WO2004059746A1

WO2004059746A1 - Integrated photovoltaic module for concentrating system and production method thereof

Info

Publication number: WO2004059746A1
Application number: PCT/FR2003/003865
Authority: WO
Inventors: Pierre Gidon
Original assignee: Commissariat A L'energie Atomique
Priority date: 2002-12-24
Filing date: 2003-12-22
Publication date: 2004-07-15
Also published as: FR2849276A1

Abstract

The invention relates to a photovoltaic module which is intended to co-operate with a light-concentrating system. The inventive module comprises a substrate (11) and a layer of semiconductor material (12) which is separated from the substrate by an electrically-insulating layer (13). A plurality of elongated bands (15) is disposed in the aforementioned layer of semiconductor material, each band comprising one inlet face that is designed to receive the light. Moreover, each of said bands is bordered by electrically-insulating walls (14) which extend to the above-mentioned electrically-insulating layer and each band comprises doped areas (D1 and D2). In addition, the module also comprises intermediate metallic tracks (17) which are used for the in-series electrical connection of the elongated bands of semiconductor material and which are disposed parallel to the electrically-insulating walls. The invention further comprises end metallic tracks (18) forming output electric terminals which are connected to the end bands of the plurality of semiconductor bands.

Description

Integrated photovoltaic module for concentration system and a manufacturing process

Technical area.

The invention relates to the field of concentrated photovoltaic systems, generally in the case of sunlight. It targets in particular components for a concentrated solar cell panel for low-cost electricity generation. A concentrating cell is a photovoltaic cell placed at the center of a light concentrating system. Concentration is ensured by lenses or mirrors. These cells may be preferred to cells without a concentrator since it is known that the efficiency of a cell increases with the incident light power. In addition, a single large-area photovoltaic cell can cost more than a small photovoltaic cell combined with a large-area concentrator.

These concentration cells are assembled to form panels which must in practice be provided with a system for monitoring the sun due to the directivity of the concentration system. The invention relates more particularly to a photovoltaic cell to be placed under a concentration system of any suitable type, the influence of which for the distribution of illumination on the photovoltaic element will simply be taken into account.

State of the art.

There are already many cells with concentration, with different performances and structures. To make a cell as economical as possible and at maximum efficiency, it has been shown that concentrations between 100 and 1000 are required. With conversion yields usually ranging from 20% to more than 30%, the electric currents generated will from a few A / cm ² to a few tens of A / cm ² . The voltages generated by the photovoltaic cells are typically 0.7V and at most a few volts. Under these conditions, in order not to lose power in the electrical interconnection lines, it is necessary that they have a very low resistance.

For example, a line that should not lose more than 0.5% of the power generated at 5A at 0.7V must have a resistance of less than 700 micro-Ohms. Such resistances require large and thick conductors and contacts, between conductors, very well made.

However, such wide conductors cannot be placed on the active surface of the solar cell without providing a shadow detrimental to its performance. It is therefore conventional to put the connections on the face opposite to the illuminated face, but unfortunately this requires the photo-charges to pass through the thickness of the cell, thereby promoting their recombinations. At the same time, it is very restrictive for their realization, since it naturally leads to choosing to make thin cells and therefore fragile.

Another solution is to multiply the number of cells and concentrators side by side on the same surface and thus reduce their size. The current in each conductor linked to each cell is reduced by the same amount, and the conductor can then have a higher resistance, and therefore be smaller. It can thus be placed on an illuminated surface recovering the charges where they are created without making too much shade.

Unfortunately, these small cells are very sensitive to the following problem: concentrated solar cells are degraded by side effects. The length / surface ratio of the edges is unfavorable for small cells, since the edge becomes very important; however, it is generally poorly controlled. Indeed, the cells of small sizes have the advantage of being able to be made collectively on the same plate, which lowers their cost, but requires cutting the plate to separate them at the end of their production. However, it can be shown that this cut locally creates a highly recombinant surface for the photo-charges. There are different techniques to overcome this drawback, such as for example that of leaving a guard zone between the photosensitive surface and the edge. The recombinant effect of the edge is then minimized; but the surface of the cell, therefore its cost, is very greatly increased (for example, a guard of only 1 mm around a 2x2 mm cell multiplies its surface and its cost by 4). The other methods are also more expensive. The problem of the cost of solar techniques, vis-à-vis other energy production systems, is a parameter which leads all experts to favor the simplest and most rapidly constructed structures possible. Complex structures are even considered only if it can be expected that their performance will be much higher than that of simple structures. The common view is that the cheapest cell should be brought together with the cheapest concentrator in a cheapest panel structure, in order to hope to produce the cheapest energy. This is why, at present, only single cells or cells with multiple junctions and stacked to be put under concentration systems are known. Another disadvantage of small cells is that they require a large number to cover the same area. At a given yield, the surface of the panel is fixed by the power to be supplied. Small cell panels become expensive by the multiplication of mounting operations to achieve them. It is therefore preferable to make concentration panels with a minimum number of cells. The size of the cells must then be maximum, which is contradictory with the previous constraints imposed by the electrical resistance.

However, it is a good compromise between these apparently contradictory criteria (with in particular a good conversion efficiency, low electrical losses, for space and a reasonable cost) that the invention aims to provide. Statement of the invention.

The invention proposes for this purpose a photovoltaic module, intended to cooperate with a light concentration system, comprising

a substrate, a layer of semiconductor material separated from the substrate by an electrically insulating layer, a plurality of elongated strips each having an inlet face adapted to receive light being formed in this layer of semiconductor material and being each bordered by electrically insulating walls extending to the electrically insulating layer, each strip comprising doped zones,

- intermediate metal tracks electrically connecting the elongated strips of semiconductor material in series and arranged parallel to the electrically insulating walls, and

- extreme metal tracks constituting electrical output terminals connected to the extreme strips of the plurality of strips of semiconductor material.

The invention thus proposes a new photovoltaic structure for concentration which rather resembles, by its design, that of a miniature photovoltaic module (or of a miniature solar panel element). However, this miniature module is monolithic, like a photovoltaic cell. This is why we can say that it is an Integrated Photovoltaic Module (abbreviation MPI used below) in the image of what are integrated electronic circuits for electronics.

It should be pointed out that the fact of designing and producing a monolithic photovoltaic structure as complex in miniature to put it at the center of a concentration system has in particular the consequence that the cost (and the difficulty) of unit production of this MPI is higher than the cost of unit production of a conventional concentration cell, on the other hand the cost of the final solar panel is very greatly reduced. According to the preferred arrangements of the invention, possibly combined: - Said intermediate metal tracks are arranged along the entry face of the elongated strips, each covering one of the electrically insulating walls bordering the elongated strips of semiconductor material and electrically connecting doped zones of different types within the strips elongated separated by this wall,

- this module can also include secondary metal tracks arranged at a distance from the electrically insulating walls, connected to one of the intermediate or extreme metal tracks and to doped zones of the same type within the underlying strip, - each of the secondary metal tracks is arranged between two secondary metal tracks or between an intermediate or extreme track, and another secondary metal track connected to doped zones of the same other type within said underlying strip,

each of the secondary tracks is arranged between two metal tracks connected to one another and from which this secondary track is isolated,

- at least some of the metal tracks are covered by a micro-light concentrator,

- said micro-concentrator has a triangular prism section,

- each micro-concentrator is an integral part of the metal track it covers,

- the micro-concentrators can be associated or replaced functionally by micro-lenses of the cylindrical type arranged on the free face of the elongated strips

- the entry face of each elongated strip is covered with a transparent and electrically insulating layer having orifices allowing contact between the semiconductor material and the metal tracks, - this transparent layer behaves like a passivation layer in contact with the semiconductor material, - the transparent layer is formed by a stack of sublayers,

- the transparent layer comprises silica,

this layer is at least partly formed of a material chosen from the group comprising Si3N4, SiOxNy, MgF2,

- the entry face covered by the transparent layer is textured,

- the entry face covered by the transparent layer is shaped into a plurality of adjacent pyramids, - the elongated strips comprise doped zones along each of the walls which border them laterally,

the elongated strips have doped zones along the electrically insulating layer and / or the light entry face,

- the elongated strips have a constant section over their entire length,

- the elongated strips have a width which is minimum in a central zone and maximum near the extreme metal tracks,

- the elongated bands have a width which is maximum at the midpoint of their ends, and minimum at these ends, - the elongated bands have a width which is minimum at midway of its ends, and which increases progressively (is continuously variable ) between this minimum and these ends,

- the walls delimiting these elongated bands are in a broken line, - the walls delimiting these elongated bands are curved,

- the elongated bands have the form of loops,

- the elongated bands form concentric rings,

- secondary tracks are connected to a conductive track running along an elongated annular strip, - the secondary tracks are substantially radial from this track, 004/059746

- this module has a rectangular or square shape, or even a circular or oval shape or in a regular polygon,

- At least one light conversion cell is further provided in the substrate, under the electrically insulating layer, for a range of wavelengths different from that for which the elongated strips are adapted to convert light; in particular in this case, the module can be turned over, being adapted to receive light from any of its faces.

- components are produced in the layer of semiconductor material; these are advantageously isolated from microcells; it may in particular be a protective diode (s), and / or a switch (s) and / or an inverter (s), etc.,

- The semiconductor material of the elongated strips is silicon, germanium or Si _x Geι- _x ., even if other materials can be chosen as required.

According to another aspect of the invention, it proposes a method of manufacturing a photovoltaic module intended to cooperate with a concentration system, according to which,

A plate is produced comprising a substrate and a layer of semiconductor material separated from the substrate by a buried electrically insulating layer,

Slots are etched in the layer of semiconductor material extending up to the buried layer so as to delimit elongated strips, and the sides of these strips are coated with an electrically insulating layer, for example of oxide,

• these slots are filled so as to produce walls separating the elongated strips,

• doped areas are produced in each of these elongated strips, at least near each wall, • metallic tracks are formed on each wall on this wall by connecting doped areas located on either side of this wall, • extreme tracks are formed on the extreme walls forming output terminals.

Objects, characteristics and advantages of the invention appear from the following description, given by way of nonlimiting illustrative example, with reference to the appended drawings in which:

• Figure 1 is a sectional view of a photovoltaic module according to a first, simple embodiment of the invention, Figure 2 is a top view, Figure 3 is a sectional view of another module according to another embodiment, Figure 4 is a sectional view of an alternative embodiment, Figure 5 is an alternative embodiment, Figure 6 is yet another alternative embodiment, Figure 7 is a sectional view of yet another module according to another embodiment, FIG. 8 is a top view, FIG. 9A is a top view of another embodiment, FIG. 9B is a top view of an alternative embodiment, Figure 9C is a top view of a variant of Figure 9B, Figure 9D is a top view of yet another alternative embodiment, Figure 10A is a top view of yet another embodiment, Figure 10B is a top view of an alternative embodiment, Figure 1 0C is a sectional view along line XX of Figure 10B, Figure 11 is a sectional view of yet another module according to another embodiment, Figure 12 is a top view of yet another module according to yet another embodiment, FIGS. 13 to 23 are sectional views showing steps in the manufacture of a module conforming to that of FIGS. 1 and 2, FIGS. 24 to 31 are sectional views representing steps in the manufacture of a module conforming to that of FIG. 3,

• Figures 32 to 44 are sectional views showing steps in the manufacture of a module according to that of Figure 11. Figures 1 and 2 show a particularly simple version of a photovoltaic module according to the invention.

This module designated as a whole under the reference 10, comprises a substrate 11, a layer 12 of semiconductor material separated from the substrate by an electrically insulating layer 13, electrically insulating walls 14 extending through the layer 12 from its upper surface up to the buried layer 13 so as to delimit in this layer 12 strips or blocks of semiconductor material 15., and metal tracks.

This module is intended to cooperate with a light concentration system (not shown) of any suitable known type, such that light arrives on the module 10 from above. The upper face of the layer 12 and therefore of the strips 15, therefore constitutes a light entry face.

The strips or blocks 15 constitute elementary solar cells also called micro-cells in the following, and comprise for this purpose, in a conventional manner, doped zones, here represented in the upper corners. One type of doping (p or n type) is designated by the reference D1 and the other (n or p type) by the reference D2.

It should be noted that the strips or blocks are here electrically insulated from below (by the buried layer 13), on the sides (by the walls 14) but also at their ends by walls 16 which define with the walls 14 extremes a rectangle containing all the bands 15.

The crosses which appear in FIG. 2 are positioning marks.

The metal tracks, therefore electrically conductive, are, in the example shown, advantageously materialized on the upper surface of the layer 12 and can be divided into two groups: * intermediate metal tracks 17 which electrically connect in series the elongated strips 15 and arranged parallel to the electrically insulating walls 14,

* extreme metal tracks 18 which constitute electrical output terminals connected to the extreme bands of the plurality of bands of the module.

As is apparent from FIGS. 1 and 2, the intermediate metal tracks 17 are advantageously arranged along the strips 14 by covering, here substantially over their entire length, each one of the electrically insulating walls 14 bordering the elongated strips by connecting them respective doped zones of different types within the bands separated by the wall considered. By way of example, FIG. 1 has two arrows which designate, on either side of the leftmost track 17, areas through which this track 17 is in electrical contact with, respectively, a D2 doped area of the strip 15 to the left of this wall 14, and a doped area D1 of the strip 15 located to the right of this same wall 14.

Also advantageously, the intermediate metal tracks 17 are covered by a micro-light concentrator, here in the form of a triangular prism; the tracks and their micro-concentrators advantageously form an integral part of one another.

Also advantageously, the upper surface of the micro-cells 15 is covered by a transparent layer 19, in practice electrically insulating which has at least orifices allowing contact between the semiconductor material and the metal tracks (see the aforementioned zones designated by the arrows on either side of the metal track 17 furthest to the left in FIG. 1). This layer 19 also plays a passivation role in practice.

It can be noted that each of the micro-cells is thus bordered on each of its faces by electrically insulating walls or layers forming a sort of box enclosing this micro-cell.

In general, it can be noted that the MPI module (for Integrated Photovoltaic Module, see above) represented in FIGS. 1 and 2 has a structure recognizable in that it comprises layers and insulating walls present in a semiconductor material, on a monolithic support and that the assembly is adapted to be placed under a high concentration system (typically: more than 10 times). These layers and walls define semiconductor blocks insulated from each other. Each semiconductor block or strip is regionally doped to form a photodiode and at least one of its faces transmits light. The metal tracks interconnect the micro-cells. Only well-defined openings can be made in the insulating layers to bring the semiconductor and the metal into contact. In the simple MPI of FIGS. 1 and 2, the openings are chosen such that the "plus" contacts of a micro-cell are electrically connected to the "minus" contacts of its neighbor. This monolithic structure is characterized in that it behaves like the placing in series of micro-cells. The following direct advantages can be noted: because of the microtechnologies available and chosen for the production, the walls insulating each block from its neighbors are of high quality. They only lead to a weak recombination of the photo-charges created in the semiconductor during a conversion of incident light. The voltage delivered by the MPI is a function of the number of microcells placed in series. Relative to a conventional concentration cell, the same size as the MPI, the current is divided by a factor equal to the number of micro-cells in series. The resistance opposed by the lines is therefore less important. Since the micro-cells already have a well-defined edge, the MPI associating them does not need any particular attention during its cutting.

In fact, even if it is preferred that it be on the front face (see above), the conductors can be placed both on the rear face and on the front face, since they are compact. By way of example, the module of FIGS. 1 and 2 is formed by a succession of micro-cells produced in semiconductors on insulator (Silicon On Isolator - for example Si / SiO2). The micro-cells are separated between them by insulating walls made of SiO2 and poly-silicon. The insulating and transparent layer located on the upper part of the semiconductor is made of SiO2 (transparency is necessary for the entry of light into the semiconductor; it is only useful in the wavelength range for which one seeks to convert light into photo-charges).

The horizontal insulating layers or the vertical insulating walls of the micro-cells can have a complex structure with several layers.

The SOI support substrate can be of any kind (typically silicon, but it can also be a glass, or a metal). The fact of giving the tracks connecting the micro-cells in series (see at the extreme tracks, a form of micro-concentrator makes it possible to minimize the light losses since the light intercepted by these tracks is thus returned towards the entry face of the 'one of the adjacent micro-cells.

Several MPIs of this type can be manufactured simultaneously side by side; but the final cutting edge which corresponds to the edge of the diagram in FIG. 2 is not critical for the performance of each cut module.

In this figure 2, the walls which exist under the metallizations are not visible (because hidden) except at their ends. The doped zones which, according to the preceding diagram, would be impossible to see appear nevertheless for comprehension or as a variant by a greater extent.

This module is integrated in the sense that it is the placing in series of concentration cells in a single photovoltaic component.

Variants The module in Figures 1 and 2 represents a particularly simple module. Various types of variants are presented in the following, each variant providing particular additional qualities, without removing the basic qualities. Of course these various variants can be combined together to benefit from several advantages at the same time. Certain variants complicate the MPI but have the consequence of simplifying the overall panel using them. This is the main advantage of MPI, the more functions it integrates, usually carried over individually to the less the panel, which has several panels, will be complicated and expensive.

Simple variants of this structure can be envisaged, for example concerning the nature and extent of the dopings or their positions, the structure of the walls, or even the overall shape of the cell.

1 / Variants concerning doping.

Thus, there can be an extension of the doped zones which is not usual for those skilled in the art. Indeed, although each semiconductor block is taken in a monolithic system, it can be doped on all its faces.

This is how the diagram in FIG. 3 shows a module 10 ′, the elements of which similar to those of FIGS. 1 and 2 are designated by the same reference number, but affected by the “prime” index. It is observed that this module is doped, not only at the location of the corner areas D1 'and D2', but also in the vicinity of the side walls 14 'and the lower 13' and upper 19 'surfaces of the semiconductor; these additional doped zones are denoted 20 ′. Of course, there may be only doping along some of the walls, for example only along the vertical walls, or even only along a single horizontal surface. Several types of doping are possible. If the semiconductor is of a type (P or N), the doping will be of the opposite type (N or P respectively). If the semiconductor can be considered as intrinsic (very low P or N doping), then the edge can be more strongly doped P or N depending on the desired result. The aim of these dopings of the faces is mainly to reduce the influence of the surface effects and consequently to obtain high quality block edges recombining the photo-charges very weakly.

2 / Variants concerning the light entry faces of the microcells.

The metal tracks may not form an exact micro-concentrator, in which case micro-lenses (preferably of the cylindrical type and extending over the length of the metal tracks) are advantageously arranged above the micro-cells to group the light on each micro-cell next to the non-transparent metal.

Furthermore there is advantageously an anti-reflection treatment of the upper faces of the micro-cells, to minimize reflections. Different conventional methods exist and are applicable to these cells. Three methods are considered here, using anti-reflective layers, texturing, or pyramids.

In the diagram of FIG. 4, where elements similar to those of FIGS. 1 and 2 are assigned the same reference numbers, increased by the number 100, the transparent layer 19 has been replaced by a stack of several transparent layers. The layer 119A in contact with the semiconductor is responsible for passivating its surface. The other layers 119B and 119 C (there are a total of three sublayers here) have well chosen thicknesses and indices to minimize the reflection of light. We can cite to make materials such as Si3N4, SiOxNy, MgF2, ...

The diagram in FIG. 5 represents a module 210 whose elements similar to those of module 10 are assigned the same reference number, increased by the number 200. It comprises micro-cells 215 with a surface texturing designated by the arrow T. The surface has a random structure in all directions, very fine (less than a micron), which does not reflect light. The transparent, insulating passivation layer 219 covers this surface.

The diagram in FIG. 6 represents a module 310 whose elements similar to those of module 10 are assigned the same reference number, but increased by the number 300. It comprises micro-cells whose entry surface has a regular pyramidal structure designated by arrow P. It is known that these pyramids, in combination with a passivating and simple anti-reflective layer 319, transmit light very well in the semiconductor. These variants are only improvements for the entry of light into the semiconductor. They are classic on other cells solar, but they apply to a module of the invention without disturbing the basic structure.

3 / variant concerning the links between tracks - interdigitated structure

A particularly interesting variant is to make micro-cells larger than in the above examples, but to keep the density of the contacts and metal tracks on the whole cell in order to collect the charges well. This amounts to making a slightly more complex MPI because each micro-cell has an interdigitated cell structure.

To keep it legible, the diagrams in Figures 7 and 8 represent a very simple case, where there are only two micro-cells in the MPI marked 410. Elements similar to those of Figures 1 and 2 are designated by reference figures which are deduced therefrom by adding the number 400.

The intermediate metal tracks 417 which are above a wall 414 always connect a doping D1 of a micro-cell 415 to the complementary doping D2 of the next micro-cell. The other metal tracks 421, called secondary tracks, are in contact with only one of the dopings, inside a micro-cell. The top view shows that at the change of micro-cell, they change doping.

As we can see in the diagram below, on a micro-cell, the metal tracks are on different dopings and are distributed alternately in the manner of interdigital combs, as in a conventional cell.

It can be noted that each of the secondary metal tracks is arranged between two secondary tracks or between an intermediate or extreme track, and another secondary metal track connected to doped zones of the same other type within said underlying strip. Similarly, another way of describing this advantageous configuration is to say that each of the secondary tracks is arranged between two metal tracks connected to one another and from which this secondary track is isolated. The advantage of this variant is that it allows the number of microcells to be adjusted during construction to choose the output voltage of the MPI, regardless of the problems of collecting photo-charges and size. of the illuminated surface (there are no more masking problems than in the case of module 10 in FIGS. 1 and 2 (there are as many metal tracks) while allowing a different voltage to be obtained at the output, between terminals 418. 4 / Variants concerning the shape of the micro-cells and of the module In the above examples, the micro-cells have, when viewed from above, identical rectangular sizes, in other words they are rectilinear and have a constant section over their entire length. If the illumination is uniform, all these micro-cells will output the same photogenerated current and there will be no difficulty. An MPI is intended to be placed under a concentration system. The concentration system will make a clear or blurred and / or distorted image of the sun as the case may be. This form is necessarily different from uniform lighting on a rectangle, the only case considered so far in this description of this structure. ue each photovoltaic micro-cell behaves like a current generator in parallel with a diode. When the micro-cells are connected in series, the current generators limit the maximum current to the value they generate. Consequently: if the illumination is inhomogeneous, the current delivered by the MPI will be the current generated by the least illuminated of the micro-cells, from where a consequent loss of efficiency, even a complete cancellation if a micro-cell, at edge for example, is not lit.

It is therefore recommended to adapt the shape of the MPI and its microcells to equalize the currents therein. It is equally recommended that this adaptation does not hinder that all the incident light enters the photovoltaic material.

If conventional solar panels include the placing of solar cells in series, none of the known panels sees the shape of its cells vary to meet the requirements of concentrated light. In addition, commonly, for the entire panel 10% and more of the incident light is lost.

This is why, in advantageous embodiments of 2004/059746

17

the invention, the shape of the micro-cells is modified to take account of the non-uniformity of illumination of the module. This modified form is chosen according to the characteristics of the concentration system.

We will consider here some special cases to illustrate how the image or the shape of the spot of concentration can justify an adaptation of the shape of the micro-cells and show that this structure lends itself to this choice while keeping its collection characteristics almost complete with incident light with illuminated face contacts and always compatible with the various anti-reflection.

a) Image of the sun perfectly clear and stable

A sharp image of the sun is a uniformly lit disc. It is a difficult case to obtain by the quality of the optics necessary but perfectly possible and consider if you want a bill of high concentration and exactly known.

The contour of the MPI can approach very closely that of a circle by a polygonal cut. In the module 510 of FIG. 9A, we have chosen the hexagon. The separation, until now linear of the 515 micro-cells, is here deformed to adjust to the circular shape of the image of the sun and define elements whose illuminated surface is almost identical. Here, we have chosen walls (and tracks) 517 in broken lines, but smoother shapes are possible.

The rings symbolizing the image of the sun of uniform intensity. The tracks with a central line represent the top view of the micro-concentrators interconnecting the doped areas of the micro-cells and reflecting the light towards the photovoltaic material. The elongated strips 515 are wider in their central portion than at their ends. The extreme tracks 518 are wider for connection to the external charging circuit.

b) Blurred and stable image of the sun

More generally, we are looking for an inexpensive optical system and simple. The image may be blurred as long as it is the desired size. The image is then often concentric with a maximum of intensity in its center and a decrease towards the edges. The concentration factor is variable. If we keep a square-shaped MPI, FIGS. 9B and 9C show two diagrams which show that the surfaces of each micro-cell can be adapted to the intensity distribution to equalize the currents without changing the structure of the MPI.

The width of the micro-cells is minimal where the light intensity is maximum. The width is chosen as a function of the light intensity delivered by the optical system in order to obtain quasi-equality of the currents of the micro-cells. It is a constant principle that we will find for all the cells described now.

The rings tightening against each other in the center of Figure 9C want to represent an increase in intensity in the center of the MPI. The micro-cells are 6 in number in these diagrams but can go to large numbers (several hundreds) in the production.

In FIGS. 9B and 9C, elements similar to those of FIG. 9A are assigned the same reference number, but with an index "prime" or "second".

c) Blurred image of the sun, distorted and unstable in one direction

It may be that, to minimize the cost of the assembly, it is preferable to achieve a greater concentration in one direction than in the other. The spot is then non-circular. It has fuzzy edges. If the tracking system, again to minimize costs, is imprecise in one direction then the MPI can be longer than the spot illuminated in this direction so that the light spot oscillating on its surface does not come out if the micro -cells and their connections by micro-concentrators are elongated according to the direction of oscillation, the currents will remain constant despite the variation in position. This is illustrated by the module 510A of FIG. 9D where the tracks 517A, rectilinear, delimit bands 515A which are all the wider as one moves away from the center towards the extreme tracks 518A. The two series of ellipses aim to symbolize two possible positions of the light spot. The micro-cells are narrower at the center of the MPI because the light spot is more intense.

d) Blurred image of the sun, distorted and rotating around a center

The sun tracking system can generate an image whose shape is stable but irregular. The apparent rotation of the sun in the sky can cause this irregular spot of light to rotate around a stable axis. An MPI with concentric microcells 615 or 615 'around the axis in this case keeps the currents constant and identical. In modules 610 and 610 'of FIGS. 10A to 10D, the micro-concentrator 619 or 619' which goes to the center connects a central contact point at the edge so that the connection system does not mask part of the MPI. The tracks are noted 617 or 617 '. The other contact is a contact taken at the edge, as in the previous examples, but the two right and left edges 618 and 618 'constitute the same contact.

The current generated goes from the edge to the center or from the center to the edge depending on the polarization of the semiconductor dopants. Regular ellipses have been shown to represent a centered spot of light. Dissymmetrical forms are entirely compatible with these MPIs. The distribution of the surface of the micro-cells can be very complex if the maximum intensity is not aligned with the axis of rotation, but a very well adapted MPI is achievable with an equality of the micro-cell currents.

The deformation of the light spot can induce that some of the micro-cells then have a very large surface, or zones of large width. To properly recover the current it may be useful to increase the surface of the contacts or their positions. This is what FIGS. 10B and 10C represent, which show what we have already described in the representations where the micro-cells did not have complex contours. Additional electrodes 620 ′ (opposite FIG. 10A) are provided on the peripheral micro-cell since it has very large areas. These additional electrodes are not vertical to a wall delimiting the micro-cells but join point or route contacts necessary for less resistive current collection. They are here substantially radial.

The micro-concentrator in the middle of the module 610 'connects a contact in the center of the central cell of the MPI to the edge passing over the other cells without contacting them, as shown by the continuous insulating layer which separates it from the semiconductor. Current flows between this central contact and the edge contact (s).

Even more complex patterns can be described if the spot deforms in a known manner during movement. This is valid for both a rotating image and a linear oscillation. The various preceding modules are made with a few microcells, but the actual systems can range to several hundred.

The versatile adaptation of this structure to the constraints of the concentration of sunlight is the strong point of MPI.

Thus, nothing is opposed to very varied forms for the modules such as rectangular, trapezoidal, circular or hexagonal, regular polygonal ..., nor to fluctuations in the shape of the micro-cells (rectilinear, curve, loop ... ) within a module. The adjacent strips can thus be parallel (in the case of parallel strips) or approximately parallel (in the case of variable sections) but also concentric, with widths equal or not from one strip to the other.

Whatever the form chosen for an MPI, it is always possible to design it with micro-cells. This maximizes compatibility with the concentration system and manufacturing techniques.

5 / Variants concerning the substrate carrying the micro-cells. A more complex variant of MPI is the possible association with an SOI support material (or substrate) which is active. The most useful case is a material that can support another type of solar cell. This case is interesting because the absorption of light is a function of materials and wavelengths. Silicon, for example, is transparent in the infrared (lambda wavelength> 1.2 μm). Light at these wavelengths reaches the support and can be collected there, as long as the material (s) constituting this substrate lends itself to it (for example a support made of germanium or SiGe for an upper silicon cell). In fact, one can choose any other semiconductor with electronic band gap having a width less than the electronic band band of the semiconductor which forms the upper micro-cells. This complementary cell does not a priori benefit from the advantages produced by the micro-cells of the upper cell, but its combination with cells on the surface can make it possible to optimize the amount of light converted into photo-charges.

FIG. 11 represents an example of a diagram representing a module 710 having such a structure. Elements similar to those of module 10 bear the same reference numbers, after adding the number 700. In fact, the upper part of this module is similar to that of this module 10, but the cell produced on the lower face marked 730 is here of a classic diagram with tracks 731 forming interdigitated combs and multiple doping zones including those denoted D1 "A and D 2" A. There may be passivation of the lower surface, as designated by the reference 740.

According to a variant embodiment not shown, the structure can also be swapped, with the large cell above the small ones, the concentrators remaining on the upper face. The large cell must in this case be made with a semiconductor with a band gap wider than that of the micro-cells.

Much more complex structures can also be envisaged, such as the stacking of structures with walls delimiting microcells. Current production techniques already allow it, even if their complexity (and consequently their cost) today leads to a significant cost.

6 / Realization of related components next to the micro-cells This is possible due to the existence of insulating walls. Components such as resistors, diodes, capacitors and transistors are then integrated separately or in groups in one or more zones of semiconductor isolated by the walls next to the micro-cells. For example, it may be worthwhile to provide protection against internal overvoltages. Indeed, it can happen events where the lighting becomes very inhomogeneous, typically during the phases of installation or pointing of the panel, or even for a problem of tracking of the sun. Part of the micro-cells then receives light, while the other part does not (or at least much less). The current generated by a micro-cell feeds the next micro-cell which behaves like a reverse diode. If the micro-cell is not lit, it refuses the passage of current. The voltage rises at its terminals. As there are in practice many micro-cells in series, a micro-cell can be subjected to the voltage generated by several micro-cells and reach its breakdown voltage. In addition, a poorly lit micro-cell is enough for the MPI to no longer deliver current, therefore more power. But it is easy to remedy this by placing a non-photovoltaic diode head to tail on each micro-cell, according to a solution known in conventional cells. FIG. 12 is a diagram of a module 810 in accordance with the principle which has just been proposed. The elements which are similar to those of the module 10 of FIGS. 1 and 2 are designated by the same reference numbers, after addition of the number 800. This module 810 thus comprises elongated strips 815 bordered by end walls 816 and walls side 814 of which only the ends are visible, these walls being over most of their length capped by intermediate metal tracks 817 or extremities 818. In fact there are three walls 816, including two walls in the upper part of FIG. 12, delimiting with the end of the walls 814 small boxes 840 in which diodes are produced by virtue of the production of doped zones D'1 and D'2 in the opposite direction to zones D1 and D2. The reference 850 designates additional connections to the outside and the reference 870 designates an additional internal connection between the left terminal 818 to part of this side part (right) produced on the module outside the electrically insulating walls 814 or 816.

The integration is such that the connection between the protective diodes and the micro-cells is made directly by extending the metal tracks 817 on the walls. The contact openings between the metal of the tracks and the semiconductor at the diode can be made in a similar way to what has been proposed for the micro-cells. The polarization of the diode is reversed as indicated by the inversion of the digits 1 and 2 after the letter D signifying the doped zones. To better protect non-photovoltaic diodes from incident light than that shown in this diagram, it is possible to provide a larger metallization which masks their surface. 11 Variant concerning the periphery of the module. The diagram of FIG. 12 comprises an additional variant by the addition of an electronic circuit 860 in a semiconductor zone insulated by walls, next to the elongated strips forming the micro-cells.

This circuit can have various functions and may or may not be associated with components external to the MPI. It can also be carried out in several zones isolated by walls. Its function is for example a simple diode, holding the reverse high voltage, in series to protect the MPI from an imbalance between the different MPIs of a panel. It can also be a simple transistor switch but also conventional choppers per se allowing to control the power emitted by this MPI with respect to the external circuit or even an inverter to obtain an oscillating voltage and an alternating current. The number of possible variants depends on the function to be performed. The realization depends on the function and can lead to additional steps, copied from the already known manufacturing steps for these components outside of their association with these solar micro-cells. The principle and the advantages of the basic MPI are not modified but supplemented.

Benefits If we assume an equal illumination between all the micro-cells, they will have the same voltage and deliver the same current. If there are N microcells in the integrated photovoltaic module, the voltage recovered at its ends will be N times the voltage of a micro-cell and the current equal to that of a micro-cell. The voltage-current couple at given power is adjustable during construction by the number N of micro-cells in view of the load for which the MPI is intended.

For example, we will pay close attention to the power lost in the connecting conductors. The power lost by the joule effect in the resistance of the lines decreases in N ² because P = RI ² . This significant reduction in connection losses allows us to use thinner and therefore less expensive power lines.

Limiting the size of the concentration cell to limit the currents delivered is pushed back to the largest areas for an MPI. Consequently, the number of MPIs to be mounted under concentrators is therefore reduced with equal panel area (power). The manufacturing cost of the final panel is reduced.

Another peculiarity of the MPI according to the invention is that, by manufacture, each micro-cell or other internal component has a well-controlled edge and that, consequently, the MPI does not need special attention for its cutting. There is no need to leave a guard zone. It is of course advisable to take care that the sawing at the end of the production loses the least possible surfaces to reduce the costs.

There are several methods to obtain this edge which can be made of several layers and thus be of high quality.

The MPI is compatible with various shapes, in particular hexagonal, to maximize agreement with the shape of the spot of light concentration.

The MPI is compatible with the various anti-reflective techniques, such as layers of well chosen thickness, but also texturizations or pyramidal surfaces. This structure is also compatible with various variants. In particular, the surface covered by each micro-cell is not necessarily a constant and can be chosen to optimize performance. In particular, the surface is chosen as a function of the uniformity of illumination.

The integration of insulating walls of the micro-cells allows the production of other complementary components isolated from the micro-cells in the semiconductor material but electrically connected by metallization. These components can be used to protect the MPI or optimize its operating point. The protection makes it possible to keep a non-zero efficiency for MPI, even when the panel which carries them is not exactly oriented.

The components and associated circuits in the integration allow optimization of compatibility with external electrical loads.

The fact of producing these components at the same time as the micro-cells in the MPI makes it possible to reduce the cost of mounting the panel since it will not be necessary to bring them back to the final assembly.

Numerical examples

Under a standard AM 1, 5 solar illumination, it reaches the ground at most 0.1 W / cm ² . If the concentration system (for example a lens) is a square with a side of 50 cm, then the power to be collected is 250W. If we place a square photovoltaic cell with a side of 2 cm, the concentration factor is 625.

When choosing to make the MPI in silicon, the range of absorbed wavelengths goes from 0.3 to 1.12 μm. To properly absorb this flux, it is necessary to maximize the thickness; on the other hand, the difficulty of manufacturing the walls increases with the thickness; this results in an arbitrary thickness choice in the range 10 to 100 μm for the silicon on insulator part.

Except in the case where one is interested in making a complementary cell in the substrate, the thickness of the electrically insulating layer is advantageously chosen to make a reflective layer and return the unabsorbed light into the cell. For a layer of SiO2 formed between a substrate and an upper layer both made of silicon, a minimum thickness chosen from the range 0.10 to 0.30 μm depending on the wavelength to be preferred is sufficient. Thicknesses which are integer multiples of this minimum thickness retain the reflective power of the layer and may be necessary for problems of technological compatibility and influence of the electric fields from the rear face.

If we choose to carry out this MPI with N = 352 micro-cells 56.8 microns wide, the expected yields of the micro-cells are greater than 24%: there is therefore approximately an electrical power P = 60.5 W recovered at pass. The voltage of a micro-cell, in charge, at maximum power, is close to that of a conventional cell, ie U = 0.68V. The MPI therefore delivers approximately I = 253mA under N * U = 239V (P = N * U * I). If it is fixed not to lose more than 0.1% of the power in the conductors, then these must not have a resistance greater than 0.945 Ohm. The resistivity of copper being close to 1.7 micro-Ohms. cm, a wire length of 1 meter to reach the next MPI must have a diameter greater than 0.15 mm, which poses no technical problem.

For comparison, a cell with a conventional concentration, of the same size, should output, if it existed, 89A at 0.68V. The conductors, which should not lose more than previously, should not have a resistance greater than 7.64 microOhm. The 1 m connecting copper wire should have a diameter greater than 53 mm (much larger than the cell!), Which is unrealistic and too expensive.

Examples of embodiment of modules according to the invention FIGS. 13 et seq. Schematically describe the steps of an example of a method of manufacturing the structure of a module according to the invention. The example shown here uses silicon and usual equipment in a microelectronic production chain. Each step mentioned below is a process known per se. We start (Figure 13) of a silicon wafer 900 in which an electrically insulating layer 901 has been formed. It is a commercial product known under the name of SOI (Silicon On Insulator).

We then proceed (Figure 14) to a thermal oxidation on the surface or to the deposition of a layer of SiO2, hence an oxide layer 902 which allows to passivate this free surface of the upper layer.

We then proceed (Figure 15) to the following steps:

- photolithography of a resin mask,

- anisotropic etching of the oxide, - anisotropic etching of the silicon until it stops on the buried insulating layer 901,

- elimination of the resin.

The upper layer of the original plate 900 is thus cut into elongated strips 903. An engraving not visible on this cut made simultaneously defines the length of the strips.

We then proceed (Figure 16) to a thermal oxidation of the sides of these elongated strips, hence the appearance of lateral layers of oxide 904.

Polysilicon is deposited (FIG. 17) until the slots remaining between the lateral layers 904 are filled (reference 905). The polysilicon deposited on the surface is then removed (for example by polishing, so that only the deposits filling said slots.

FIG. 18 represents a step during which a photolithography of a resin mask is carried out, an etching of the upper oxide 902, and elimination of the resin. Thin slits 906 are thus obtained on either side of the walls 905.

We proceed (FIG. 19) to a photolithography step of a resin mask 907 so as to cover the slots 906 situated on a given side of the walls, and a doping material (for example boron) is implanted to form a first type of D1 doped zones (for example of the "P" type) in the elongated strips under the other slots (not covered by the resin).

The resin deposited during the previous step is eliminated, and another resin mask 908 is deposited by photolithography (FIG. 20). sort of covering the previously doped slots, leaving the other slots bare, and implanting another doping material (for example phosphorus) to form a second type of D2-doped areas (for example of the "N" type) in the bands elongated under said other slots. The resin 908 is eliminated (FIG. 21), and an activation and diffusion annealing of the dopants is advantageously carried out; preferably a slight partial chemical deoxidation of the surfaces is carried out for the future production of electrical contacts.

According to FIG. 22, a thick layer of material 910 is deposited, for example in AISi, then a photolithography of a resin 911 is carried out with sloping edges.

According to FIG. 23, an anisotropic etching of the thick layer 910 is finally carried out, preferably a dry etching, and one ends by a chemical etching, and the residues of the resin are eliminated, which results in prisms 912.

This sequence of technological steps is only one possible example to create this structure in accordance with the module of Figures 1 and 2.

For the realization of the variants, the sequence of steps is very close to the above. However, the variants described above in 1 / (see Figure 3) and in 5 / (see Figure 11) may have clear differences.

The steps for producing variant 1 / are shown in FIGS. 24 to 31.

We start (Figure 24) from a commercial SOI plate 920 comprising a buried oxide layer 921 bordered by a buried doped layer 922. We penetrate (Figure 25), by implantation or by diffusion in an oven, of a dopant at the surface (layer 923) and a thermal oxidation and / or deposition of a layer of SiO2 is carried out, so as to obtain a passivation layer 924 of the surface.

As indicated above, we proceed (FIG. 26) to a photolithography of a resin mask, to an anisotropic etching of the oxide layer 924, then to an anisotropic etching of the silicon until stopping on the buried insulating layer. 921. Elongated bands 925 are obtained. According to FIG. 27, a dopant is made to diffuse in an oven, under a controlled atmosphere, so as to dope the flanks of these bands (reference 926) and then these flanks are thermally oxidized, so as to obtain a lateral oxide layer 927. From now on, the steps are identical to the previous case.

Only the appearance of the diagrams changes:

- In Figure 28, there is deposit of polysilicon until filling (reference 928) the slots separating the strips 925, then elimination of the polysilicon thus deposited on the layer 924; - In Figure 29, there is photolithography of a resin mask and etching of the upper layer 924 so as to form slots 929 along the walls 928, then elimination of the resin. As before, there is then (without the steps being shown): - photolithography of a resin mask,

- implantation (Boron to form P + zones),

- elimination of the resin,

- photolithography of a resin mask,

- implantation (phosphorus to form N + zones), - elimination of the resin

- activation and diffusion annealing of dopants, and

- slight partial (chemical) deoxidation of surfaces for future contacts.

This leads to the configuration of FIG. 30 with doped zones D1 'and D2 \

Then, a thick layer of AISi is deposited, photolithography of a thick resin with sloping edges, anisotropic etching of the thick layer, by dry etching, then chemical finishing treatment, and elimination of resin residues, which gives the configuration of FIG. 31, with prisms 930 covering the walls 928, that is to say the configuration of FIG. 3. The realization of variant 5 is more complex due to the interventions to be carried out on both sides. Here are the main steps as an example.

The first step concerns the SOI material which in this case is particular. It is for example a structure Si / SiO2 / SiGe or Si / SiO2 / Ge.

The first steps which follow are identical to the above, without it being useful to represent them again. These steps include:

- thermal oxidation and / or SiO2 deposition for surface passivation, - photolithography of a resin mask,

- anisotropic etching of the oxide,

- anisotropic etching of the silicon until it stops on the buried insulating layer,

- elimination of the resin, - thermal oxidation of the sides of the etched silicon.

FIG. 32 takes up the configuration of FIG. 17, with elements similar to those of this figure which are designated by reference signs which are deduced therefrom by adding the "prime" index.

According to FIG. 33, a photolithography of a resin mask 950 is carried out on the rear face, then implantation on the rear face, through slots 951, so as to form doped zones of a first type DD1.

According to FIG. 34, the resin 950 is eliminated and a photolithography of another mask of resin 952 is carried out, so as to cover the areas DD1, then implantation, through wide slots 953, so as to form doped areas of another type DD2.

According to FIG. 35, there is elimination of the resin 952 and deposition of a layer 954 of SiO2 on the rear face, for example by PECVD.

According to FIG. 36, there is photolithography of a resin mask and etching of the upper oxide and elimination of the resin (cf. FIG. 18). According to FIGS. 37 and 38, the operations of FIGS. 19 and 20 are carried out. According to FIG. 39 (analogous to FIG. 21) an activation and diffusion annealing of the dopants is carried out, both in the upper part and in the lower part.

According to FIG. 40, a photolithography step of a resin mask is carried out on the rear face, an etching of the oxide on the rear face then an elimination of the resin, which reveals, in the layer 954, slots 955 opposite each DD1 or DD2 doped zone.

According to FIG. 41, a slight partial deoxidation is carried out, chemically, of the surfaces for the future electrical contacts, and deposition of a thick layer 956 and 957 on the upper and lower faces, respectively.

In FIG. 42, there is photolithography of a mask 958 on the rear face and etching of the layer 957 through this mask.

According to FIG. 43, there is photolithography of a thick resin on the layer 956, with sloping edges, then anisotropic etching of this thick layer.

After elimination of the resins and chemical finishing of the engravings on both sides, we obtain the configuration of Figure 44, which corresponds to Figure 11. As regards the type 21 variants, they add steps in the first steps of the without deeply modifying them. The constraint on the choice of the additional stages comes from the "hot" stages which will follow. The added materials must withstand the annealing temperatures of the following steps. Various solutions exist, in particular with Si3N4, to be chosen from all the possible options depending on the design of the concentration system. Certain concentration systems can impose the presence of a lens stuck on the cell or on the MPI. This is advantageous because the medium above the MPI has an index close to silica. We can therefore add as many layers of silica as necessary to passivate and protect these layers or structures from the rest of the process.

The realization of variants 3 /, AI, 6 / and 7 / requires above all modifications of the photolithography masks but does not add any step. Optimization can of course make the process more complex, but the structure obtained remains that described.

Claims

CLAIMS 1. Photovoltaic module, intended to cooperate with a light concentration system, comprising

- a substrate (11, 11 ', 111, 211, 311, 411, 711), - a layer of semiconductor material (12, 12', 112, 212, 312,

412, 712) separated from the substrate by an electrically insulating layer (13, 13 ', 113, 213, 313, 413, 713), a plurality of elongated strips (15, 15', 115, 215, 315, 415, 515, 615, 715, 815) each having an entry face adapted to receive light being provided in this layer of semiconductor material and being each bordered by electrically insulating walls (14, 14 ', 114, 214, 314, 414, 714) extending to the electrically insulating layer, each strip comprising doped zones (D1, D2, D1 ', D2'), intermediate metal tracks (17, 17 ', 117, 217, 317, 417 , 517, 617, 717, 817) electrically connecting in series the elongated strips of semiconductor material and arranged parallel to the electrically insulating walls, and

- extreme metal tracks (18, 18 ', 118, 218, 318, 418, 518, 618, 718, 818) constituting electrical output terminals connected to the extreme strips of the plurality of strips of semiconductor material. 2. Module according to claim 1, characterized in that said intermediate metal tracks are arranged along the entry face of the elongated strips, each covering one of the electrically insulating walls bordering the elongated strips of semiconductor material and by electrically connecting doped zones of different types within the elongated strips separated by this wall.

3. Module according to claim 1 or claim 2, characterized in that it further comprises secondary metal tracks (421) arranged at a distance from the electrically insulating walls, connected to one of the intermediate or extreme metal tracks and to doped areas of the same type within the underlying band.

4. Module according to claim 3, characterized in that each of the secondary metal tracks is arranged between two metal tracks secondary or between an intermediate or extreme track, and another secondary metal track connected to αopeope zonesee rauire type within αe said underlying strip.

5. Module according to any one of claims 2 to 4, characterized in that each of the secondary tracks is arranged between two metal tracks connected to each other and from which this secondary track is isolated.

6. Module according to any one of claims 1 to 5, characterized in that at least some of the metal tracks (17, 17 ', 117, 217, 317, 417, 517, 617, 717, 817) are covered by a micro light concentrator. ^"• ',

7. Module according to claim 6, characterized in that said micro-concentrator has a triangular prism section.

8. Module according to claim 6 or claim 7, characterized in that each micro-concentrator is an integral part of the metal track which it covers.

9. Module according to any one of claims 6 to 8, characterized in that the micro-concentrators cooperate with microlenses of cylindrical type arranged on the free face of the elongated strips.

10. Module according to any one of claims 1 to 9, characterized in that the entry face of each elongated strip is covered with a transparent and electrically insulating layer (19, 19 ', 119A to 119C, 219, 319 , 419, 719) having orifices allowing contact between the semiconductor material and the metal tracks.

11. Module according to claim 10, characterized in that this transparent layer behaves like a passivation layer (119A) in contact with the semiconductor material.

12. Module according to claim ^" 10 or claim 11, characterized in that the transparent layer is formed of a stack of sublayers (119A to 19C).

13. Module according to any one of claims 10 to 12, characterized in that the transparent layer comprises silica.

14. Module according to any one of claims 10 to 13, characterized in that this layer is at least partly formed of a material chosen from the group comprising S-3N4, SiOxNy, MgF2.

15. Module according to any one of claims 10 to 14, characterized in that the entry face covered by the transparent layer

(219) is textured.

16. Module according to any one of claims 10 to 15, characterized in that the entry face covered by the transparent layer (319) is shaped into a plurality of adjacent pyramids.

17. Module according to any one of claims 1 to 16, characterized in that the elongated strips have doped zones (20 ') along each of the walls which border them laterally.

18. Module according to any one of claims 1 to 17, characterized in that the elongated strips comprise doped zones (20 ') along the electrically insulating layer.

19. Module according to any one of claims 1 to 18, characterized in that the elongated strips comprise doped zones (20 ') along the light entry face.

20. Module according to any one of claims 1 to 19, characterized in that the elongated strips (15, 15 ', 115, 215, 315, 415,

715, 815) have a constant section over their entire length.

21. Module according to claim 20, characterized in that the elongated strips (515A) have a width which is minimum in a central zone and maximum near the extreme metal tracks.

22. Module according to any one of claims 1 to 19, characterized in that the elongated strips (515) have a width which is maximum halfway between their ends, and minimum at these ends.

23. Module according to any one of claims 1 to 19, characterized in that the elongated strips (515 ', 515 ") have a width which is minimum halfway between their ends, and which is continuously variable.

24. Module according to claim 22 or claim 23, characterized in that the walls (517, 517 ') delimiting these ^" elongated strips are in broken line.

25. Module according to claim 22 or claim 23, characterized in that the walls (517 ") delimiting these elongated strips are curved.

26. Module according to any one of claims 1 to 20, characterized in that the elongated strips (615, 615 ') have the form of loops.

27. Module according to claim 26, characterized in that the elongated bands form concentric rings.

28. Module according to claim 27, characterized in that secondary tracks (620 ') are connected to a conductive track along an elongated annular strip.

29. Module according to claim 28, characterized in that the secondary tracks are substantially radial from this track.

30. Module according to any one of claims 1 to 29, characterized in that it has a shape of regular polygon (510).

31. Module according to claim 30, characterized in that it has a square shape.

32. Module according to any one of claims 1 to 29, characterized in that it has a circular or oval shape.

33. Module according to any one of claims 1 to 32, characterized in that at least one light conversion cell (711) is further provided in the substrate, under the electrically insulating layer, for a range of lengths d 'wave different from that for which the elongated strips are adapted to convert light.

34. Module according to claim 33, characterized in that this module is adapted to receive light from any of its faces.

35. Module according to any one of claims 1 to 34, characterized in that electronic components are produced in the layer of semiconductor material (840, 860).

36. Module according to claim 35, characterized in that these electronic components are electrically isolated from the elongated strips.

37. Module according to claim 35 or claim 36, characterized in that at least one of the components is a protection diode.

38. Module according to any one of claims 35 to 37, characterized in that at least one of the components is a switch.

39. Module according to any one of claims 35 to 38, characterized in that at least one of the components is an inverter.

40. Module according to any one of claims 1 to 39, characterized in that the semiconductor material of the elongated strips is chosen from the group formed of silicon, germanium or Si _x Ge-ι- _x .

41. Method for manufacturing a photovoltaic module intended to cooperate with a concentration system, according to which,

A plate (900) is produced comprising a substrate and a layer of semiconductor material separated from the substrate by an electrically insulating buried layer (901),

• slots in this layer of semiconductor material are etched extending to the buried layer so as to delimit elongated strips, and the sides of these strips are coated with an electrically insulating layer (904),

• these slots are filled so as to produce walls (905) separating the elongated strips, • doped zones (D1, D2) are produced in each of these elongated strips, at least near each wall,

• metallic tracks (912) covering this wall are produced on each wall by connecting doped zones located on either side of this wall, • extreme tracks are formed on the extreme walls forming output terminals.