EP0625286A1

EP0625286A1 - Integrated-circuit stacked-cell solar module

Info

Publication number: EP0625286A1
Application number: EP93902041A
Authority: EP
Inventors: Wilhelm Dr. Kusian; Joze Prof. Dr. Furlan; Wolfgang Dr. Riedl; Hans Dr. Pfleiderer
Original assignee: Siemens AG
Current assignee: Siemens AG
Priority date: 1992-02-04
Filing date: 1993-01-22
Publication date: 1994-11-23
Also published as: US5527716A; WO1993015527A1; JPH07503105A

Abstract

Pour un module solaire à cellules empilées, il est proposé de disposer en alternance les cellules solaires en mode pin, nip, pin, etc., de structurer l'empilement de cellules solaires en forme de bandes et de réaliser le câblage par des structures d'électrode en peigne qui relient les côtés p des cellules solaires d'un empilement au côté n des cellules solaires de l'empilement voisin. Le module solaire pouvant être réalisé sous une forme intégrée est monté en parallèle à l'intérieur d'un empilement et en série entre les empilements. Par une répétition simple des opérations de fabrication, on peut ainsi réaliser des empilements de deux cellules solaires ou plus présentant une moindre altération à la lumière par rapport aux modules solaires connus.For a solar module with stacked cells, it is proposed to alternately arrange the solar cells in pin, nip, pin mode, etc., to structure the stack of solar cells in the form of strips and to carry out the wiring by structures d comb electrode which connects the sides p of the solar cells of a stack to the side n of the solar cells of the neighboring stack. The solar module, which can be produced in an integrated form, is mounted in parallel inside a stack and in series between the stacks. By a simple repetition of the manufacturing operations, it is thus possible to produce stacks of two or more solar cells having less light alteration compared to known solar modules.

Description

Integrated stacked solar cell module.

Under the influence of light, solar cells and solar modules made of amorphous silicon (a-Si: H) show a considerable drop in electrical power, which can be up to 40 percent of the original power. This occurs during the operation of the solar cell aging process can be minimized by annealing at 180 ^* C undone, but relies Lichtein¬ effective again.

One way to circumvent this aging effect, also known as the Staebler-Wronski effect in solar cells, is to use mixed semiconductors with a smaller band gap for the production of the active i-layer. It is also possible to manufacture pin solar cells from amorphous silicon with thinner i-layers that hardly show any aging below a layer thickness of approx. 300 nm. However, such a thin i-layer absorbs the red portion of the sun spectrum only incompletely, so that a short-circuit current which is too low is obtained with these cells.

Stack cells offer one possibility for producing solar cells with thinner i-layers but high red absorption. Such a stacked cell, made up of two or more individual cells, each of which is thin, absorbs a sufficient part of the sunlight. A double cell has the structure pin-pin, a triple cell pin-pin-pin and so on. The i-layers of the partial cells of a stack can consist of one and the same semiconductor material, or can be produced from different semiconductor materials, for example from silicon / carbon or silicon / germanium alloys.

Tandem or generally stacked cells do indeed show one higher stability against light aging with constant or even higher efficiency. The cells of a stack are optically connected in series and, in the simplest case, electrically connected in series. The photo voltages of cells lying one above the other add up, while the photo currents must match. The maximum achievable photocurrent in such a stack cell is determined by the photocurrent of the weakest single cell.

10

Current adjustment or "current matching" is, however, difficult to carry out, particularly in the production of large-area solar modules. Fluctuations in layer thickness during the deposition of the active semiconductor layer increasingly occur with a substrate area that becomes larger and influence the photocurrent.

In order to avoid the problem of current adaptation, EP 0 326 857 proposes for the first time to use stacked cells with a 0 pinip or nipin structure and to connect them integrated in a module. In this arrangement, the two superimposed solar cells of the tandem cell are electrically connected in parallel, with the photocurrents of the sub-cells adding up. Only a voltage adjustment is required

25 of the partial cells in order to have identical photo voltages. However, since there is only a slight dependence of the photo voltage on the photocurrent in the working area of a solar cell, the voltage adjustment of a tandem or stacked cell with parallel connection can be carried out much more easily than

30 a current adjustment with series connection.

A disadvantage of the known tandem solar module with parallel connection is the complicated structure compared to a simple thin-film solar module, which requires flank insulation in the separating trenches between the 35th stack cells and in which the contact is made via layers in the layers for the connection.

The object of the present invention is therefore to provide a to provide a complex structure for a stacked cell solar module, which can be easily produced using known thin-layer structuring techniques and thereby ensures technologically simple and reliable interconnection even with a minimal loss of active semiconductor area.

This object is achieved according to the invention by a multilayer solar module, in which a plurality of solar cell stacks, each with n thin-film solar cells, are arranged next to one another in strips in a row, with the following features:

a) the solar cells each have a pn junction or a pin structure,

b) the direction of the semiconductor transition is reversed from one solar cell layer to the one above it,

c) the electrodes of the solar cells are transparent layers and, together with their interconnection, viewed in cross section perpendicular to the strips, form regular comb-like structures, each having n + 1 teeth, which alternate between the two stacks on both sides of the perpendicular to the substrate Back are arranged

d) the n + 1 electrode layers of a stack are alternately formed by the teeth of the comb-like structures adjacent to the stack on both sides, these interlocking without contact,

wherein each of the comb-like structures electrically connects the p-sides of the solar cells of a stack with the n-doped sides in the adjacent stack and thus electrically connects all the solar cells of a stack in parallel and electrically connects the stacks in series via the module, and where n = ≥2.

Further refinements of the invention and a method for to take .

The module according to the invention has only regular structures and therefore a simple structure, the manufacture of which does not require any additional technological effort compared to thin-film solar modules and is therefore compatible with known modules in terms of production technology.

The possible number n of thin-film solar cells stacked in the stack is variable in the structure according to the invention and is at least 2. As a result, the individual solar cells can be made as thin as desired and light aging can thus be reduced. At the same time, high light absorption is guaranteed, since the lower absorption of a thinner layer can be compensated for by a higher number of solar cells in the stack. The total layer thickness of the n thin-film solar cells in the stack is ultimately responsible for the total absorption.

In the strip-shaped solar cell stack, the individual solar cells can be arranged exactly one above the other, so that due to the connection in the module, only a minimal loss of active semiconductor area is required, which corresponds to only one trench width per strip or per stack, ie not more than in known thin-film modules .

In principle, the new module structure is suitable for all integrated, structurable semiconductor materials and thus for almost all semiconductors suitable for thin-film solar cells. In particular, these are amorphous layers made of silicon (a-Si: H) or its alloys, for example with germanium (a-Si / Ge: H) and cells made of pure germanium (a-Ge: H) or polycrystalline chalcopyrite materials, for example from copper indium diselenide or copper gallium diselenide. The module structure according to the invention is always particularly advantageous when semiconductor layers which are as thin as possible are desired, that is to say all materials which, for example, show light aging or in which the production of thicker layers is technological is difficult.

In the solar module according to the invention, a voltage adjustment is necessary in order not to get any losses in the module performance. This does not necessarily require identical materials for the solar cells in the stack, but makes it desirable and advantageous.

Thus, the light can be almost completely absorbed in the usable part of the spectrum, without the module losing power. High photocurrents and thus high efficiency are obtained.

Although the side of the stack structure facing away from the substrate can also serve as the light incidence side, the substrate side is nevertheless preferred. Glass is therefore usually selected as the substrate material, while known TCO materials can be used for the transparent conductive electrodes, for example doped metal oxides such as fluorine-doped tin oxide (Sn0 ₂ : F), boron-doped zinc oxide (Zn0: B) or indium tin oxide (ITO) and others. The top electrode layer can also consist of TCO or a metal layer in order to reflect light that is not absorbed in the stack into the solar cells.

Although the number n of solar cell layers in the stack can be increased as desired, in practice a minimum number of cells is preferred in order to limit the manufacturing outlay for the many layers required. In addition, the advantageous effects of the invention are fully pronounced even with a tandem module (n = 2) and more than two layers are therefore rarely required.

The invention is explained in more detail below with the aid of an exemplary embodiment and the associated eight figures. Show at

Figures 1 to 5 different process stages in the manufacture Position of a stack module with two solar cell layers in a schematic cross section, during

FIG. 6 shows a module according to the invention with three solar cell layers.

The equivalent circuit diagrams for modules with n = 2 and n = 3 are shown in FIGS.

1: A first electrode layer ESI is deposited over the entire surface of a glass substrate S, for example 2 mm thick, for example by PECVD (plasma enhanced chemical vapor deposition) or by sputtering of TCO materials. The first electrode layer is then structured with a laser, narrow trenches G1 being produced parallel to one another by removing electrode material, so that electrode strips of, for example, 1 to 5 cm in width remain. The width of the trenches Gl is chosen to be as small as possible and is 50 μm, for example, in the case of a laser used.

FIG. 2: A first thin-film solar cell DS1, in the present case an a-Si: H pin diode, is now generated over the entire surface of the strip of the first electrode layer ESI, a PECVD method being used. The structuring of the first thin-film solar cell DS1 is likewise carried out using a laser, trenches G2 being produced parallel to the trenches G1, which are laterally offset by at least one trench width, so that the surface of the first electrode layer ESI is exposed at the bottom of the trenches G2 is.

FIG. 3: A second electrode layer ES2 is now applied over the entire surface of the solar cell DS1 and structured in the same way with a laser, trenches G3 being produced offset by at least one further trench width in the same direction with respect to G2 and thereby exposing the surface of the thin-film solar cell DS1 becomes. FIG. 4: In the next step, a second thin-film solar cell DS2 is produced over the entire surface, which has an inverse structure, here nip, with respect to the direction of its semiconductor junction with respect to the first solar cell DS1. This is followed by a further laser structuring, trenches G4 being produced which are laterally offset in the other direction against the trenches G2 and thus lie directly over the trenches G2.

FIG. 5: As the last step in the production of a finished tandem solar cell module, a third TCO electrode layer ES3 is generated and structured in a manner analogous to that of the first electrode layer ESI.

The exemplary structure shown in FIG. 5 for a tandem solar cell module according to the invention can be varied further within the scope of the invention. For example, solar cells of different thicknesses can be arranged one above the other in the stack, the second solar cell DS2 preferably having a higher layer thickness. In the exemplary embodiment, the first solar cell DS1 could have a thickness of 300 nm, for example, and the second solar cell DS2 could have a thickness of 300 to 500 nm. The lower limit of a possible layer thickness is determined not by physical but by technological parameters. A photocurrent is observed in a pin solar cell from a layer thickness of approx. 25 nm. Due to layer thickness fluctuations or unevenness in the structure of the substrate or one of the layers deposited thereon, however, in thin layers there is a risk of a short circuit between two electrode layers in a stack, which could reduce the performance of the module. Larger, technologically safe layer thicknesses are therefore preferred.

The thickness of the electrode layers is designed according to their conductivity. When using boron-doped zinc oxide, a thickness of approx. 1 μm is sufficient. However, a larger structure width or a larger distance between two adjacent trenches G in a layer can also have a higher conductivity. of an electrode layer ES and thus require a higher layer thickness.

The structure width is generally determined by the size of the selected substrate. So many stripe-shaped solar cell stacks are produced that the cell voltages of the electrically connected solar cell stacks add up to a total voltage of 12 to 14 volts at the operating point. Since the voltage of a single cell made of a-Si: H is about 0.6 to 0.8 volts, about 20 or more strips are required for the desired final voltage. The width of a strip can be up to 5 cm.

FIG. 6 shows a stacked cell solar module according to the invention with three layers of thin-film solar cells in the stack. For this purpose, a further thin-film solar cell DS3 with a pin structure is deposited on the tandem cell structure (see FIG. 5) and structured in the same way as the first and second thin-film solar cell DS1 or DS2. The back electrode is formed by a fourth electrode layer ES4, which is produced in the same way as the second electrode layer ES2 and structured congruently with it in order to keep the loss of active semiconductor area low.

It can be clearly seen in this illustration how the electrode strips ES form comb-like structures KS in cross section. By the original trenches G2 and G4 axis formed of electrode material represents the comb back of which ab¬ alternately to both sides facing teeth sc the electrode ^'hichten represent for the solar cell. In other words, the electrodes of a solar cell stack SZ are alternately formed by a first comb-like structure KS1 and a second comb-like structure KS2, which engage in one another or in the solar cell stack SZ without contact (to avoid short-circuits). The outer comb structures EKS at the edge of the module are constructed analogously to the middle comb-like structures KS, but are only one solar cell Stacks are adjacent and therefore have teeth or electrode strips ES pointing only to one side. Contacting for connecting the module to a consumer or a storage unit also takes place at the end structures EKS.

If a metal layer is applied over the entire surface as the top electrode layer ES3 (n = 2), direct structuring from above by means of a laser is not possible or is only possible with great difficulty, without at the same time the solar cell layer DS2 exposed underneath in the region of the trenches G5 to change structurally, with the risk of a short circuit of the solar cell. For this structuring step, it is alternatively possible to use a laser to irradiate from the substrate side through the glass substrate, as is proposed, for example, in the earlier European application 91 11 70 61.1. Due to strong local heating at the interface between the second solar cell DS2 and the second electrode layer ES2 in the focus of the laser used for structuring, the second solar cell layer and the third electrode layer ES3 lying above it come off in the area of the focus. For this step, however, it is necessary that the trench width G2 is significantly larger than the trench width G4, so that the laser beam incident through the substrate S only an interface between a solar cell layer DS and an electrode layer ES (here ES2 / DS2) "sees". For this process variant, the trench width G2 must therefore be calculated at least from the sum of the trench widths G4 and G5 plus a tolerance value which results from the tolerance of the cutting device (laser) and the adjustment accuracy of the device.

With continued layer stacking (n greater than 2), however, structuring through the substrate S is no longer possible. As an alternative to the conventional laser structuring method from "above", as with all other structuring steps, a lifting technique can be used. For this purpose, before the generation or deposition of a layer to be structured in the area of the later trenches G, a capable structuring paste applied, for example with the help of screen printing. After the layer to be structured has been produced above the printed paste, the paste is mechanically detached, the layer to be structured above being lifted off in the region of the paste. Each laser structuring step can therefore be replaced in the method by the steps "printing" of a structuring paste in the area of the structuring lines (trenches) and "lifting off" of structuring paste including the layer areas above it, which are necessary for the lifting technique. These two steps are technologically easy to carry out and therefore represent a real alternative to laser structuring.

FIG. 7 shows an equivalent circuit diagram for the stacked cell solar module shown in FIG. 5. The solar cells are shown as diodes in accordance with the direction of their semiconductor transition, while the connecting lines drawn represent the comb-like electrode structures.

FIG. 8 shows the equivalent circuit diagram for a stacked cell solar module consisting of three solar cell layers (see FIG. 6). It can be clearly seen that the solar cells (diodes) are connected in parallel in the stack, but are connected in series via the stack.

Claims

1. Multi-layer solar module in which a plurality of solar cell stacks, each with n thin-film solar cells, are arranged next to one another in strips on a substrate, with the following features:

a) the solar cells each have a pn junction or a pin structure,

c) the electrodes of the solar cells are transparent layers and, together with their interconnection when viewed in cross section perpendicular to the strips, form regular comb-like structures, each having (n + 1) teeth, which alternate between two stacks on both sides of the perpendicular to the substrate Back are arranged

wherein each of the comb-like structures electrically connects the p-sides of the solar cells of a stack with the n-doped sides in the adjacent stack, thus electrically connecting all the solar cells of a stack in parallel and electrically connecting the stacks in series via the module, and wherein n ^ 2.

2. Solar module according to claim 1, in which the layer thicknesses of the thin-film solar cells in the stack add up to a total layer thickness which shows an absorption in the sensitivity range of the solar cell material of a maximum of more than 80 percent.

3. Solar module according to claim 1 or 2, in which for the solar cell len a semiconductor material is used, which can be deposited over a large area and can be structured in an integrated manner.

4. Solar module according to claim 3, whose solar cells consist of amorphous silicon.

5. Solar module according to one of claims 1 to 4, in which the substrate is transparent and all electrode layers, optionally with the exception of the uppermost electrode layer, consist of transparent conductive oxide (TCO).

6. Solar module according to one of claims 1 to 5, in which each stack consists of two solar cell layers (n = 2).

7. Method for producing an integrated interconnected stacked cell solar module with the following steps

a) Creation of a transparent electrode layer over the entire surface on a transparent substrate

b) strip-like structuring of the first electrode layer

c) Generating a whole area of a first thin-film solar cell with a pin or pn structure over the stripe-shaped structured first electrode layer

d) structuring of the first thin-film solar cell parallel and laterally offset to the strips of the first electrode layer in such a way that the surface of the strips of the first electrode layer are exposed in the trenches separating the strip-shaped solar cells

e) deposition of a second transparent electrode layer over the entire surface

f) strip-shaped structuring of the second electrode layer parallel and laterally offset in the same direction to the strip-shaped first thin-film solar cell g) deposition of the entire surface of a second thin-film solar cell with a structure inverse to the first thin-film solar cell with respect to the direction of the semiconductor transition

h) structuring of the second thin-film solar cell corresponding to the first thin-film solar cell

i) Generating a third electrode layer over the entire surface

j) structuring of the third electrode layer corresponding to the first electrode layer.

8. The method as claimed in claim 7, in which at least one further thin-film solar cell and at least one further electrode layer are produced and structured above the third electrode layer by repeating steps g) to j), each further thin-film solar cell having a structure which is inverse to that of the solar cell underneath owns and where the structuring of all solar cell layers is identical, while the structuring of an electrode layer x takes place in accordance with the structuring of the electrode layer x - 2.

9. The method according to claim 7 or 8, in which a laser is used for structuring the solar cells and the electrode layers.

10. The method according to claim 7 or 8, in which a lifting technique is used for structuring electrodes and / or .solar cell layers.

11. The method according to any one of claims 7 to 10, in which a substrate made of glass is used, doped metal oxides are used as the electrode material, the thin-film solar cells are made from amorphous silicon (a-Si: H) with a pin structure and the Number n of the superimposed solar cells is 2 to 4.

12. The method according to any one of claims 7 to 11, in which a metal layer is produced as the top electrode layer, in which the structuring of the top electrode layer, including the underlying solar cell, is carried out with a laser, through the substrate and through the layer to be structured Layer lying transparent electrode layers is irradiated.