DE102010026289B4 - Solar cell and process - Google Patents

Abstract

Solar cell (1) with a solar cell level (E) along which a side facing the sunlight (2) and a side facing away from the sunlight (3) run and along the solar cell level (E) in the solar cell level (E) a multiplicity of n-doped regions (n) and p -doped regions (p) are arranged alternately, the n- and p-doped regions (n, p) each having a doping of at least 1019 foreign atoms / cm 3 and successive p- and n-doped regions (p, n) each by an intrinsic region (i), which has a doping of less than 1015 foreign atoms / cm3, are separated from and spaced from one another, and the n, p and i regions (n, p, i) are facing away from the sunlight ( 3) extend to the side facing the sunlight (2) and the n, p and i areas (n, p, i) are arranged one on top of the other with the sequence: p area (p), i area (i), n -Area (s), conductive contact area (21), n-area (n), i-area (i), p-area (p), conductive Contact area (21), the width of each of the i-areas (i) being between 50 μm and 5000 μm and contacts (5) being applied to the conductive contact areas (21).

Description

The invention relates to a solar cell and method for producing the cell.

Semiconductor solar cells are well known in the art. In the US 2010/0052089 A1 For example, solar cells based on wafers are disclosed, which on the side facing the sunlight have a very narrow layer with a height of 0.1-1.0 μm, which is heavily n-doped. The strong n-doping means 10 ¹⁹ to 10 ²⁰ impurities per cm ³ . On the side facing away from the sunlight, a p-doped layer with a height of 100-300 μm is provided. The layer is p-doped, ie it comprises 10 ¹⁵ to 10 ¹⁶ foreign atoms per cm ³ . In addition, to increase the doping before the contact of the side facing away from the sunlight, an increase in the p-type doping may be provided, which acts as a barrier for electrons and as a low-resistance contact with holes.

From the US 2010/0052089 A1 Solar cells are also known, which have juxtaposed PIN structures, which are arranged transversely to the direction of sunshine. In this case, trenches introduced into a wafer are provided on one side with an n-doping, on another side thereof electrically insulated with a p-doping.

Solar cells usually consist of semiconductor materials in which the energy gap between valence band and conduction band is on the order of magnitude of the photon energy of the sunlight. By absorption of photons electron-hole pairs are generated, which can recombine again after a carrier lifetime. The principle of a solar cell is based on causing the generated free charge carriers before recombination through a composition of suitably doped semiconductor materials that the electrons preferentially move to n-contact and the holes to p-contact, without an externally applied voltage , The charge carriers can be tapped on the contacts deposited on the doped layers, and a current flow results. In general, electron-hole pairs are generated in the entire semiconductor material, and the charge carriers generated must reach the n- and p-contacts essentially by diffusion. However, a spatially directed separation of the electron-hole pairs takes place only in the transition region from the n-doped to the p-doped semiconductor material, the photoactive region.

A disadvantage of the known solar cells is that they are complex to produce small-scale individual solar cells of single-crystal high-purity silicon, which are used rather as concentrator cells or used for applications in which small dimensions have priority over production costs. In addition, these cells must be individually contacted to obtain a larger solar cell area. Also, there are still doped areas with increased recombination on the sun-facing side in this Salarzellenbauform.

From the DE 698 27 319 T2 For example, a solar cell with p-doped lenticular regions and n-doped lenticular regions is known, which are spaced apart from one another and are arranged exclusively on the side facing away from the sunlight.

In the WO 96/35235 A1 a solar cell with areas of horizontal layer sequence and areas with vertical layer sequence is disclosed. The regions of vertical layer sequence have n- and p-dated regions separated by metal contacts.

In the KR 100378343 B1 describes solar cells with n- and p-doped layers, which are arranged within grooves. They extend exclusively along the sunlight side facing away.

In the DE 40 39 390 C2 Solar cells are described with small n collector electrodes and small p collector electrodes.

From the US 4,305,776 A There are disclosed methods for producing ribbon-like silicon crystals.

In the US 3 422 527 A discloses a solar cell formed of stacked wafers each having a p ⁺ nn ⁺ structure. The overcoated wafers are cut. The wafers are connected in series to form the solar cell.

The object of the present invention is to provide a solar cell and a method for producing the solar cell, which overcome the above-mentioned disadvantages.

In its first aspect, the object is achieved by a solar cell having the features of claim 1.

The invention makes use of the knowledge that an arrangement of p-doped regions and n-doped regions in a wafer transversely, preferably perpendicular to the side facing the sunlight increases the efficiency of the solar cell. Advantageously, sunlight incident on the side facing the sunlight no longer strikes, as in conventional solar cells, an n-doped layer which must first be penetrated until it has penetrated reaches the photosensitive area. According to the invention, the n-doped regions and p-doped regions are provided alternately and are completely separated from each other by an intrinsic region.

The electric field between the space charges advantageously extends over the entire intrinsic region starting from the p-doped region to the n-doped region and allows a targeted drift movement of the charge carriers generated by the sunlight up to the preferably heavily n- and p-doped regions. A further advantage is the increased charge carrier lifetime in the intrinsic semiconductor material, since there are fewer recombination possibilities due to the dopants.

The invention is based on a semiconductor structure known as PIN photodiode, an electronic component for light detection, or as a particle detector in accelerator experiments. Typically, this doping structure is also used for amorphous silicon solar cells. If the thickness of the amorphous solar cell is sufficient, the charge carrier lifetime would be too short for a simple pn junction in order to be able to sufficiently collect the charge carriers generated by sunlight at the p and n contacts. The advantage of the invention is based on the fact that the intrinsic region provides a larger distance for the targeted drift movement. The gradient of Fermi energy extends over the entire intrinsic region.

Depending on the properties of the semiconductor material, the designs present in this invention must be dimensioned so that the advantages of the intrinsic region can be utilized, ie. H. The drift length of the charge carriers should be large enough to reach the respective n- and p-contacts.

Advantageously, the light incident on the solar cell now directly strikes the intrinsic region, which here is the photoactive region, and causes electron hole pair formation in the photoactive region. Due to the spacing of the n-doped and the p-doped regions, a positive space charge zone forms in the n-doped region toward the intrinsic region. Conversely, a negative space charge zone forms in the p-doped region towards the intrinsic region. Due to the forming space charge zones, an electric field develops in the intrinsic region and through the intrinsic region, which leads to a directed drift movement of the freely movable charge carriers. The free charge carriers generated by the sunlight are thus directed in addition to the diffusion movement directed into the respective doped region and can be tapped there outside via an electrical conductor in contact with the doped region in contact.

Conveniently, each n-doped region of two p-doped regions is adjacent transversely to the solar cell plane and each p-doped region of two n-doped regions adjacent to It forms an alternating sequence of n- and p-doped regions in the solar cell plane each separated by an intrinsic region. This produces a doping sequence in the solar cell plane: intrinsic region, n-doped region, intrinsic region, p-doped region, intrinsic region, n-doped region, etc. Several hundred or a thousand such regions can follow one another.

The intrinsic region is characterized by the same charge carrier concentration of electrons in the conduction band and holes in the valence band. An equal charge carrier concentration can, for. B. be prepared by compensating doping. The intrinsic region preferably has a doping of at most 10 ¹⁵ foreign atoms / cm ³ . The p- and n-doped regions are heavily doped and thus have 10 ¹⁹ or more impurities / cm ³ . Due to the low doping, preferably no doping of the intrinsic region, the intrinsic region for free charge carriers is particularly well transmissive, and a relatively strong, the drift movement promoting electric field can form through the intrinsic region.

In a preferred development of the invention, the intrinsic region extends along the entire sunlight-facing side of the solar cell. In this embodiment of the invention, the p-doped and n-doped regions do not extend to the sunlight-facing side of the solar cell. They each end a little way in the interior of the solar cell, spaced from the sunlight-facing side. As a result, the photosensitive area exposed to sunlight is particularly large.

In a further preferred embodiment of the invention, adjacent n-doped and p-doped regions are completely surrounded by the intrinsic region along a cross-section. In this embodiment of the invention, the n- and p-doped regions extend to the side of the solar cell facing away from the sunlight. In this embodiment of the invention, the sunlight can also be deflected by means of a mirror arrangement or the like to the side facing away from the sunlight, and thus the solar cell can be exposed to sunlight on both sides. It is thus produced a particularly high efficiency.

In order to form as homogeneous an electric field as possible between the n- and p-doped regions, which promotes the drift motion of the electrons, outer surfaces of adjacent n- and p-doped regions preferably extend substantially parallel to one another, whereby a kind of artillery light is generated by the two adjacent doped regions Plate capacitor forms and forms a homogeneous field between the outer surfaces of the n- and p-doped regions.

The solar cells according to the invention can be produced in different processes.

A possible manufacturing method for solving the problem set in terms of the method is achieved by the features of claim 7. In this case, a stratification with the repeating sequence sequence: p-range, i-range, n-range, conductive contact range, n-range, i range, p-range, conductive contact range is generated, wherein the conductive contact region can be introduced in various ways , Preferably, a multiplicity of wafers are each p-doped on one outer side and n-doped on an outer side opposite to it. The doped wafers are layered with equally doped outsides into a block, d. H. every second wafer is rotated 180 degrees. The wafer surfaces may be precoated with a metal layer, and the stacked contacting metal layers may be bonded with a metal contact adhesive. Then the block is cut transversely to the layering in solar cell high slices and used as an essential building block of the solar cells.

A further production method results from the features of claim 8, in which trenches are burned next to one another in a sunlight-side side of a wafer by means of a laser and then trench walls of a trench are either completely p-doped or completely n-doped. In this case, trenches with a p-doped wall are formed adjacent only by trenches with an n-doped wall in the solar cell plane, and, logically, trenches with an n-doped wall are spaced adjacent and contacted by trenches with a p-doped wall.

This production method is known in another application by the term "laser grooved buried contacts" and thus tested and thus executable.

In other manufacturing methods, p-type lines and n-type lines are first provided. P-type leads are electrical conductors that include doping elements or are encased by doping elements or conductors that are surrounded by p-type semiconductor material. N-type leads are electrical conductors that include doping elements or are encased by doping elements or conductors that are surrounded by n-doped semiconductor material. Preferably, intrinsic semiconductor material is deposited around the n and p lines in different processes.

First, a plurality of n and p lines are arranged side by side alternately, each of which is preferably formed completely either as a p-type line or as an n-type line. The p-type line comprises at least one p-type element of the third main group, preferably aluminum or boron, but also indium or gallium. Each of the n-type lines comprises an n-type element which may be a fifth main group element, preferably phosphorus or else arsenic or antimony. It is also possible to use nitrogen for doping, for example, silicon carbide. The p and n lines are then surrounded by a semiconductor material, and by heating the doping atoms diffuse into the surrounding material or it forms an alloy, if necessary, is then formed. Forming is understood to mean that the lines are optionally exposed to a brief impulse, which causes an integration of the doping atoms in the surrounding semiconductor material and thus a corresponding doping in the region of the semiconductor material surrounding the conduit.

The semiconductor material preferably grows epitaxially on the n and p lines. In epitaxial processes, semiconductor material, in particular silicon in the form of a process gas silane, in particular dichlorosilane and silicon tetrachloride, is brought to the n and p lines in gaseous form. The temperature of the gaseous semiconductor material is about 600 ° C-1200 ° C. This makes it possible to provide p and n lines with less heat resistance, i. H. from a melting point of 600 ° C to use. In a so-called CVP (Chemical Vapor Deposition) process, the process gases are passed over the heated substrates, and crystal layers grow on the lines. It is also conceivable to use process liquids, from which the semiconductor material lays around the p and n lines.

In another manufacturing process, high temperature resistant n and p lines are selected. The semiconductor material is placed around the n and p lines either as a powder, and the powder is then melted, or it is immediately poured around the lines as a liquid semiconductor material and then each cooled. The semiconductor material is cooled so that it crystallizes on cooling. This is also called alloying. Known and applicable processes are SSP (Silicon Sheeds from Powder), LLC (US Pat. Layered laser crystallization) and crystallization or recrystallization with the help of an electron beam.

The p-line should be current-conducting and high-temperature resistant. High-temperature resistant here means that the p-line should tolerate temperatures well above the melting point of the semiconductor material used. For silicon, the melting temperature is 1,410 ° C. Boron-coated tungsten wires, also known as boron fibers, can be used.

The n-line should be high temperature resistant and conductive. It may be in the n-lines to highly doped Siliziumcarbidfasern each with graphite core, z. B. Graphitfasergarne 10 microns diameter, which are coated with nitrogen-doped SiC layers of 0.5 microns act. Also, spinning of polycarbosilanes with subsequent pyrolysis to SiC yarn are possible for the production of SiC fiber bundles. It can also be phosphorus-containing graphite fibers or a nitrogen highly doped silicon carbide fiber.

After alloying, if this has not already been done by the preceding method step, the doping atoms are integrated into the surrounding crystal structure by means of a current surge in the lines. This is called forming.

The n and p lines can take different forms.

A development of the method makes use of rectangular in cross-section perpendicular to the longitudinal direction ladders. In this case, adjacent conductors are facing each other with their larger outer surfaces, arranged side by side in the mold, and then alloyed with the rectangular conductors and formed.

But it is also conceivable that the n- and p-lines are respectively formed needle-shaped. The ladder needles are conductively connected to each other in at least two groups. The two groups are electrically isolated from each other. Subsequently, the ladder needles, which form a kind of pincushion, are pressed onto an outside of a wafer and pressed into the wafer. Subsequently, the conductor needles are subjected to a rush current, whereby the n-doping and p-doping layers penetrate a little way into the intrinsic region of the wafer and either n-type or p-type the region around the needles.

In a particularly preferred embodiment of the production method according to the invention, a multiplicity of n and p lines arranged next to one another are drawn through liquid semiconductor material. In this case, semiconductor regions are formed between the n and p lines. Since there are a multitude of fibers available to the molten silicon to which it can attach by means of adhesion forces, the intrinsic regions between the n and p lines are easier to implement than in the industrially established EFG or ribbon string method, in which several fibers separated by a few centimeters, between which silicon regions are formed in the drawing. In the method according to the invention, the fibers are much less spaced apart depending on the material, preferably their spacing is between 50 μm and 5000 μm. Due to the small distance between the p and n lines from each other, a stable fiber composite is formed, which, compared to solar cells produced by EFG process results in a much higher stability. In addition, the layer thickness of the silicon regions can be pulled thin, because n- and p-type lines with a diameter of only 10 μm-100 μm can be used. The layer thickness of the silicon regions is below, preferably a layer thickness of 50-100 microns is selected.

It is also conceivable that the strained n and p lines are immersed in molten semiconductor material and pulled out freely floating again. The semiconductor material forms areas between the n and p lines in which it crystallizes.

Alternatively, the n and p lines are suspended by a process gas or a process fluid. The lines form a framework, and the semiconductor material forms, starting from the p and n lines, until the interstices of the lines are overgrown.

It is also conceivable to arrange the n- and p-lines staggered relative to one another so that they form a zig-zag structure or a honeycomb-shaped structure is formed in the case of multilayer p and n lines. Advantageously, the honeycomb interiors remain free of the semiconductor material. As a result, the structure is material-saving and yet stable. No carrier material is needed. Reflected or absorbed sunlight gets caught in the structure more easily.

Preferably, the n and p lines are completely in the intrinsic region. This solar cell can then be used particularly well on both sides.

The outer surfaces of the solar cell can increase the usual efficiency Surface treatment, such as passivation or an anti-reflection coating.

The invention will be described with reference to five embodiments in FIGS. Show

1 a schematic sectional view of a first embodiment of a solar cell according to the invention,

2a a schematic, perspective view of a layered block,

2 B a sectional view and schematic representation of the block in 2a .

2c a solar cell that comes from a block in 2a sawed off disc,

3 a wafer having diffusion-incorporated half-lobe doping regions,

4 a wafer with doping regions introduced by the laser grooved buried contact method,

5 a wafer with doping-coated needles doped regions,

6a a view of the sunlight side wafer side with p and n-doping regions 3 and 5 .

6b a plan view of the side facing away from the sun with trenches according to 4 .

6c a plan view of the side facing away from the sun with worm-shaped trenches in accordance with 4 .

6d a plan view of the side facing away from the sun with comb-shaped interlocking trenches in 4 .

7a a sectional view of completely enclosed by the intrinsic layer p- and n-doped regions,

7b a perspective view of a p- and n-doped wafer in 7a .

8a FIG. 2 shows a sectional view of a p- and n-doped wafer with intrinsically completely surrounding intrinsic layer with p-type and n-type doping regions which are elongated in cross-section, FIG.

8b a perspective view of the embodiment in 8a .

9a a schematic view of a drawing process with juxtaposed lines for producing a solar cell,

9b a sectional view taken along the line IX-IX in 9a .

9c a sectional view of zigzag lines in 9a .

10a a schematic view of stretched in a frame n and p lines,

10b a sectional view taken along the line X _b -X _b in 10a ,

The following drawings are not true to scale and not to scale.

In the 1a illustrated sectional view perpendicular to the solar cell plane E shows a first embodiment of a solar cell according to the invention 1 , The solar cell 1 is flat and flat and has outer peripheral dimensions of about 10 × 10 cm and a thickness of between 100 μm and 300 μm. The solar cell 1 has a sunlight-facing side 2 and a side facing away from sunlight parallel to it 3 on. The incident light rays 4 meet on the sunlight-facing side 2 ,

Along the solar cell E are a plurality of n-doped regions and a plurality of p-doped regions, the n-doped region n has a doping of 10 ¹⁹ pentavalent impurities / cm ³ or more, and the p-doped regions p of 10 ¹⁹ trivalent foreign atoms / cm ³ or more. Between adjacent n- and p-doped regions i-regions, intrinsic regions i are provided. Here, intrinsic regions i are understood to mean regions with dopings of less than 10 ¹⁵ foreign atoms / cm ³ .

The i-regions i, the n-regions n and the p-regions p all extend over the entire height H of the solar cell 1 , Each of the regions i, n, p extends from the side away from the sunlight 3 to the sunlight-facing side 2 , The height of the differently doped regions H _i , H _p , H _n extends for each of the plurality of individual regions i, n, p over the entire height H of the solar cell 1 , The extent of each of the regions p, i, n in the solar cell plane E along the entire height H is constant. A width B _{i of} each of the intrinsic layers i is between B _i = 50 μm to 5000 μm, preferably up to about 1 cm, depending on the drift length of the charge carriers in FIG intrinsic semiconductor material, a width B _{p of} the p-regions is about B _p = 1-3 microns, a width B of the n-regions is also about B _n = 1-3 microns. The lengths of the individual regions L _i , L _p , L _n extend over the entire longitudinal extent L of the solar cell 1 , The widths B _i , B _p , B _n remain along the entire in 1 arranged perpendicular to the plane of the longitudinal extent L _i , L _p , L _n substantially constant. The n- and p-doped regions n, p run parallel to each other in the longitudinal direction L. The n- and p-doped regions are spaced apart from each other by one of the intrinsic regions i.

The n- and p-doped regions n, p are along their longitudinal direction L _n , L _p with current contacts on an outer side 2 . 3 Mistake. In the in 1 shown sectional view along the entire width B of the solar cell 1 several hundred of the sequences of the structure: n-range n, i-range i, p-range p, i-range i, ie the sequence nipi arranged directly next to each other. The regions i, n, p are layered along the width B. The layer sequence is transverse to the direction of the sun's rays 4 arranged.

The photoactive areas of the solar cell 1 are the intrinsic regions i. Advantageously, the intrinsic regions i are directly the sun's rays 4 exposed, and they do not need to pass through an n-doped layer as in conventional solar cells. Thus, the sunlight hits 4 particularly effective and with only slight absorption losses to the intrinsic regions i, in order to generate charge carrier pairs, holes + and electrons - in the intrinsic regions i by absorption of light energy. The charge carrier pairs +, - generated by the incident light in the intrinsic regions i, spatially separate through an electric field E _el present in the intrinsic layer i. The electric field _El is probably due to the fact that a positive space charge zone is formed at the boundary layers between the i-regions i and adjacent n-regions n in the n-region n, since free electrons of the donor atoms diffuse into the intrinsic layer and stationary, positively charged donor atoms remain in the n-layer and a negative space charge zone is formed at the interface between p-region p and intrinsic region i in the p-region p, because the mobile holes in the valence band diffuse into the intrinsic region i. Due to the excess of bound electrons in the p-region p, a negative space-charge zone is formed in the p-region p. The free-electron originating from the n-region n and the holes originating from the p-region p + recombine in the intrinsic region i. A negative space charge zone therefore forms in the edge region of the p-region p to the intrinsic region i, and a positive space charge region is formed in the edge region of the n-region n. Considering the entire semiconductor, the charge neutrality is maintained. The electric field E _el forming between these two space charge zones passes through each of the intrinsic regions i along its entire width Bi. The electric field E _el is homogeneous and has substantially equidistant field lines in the solar cell plane E. The electrons drift to the adjacent n-region n and the holes + drift to the adjacent p-region p. The n-region n is advantageously semipermeable, ie here, it has for electrons - a large conductivity and for holes + a small conductivity. As a result of the homogeneous electric field E _el forming in the intrinsic region i, the charge carriers +, - are directed into the respective space charge zone and ultimately to the contacts. In this case, in particular the high doping is very helpful in order to achieve metal-semiconductor contacting of low resistance for the electrical current both on the n-side (tunnel effect) and on the p-side (ohmic contact).

The solar cell 1 in 1 can by a in 2a to 2c schematically illustrated method can be produced. First, conventional wafers made of intrinsic semiconductor material, for example silicon or germanium, are n-doped on their outer side lying in a wafer plane and p-doped on an outer side facing the same. It will be the usual PIN structures 20 educated. In this case, conventional doping methods are used for doping the wafers. The PIN structures 20 become on the doped outer surface with a metal layer 21 provided, and then metal layers 21 glued over equally doped outer surfaces by means of metal contact adhesive to each other. In particular, in the case of rough wafer surfaces, as arise in particular during drawing processes, a metal layer, for example, is advantageously produced between the doped wafers. As aluminum and / or silver introduced, and the stratification is then heated so that only the metal layer melts and so compensates for the roughness and a firm contact is formed. Since aluminum itself is p-doping, the wafer material would only need to be n-doped.

This creates the in 2 B Layer shown p-, i-, n-, metal layer 21 , n, i, p, metal layer 21 , p-doped regions along the height of in 2a illustrated blocks 22 , Between the n-areas a silver layer and between the p-areas an aluminum layer is arranged. The in 2a illustrated block 22 is then sawn vertically into narrow slices along its vertical extent, the basic building block being a solar cell 1 form. A thickness of the block 22 sawed-off washers corresponds to the height H of the solar cell 1 , the discs are the basic building block of the solar cell 1 , The basic building block of the solar cell 1 is in 2c shown. At him then still contacts the metallization 21 applied. The outer surfaces of the solar cell 1 can be further treated by passivation and / or an "anti-reflection coating". The whole solar cell 1 is in a frame 100 bordered or fixed on a carrier material. The p, i, n, metal layer 21 , n, i, p, metal layer 21 p-type sequence sections extend immediately one after the other along the width B of the solar cell 1 , The solar cell 1 is with its area layering transverse to the direction of sunlight 4 arranged so that the light falls directly on the intrinsic areas i. By appropriate mirror arrangement, both the sunlight-facing side 2 the solar cell 1 as well as the sunlight side facing away 3 the solar cell 1 the sunlight 4 be exposed and thus particularly effectively cause the charge carrier collection of the electron-hole pair formation.

3 shows a second, not belonging to the invention embodiment of a solar cell 1 , The p-doped regions p and the n-doped regions n are each formed in a semi-lenticular shape and on the side facing away from the sunlight 3 the solar cell 1 arranged. Also in this embodiment, the dopants have the orders of magnitude mentioned in the first embodiment.

First, wafers are provided with a height of about 200 microns. The side facing the sunlight 2 the wafer is not treated, the side facing away from the sunlight 3 of the wafer is partially p- and n-doped. The p- and n-doped regions p, n are spaced from each other by an intrinsic region i. The material of the intrinsic region i corresponds to the untreated wafer material. The distance between the p- and n-doped regions p, n is approximately 300 μm.

P and n dopants in the wafer can be converted to the sunlight-facing side by conventional photolithographic techniques 3 of the wafer.

Boron, indium, aluminum or gallium or combinations of such elements are chosen as p-doping element. The n-doping elements may be nitrogen, phosphorus, arsenic or antimony. Depending on the duration of the doping process and the concentration of the doping gas, a differently-colored and extended half-lens-shaped region arises in the treated wafer side 3 , Subsequently, the focus is on each of the n- and p-doped regions on the side facing away from the sunlight 3 an electrically conductive contact 5 applied. The n-doped and p-doped regions may be circular on the sunlight-facing side 3 be arranged, like 6s shows. However, they can also take the form of oblong, parallel areas ( 6b ) are formed. You can also accept other arrangements, such as helical ( 6c ) or comb-shaped arrangements ( 6d ).

In a further embodiment of a production method of a solar cell 1 according to 3 become the p-type dopants and n-type dopants by means of an ion gun in the sunlight away side 3 shot of the wafer.

4 shows a third, not belonging to the invention embodiment of a solar cell 1 , The n and p regions n, p are distinguished by the fact that they each extend up to three quarters along the height H of the solar cell 1 , starting from the side facing away from the sunlight 3 , in the solar cell 1 extend into it. As a result, homogeneous electric fields E _el are also formed in the intrinsic material i, at least over larger areas, between the n and p regions n, p.

A possible method for the introduction of trenches 40 in the sunlight side facing away 3 of the wafer takes place in the well-known "Laser Grooved Buried Contact" method. First, a wafer formed of intrinsic material is again provided. Then turn into the side facing away from the sun 3 trenches 40 melted by a laser. The trench structure can be found in the 6b to 6d take the form shown. In the embodiments of the 6c and 6d only two external connections are necessary.

It will first be a ditch 40 or trench system in the sunlight side facing away 3 brought in. After that, the walls become 41 the trenches 40 p-doped or n-doped. The doping can likewise be effected by means of photolithographic processes. The entire wall 41 a trench 40 is either completely p-doped or fully n-doped. Then the trenches 40 with metal 42 filled and each with a laterally protruding from the wafer connection 5 Mistake.

5 shows a fourth, not belonging to the invention embodiment of a solar cell 1 , The n and p regions n, p do not extend over the entire height H of the solar cell 1 but approximately along three quarters of the height H, starting from the sunlight side facing away 3 , The p- and the n-doped regions p, n have an approximately circular diameter in the solar cell plane E, which is smaller than the trench width of the trenches 40 of the 4 ,

There are needles 50 spaced apart provided with a positive doping or negative doping coating. At the negative doping needles 50 Tungsten wire can be coated with boron.

The coated needles 50 For example, they can be pressed into a wafer. The needles inserted into the wafer 50 are then briefly energized, whereby a portion of the coating migrates into the intrinsic material i and there in an area around the needle 50 doped around the wafer. In the cross-section in the solar cell plane E, approximately circular n- or p-doped regions n, p are formed. This process step is known in other context as 'forming'.

In addition to forming and alloying is in another process for producing the solar cell 1 according to 5 a melting method is applicable, in which a pincushion with p- and n-doping coatings is coated with silicon powder. The p-doping needles 50 and the n-doping needles 50 are each conductively connected and the two groups of pins insulated from each other. The silicon powder is then melted and cooled, and thus a polycrystalline silicon structure can be deposited on the two-part pincushion.

A fifth embodiment of a solar cell 1 is in the 7a and 7b shown in cross-section perpendicular to the solar cell plane E. The solar cell 1 consists essentially of an intrinsic region i, which completely encompasses the n- and p-doped regions n, p, into the elongated p-doped and n-doped regions p, n in, which run approximately in the longitudinal direction L along the height H. the semiconductor are introduced. The n- and p-doped regions n, p surround a wire or fiber-shaped conductor 70 , The solar cell 1 of the 7a is two-sided with sunlight 4 applicable. The electric field E _el pervades arcuate field lines wide areas of the intrinsic region i.

The solar cell 1 of the 7a can be produced, for example, in such a way that boron fibers each having a tungsten core and n-doped silicon carbide fibers, each with a graftite core, are arranged alternately and are clamped horizontally in a production mold and are covered with semiconductor material. After curing, the conductors are subjected to an impulse if necessary, whereby the doping migrates into the respectively adjacent intrinsic region. 7b shows the solar cell the 7a in a perspective view with multiple contacts 5 , The path length through the doped areas to the wire 70 is for the charge carriers -, + comparatively short. Thus, the recombination rate is low, and the solar cell 1 Overall, it has a high degree of efficiency.

However, it is also conceivable to grow the semiconductor material between the lines by means of epitaxial growth.

8a . 8b show an embodiment according to the 7a . 7b , but with the height H of the solar cell 1 The p-type and n-type regions p, n each extend over at least half the height extent H of the solar cell 1 , but not up to the sun-facing and sun-facing side 3 . 2 to reach. The n- and p-regions n, p are formed in cross-section substantially rectangular and arranged parallel to each other. As a result, electric fields E _el are formed between the doped regions n, p, with field lines extending substantially parallel to one another at least over a wide range, which generate a drift of the charge carriers, +, in a particularly favorable manner. A manufacturing method corresponds to that of the embodiment in FIGS 7a . 7b , but are not round in cross-section wires, but used in cross-section elongated, rectangular conduction bands.

9a shows a drawing process for the production of the solar cell according to the invention 1 , In this case, uniformly spaced n and p lines n, p through a silicon bath 90 drawn. Due to the high surface tension of the molten silicon, intrinsic silicon regions i are formed between the n and p lines n, p, which cool and harden. It will be a slat 91 from the silicon bath 90 drawn. As n and p lines n, p conductive and high temperature resistant materials are used. The melting point of the p and n lines p, n should be above the melting point of the silicon. Intrinsic silicon i has a melting point of about 1,410 degrees Celsius. Borfaser with a wire-shaped graphite or Wofram core can be used as the p-type line p, and the n-type line n can be a nitrogen-doped silicon-carbide fiber. The high doping carries electrical current conductive character, so that no extra conductive core is needed. The materials are high temperature resistant and conductive. At the in 9 As shown, the n- and p-lines n, p are completely wetted completely by the intrinsic silicon when being pulled through transversely to the longitudinal direction L. The n and p lines n, p thus run completely in the intrinsic region i of the later produced solar cell.

9b shows in a sectional view of the drawn lamella 91 , p and n lines n, p are arranged alternately in cross section. The n- and p-lines p, n are circular in cross section and formed in each case 180 microns in diameter. The distance between adjacent lines n, p is about 500 μm.

9c shows an alternating arrangement of n- and p-doped lines n, p in a cross-section perpendicular to the line direction. In contrast to the arrangement of the lines n, p in 9b However, the n and p lines n, p are zigzagged offset from each other. The zigzag arrangement of the n and p lines n, p results in a high mechanical stability of the wafer, so that the intrinsic areas i between the p and n lines p, n can be pulled particularly thin. In addition, there is the advantage that due to the zigzag shape multiple reflections are possible, resulting in less light losses due to reflection from the side facing sunlight 2 occur. The zig-zag shape provides a particularly suitable form of texturing the sun-facing side 2 ,

Both the zigzag structure of 9c as well as the planar structure of the 9a can in the drawing process by a silicon bath 90 pulled through and so form a wafer or in a silicon bath 90 immersed and pulled out freely floating, so that wet the spaces between the n- and p-lines n, p due to the high surface tension of the liquid silicon and form the intrinsic regions i.

The 10a and 10b show a further embodiment of a wafer which can be produced by the dipping method or also by means of an n- and p-type conduction structure n, p. These are in a frame 100 n- and p-lines n, p clamped parallel. The 10b shows a cross section along the line X _b -X _b of 10a , In 10b It can be seen that alternately planes E of n-lines n and p-lines p offset from each other are provided, so that forms a hexagonal honeycomb structure in cross-section. If the lines n, p through a silicon bath 90 are drawn through, forms an intrinsic silicon film i between the next adjacent n- or p-lines n, p, while the interiors of the honeycomb structure remain free. The honeycomb structure makes it possible to produce particularly stable wafers, and moreover, very little silicon is required since the honeycomb structure is substantially hollow. Nevertheless, the absorption rate due to the zigzag shape of the sun-facing side 2 as well as the multi-layered construction. The diameter of the lines n, p is about 10 microns. As a result, a thickness of the silicon film of preferably less than 10 μm is formed. Due to the very small distances between the n and p lines n, p, particularly high-quality monocrystalline silicon is produced.

However, it is conceivable on the line structures n, p the 10a . 9a and 9c Growing silicon in the epitaxy process. Maybe after the epitaxial growth again formed. Conveniently, in the epitaxial process, graphite fibers are first surrounded by heavily doped molten silicon to form a heavily doped silicon carbide crystal having a preferred layer thickness on the order of 1 micron on which semiconductor material can grow in the epitaxial growth process.

LIST OF REFERENCE NUMBERS

1

solar cell

2

sunlight-facing side

3

sunlight side away

4

incident light / sunbeams

5

connection

7

wires

20

PIN structures

21

metal layer

22

block

40

trenches

41

walls

42

meter

50

needles

70

ladder

90

silicon bath

91

lamella

100

frame

B

width

B _i , B _n , B _p

Widths of the individual areas

e

solar cell level

E _el

electric field

H

height

H _i, H _n, H _p

Heights of the individual areas

L

longitudinal extension

L _i , L _n , L _p

Lengths of the individual areas

n

n-doped regions

p

p-doped regions

i

intrinsic areas

+

holes

-

electrons

Claims (4)

Solar cell ( 1 ) with a solar cell plane (E) along which a sunlight-facing side (E) 2 ) and a side facing away from the sun ( 3 ) and along the solar cell plane (E) in the solar cell plane (E) a plurality of n-doped regions (n) and p-doped regions (p) is arranged alternately, wherein the n- and the p-doped region (n, p ) each have a doping of at least 10 ¹⁹ foreign atoms / cm ³ and successive p- and n-doped regions (p, n) in each case by an intrinsic region (i) having a doping of less than 10 ¹⁵ foreign atoms / cm ³ , separated from each other and spaced from each other and themselves the n, p and i ranges (n, p, i) from the sunlight side ( 3 ) to the sunlight side ( 2 ) and the n, p and i regions (n, p, i) are arranged on top of each other with the sequence: p-region (p), i-region (i), n-region (n), conductive contact region ( 21 ), n-range (s), i-range (i), p-range (p), conductive contact area ( 21 ), wherein the width of each of the i-regions (i) is between 50 μm and 5000 μm and at the conductive contact regions ( 21 ) Contacts ( 5 ) are applied. Solar cell ( 1 ) according to claim 1, characterized in that along a cross-section transverse to the solar cell plane (E) each n-doped region (s) of two p-doped regions (p) is adjacent. Solar cell ( 1 ) according to claims 1 to 2, characterized in that mutually facing outer surfaces of adjacent n- and p-doped regions (n, p) are substantially parallel to each other and thus form a homogeneous electric field (E _el ) between them. Process for producing a solar cell ( 1 ) according to claims 1 to 3, in which regions are arranged on top of one another with the sequence: p-region (p), i-region (i), n-region (n), conductive contact region ( 21 ), n-range (s), i range (i), p-range (p), conductive contact area ( 21 ) by doping the intrinsic regions (i) with at most 10 ¹⁶ foreign atoms / cm ³ and doping the p and n regions (p, n) each with at least 10 ¹⁹ foreign atoms / cm ³ and a multiplicity of wafers in each case an outside ( 2 . 3 ) is p-doped and on an opposite outer side ( 2 . 3 ) n-doped and the doped wafers with identically doped outer sides ( 2 . 3 ) facing each other to a block ( 22 ) are stacked conductive and cut transversely to the stratification in solar cell high slices.

Images