated contact on each of the surfaces of the substrate. Surface passivation is a major issue in the photovoltaic sector, making it possible to limit recombinations between minority and majority carriers, and therefore to increase the number of collected carriers, which results in a significant improvement in efficiency. By "passivation" is meant the neutralization of electrically active defects on the surfaces of the substrate. Indeed, a surface of a crystalline silicon substrate has a density of defects (e.g. dangling bonds, impurities, crystal discontinuity, etc.) which can lead to significant losses linked to the surface recombination of carriers in the case of a photovoltaic application.

[0002] More precisely, the passivated contact can be a structure of the {oxide / polysilicon} type, comprising: an oxide film, preferably ultra-thin (tunnel oxide), such as silicon oxide, formed on the surface of the substrate; a layer of polysilicon, heavily doped (n-type or p-type), deposited on the oxide film.

[0003] Such a passivated contact based on polysilicon is a promising approach compared to a passivated contact based on amorphous silicon. Indeed, a passivated contact of the {oxide / polysilicon} type forms a high-temperature heterojunction, which can withstand a maximum temperature of around 1000°C for a photovoltaic cell during manufacture, and a maximum temperature of around 450°C for a finished photovoltaic cell. Conversely, a passivated contact of the {oxide / amorphous silicon} type forms a low-temperature heterojunction, which can withstand a maximum temperature of around 250°C. Thus, being able to work at a temperature above 250°C opens up wider possibilities for passivating the edges of the photovoltaic cell after laser cutting.

[0004] In addition, the polysilicon layer of the passivated contact of the {oxide / polysilicon} type has impurity trapping properties (" getter» in English), which allows us to consider the use of materials with a low carbon footprint.

[0005] The fact that polysilicon has a higher electrical conductivity than amorphous silicon allows for the broadening of specifications for the electrical properties of transparent conductive oxide deposited on polysilicon, which therefore also facilitates the use of transparent conductive oxide without indium.

[0006] The invention therefore finds its application in particular in the manufacture of high-temperature heterojunction silicon (SHJ-HT) photovoltaic cells. State of the art

[0007] The polysilicon layer of the passivated contact has the disadvantage of being absorbent in ultraviolet light. A compromise must therefore be sought for the thickness of the polysilicon layer in order to ensure the role of electrical contact passivating the surface of the substrate, without significantly affecting the conversion efficiency of the photovoltaic cell. The person skilled in the art therefore seeks the thinnest possible polysilicon thickness in order to limit absorption losses. More specifically, the person skilled in the art seeks the thinnest possible thickness for the n-type doped polysilicon layer, denoted poly-Si (n+), on the front face of the photovoltaic cell intended to be exposed to solar radiation, with good passivation quality.

[0008] There are two main methods to form a passivated contact of the {oxide / poly-Si (n+)} type for SHJ-HT type photovoltaic cells: (i) deposition of a layer of amorphous silicon by plasma-enhanced chemical vapor deposition (PECVD for “ Plasma Enhanced Chemical Vapor Deposition » in English) with implantation of dopants, followed by annealing to crystallize the amorphous silicon layer and form a polysilicon layer; this is called in-situ doping; (ii) deposition of the polysilicon layer by low-pressure chemical vapor deposition (LPCVD for " Low Pressure Chemical Vapor Deposition » in English), followed by diffusion of dopants, either by gaseous means or by implantation of the dopants; this is called ex-situ doping.

[0009] Method (i) allows to obtain a poly-Si (n+) layer with a thickness greater than 20 nm with good passivation quality. On the other hand, a poly-Si (n+) layer with a thickness less than 20 nm, obtained by method (i), leads to a significant degradation of the passivation.

[0010] Method (ii) allows to obtain a poly-Si (n+) layer with a thickness of 15 nm with good passivation quality. On the other hand, a poly-Si (n+) layer with a thickness strictly less than 15 nm, obtained by method (ii), leads to a significant degradation of performance, which can be explained by too large a quantity of dopants introduced into the substrate.

[0011] The person skilled in the art is looking for a solution to obtain a poly-Si (n+) layer strictly less than 15 nm while maintaining good passivation quality.

[0012] A solution known from the state of the art, in particular from the document Y. Larianova et al., “Ultra-Thin Poly-Si Layers: Passivation Quality, Utilization of Charge Carriers Generated in the Poly-Si and Application on Screen-Printed Double-Side Contacted Polycrystalline Si on Oxide Cells”, RRL Solar, vol. 4, n°10, 2020, consists of: deposit a 25 nm thick polysilicon layer by LPCVD, introduce the n-type dopants into the polysilicon layer by ion implantation, chemically etch the polysilicon layer until a polysilicon thickness strictly less than 15 nm is obtained.

[0013] Chemical etching is controlled in a dilute NH 4 OH bath. The bath temperature must be precisely controlled, as well as the NH 4 OH concentration and the immersion time of the plates.

[0014] Such a state-of-the-art solution is therefore not entirely satisfactory due to the restrictive control that must be exercised over the diluted NH 4 OH bath.

[0015] Document FR 3 071 358 A1 also describes a method for manufacturing a photovoltaic cell in which a layer of amorphous silicon is crystallized by thermal annealing, this step being followed by thermal oxidation which allows the formation of an oxide layer, so that this oxide layer consumes a certain quantity of the polysilicon layer previously obtained. Statement of the invention

[0016] The invention aims to remedy all or part of the aforementioned drawbacks. To this end, the invention relates to a method for manufacturing a photovoltaic cell, comprising the steps: a) providing a structure comprising: a crystalline silicon-based substrate, having first and second opposed surfaces; first and second oxide films, preferably of the tunnel oxide type, formed respectively on the first and second surfaces of the substrate; first and second polysilicon layers, formed respectively on the first and second oxide films, and respectively comprising n-type dopants and p-type dopants having first and second surface concentrations respectively; the n-type dopants preferably being phosphorus atoms, the p-type dopants preferably being boron atoms; the first and second polysilicon layers each having a thickness greater than or equal to 15 nm, preferably greater than or equal to 20 nm;b) applying a thermal anneal to the structure, under an atmosphere, at an annealing temperature above a minimum temperature beyond which the n-type dopants and the p-type dopants can be electrically activated; c) starting an injection of an oxidizing agent into the atmosphere during step b), so as to form first and second oxide layers, respectively on the first and second polysilicon layers; the injection being started at a first instant defining, with the annealing temperature, a first thermal budget adapted to electrically activate the n-type dopants and the p-type dopants;d) stopping the injection of the oxidizing agent at a second instant defining, with the annealing temperature, a second thermal budget adapted to the first surface concentration so that: the first and second oxide layers, formed between the first and second instants, respectively have first and second thicknesses; the first thickness being predetermined; the formation of the first and second oxide layers, according to the first and second thicknesses, consumes a quantity of polysilicon between the first and second instants adapted so that the first polysilicon layer has a thickness strictly less than 15 nm at the end of step d).;

[0017] Thus, such a method according to the invention makes it possible to obtain a first layer of polysilicon, doped with n-type, denoted poly-Si (n+), with a thickness strictly less than 15 nm at the end of step d) without requiring a temperature-controlled chemical etching bath as in the prior art. Indeed, the reduction in the thickness of the poly-Si (n+) layer is carried out by a consumption of material due to the oxidation reaction taking place between the first and second instants. Such a method according to the invention is therefore much easier to implement than that of the prior art.

[0018] It should be noted that the second p-doped polysilicon layer, denoted poly-Si (p+), is also consumed during the oxidation reaction occurring between the first and second instants. However, the inventors have observed experimentally that the reduction in the thickness of the poly-Si (p+) layer remains limited compared to that of the poly-Si (n+) layer due to the significant difference between the first and second surface concentrations. Furthermore, the thickness of the poly-Si (p+) layer on the rear face is less critical than the thickness of the poly-Si (n+) layer in terms of UV absorption, since the second surface of the substrate is not intended to be exposed to direct solar radiation.

[0019] The inventors have also experimentally observed that the injection of the oxidizing agent must be carried out after the electrical activation of the n-type dopants and the p-type dopants. If the injection of the oxidizing agent starts before the electrical activation of the n-type dopants and the p-type dopants, the performance of the cell in terms of passivation is significantly degraded, which could suggest premature destruction of the dopants on the surface of the corresponding polysilicon layer without having been able to migrate into substitutional sites.

[0020] The injection of the oxidizing agent is stopped at the second defined time when the desired thickness of the first n-type doped polysilicon layer is obtained.

[0021] The invention also relates to a method for manufacturing a photovoltaic cell, comprising the steps: a') providing a structure comprising: a crystalline silicon-based substrate, having first and second opposite surfaces; first and second oxide films, preferably of tunnel oxide type, formed respectively on the first and second surfaces of the substrate; first and second amorphous silicon layers, formed respectively on the first and second oxide films, and respectively comprising n-type dopants and p-type dopants having first and second surface concentrations respectively; the n-type dopants preferably being phosphorus atoms, the p-type dopants preferably being boron atoms; the first and second amorphous silicon layers each having a thickness greater than or equal to 15 nm, preferably greater than or equal to 20 nm;b) applying a thermal anneal to the structure, under an atmosphere, at an annealing temperature higher than a minimum temperature beyond which the first and second amorphous silicon layers can be crystallized; c) starting an injection of an oxidizing agent into the atmosphere during step b), at a first time defining, with the annealing temperature, a first thermal budget adapted to crystallize the first and second amorphous silicon layers so as to form first and second polysilicon layers; step c) being carried out so as to form first and second oxide layers, respectively on the first and second polysilicon layers;d) stopping the injection of the oxidizing agent at a second instant defining, with the annealing temperature, a second thermal budget adapted to the first surface concentration so that: the first and second oxide layers, formed between the first and second instants, respectively have first and second thicknesses; the first thickness being predetermined; the formation of the first and second oxide layers, according to the first and second thicknesses, consumes a quantity of polysilicon between the first and second instants adapted so that the first polysilicon layer has a thickness strictly less than; 15 nm at the end of step d).

[0022] Thus, such a method according to the invention makes it possible to obtain a first layer of polysilicon, doped with n-type, denoted poly-Si (n+), with a thickness strictly less than 15 nm at the end of step d) without requiring a temperature-controlled chemical etching bath as in the prior art. Indeed, the reduction in the thickness of the poly-Si (n+) layer is carried out by a consumption of material due to the oxidation reaction taking place between the first and second instants. Such a method according to the invention is therefore much easier to implement than that of the prior art.

[0023] It should be noted that the second p-doped polysilicon layer, denoted poly-Si (p+), is also consumed during the oxidation reaction occurring between the first and second instants. However, the inventors have observed experimentally that the reduction in the thickness of the poly-Si (p+) layer remains limited compared to that of the poly-Si (n+) layer due to the significant difference between the first and second surface concentrations. Furthermore, the thickness of the poly-Si (p+) layer on the rear face is less critical than the thickness of the poly-Si (n+) layer in terms of UV absorption, since the second surface of the substrate is not intended to be exposed to direct solar radiation.

[0024] The inventors have also experimentally observed that the injection of the oxidizing agent must be carried out after the electrical activation of the n-type dopants and the p-type dopants (performed during the crystallization of the first and second amorphous silicon layers). If the injection of the oxidizing agent starts before the electrical activation of the n-type dopants and the p-type dopants, the performance of the cell in terms of passivation is significantly degraded, which could suggest premature destruction of the dopants on the surface of the corresponding amorphous silicon layer without having been able to migrate into substitutional sites.

[0025] The injection of the oxidizing agent is stopped at the second defined time when the desired thickness of the first n-type doped polysilicon layer is obtained.

[0026] This method according to the invention differs from the previous subject in that the n-type dopants and the p-type dopants are incorporated beforehand into a layer of amorphous silicon, which is crystallized during the thermal annealing of step b) so as to form a layer of polysilicon. This is called in-situ doping.

[0027] The method according to the invention may comprise one or more of the following characteristics.

[0028] According to a characteristic of the invention, the thermal annealing applied during step b) comprises: a first phase, comprising a rise in temperature until reaching the annealing temperature; a second phase, comprising a temperature plateau equal to the annealing temperature; a third phase, comprising a fall in temperature from the annealing temperature; method in which step c) is performed so that the injection of the oxidizing agent is started during the second phase, and step d) is performed so that the injection of the oxidizing agent is stopped during the second phase.

[0029] Thus, the inventors observed experimentally that the first and second instants were advantageously chosen during the second phase (the first instant being chosen after the electrical activation of the dopants) in order to obtain a satisfactory quality of the passivation of the substrate surfaces. Indeed, other configurations No. 1 to 3 (detailed below) lead to a significant degradation of the passivation: configuration n°1: the first instant is chosen during the first phase, and the second instant is chosen during the second phase; configuration n°2: the first instant is chosen during the second phase (before the electrical activation of the dopants), and the second instant is chosen during the second phase; configuration n°3: the first and second instants are chosen during the third phase.

[0030] According to a characteristic of the invention, step c) is carried out so that the oxidizing agent comprises dioxygen O 2 .

[0031] Thus, one advantage provided is the ease of injecting this gas into a furnace.

[0032] According to a characteristic of the invention, the atmosphere under which the thermal annealing is applied during step b) comprises nitrogen N 2 or argon Ar.

[0033] According to a characteristic of the invention, step d) is carried out so that the first and second thicknesses are greater than or equal to 2 nm.

[0034] Thus, an advantage provided by such thicknesses is to allow consumption of polysilicon material.

[0035] According to a characteristic of the invention, the method comprises a step e) consisting of removing the first and second oxide layers formed between the first and second instants; step e) being carried out after step d), preferably by a hydrofluoric acid HF solution.

[0036] According to a characteristic of the invention, the method comprises a step f) consisting of forming first and second layers of transparent conductive oxide, respectively on the first and second layers of polysilicon; step f) being carried out after step e).

[0037] Thus, one advantage provided by transparent conductive oxide layers is in particular to ensure electrical contact between an electrode (for example metallic) and the substrate. Transparent conductive oxide layers, with a suitable thickness, can also act as an anti-reflective layer. The anti-reflective layer makes it possible to reduce optical losses linked to the reflection of light radiation, and therefore to optimize the absorption of light radiation by the substrate.

[0038] According to a characteristic of the invention, the method comprises a step g) consisting of forming electrodes on the first and second layers of transparent conductive oxide.

[0039] According to a characteristic of the invention: step a) is performed so that the n-type dopants are phosphorus atoms and the p-type dopants are boron atoms; step b) is performed so that the annealing temperature is between 700°C and 900°C.

[0040] According to a characteristic of the invention, step b) is carried out so that the annealing temperature is between 700°C and 900°C. Definitions

[0041] By "substrate" is meant a physical, self-supporting support intended for the manufacture of a photovoltaic cell. The substrate may be a wafer (" wafer»in English) cut from a crystalline silicon ingot. By "based on", we mean that crystalline silicon is the main and majority material making up the substrate. By "crystalline silicon", we mean the multicrystalline form or the monocrystalline form of silicon, therefore excluding amorphous silicon. By "tunnel oxide", we mean an oxide thin enough to allow the circulation of an electric current within it by tunneling. A tunnel oxide can also act as a diffusion barrier for n-type dopants and p-type dopants in the substrate. The term "polysilicon" refers to polycrystalline silicon, as in common usage. However, this term also refers to nanocrystalline silicon, or microcrystalline silicon, in the application.By "n-type dopant" is meant species which, when introduced into a matrix of a crystalline silicon-based semiconductor material, donate an electron to the conduction band. By "p-type dopant" is meant species which, when introduced into a matrix of a crystalline silicon-based semiconductor material, accept an electron from the conduction band. By "thermal budget" is meant an input of energy of a thermal nature, determined by the choice of a value for the annealing temperature and the choice of a value for the annealing duration. By "oxidizing agent" is meant a substance, preferably in gaseous form, capable of oxidizing the surface of a polysilicon layer by a transfer of atoms. By "electrically activate" is meant an input of energy of a thermal nature to cause the migration of dopants into substitutional sites in which they will be capable of generating carriers.By "predetermined" is meant that the first thickness is determined prior to the implementation of a method according to the invention. By "thickness" is meant a dimension along the normal to the first surface (or to the second surface) of the substrate. By "transparent conductive oxide" (TCO for ". Transparent Conductive Oxide » in English), means an oxide that is transparent in all or part of the solar spectrum, and electrically conductive. For example, the transparent conductive oxide may have a transmittance greater than or equal to 60% (preferably greater than or equal to 80%) over the spectrum [300 nm, 1200 nm]. The values ​​X and Y expressed using the expressions “between X and Y” or “between X and Y” are included in the defined range of values. Brief description of the drawings

[0042] Other features and advantages will become apparent in the detailed description of various embodiments of the invention, the description being accompanied by examples and references to the attached drawings. Figure 1 (1a to 1f) comprises schematic sectional views, illustrating steps of a first mode of implementation of a method according to the invention. Figure 2 (2a to 2f) comprises schematic sectional views, illustrating steps of a second mode of implementation of a method according to the invention. Figure 3 is a graph representing the evolution of the temperature (on the ordinate, arbitrary unit) at which step b) is executed as a function of the execution time (on the abscissa, arbitrary unit) of step b).

[0043] It should be noted that the drawings described above are schematic, and are not necessarily to scale for the sake of readability and to simplify their understanding. The sections are made according to the normal to the first surface (or to the second surface) of the substrate. Detailed description of the implementation methods

[0044] Elements that are identical or provide the same function will have the same references for the different embodiments, for the sake of simplification. The first and second polysilicon layers will respectively have the references “3a” and “3b” before the first instant. The first and second polysilicon layers will respectively have the references “3'a” and “3'b” after having undergone material consumption between the first and second instants. Case 1: the polysilicon layers have undergone ex-situ doping

[0045] As illustrated in the figure 1 , an object of the invention is a method of manufacturing a photovoltaic cell, comprising the steps: a) providing a structure comprising: a substrate 1 based on crystalline silicon, having first and second opposite surfaces 10, 11; first and second oxide films 2a, 2b, preferably of tunnel oxide type, formed respectively on the first and second surfaces 10, 11 of the substrate 1; first and second polysilicon layers 3a, 3b, formed respectively on the first and second oxide films 2a, 2b, and respectively comprising n-type dopants and p-type dopants having respectively first and second surface concentrations; the n-type dopants preferably being phosphorus atoms, the p-type dopants preferably being boron atoms; the first and second polysilicon layers 3a, 3b each having a thickness greater than or equal to 15 nm, preferably greater than or equal to 20 nm; step a) is illustrated in Figure 1a; b) applying a thermal anneal to the structure, under an atmosphere A, at an annealing temperature TA above a minimum temperature T min beyond which the n-type dopants and the p-type dopants can be electrically activated; step b) is illustrated in Figure 1b ; c) starting an injection of an oxidizing agent 4 into the atmosphere A during step b), so as to form first and second oxide layers 5a, 5b, respectively on the first and second polysilicon layers 3a, 3b; the injection being started at a first instant t 1 defining, with the annealing temperature TA , a first thermal budget adapted to electrically activate the n-type dopants and the p-type dopants; step c) is illustrated in Figure 1c; d) stopping the injection of the oxidizing agent 4 at a second time t 2 defining, with the annealing temperature TA , a second thermal budget adapted to the first surface concentration so that: the first and second oxide layers 5a, 5b, formed between the first and second times t 1 , t 2 , respectively have first and second thicknesses, the first thickness being predetermined; the formation of the first and second oxide layers 5a, 5b, according to the first and second thicknesses, consumes a quantity of polysilicon between the first and second times t 1 , t 2 , adapted so that the first polysilicon layer 3'a has a thickness strictly less than 15 nm at the end of step d); the situation at the end of step d) is illustrated in Figure 1d . Step a)

[0046] The substrate 1 of the structure provided during step a) is advantageously n-type doped. The first surface 10 of the substrate 1 is intended to be exposed to light radiation.

[0047] Step a) is advantageously performed so that the first and second surfaces 10, 11 of the substrate 1 are textured in order to reduce the reflection coefficient and the optical losses in the photovoltaic cell. The first and second surfaces 10, 11 of the substrate 1 preferably comprise inverted pyramid patterns, arranged to create a surface roughness. The texturing is preferably performed by a chemical attack based on potassium hydroxide KOH.

[0048] By way of non-limiting example, the substrate 1 may have a thickness of the order of 150 µm.

[0049] Step a) is advantageously carried out so that the first and second oxide films 2a, 2b are of the tunnel oxide type. Step a) is advantageously carried out so that the first and second tunnel oxide films 2a, 2b have a thickness less than or equal to 3 nm, preferably less than or equal to 2 nm.

[0050] The first and second oxide films 2a, 2b are advantageously silicon oxides. By "silicon oxide" is meant silicon oxide of formula SiO 2 (silicon dioxide) or its non-stoichiometric SiO x derivatives.

[0051] Step a) may include the following steps: a 1 ) providing a structure comprising: a substrate 1 based on crystalline silicon, having first and second opposite surfaces 10, 11; first and second oxide films 2a, 2b, preferably of tunnel oxide type, formed respectively on the first and second surfaces 10, 11 of the substrate 1; a 2 ) forming first and second polysilicon layers 3a, 3b, respectively on the first and second oxide films 2a, 2b; a 3 ) implanting the n-type dopants and the p-type dopants, respectively in the first and second polysilicon layers 3a, 3b.

[0052] Step a 2 ) is advantageously carried out by low pressure chemical vapor deposition LPCVD.

[0053] Step a 3 ) is advantageously carried out by a plasma immersion ion implantation technique. When the n-type dopants are phosphorus atoms, the implantation of the phosphorus atoms is preferably carried out under an atmosphere comprising phosphine PH 3 . When the p-type dopants are boron atoms, the implantation of the boron atoms is preferably carried out under an atmosphere comprising diborane B 2 H 6 or boron trifluoride BF 3 . Step a 3 ) is advantageously carried out so that the phosphorus atoms and the boron atoms have a volume density greater than 10 20 < at. / cm 3 < . Thus, an advantage provided is to create a strong field effect conducive to good passivation of the first and second surfaces 10, 11 of the substrate 1, as well as to form a good quality electrical contact zone.

[0054] Other techniques for forming dopants in the first and second polysilicon layers 3a, 3b are possible: ion beam implantation, use of a sol-gel process, etc. Step b)

[0055] By way of non-limiting example, step a) may be performed such that the n-type dopants are phosphorus atoms, and the p-type dopants are boron atoms. Step b) may then be performed such that the annealing temperature TA is between 700°C and 900°C. Such an annealing temperature TA makes it possible to electrically activate the phosphorus atoms and the boron atoms concomitantly. The electrical activation of the phosphorus atoms and the boron atoms may occur in the polysilicon from 700°C. An annealing temperature TA above 900°C risks promoting excessive diffusion of the phosphorus atoms and the boron atoms in the substrate 1, through the first and second polysilicon layers 3a, 3b respectively, which may be detrimental to the properties of the passivated contacts. Case 2: the polysilicon layers have undergone in-situ doping

[0056] As illustrated in the figure 2, an object of the invention is a method of manufacturing a photovoltaic cell, comprising the steps: a') providing a structure comprising: a substrate 1 based on crystalline silicon, having first and second opposite surfaces 10, 11; first and second oxide films 2a, 2b, preferably of tunnel oxide type, formed respectively on the first and second surfaces 10, 11 of the substrate 1; first and second amorphous silicon layers 3c, 3d, formed respectively on the first and second oxide films 2a, 2b, and respectively comprising n-type dopants and p-type dopants having respectively first and second surface concentrations; the n-type dopants preferably being phosphorus atoms, the p-type dopants preferably being boron atoms; the first and second amorphous silicon layers 3c, 3d each having a thickness greater than or equal to 15 nm, preferably greater than or equal to 20 nm; step a') is illustrated in Figure 2a; b) applying a thermal annealing to the structure, under an atmosphere A, at an annealing temperature TA higher than a minimum temperature T min beyond which the first and second layers of amorphous silicon 3c, 3d can be crystallized; step b) is illustrated in Figure 2b ; c) starting an injection of an oxidizing agent 4 into the atmosphere A during step b), at a first instant t 1 defining, with the annealing temperature TA , a first thermal budget adapted to crystallize the first and second layers of amorphous silicon 3c, 3d so as to form first and second layers of polysilicon 3a, 3b; step c) being executed so as to form first and second layers of oxide 5a, 5b, respectively on the first and second layers of polysilicon 3a, 3b; step c) is illustrated in Figure 2c; d) stopping the injection of the oxidizing agent 4 at a second time t 2 defining, with the annealing temperature TA , a second thermal budget adapted to the first surface concentration so that: the first and second oxide layers 5a, 5b, formed between the first and second times t 1 , t 2 , respectively have first and second thicknesses, the first thickness being predetermined; the formation of the first and second oxide layers 5a, 5b, according to the first and second thicknesses, consumes a quantity of polysilicon between the first and second times t 1 , t 2 adapted so that the first polysilicon layer 3'a has a thickness strictly less than 15 nm at the end of step d); the situation at the end of step d) is illustrated in figure 2d . Step a')

[0057] The substrate 1 of the structure provided during step a') is advantageously n-type doped. The first surface 10 of the substrate 1 is intended to be exposed to light radiation.

[0058] Step a') is advantageously carried out so that the first and second surfaces 10, 11 of the substrate 1 are textured in order to reduce the reflection coefficient and the optical losses in the photovoltaic cell. The first and second surfaces 10, 11 of the substrate 1 preferably comprise inverted pyramid patterns, arranged to create a surface roughness. The texturing is preferably carried out by a chemical attack based on potassium hydroxide KOH.

[0059] By way of non-limiting example, the substrate 1 may have a thickness of the order of 150 µm.

[0060] Step a') is advantageously carried out so that the first and second oxide films 2a, 2b are of the tunnel oxide type. Step a') is advantageously carried out so that the first and second tunnel oxide films 2a, 2b have a thickness less than or equal to 3 nm, preferably less than or equal to 2 nm.

[0061] The first and second oxide films 2a, 2b are advantageously silicon oxides. By "silicon oxide" is meant silicon oxide of formula SiO 2 (silicon dioxide) or its non-stoichiometric SiO x derivatives.

[0062] Step a') may include the following steps: a' 1 ) providing a structure comprising: a substrate 1 based on crystalline silicon, having first and second opposite surfaces 10, 11; first and second oxide films 2a, 2b, preferably of tunnel oxide type, formed respectively on the first and second surfaces 10, 11 of the substrate 1; a' 2 ) forming first and second layers of amorphous silicon 3c, 3d, respectively on the first and second oxide films 2a, 2b; a' 3 ) implanting the n-type dopants and the p-type dopants, respectively in the first and second layers of amorphous silicon 3a, 3b.

[0063] Step a' 2 ) is advantageously carried out by plasma-assisted chemical vapor deposition PECVD.

[0064] Step a' 3 ) may be carried out by a plasma immersion ion implantation technique. When the n-type dopants are phosphorus atoms, the implantation of the phosphorus atoms during step a' 3 ) is preferably carried out under an atmosphere comprising phosphine PH 3 . When the p-type dopants are boron atoms, the implantation of the boron atoms during step a' 3 ) is preferably carried out under an atmosphere comprising diborane B 2 H 6 or boron trifluoride BF 3 . Step a' 3 ) is advantageously carried out so that the phosphorus atoms and the boron atoms have a volume density greater than 10 20< at. / cm 3< . Thus, an advantage provided is to create a strong field effect conducive to good passivation of the first and second surfaces 10, 11 of the substrate 1, as well as to form a good quality electrical contact zone.

[0065] Other techniques for forming dopants in the first and second layers of amorphous silicon 3c, 3d are possible: ion beam implantation, use of a sol-gel process, etc. Step b)

[0066] By way of non-limiting example, step b) may be carried out such that the annealing temperature TA is between 700°C and 900°C. Such an annealing temperature TA makes it possible to crystallize the first and second layers of amorphous silicon 3c, 3d. The crystallization of the first and second layers of amorphous silicon 3c, 3d is accompanied by the electrical activation of the n-type dopants and the p-type dopants by diffusion. Characteristics common to the objects of the invention Step b)

[0067] As illustrated in the figure 3 , the thermal annealing applied during step b) comprises: a first phase P 1 , comprising a rise in temperature P 10 (for example according to a temperature ramp of 1°C to 5°C per minute) until reaching the annealing temperature TA; a second phase P 2 , comprising a temperature plateau P 20 equal to the annealing temperature TA; a third phase in P 3 , comprising a fall in temperature P 30 from the annealing temperature TA , until reaching an ambient temperature T amb .

[0068] Atmosphere A under which the thermal annealing is applied during step b) advantageously comprises nitrogen N 2 or argon Ar.

[0069] The thermal annealing applied in step b) is a global thermal annealing in the sense that it is applied to the entire structure. It is therefore not a localized thermal annealing applied to a part of the structure, for example using a laser.

[0070] Step b) is preferably carried out in an oven equipped with means 40 for injecting an oxidizing agent 4. Step c)

[0071] Step c) is advantageously carried out so that the injection of the oxidizing agent 4 is started during the second phase P 2 ., after the electrical activation of the n-type dopants and the p-type dopants.

[0072] Step c) is advantageously carried out so that the oxidizing agent 4 comprises dioxygen O 2 . Step d)

[0073] The second instant t 2 defines, with the annealing temperature TA , a second thermal budget adapted to the first surface concentration so that the first and second oxide layers 5a, 5b, formed between the first and second instants t 1 , t 2 , respectively have first and second thicknesses, the first thickness being predetermined. More precisely, the first thickness is predetermined as a function of the desired thickness of the first polysilicon layer 3'a at the end of step d) and therefore of the consumption of the corresponding quantity of material (polysilicon). The second thermal budget is adapted to the first surface concentration to form the first oxide layer 5a according to the first predetermined thickness. The second thickness depends on the second thermal budget (and therefore on the first predetermined thickness) and on the second surface concentration.

[0074] Step d) is advantageously carried out so that the injection of the oxidizing agent 4 is stopped during the second phase P 2 . By way of non-limiting example, the time elapsing between the first and second instants t 1 , t 2 may be of the order of a few minutes.

[0075] According to an alternative, step d) can be executed so that the injection of the oxidizing agent 4 is stopped during the third phase P 3 , as soon as the second thermal budget is adapted to obtain the first predetermined thickness for the first oxide layer 5a, and thereby the desired thickness for the first polysilicon layer 3'a.

[0076] Step d) is advantageously carried out so that the first and second thicknesses are greater than or equal to 2 nm, preferably between 5 nm and 30 nm. By way of non-limiting example, the first predetermined thickness may be between 10 nm and 15 nm. By way of non-limiting example, the second thickness may be between 3 nm and 5 nm. By way of non-limiting example, the first polysilicon layer 3'a may have a thickness of 6 nm (± 1 nm) at the end of step d). Step e)

[0077] As illustrated in figures 1e And 2e, the method advantageously comprises a step e) consisting of removing the first and second oxide layers 5a, 5b formed between the first and second times t 1 , t 2 ; step e) being carried out after step d), preferably by a solution of hydrofluoric acid HF, more preferably diluted. Step e) can also be carried out by hydrofluoric acid vapors HF to reduce the chemical inputs. According to another possibility, step e) can be carried out by plasma etching, preferably with a fluorinated plasma to reduce the chemical inputs. Stage f)

[0078] As illustrated in figures 1f And 2f , the method advantageously comprises a step f) consisting of forming first and second layers of transparent conductive oxide 6a, 6b, respectively on the first and second layers of polysilicon 3'a, 3'b. Step f) is carried out after step e).

[0079] The first and second conductive transparent oxide layers 6a, 6b are advantageously made of a material chosen from CuO, NiO, TiO, a tin-doped fluorine oxide, tin oxide SnO 2 , zinc oxide ZnO; SnO 2 and ZnO being preferably doped with fluorine and aluminum respectively. Step g)

[0080] As illustrated in figures 1f And 2f , the method advantageously comprises a step g) consisting of forming electrodes E on the first and second conductive transparent oxide layers 6a, 6b. More precisely, step g) may consist of forming at least one electrode E on the first conductive transparent oxide layer 6a, and at least one electrode E on the second conductive transparent oxide layer 6b. Step g) advantageously comprises a metallization step, preferably carried out by screen printing. Each electrode E is preferably made of silver and / or aluminum.

[0081] The invention is not limited to the embodiments disclosed. Those skilled in the art are able to consider their technically effective combinations and to substitute equivalents for them.

Claims

1. Process for manufacturing a photovoltaic cell, comprising the following steps: a) providing a structure comprising: - a crystalline silicon-based substrate (1) having opposite first and second surfaces (10, 11); - first and second oxide films (2a, 2b), preferably of tunnel-oxide type, formed on the first and second surfaces (10, 11) of the substrate (1), respectively; - first and second polysilicon layers (3a, 3b) that are formed on the first and second oxide films (2a, 2b), respectively, and that comprise n-type dopants and p-type dopants, respectively, having first and second concentrations per unit area, respectively, the n-type dopants preferably being phosphorus atoms and the p-type dopants preferably being boron atoms, the first and second polysilicon layers (3a, 3b) each having a thickness greater than or equal to 15 nm and preferably greater than or equal to 20 nm; b) applying, under an atmosphere (A), a thermal anneal to the structure at an annealing temperature (TA) greater than a minimum temperature (Tmin) of electrical activation of the n-type dopants and p-type dopants; c) initiating an injection of an oxidant (4) into the atmosphere (A) in step b), so as to form first and second oxide layers (5a, 5b) on the first and second polysilicon layers (3a, 3b), respectively, the injection being initiated at a first time (t1) defining, with the annealing temperature (TA), a first thermal budget configured to electrically activate the n-type dopants and p-type dopants; d) stopping the injection of the oxidant (4) at a second time (t2) defining, with the annealing temperature (TA), a second thermal budget configured for the first concentration per unit area, in such a way that: - the first and second oxide layers (5a, 5b) formed between the first and second times (t1, t2) have first and second thicknesses, respectively, the first thickness being predetermined; - the formation of the first and second oxide layers (5a, 5b), with the first and second thicknesses, consumes an amount of polysilicon between the first and second times (t1, t2) that is configured so that the first polysilicon layer (3'a) has a thickness strictly less than 15 nm at the end of step d).

2. Process for manufacturing a photovoltaic cell, comprising the following steps: a') providing a structure comprising: - a crystalline silicon-based substrate (1) having opposite first and second surfaces (10, 11); - first and second oxide films (2a, 2b), preferably of tunnel-oxide type, formed on the first and second surfaces (10, 11) of the substrate (1), respectively; - first and second amorphous-silicon layers (3c, 3d) that are formed on the first and second oxide films (2a, 2b), respectively, and that comprise n-type dopants and p-type dopants, respectively, having first and second concentrations per unit area, respectively, the n-type dopants preferably being phosphorus atoms and the p-type dopants preferably being boron atoms, the first and second amorphous-silicon layers (3c, 3d) each having a thickness greater than or equal to 15 nm and preferably greater than or equal to 20 nm; b) applying, under an atmosphere (A), a thermal anneal to the structure at an annealing temperature (TA) greater than a minimum temperature (Tmin) of crystallization of the first and second amorphous-silicon layers (3c, 3d); c) initiating an injection of an oxidant (4) into the atmosphere (A) in step b), at a first time (t1) defining, with the annealing temperature (TA), a first thermal budget configured to crystallize the first and second amorphous-silicon layers (3c, 3d) so as to form first and second polysilicon layers (3a, 3b), step c) being executed so as to form first and second oxide layers (5a, 5b) on the first and second polysilicon layers (3a, 3b), respectively; d) stopping the injection of the oxidant (4) at a second time (t2) defining, with the annealing temperature (TA), a second thermal budget configured for the first concentration per unit area, in such a way that: - the first and second oxide layers (5a, 5b) formed between the first and second times (t1, t2) have first and second thicknesses, respectively, the first thickness being predetermined; - the formation of the first and second oxide layers (5a, 5b), with the first and second thicknesses, consumes an amount of polysilicon between the first and second times (t1, t2) that is configured so that the first polysilicon layer (3'a) has a thickness strictly less than 15 nm at the end of step d).

3. Process according to Claim 1 or 2, wherein the thermal anneal applied in step b) comprises: - a first phase (P1), comprising a temperature rise (P10) until the annealing temperature (TA) is reached; - a second phase (P2), comprising a temperature plateau (P20) at a temperature equal to the annealing temperature (TA); - a third phase (P3), comprising a temperature drop (P30) from the annealing temperature (TA); and wherein step c) is executed in such a way that the injection of the oxidant (4) is initiated during the second phase (P2), and step d) is executed in such a way that the injection of the oxidant (4) is stopped during the second phase (P2).

4. Process according to any of Claims 1 to 3, wherein step c) is executed in such a way that the oxidant (4) contains dioxygen O2.

5. Process according to any of Claims 1 to 4, wherein the atmosphere (A) under which the thermal anneal is applied in step b) contains dinitrogen N2 or argon Ar.

6. Process according to any of Claims 1 to 5, wherein step d) is executed in such a way that the first and second thicknesses are greater than or equal to 2 nm.

7. Process according to any of Claims 1 to 6, comprising a step e) of removing the first and second oxide layers (5a, 5b) formed between the first and second times (t1, t2), step e) being executed after step d), preferably using a solution of hydrofluoric acid HF.

8. Process according to Claim 7, comprising a step f) of forming first and second transparent-conductive-oxide layers (6a, 6b) on the first and second polysilicon layers (3'a, 3'b), respectively, step f) being executed after step e).

9. Process according to Claim 8, comprising a step g) of forming electrodes (E) on the first and second transparent-conductive-oxide layers (6a, 6b).

10. Process according to Claim 1 and any of Claims 3 to 9 in combination with Claim 1, wherein: - step a) is executed in such a way that the n-type dopants are phosphorus atoms and the p-type dopants are boron atoms; - step b) is executed in such a way that the annealing temperature (TA) is between 700°C and 900°C.

11. Process according to Claim 2 and any of Claims 3 to 9 in combination with Claim 2, wherein step b) is executed in such a way that the annealing temperature (TA) is between 700°C and 900°C.