EP0763260A1

EP0763260A1 - Thin film solar cell and process for producing the same

Info

Publication number: EP0763260A1
Application number: EP95920751A
Authority: EP
Inventors: Claus Beneking; Bernd Rech
Original assignee: Forschungszentrum Juelich GmbH
Current assignee: Forschungszentrum Juelich GmbH
Priority date: 1994-06-01
Filing date: 1995-05-27
Publication date: 1997-03-19
Also published as: DE4419273A1; DE4419273C2; WO1995033284A1

Abstract

A thin film solar cell based on amorphous, hydrogenated silicon and/or its alloys is disclosed, as well as a process for producing such solar cells. The solar cell is made of a series of layers that comprise a non-doped layer, a buffer layer adjacent and joined to the non-doped layer at the light input side, a p-doped layer adjacent and joined to the latter at the light input side, and a n-doped layer adjacent and joined to the other side of the non-doped layer. The object of the invention is to create a solar cell with increased long-term stabilised effectiveness. For that purpose, a compensation layer is provided in the solar cell between the p-layer and the buffer layer.

Description

Thin film solar cell and process for its manufacture

1. Technical field

The invention relates to a thin-film solar cell based on the amorphous, hydrogenated silicon and / or its alloys.

2. State of the art

Thin-film solar cells based on the amorphous hydrogenated silicon (a-Si: H) and its alloys are produced in a known manner with the doping sequence nip or pin by layer deposition on suitable opaque or transparent substrates, also as stacked solar cells in a cascade arrangement (so-called Stacked cells, eg pinpin) with identical or different band gaps of the sub-cells used.

Transparent, conductive oxide layers (so-called TCO layers, such as ITO, Sn0 _{2 /} ZnO, Ti0 ₂ ) are generally used as contact layers on the light entry side / likewise transparent conductive oxide or metal layers or a combination on the side facing away from the light from both.

The incidence of light advantageously takes place from the p-doped side. The band gap of the p-doped layer is chosen to be higher than that of the i-layer in order to improve the entry of light into the undoped i-layer which is relevant for the light conversion and for electrical improvement. This can e.g. can be achieved with the aid of alloys of a-Si: H with N, C or O.

The efficiency y of a solar cell can be calculated as the product of the parameters of the current / voltage characteristic open terminal voltage _oc measured under sunlight, the short-circuit current density 'sc and the fill factor FF divided by the intensity of the

Sunlight I_ _s write:

V ^{= (v} oc * 3 _sc * FF) / I _S s- (1)

Various authors have described that, above all, the open terminal voltage V _oc , but also j _sc and FF of a-Si pj - n or nip solar cells can be changed by inserting an additional, thin, so-called buffer layer (English: " buffer (layer) "or" pi interface _ayer ") into the pi interface (ie between the p and i layers). This results in an overall significantly higher efficiency (Fig. La).

In stacked solar cells, e.g. With p-i-n-p-i-n structure, buffer layers can be used in several or even each sub-cell (FIG. 1b).

A buffer layer is a layer with a higher band gap than the subsequent i-layer, but the buffer layer, in contrast to the p-layer, is nominally undoped. The higher band gap can, as with the p layer, be achieved by alloying the a-Si: H with, for example, N, C or 0, but also by changing the material of the amorphous silicon without alloy. The increased band gap reduces the recombination in the p-i interface area and thereby increases the open terminal voltage.

Buffer layers with a graded band gap have also been described. "Nominally undoped" means that there is no active doping, even if due to carryover of dopants, e.g. from the p-deposition, one after the p-layer in the same

Reaction chamber prepared buffer layer can be slightly doped. An intentional low doping of the buffer layer is also known. Literature: RR Arya et al. describe an a-SiC buffer layer in which the carbon content is graded downwards from the p-doped a-SiC layer to the undoped a-Si i layer [1]. KS Lim et al. describe the same buffer structure, but with weak p-yp doping [2]. The best results so far have been achieved with "high quality" buffer layers in which the material used for the buffer is nominally undoped and has a photoconductivity with values above 1 * 10 (Ohm * cm) -l.

From T. Yoshida et al. a buffer made of a-SiC is known, the a-SiC material having a high photoconductivity. It was prepared using a high hydrogen dilution during plasma deposition [3]. Furthermore, from S. Fujikake et al. a nominally undoped a-SiO buffer is known, the a-SiO material being highly photoconductive and prepared using Sin _* , CO-, and H2 [4]. In addition to these special alloys, a-Si material with an increased band gap can also be used as a "high quality" buffer.

Such buffer layers, in particular those with "high quality", are successfully used to increase the efficiency of a-Si solar cells. It has been found, however, that the use of these buffers can induce increased aging (decrease in efficiency under long-term exposure to light).

In this context, it is on the one hand from Y.-M. Li et al. known that no additional aging was found when using undoped high-quality a-SiC buffer layers [51-

On the other hand, for example, P. Lechner et al. , but also by C. Beneking et. al. a significant increase in light aging was found in cells with high-quality buffer layers [6,7].

The prior art described in this regard is summarized in FIG. 2 for clarification: If the initial efficiency of a pin solar cell made of a-Si: H is equal to one, an initial efficiency greater than one (eg 1.1) is achieved by inserting the buffer layer.

This advantage can be maintained in the course of light aging ("good result", upper curve), but the buffer can also cause such an additional degradation that the initial advantage is more than lost.

In summary, it turned out to be a disadvantage that the "good result" described above (FIG. 2) cannot be achieved reproducibly with the structure p-buffer (high quality) -i-n.

3. Object of the invention

The object of the invention is therefore to provide a solar cell in which an increase in the initial efficiency is achieved with a simultaneous lower relative degradation and a significant increase in the (long-term) stabilized efficiency is achieved. It is also an object of the invention to provide a manufacturing method for such solar cells.

. solution

To solve the problem, the solar cell of the type mentioned at the outset has an additional, relatively thin, low p-doped layer as a compensation layer between the p and buffer layers (high quality) (FIG. 3 a). This measure can also be used multiple times in ^stacked cells (FIG. 3b).

The doping should be small compared to the doping of the p-layer, so that this layer does not act as an extension of the p-layer. In the case of a stack cell, the compensation layer can also be provided several times (FIG. 3b).

It was recognized that in p-i-n and p-buffer-i-n solar cells, charge carriers are generated in the i-layer of the solar cell by light irradiation and transported to the p- or n-layer by the built-in electric field. The efficiency of charge carrier separation depends on the strength of the built-in electric field and the defect density of the i-layer.

It was recognized that defect states in the part of the i or buffer layer adjacent to the p layer lie above the Fermini level and are emptied by electrons. Especially in the case of the p-buffer i-n cells, there are many defect states. A positive space charge zone is formed which, together with the negative space charge of the acceptors, forms an electrical double layer in the p-layer, by means of which the electric field in the i-layer is broken down. After degradation, the field is no longer sufficient for good charge separation, especially in cells with a buffer, which is why the efficiency in cells with a buffer decreases sharply over time.

It was recognized that by inserting a low p-doped layer at the beginning of the buffer, the positive space charge of the defect states is compensated for (hence "compensation layer").

This reduces the formation of the undesired electrical double layer and thus also the field degradation in the i-layer. This means that even after cells with a buffer have aged, there is still enough field in the i-layer to separate the light-generated charge carriers. The light-induced decrease in efficiency is therefore less.

The invention is explained in more detail below on the basis of exemplary embodiments. Show it:

FIG. 1 a shows a schematic representation of a solar cell layer structure with a pin structure without a buffer layer (i), or with a p-buffer-in structure with a buffer layer in the pi transition region (ii) to increase efficiency; Figure lb Schematic representation of a layer structure of a

Stack cell with p-i-n-p-i-n structure without buffer layer (i), or with p-buffer 1-i-n-p buffer 2-i-n structure with additional buffer layers (ii);

Figure 2 Relative efficiency of a-Si: H solar cells, normalized to the initial efficiency of a p-i-n cell without a buffer layer as a function of the irradiation time in

Hours for a pin structure without puf (a), for a p-buffer-in structure without inducing additional degradation by the buffer layer ("good result") (b) and for a p-buffer-in structure with additional Degradation through the buffer layer

("poor result") (c);

Figure 3 Schematic representation of a simple (a) and

Stacked solar cell (b) made of amorphous silicon with a "p _" compensation layer between the p and i layers;

Figure 4 Light aging of stacked solar cells with and without

Compensation layer between p and buffer layers;

5. embodiments; Test descriptions

The following solar cells were prepared:

5.1 single cells with the following layer structure:

a) p-i-n b) p-buffer-i-n c) p-compensation layer buffer-i-n

The p, i, and n layers were deposited in these three solar cells under comparable deposition conditions. The Layer thicknesses of these layers were chosen identically for the three solar cells (p: 10 nm, i: 500 n, n: 25 nm). The deposition parameters and layer thicknesses of the p, compensation and buffer layers are given in Table 1.

For comparison, each solar cell in Table 2 is in the unaged (before aging) and aged (after 300 h illumination under an AMI .5 spectrum, at 50 ° C and open clamps)

- the measured open circuit voltage V _oc in milli-volts,

- the short-circuit current density j _sc in milli-amperes per cm,

- the fill factor FF in%, and

- the efficiency y in%

specified.

5.2 Multiple cells

The compensation layer described above was also used to improve the efficiency of stacked cells.

The layer structure of this solar cell is p-compensation layer-buffer-i (60 nm) - n - p - compensation layer - buffer - i (330 nm) - n.

The comparison solar cell was also deposited in a stacked cell structure (p-buffer-i-n-p-buffer-i-n). The efficiency of both cells as a function of the irradiation time is shown in FIG. 4.

The exemplary embodiments show the result:

1. The increase in efficiency in the unaged condition by inserting a buffer layer (Table 2) from y = 8.0 to y = 9.0. 2. The increased degradation of p-buffer in cells without compensation layer (Table 2) from y = 9, l to y = 6.2.

3. The significant improvement in stability with a high initial efficiency by inserting the compensation layer, both in single and in stack cells (Table 2 and Figure 4) from y = 9.0 to y = 7.1 compared to y = 9, l to y = 6.2, or in the case of stacked cells from y = 10.3 to y = 9.0 in comparison to y = 9.9 to y = 7.5.

Literature:

[1] R.R. Arya, A. Catalano, and R.S. Oswald, Appl. Phys. Lett. 49 (1986) 1089

[2] K.S. Lim et al., J. Appl. Phys. 56 (1984) 538

[3] T. Yoshida, K. Maruya a, O. Nabeta, Y.

Ichikawa, H. Sakai, and Y. Uchida, Proc. 19 ^tn IEEE-PVSC, New Orleans, Louisana, 1987, p. 1095

[4] S. Fujikake et al., MRS Proc. Vol. 258 (1992) p. 875

[5] Y.-M. Li et al., Proc. MRS Spring Meeting, San Francisco, 1994, in press

[6] P. Lechner et al., MRS Proc. Vol. 149 (1989) p. 583

[7] C. Beneking et al. , Proc. 12 ^th EC PVSEC, Amsterdam, 1994, in press Table 1: Process parameters for p, compensation and buffer layers

p-layer compensation layer buffer

Gas flows TMB (2% in He): 7.5 TMB (2% in He): 2

(sccm) SiH4: 21 SiH4: 50 SiH4: 6

CH4: 25 CH4: 30 CH4: 2

H2: 1 19 H2: - H2: 170

Heater temperature 285 285 285

(° C)

Deposition pressure 1000 800 1500

(mTorr)

RF power 4 3.6 5

(W)

Deposition time 45 15 300

00

Layer thickness 10 2 10

(nm)

Table 2: Parameters of the test solar cells before and after aging (300 h, 1 sun, 50 ° C, open clamps).

Before aging p-i-n p-buffer-i-n p-compensation layer-buffer-i-n y (%) 8.0 9.1 9.0

Voc (mV) 820 885 885

Isc (mA / cm2) 13.8 14.5 14.3

FF (%) 70.4 70.6 71.2

After aging y (%) 6.3 6.2 7.1

Voc (mV) 812 845 860

Isc (mA / cm2) 13.6 13.2 13.9

FF (%) 57.4 55.5 59.4

Claims

Patent claims

1. Thin-film solar cell based on the amorphous, hydrogenated silicon and / or its alloys with a

a) an undoped layer (i),

b) one of these adjacent to the light entry side and connected to it puffs up,

c) one of these p-doped layer (p) and adjacent to the light entry side and connected to it

d) an n-doped layer (s) connected to this on the other side of the undoped layer

showing layer sequence (p-buffer-i-n),

d a d u r c h g e k e n n z e i c h n e t, d a ß

e) a compensation layer (p_) is provided between the p and buffer layers.

2. Thin-film solar cell according to claim 1,

d a d u r c h g e k e n n z e i c h n e t, d a ß

several layer sequences (p-buffer-in) are connected to one another and one or more compensation layers (p_) are provided.

3. Process for the production of a thin-film solar cell based on the amorphous, hydrogenated silicon and / or its alloys with a

a) an undoped layer (i),

b) a buffer layer adjacent to and connected to the light entry side,

c) one of these adjoining the light entry side and connected to it, μ-doped layer (p) and

showing layer sequence (p-buffer-i-n),

d a d u r c h g e k e n n z e i c h n e t, d a ß

e) a compensation layer (p_) is provided between the p and buffer layers.