FR2464564A1 - AMORPHOUS SILICON SOLAR BATTERY - Google Patents

Abstract

SOLAR BATTERY CHARACTERIZED IN THAT IT INCLUDES: A PLURALITY OF STRIPS 34 OF TRANSPARENT CONDUCTIVE OXIDE ON THE SECOND OPPOSITE MAIN SURFACE; A PLURALITY OF HYDROGEN AMORPHOUS SILICON SOLAR CELLS 20, 21, 22 AT TANDEM JUNCTION, MADE ABOVE A MAIN PART OF EACH OF THESE STRIPS OF TRANSPARENT CONDUCTIVE OXIDE AND IN ELECTRICAL CONTACT WITH THESE BANDS, SOLAR CELLS HYDROGEN 38, 42 AMORPHOUS SILICON SEMICONDUCTORS HAVING DIFFERENT CONDUCTIVITY TYPE REGIONS, THOSE LAYERS BEING SEPARATED BY A TUNNEL 40 JUNCTION, SAID SOLAR CELLS HAVING A WIDTH SUCH AS LOST THICKENED THICKEN THE LOSS OF ENERGY RESULTING FROM THE INCORPORATION OF A METAL GRID ELECTRODE IN THE TRANSPARENT CONDUCTIVE OXIDE AND MEANS FOR SERIAL CONNECTION OF SAID CELLS.

Description

The present invention relates to solar silicon batteries

amorphous. It relates more particularly to a solar battery comprising a series of strips of two or more layers of hydrogenated amorphous silicon arranged in a tandem battery configuration, in which the strips of hydrogenated amorphous silicon are connected in series. It is known that photovoltaic devices, such as in particular solar cells, are capable of transforming solar radiation into usable electrical energy. The transformation into electrical energy results from the photovoltaic effect well known in the field of solar cells. Solar radiation striking a solar cell and absorbed by an active region generates electrons and holes

or gaps. The electrons and the holes are separated by an electric field

  internal stick, for example a straightening junction in the cell

solar.

  A rectifier junction can be generated in a solar cell by an active semiconductor layer having regions

p-type, intrinsic, and n-type hydrogenated amorphous silicon.

  The electrons generated in the intrinsic region by absorption of solar radiation from the appropriate band gap produce electron-hole pairs. The separation of electron-hole pairs of electrons flowing to the n-type conductivity region and holes flowing to the p-type conductivity region produces the phototension and photocurrent of the cell. The total performance of the solar cell is maximized by increasing the total number of photons of different energy and wavelength, which are absorbed by the

semiconductor material.

American patent application No. 031 460 filed on April 19, 1979 in the name of Joseph John HANAK describes a tandem junction structure for an improved amorphous silicon solar cell. The structure described in this application comprises a series of at least

two layers of hydrogenated amorphous silicon arranged in a confi

  Battery treatment in tandem with an optical path, a tunnel junction ensuring electrical interconnection. The layers of hydrogenated amorphous silicon have regions of different conductivity

  in order to achieve an internal electric field in each semi-layer

  conductive. The layers can have the same inter-

  say, or according to a preferred embodiment of the different forbidden bands, in order to more completely absorb the distribution of the photons of the different energies of the solar spectrum. Therefore a solar cell structure according to the cited US patent application

above shows increased performance by absorption of a por-

  more important of the solar spectrum. However, a grid electrode is required in solar cells with a large area in order to collect the photocurrent. The gate electrode can protect the active region from some of the available solar radiation

  up to around 10%. Furthermore, given that the dimen-

  sions of the solar cell and the current of this cell increases, the complexity of the grid electrode also increases which constitutes a practical limitation to the dimensions of a large surface solar cell. It is therefore therefore highly desirable to produce a structure which maximizes the absorption of solar radiation of different energy and wavelength without the screen effect and the

size limitations of the gate electrode.

  An amorphous silicon solar battery according to the invention comprises a plurality of bands of tandem junction solar cell connected in series, said tandem junction solar cells comprising a plurality of layers of hydrogenated amorphous silicon separated by a

  tunnel junction, arranged in a tandem stack configuration.

  The thickness of the layers of amorphous silicon is adjusted so as to

  obtain maximum efficiency and equalize the current in each layer.

  There is no basic limit to the length of the solar cell strips. The width of the strips is chosen so that a gate electrode is not necessary to collect the current generated by

the solar battery.

The band gap or band gap of the layers of hydrogenated amorphous silicon in the bands of the solar cell - at tandem junction can vary in a range from 1.5 eV to about -1, eV by adjusting the concentration of hydrogen in the layers of hydrogenated amorphous silicon.

Other characteristics and advantages of this invention are

  will derive from the description given below with reference to the single figure of

  attached drawing which is a sectional view of a high-voltage solar battery

tension according to this invention.

  Referring to this figure, there are designated by the reference 10

  a high voltage solar battery, connected in series, with tandem junction.

  The solar radiation 100 striking on the surface of the solar battery 10 constitutes a reference point for the incident surface of each layer or region of the solar battery. The solar battery 10 comprises a transparent substrate 32 of materials such as ordinary window glass or a borosilicate glass. A plurality of strips 34 of a transparent conductive oxide, such as indium tin oxide, or other similar materials, are formed on the substrate 32. The strips 34 constitute the upper electrode for a plurality of cells tandem junction hydrogenated amorphous silicon solar panels, connected in series 20, 21, 22 etc. Indium oxide bands and

of tin should be as thin as possible to ensure a trans-

  maximum mission of solar radiation. However, the thickness must be adjusted to obtain a layer resistivity of the order of 100 μl per square or less. Preferably the layer resistivity is of the order of 10.2 per square. The thickness of the indium oxide layer and

  tin can be chosen by taking advantage of its anti-reflective properties

  boring. Since the solar cells 20 to 22 are equivalent, only the solar cell 20 will be described in detail, the solar cells 21 and 22 comprising the same elements provided, in the drawing, with the same

references.

  The solar cell 20 comprises a layer of cermet 36 in electrical contact with the layer of indium and zinc oxide 34. This layer of cermet 36 is made from materials such as PtSiO2 containing of the order of 7 to 15 % by volume of platinum and having a thickness ranging from approximately 2 to 10 nanometers. Alternatively, the cermet layer 36 can be made from a dielectric material such as TiO2 plus a metal with high extraction power, as described

in U.S. Patent No. 4,167,015.

  An active layer 38 of hydrogenated amorphous silicon, incident to solar radiation, is deposited on the cermet layer 36 and is in electrical contact with this layer 36. The active layer 38 comprises regions 38a, 38b and 38c of different conductivity types. A first region 38a is formed by a layer of hydrogenated amorphous silicon doped with p-type conductivity modifiers. such

than boron for example or any other suitable p-type doping agent.

  Said region has a thickness of the order of 10 to 40 nanometers and preferably about 37.5 nanometers. The region 38b of intrinsic hydrogenated amorphous silicon, having a thickness of the order of 30

  at 300 nanometers, is deposited on the region 38 a. We found that if-

intrinsic or undoped hydrogenated amorphous silicon exhibited a conduc-

  slightly n-type activity, as mentioned in US Pat. No. 4,064,521. A region 38c of n + type hydrogenated amorphous silicon, having a thickness of about 10 to 40 nanometers approximately, is deposited on region 38b, of so as to be contiguous to this region. The

  preferred thicknesses indicated above of the silicon layers

  phe of type p + and n + respectively relate to real materials

read with dopant concentrations of 0.1% B2H6 in SiH4 and 0.2% PH3 in SiH4 respectively. Concentration

  doping gas can affect the optimal thickness of these layers.

A second active layer is designated by 42a, 42b and 42c. This

  second active layer 42 comprises regions 42a, 42b and 42c of sili-

  Amorphous hydrogenated cium of a different type of conductivity. Region 42a is similar to region 38a and incorporates a suitable p-type doping agent. The region. 42 b is similar to region 38b and region 4Zc is similar to region 38c. Regions 42a, 42b and 42c can be deposited at high temperature to produce a layer

  with a lower hydrogen concentration and an inter-

  said to be weaker than the active layer 38. The thickness of the second active layer 42 must be adjusted so that the current produced by said layer is substantially equal to the current produced by the first active layer 38 since the total current of the solar cell 20 will be limited to the lower current either of the active layer 38 or of the layer

active 42.

The tandem junction solar cell 20 is not limited to two active layers. The solar cell can contain a plurality of active layers. Each pair of active layers is separated by junctions

tunnel or through a cell interconnect layer. Preferably the cellu-

  solar 20 has 2 to 5 active layers, each of these layers

active being separated by a tunnel junction or by an inter-

cell connection which acts as a tunnel junction.

  A cell interconnect layer 40 is located between the active semiconductor layers 38 and 42. The cell interconnect layer 40 provides a single electrical path through the first

active layer 38 and from the second active layer 42 towards the post-contact

  laughing 44. The cell interconnection layer 40 also allows the transmission of solar radiation which is not absorbed by the region

active 38, towards the second active region 42 or additional active regions

where additional absorption may occur.

  The interconnection layer 40 has a thickness ranging from approximately 2 to 15 nanometers and it consists of a PtSiO2 cermet or a thin metallic layer and a PtSiO2 cermet, or a thin metallic layer. This metallic layer can consist of a metal such as platinum, titanium, nickel and other materials

  transparent to solar radiation. If a metallic layer is used,

  than thin without the cermet, it is preferable to use a metal with high extraction power such as platinum. The performance of a tandem junction 20 solar cell is degraded if the interconnection layer

  is an insulator, despite the fact that this insulator could be sufficient

thin enough to allow electrons to pass through it by tunnel effect.

  Cell interconnection layer functions as a junction

  tunnel between active regions 38 and 42.

  Interconnection layer 40 can be omitted if region 38c and region 40a have sufficient p-type modifiers and

  n-type, respectively, to form a tunnel junction between them.

The efficiency of the transformation of light into electricity of a hydrogenated amorphous silicon solar cell according to the structure described above approaches a constant when the intrinsic region has a thickness of the order of 500 nanometers. In a tandem junction structure, any additional thickness of this region only contributes to increasing the absorption of solar radiation without increasing the performance of the cell and absorbs all the following layers of solar radiation. Consequently, the thickness of each intrinsic region must be thinner and thinner as and when

  increasing the number of layers of hydrogenated amorphous silicon empi-

lées. In addition, the thickness of each intrinsic region following the incident intrinsic region must be thicker than the previous region. The rear contact 44 which can be made of titanium, molybdenum, niobium and similar materials, which adheres to the region 42c by forming a good ohmic contact therewith, is deposited on this region. A solar cell interconnection layer 46, made of indium, tin or other materials, is arranged in ohmic contact with the posterior contact 44, this layer 46 interconnecting the solar cell 20 with the base substrate 34 of the solar cell 21. Conductors 52 and 54 are placed in contact with layers 34 and 36 respectively to extract the current generated during the illumination of the solar battery

by solar radiation 100.

Since the solar cells 20, 21 and 22 are connected in series, the current remains constant and the voltage of each cell is cumulative. The sum of the voltages makes it possible to manufacture a solar cell of any voltage desired for a specific application. The solar cell interconnection layer 46 can be small enough to constitute approximately 0.5% of the total surface of the device. Although the layer 46 made of monocrystalline materials could create problems by possibly short-circuiting the solar cell structure 20, this is not a problem with amorphous silicon since the lateral conductivity of the doped layers is so low - only for practical purposes, it does not exist, that is to say that the lateral conductivity of the amorphous semiconductor layers is similar to that of an insulator having a lateral layer resistivity greater than approximately

1010 n per square.

When manufacturing a solar battery 10, the maximum width

of the upper electrode and of the solar cells 20, 21 and 22 is deter-

  undermined by the layer resistivity of the conductive and transparent oxide layer, by the short-circuit current J of the individual stacked solar cell strips and by a factor which depends on the loss

of power or energy from each cell which is acceptable without elect

screen grid. A gate electrode is required when the power loss factor is greater than about 0.05 or about 5%. The width can be increased to a maximum, but this width must however be less than a width which would make

  gate electrode for extracting the current generated during operation

operation of the solar battery.

More specifically, the width is determined by the following formula W w = \ 3f V Oc F Ri JSN RI sc in which: Voc is the total open circuit voltage of the solar battery; Rii is the layer resistivity of the incident electrode; F is the load factor; Jsc is the short circuit current density; N is the number of bands of the solar battery; and

f is the factor relating to the percentage of power loss in the electricity

  front trode resulting from resistive heating.

  The factor f is generally on the order of 0.01 to 0.08 and preferably about 0.05.

In order to determine the width, it is assumed that only the elect

  front electrode, in the figure the electrode 34, limits the current since the posterior electrode 44 can be thick enough not to

  not take into consideration the layer resistivity.

As an example, we indicate that for Jsc = 3 mA / cm2, N = 9, Rice = 100 Q / square, Voc = 12, 5, F = 0, 6 and f = 0.05, the width of

the cell is equal to 0.65 cm.

For an interval width of 0.005 cm, which can be easily

obtain by using existing photolithographic techniques

  aunts, the total area of the solar battery that is not used

  represents about 0.7% of the total surface of the solar battery.

  This value is an order of magnitude higher than that of a structure

  ture of solar cell requiring a grid electrode to collect the photocurrent. As is clear from the formula indicated above, when the resistance of the front electrode decreases, the width of the bands can increase without affecting performance

  of the solar battery. If the width is kept constant, the perforations

mances of the solar battery increase.

The solar battery can be produced by various methods. The substrate is coated with a transparent conductive oxide layer, by all known methods and in particular by evaporation, sputtering or pyrolysis of inorganic or organometallic compounds. A transparent conductive oxide glass coated with an indium tin oxide can also be obtained commercially in the prefabricated state, in particular from the English firm Triplex Glass Co., Ltd, Kings Norten, Birmingham, England. The transparent conductive oxide is covered with a positive photoresist such as "Shipley 1350 H '. The resist is subjected to centrifugation, or it is dried and then exposed to a

light source through a photomask in order to define the rays

stripes between the bands. The device is placed in a device designed to align the mask which keeps this mask fixed and which allows the device to move in the x, y and z directions, and also

to rotate around the geometric axis perpendicular to the plane of the

  sample. Such a device is known. The network is developed in a suitable development device. The grooves are made by chemical attack of the transparent conductive oxide using a chemical attack agent consisting for example of 55 to 58% of HI at 350C.

for tin and indium oxide.

  Then, the cermet layer is made according to the lessons drawn from the American patent n ° 4,167,015 cited above. The transparent conductive oxide layer and the cermet layer can be chosen from

so as to form a quarter-wave anti-reflective coating, for example

  pl we can choose a thickness of 60 nanometers for the conductive oxide

transparent and 10 nanometer for tin and indium oxide.

  The semiconductor layers of hydrogenated amorphous silicon 38 and 42 are deposited using a luminescent discharge of silane.

or other suitable atmosphere containing silicon and hy-

  drogeneous as indicated in the aforementioned US patent No. 4,064,521 as well as in the filed US patent application No. 727,659

  September 29, 1976. Layers 38 and 42 can also be manufactured

  quées by a high frequency deposition system in which-the electrodes or coils are contained inside the deposition chamber. The

parameters suitable for high frequency discharge are a powerful

this high frequency equal to or less than 0.5 W / cm2 on a target having a

surface of the order of 160 cm, a gas pressure of the order of 20 milli-

  torr at around 50 millitorr, a silane flow rate of the order of 30 cm / mi

  nute and a temperature range of between 200 and 3500C approximately.

The p-type region of layer 38 or layer 42 is fabricated with

  an appropriate concentration of p-type dopant, boron or other do-

  suitable pants, in a quantity of the order of 0.01 to 1% relative

by volume of silane. The n-type region is made with a concentration

  tration of n-type dopants such as PH3 of the order of 0.1 to 1% of the deposition atmosphere. After performing the deposition of the active regions, the posterior electrode 44 is deposited by evaporation or spraying

high frequency cathode or by any other suitable method.

  After having carried out the application of the posterior electrode 44, the solar battery is coated with a positive or negative photoresist. Individual cells are defined by exposing the surface of the photoresist through an appropriate photomask, i.e. a positive mask

or negative and developing the resist by any known method. The rai-

Nures are etched by chemical attack in the posterior electrode with an appropriate chemical etching agent consisting for example of one part of HF, two parts of HNO3 and 7 parts of water for a titanium electrode. The photoresist layer is detached and a new photoresist layer is applied which is exposed to light

  and develops according to the process indicated above.

  The active layers and the cermet are etched by chemical attack to the transparent conductive dioxide layer, using a plasma etching agent in an atmosphere of oxygen and 4% CF4. The plasma quickly attacks and scours the active regions, but it only attacks the cermet slowly, at a speed of the order of 10 nm / minute. If the cermet film has a thickness greater than 10 nm, the cermet layer can be etched and etched by using a sputtering etching process

at high frequency in an atmosphere of CF4-02 or Ar-CF4-02.

The end point of attack or pickling is generally determined by the appearance of clear transparent grooves down the side of the

  transparent conductive oxide layer.

Finally, the photoresist is removed and the surface of the device is removed.

  plasma to remove all traces of organic molecules before depositing the interconnection layer in series. This interconnection layer is obtained by carrying out an evaporation between the

  grooves at an angle of about 45 relative to the surface, perpen-

  - specifically to the grooves and from the direction which interconnects the posterior electrode of the solar cell 20 towards the oxide layer 11

  transparent and conductive of the solar cell 21. The intercon-

serial connection can also be applied by evaporation of the layer

  over the entire surface of the solar battery. Then we remove the material

laugh. in excess and the grooves are formed by implementing the photolithography techniques described previously. After detaching the photoresist layer from the attachment of the conductors 52 and 54 by any known method, the rear part of the solar battery can be wrapped by an appropriate material such as "Apiezon W", a product marketed by the firm American James G. Biddle Co.,

Plymouth Meeting, Pa., E.U.A.

  The example below illustrates the invention. Of course, this example does not have any limiting nature, any modification and / or variant that can be envisaged by those skilled in the art without departing from the scope.

of the invention.

  Example A soda-lime glass substrate of 7.6 x 7.6 cm and having a thickness of the order of 0.16 cm, provided with a coating of tin and indium oxide with a resistance of layer of the order of 10 A per square, has been coated with a positive photoresist, consisting for example of Shipley 1350-H, produced by the American company Shipley Co., Inc.,

  of Newton, Mass., using a centric deposition process

  fugation at a speed of 4000 revolutions / minute, this photoresist then being

  dried for 1 hour at a temperature of about 750C. The photo-

resist was exposed in a Shipley developer and the soluble portion of photoresist defining the area between the bands mentioned above was removed. Then, the exposed substrate was immersed in a 55-58% hydroiodic solution at 350C to remove the exposed tin and indium oxide, leaving behind strips coated with a photoresist of tin oxide and d indium having a length of 7.6 cm and a width of 0.68 cm. The remaining photoresist was cleaned of the bands of indium tin oxide and a cermet of PtSiO2, having a metal content of the order of 12% by volume of platinum, was deposited on the substrate, by sputtering. at high frequency, until a

  thickness of the order of 23.5 nanometers. Then a semi-layer

active conductor of hydrogenated amorphous silicon, having a p + type region having a thickness of approximately 31.8 nanometers, an undoped region having a thickness of the order of 181.6 nanometers and an n + type region having a thickness d 'about 90.8 nanometers was deposited on the PtSiO cermet The hydrogenated amorphous silicon layer was formed by a high frequency capacitive luminescent discharge of silane with respect to the undoped layer and diborane gas according to a concentration of about 0.1% by volume

  which concerns the p + type region and gaseous PH3 according to a concentration

of about 0.2% by volume with regard to the region of

  type n., these concentrations being expressed relative to the silane.

A second layer of PtSiO2 cermet having a thick-

  About 10.5 nanometers, with a platinum concentration of about 12% by volume, was deposited on the first active region by high frequency sputtering. Then a second active semiconductor layer is deposited on the second cermet layer. The second semiconductor cermet layer was fabricated according to the process outlined above for the first layer. It presented a p + type region having a thickness of the order of 31.8 nanometers, an intrinsic region of 363, approximately 2 nanometers thick, adjacent to said p + type region and an n + type region.

approximately 90.8 nanometers thick contiguous to said intrinsic region.

A posterior titanium electrode having a thickness of about 200 nanometers was deposited by sputtering on the n + type region of the second active semiconductor layer. Then the device

was covered with positive photoresists and exposed through a photo-

mask to create a configuration similar to that of the bands produced in the layer of tin and indium oxide but offset from the

grooves in the said indium tin oxide layer so as to

  put the interconnection in series of the solar cell strips at junction

  tandem. The positive photoresist is the Shipley 1350 H, produced by the company

  American tee Shipley Cy. The exposed photoresist is developed and the soluble portion defining the grooves is washed out using a solvent. The structure is then etched by chemical attack to

using a solution containing one part of hydrofluoric acid, 2 parts of HNO3 and 7 parts of water in order to remove the portion of the electrode

titanium posterior to form the grooves in the device. After titanium etching, the photoresist is detached with acetone and another photoresist coating is applied which is developed in the same way as specified above. The device was placed in a plasma etching machine and the active layers located below the titanium, which have already been removed, were etched by chemical attack in a CF4 atmosphere containing 4% oxygen, that is to say say Freon DE-l00, marketed by the American company Scientific Gas Products, Inc., South Plainfield, NJ, USA The plasma pickling machine was an IPC-200 machine, produced by the American company International Plasma Corporation, Hayward, California, E.U.A. The conditions used during the etching by plasma attack were as follows: high frequency power of about 800 watts pressure CF 42 between 0.5 and 1 torr starting temperature of about 250C, and

  final temperature less than or equal to around 90 ° C.

These parameters lead to a pickling rate of amorphous silicon of the order of 200 nanometers per minute. The pickling continues through the second active layer and the cermet interconnection tunnel junction, this layer being attacked and pickled at a rate of only 10 nanometers per minute up to the layer of tin and indium oxide . The end point of pickling is determined optically since the substrate with tin and indium oxide appears transparent

as soon as the hydrogenated amorphous silicon has been removed.

At this stage of the manufacturing process, the device was checked for pinholes using light projected through this device. The pinholes were covered with "Microstop", which is a product marketed by the American company

  Michigan Chrome and Chemical Cy, Detroit, Michigan.

  The indium series interconnection layer was isotropically evaporated on the titanium layer until a thickness of about 200 nanometers was obtained. Next, the device was covered with a positive photoresist, Shipley 1350-H, and exposed through a photo mask aligned to place the third set of grooves, adjacent to the second set of grooves. The excess indium was then

removed by pickling with a chemical attack solution

  one part of concentrated hydrochloric acid, one part of H 202 at 30% and 6 parts of water by volume. The remaining photoresist and the "Microstop" were removed by washing with acetone, washing with water, washing with deionized water and finally the device was dried in an oven for a period of 1 hour. '' from 30 minutes to 1 hour at

a temperature of 1000C.

Short circuits and shunts were eliminated from the device according to the process described in US Patent No. 4,166,918. The contact conductors of a flexible copper cable were attached to the

  terminal electrodes by an epoxy-silver adhesive.

  The solar battery was illuminated by a light presenting

intensity of a sun, using a halogen projection lamp-

tungsten, such as the Sylvania ELH lamp, 300 watts at 120 volts.

  The solar battery having a total open circuit voltage (VOC) of approximately 12.6 volts, a charge factor (F) of the order of 0.56 and a short-circuit current (Jsc) of approximately 1 , 82 mA / cm with a total yield of approximately 1.42%. By using a lamp having an intensity of two suns for the illumination, the total efficiency of the cell went from 1.42% to approximately 1.45%, the load factor (F) having increased

from 0.56 to 0.57.

  It remains to be understood that this invention is not limited to the embodiment described and shown, but that it encompasses

all variants.

24 & 4564

Claims (13)

  1. Amorphous silicon solar battery comprising a substrate   transparent having a first main surface incident to the ray-   solar and a second opposite main surface, characterized in that it comprises: a plurality of strips (34) of transparent conductive oxide on said second opposite main surface; a plurality of hydrogenated amorphous silicon solar cells (20, 21, 22) with rods tandem, made over a main part of each of said bands of transparent conductive oxide and in electrical contact with the bands, said solar cells having a plurality of semiconductor layers of hydrogenated amorphous silicon (38, 42) comprising regions of different conductivity type, these layers being separated by a tunnel junction (40), said solar cells having a width such that the energy loss due to the thickness is less than the energy loss resulting from the incorporation of a metal grid electrode into the transparent conductive oxide and means for connecting   in series the so-called solar cells. 2. Solar battery according to claim 1, characterized in that said layers of hydrogenated amorphous silicon (38-42) have a region (38a-42a) of p-type conductivity which constitutes the incident region of said layer, a region of intrinsic amorphous silicon (38b-42b) contiguous to the p-type region and a region of conductivity type n + (38c-42c) adjoins the said intrinsic region. 3. Solar battery according to one of claims 1 or 2, charac-   terized in that said tunnel junction (40) is formed between the n + type region of an incident layer of amorphous silicon and the p-type region of the next layer of amorphous silicon.   4. Solar battery according to any one of the pre- claims   cédentes, characterized in that the said tunnel junction is a dry layer   adorned with a material chosen from the group comprising the cermet of PtSiO2, nickel, molybdenum or titanium. 5. Solar battery according to claim 4, characterized in that said layer in tunnel junction consists of cermet of PtSiO2. 6. Solar battery according to claim 5, characterized in that said tunnel junction layer comprises a transparent metallic layer disposed between said cermet and said n + type region of the semiconductor layer. 7. Solar battery according to claim 1, characterized in that the means for interconnecting said solar cells in series   are made in the form of a metal layer. 8. Solar battery according to claim 1, characterized in that the forbidden band energy of said semiconductor layers increases from the incident semiconductor layer to the semiconductor layers following drivers.   9. Solar battery according to any one of claims previous, characterized in that the junction solar cells tandem (20, 21, 22) have 2 to 5 semiconductor layers.   10. Solar battery according to claim 9, characterized in what the thickness of each intrinsic region of each semi-layer conductive increases from the incident semiconductor layer. 11. Solar battery according to any one of claims   previous, characterized in that the width of the tandem junction solar cell strip is determined by the following formula \ / oc W = J V   sc where V is the total open circuit voltage of the solar battery; oc J is the short-circuit current density; sc F is the load factor; RC is the layer resistivity of the transparent conductive oxide; N is the number of solar cell strips in the solar battery, and   f is a factor relating to the percentage of power loss in the electricity   frontal trode, this factor being chosen so as to be less than the loss of power or energy of a gate electrode resulting from   the screen effect of this electrode. 12. Solar battery according to claim 11, characterized   in that W is between about 0.2 cm and 5.0 cm.   13. Solar battery according to claim 12, characterized in that W is between approximately 0.2 cm and 2.0 cm.