Patent Application
20110156221
The present invention relates to a process for producing a passivation layer on crystalline silicon for the production of solar cells by transformation of a thin (10-200 nm) coating which contains perhydropolysilazanes (also referred to as PHPSs) or an organic polysilazane as main constituent and serves as hydrogen source for the volume passivation during the transformation. The transformation is carried out at temperatures of from 200 to 1000° C.
In view of the increasing scarcity of fossil energy sources, photovoltaics is of very great importance as alternative technology for the provision of energy. To achieve substantial and more rapid utilization of the technology, the production costs of photovoltaic modules have to be minimized. To reduce these costs, the efficiency of solar modules has to be increased and cheaper production technologies have to be developed.
An increase in the efficiency of industrially manufactured solar cells is achieved, inter alia, by suppression of electrical losses which very largely take place by recombination of charge carriers at defects in the silicon both in the volume and at the surfaces.
To passivate the surfaces, dielectric layers composed of thermally grown SiO2 or PECVD-SiNx are applied. Application is effected on the n-conducting silicon layer and can additionally be effected on the p-conducting silicon layer (rear side passivation). It is known that hydrogen makes it possible to deactivate volume defects such as stretched bonds or charged defects and thus improve the electrical properties of solar cells. Industrially, the volume passivation of silicon solar cells is effected by diffusion from hydrogen-rich layers.
In the case of silicon nitride and silicon dioxide layers produced industrially in CVD processes, hydrogen is incorporated during deposition on the silicon surface. Apart from hydrogen, ammonia (NH3) also serves as hydrogen source (64Widenborg, P. I., A. B. Sproul, and A. G. Aberle. Impurity and defect assivation in poly-Si films fabricated by aluminum-induced crystallisation. in Proc. 3rd WC PVSEC. 2003. Osaka.) When dissociated completely, NH3 provides three hydrogen atoms per molecule instead of the two atoms of an H2 molecule.
During heat treatment, i.e. heating of these layers to temperatures of >600° C., diffusion of the hydrogen from the layers into the silicon volume occurs. In the screen printing solar cell process, hydrogen diffusion through the hydrogen-containing layer on the Si surface takes place during firing of the contacts.
The disadvantages of applying passivation layers by CVD technologies are high costs for the complicated vacuum technique and the use of highly flammable (SiH4, CH4, H2) and toxic (NH3) gases.
The use of polysilazanes for the formation of SiOx and silicon nitride layers has been described.
JP 05243212 A describes polysilazane films having good covering properties and their use for coating semiconductor devices. Sintering and oxidation at relatively low temperatures results in formation of a passivation layer comprising a fine hygroscopic silicon oxide film.
The formation of antireflection coatings composed of polysilazane is likewise known.
Thus, JP-A-2005033063 describes polysilazane solutions which are applied to silicon single crystal substrates at normal temperatures by the spin coating method and are then fired at 600-800° C. in a nitrogen-containing atmosphere. In this way, it is possible to produce a low-reflection film having silicon nitride as main component on a solar cell.
It is therefore an object of the invention to provide a process for passivating crystalline silicon for solar cell production which allows expensive vacuum techniques to be circumvented and carry out a surface and volume passivation of silicon in a simple and economical way.
The present invention achieves the object and provides a process for passivating crystalline silicon by means of polysilazanes, by
a) coating a silicon wafer with a solution containing at least one polysilazane or a mixture of polysilazanes of the formula 1,
—(SiR′R″—NR′″)n- (1)
where R′, R″, R′″ are identical or different and are each, independently of one another, hydrogen or a substituted or unsubstituted alkyl, aryl, vinyl or (trialkoxysilyl)alkyl radical and n is an integer and n has such a value that the polysilazane has a number average molecular weight of from 150 to 150 000 g/mol,
b) subsequently removing the solvent by evaporation, resulting in polysilazane layers having a thickness of 50-500 nm remaining on the silicon wafer, and
c) heating the polysilazane layer to 200-1000° C. at atmospheric pressure in the presence of air or nitrogen, resulting in the ceramic layers liberating hydrogen during the heat treatment to effect volume passivation of the silicon.
In a preferred embodiment, the coating according to the invention contains at least one perhydropolysilazane (in which R′, R″, R′″=H).
Polysilazanes in which R′, R″, R′″ are each, independently of one another, a radical from the group consisting of hydrogen, methyl, ethyl, propyl, isopropyl, butyl, isobutyl, tert-butyl, phenyl, tolyl, vinyl and 3-(triethoxysilyl)propyl, 3-(trimethoxysilylpropyl) are particularly useful here.
In a further preferred embodiment, the coating according to the invention contains polysilazanes of the formula (2),
—(SiR′R″—NR′″)n-(SiR*R**—NR***)p— (2)
where R′, R″, R′″, R*, R** and R*** are each, independently of one another, hydrogen or a substituted or unsubstituted alkyl, aryl, vinyl or (trialkoxysilyl)alkyl radical and n and p have such values that the polysilazane has a number average molecular weight of from 150 to 150 000 g/mol.
Particular preference is given to compounds in which
Preference is likewise given to using polysilazane of the formula (3)
—(SiR′R″—NR′″)n-(SiR*R**—NR***)p—(SiR1,R2—NR3)q— (3)
where R′, R″, R′″, R*, R**, R***, R1, R2 and R3 are each, independently of one another, hydrogen or a substituted or unsubstituted alkyl, aryl, vinyl or (trialkoxysilyl)alkyl radical and n, p and q have such values that the polysilazane has a number average molecular weight of from 150 to 150 000 g/mol.
Very particular preference is given to compounds in which
R′, R′″ and R*** are each hydrogen and R″, R*, R** and R2 are each methyl, R3 is (triethoxysilyl)propyl and R1 is alkyl or hydrogen.
In general, the proportion of polysilazane in the solvent is 1-80% by weight of polysilazane, preferably 2-50% by weight, particularly preferably 3-10% by weight.
Suitable solvents are, in particular, organic, preferably aprotic solvents which contain no water and also no reactive groups such as hydroxyl or amino groups and are inert toward the polysilazane. Examples are aromatic or aliphatic hydrocarbons and mixtures thereof.
The solvents can be, for example, aliphatic or aromatic hydrocarbons, halogenated hydrocarbons, esters such as ethyl acetate or butyl acetate, ketones such as acetone or methyl ethyl ketone, ethers such as tetrahydrofuran or dibutyl ether and also monoalkylene and polyalkylene glycol dialkyl ethers (glymes) or mixtures of these solvents.
Additional constituents of the polysilazane solution can be catalysts such as organic amines, acids or metals or metal salts or mixtures of these compounds which accelerate the layer formation process. The catalysts are preferably used in amounts of from 0.001 to 10%, in particular from 0.01 to 6%, particularly preferably from 0.1 to 3%, based on the weight of the polysilazane.
Further constituents can be additives for substrate wetting and film formation or inorganic nanoparticles such as SiO2, TiO2, ZnO, ZrO2 or Al2O3.
The process of the invention makes it possible to produce an impervious, vitreous layer which owing to its freedom from cracks and pores and the high hydrogen content is excellent for passivation.
The layers produced preferably have a thickness of from 10 to 200 nm.
The layers produced have a hydrogen content at the beginning of ceramicization and this is driven from the layer with increasing ceramicization temperature.
The process of the invention makes it possible to drive the hydrogen from the amorphous polysilazane layer which has been applied in a first step over a period of from 0.1 to 30 minutes at temperatures of 200-1000° C. and convert the layer into a transparent ceramic phase. Depending on the ceramicization conditions, whether in air or a nitrogen-containing atmosphere, the ceramic phase formed has a different composition.
Thermal initiation of the transformation of the polysilazane framework in air results in the hydrogen passivation of the silicon substrate occurring with transformation into a transparent ceramic phase of the composition
SiuNvHwOxCy (4)
where u, v, w, x, y are atom % and, depending on the ceramicization temperature, phases of the composition in which x>v, where v<1 and x<1.3 and are not equal to zero, and w=2.5-0 and y<0.5, with u in each case being 1, are formed.
Thermal initiation of the transformation of the polysilazane framework in a nitrogen atmosphere results in the hydrogen passivation of the silicon substrate occurring with transformation into a transparent ceramic phase of the composition
SiuNvHwOxCy (4)
where u, v, w, x, y are atom % and, depending on the ceramicization temperature, phases of the composition in which v<1.3 and x<0.1 and also w=2.5-0 and y<0.2, with u in each case being 1, are formed.
The liberation of the passivating hydrogen can, according to the invention, occur in an air or nitrogen atmosphere in each case alone, in succession or alternatingly.
The heat input can be effected by means of a heatable furnace which is operated under nitrogen or air and can be regulated in the range 200-1000° C.
The coatings according to the invention are thus suitable not only for surface passivation but also for the hydrogen passivation of the silicon substrate. The process known hitherto from the prior art requires complicated vacuum methods for this purpose.
Spin coating
Transformation of polysilazane into a ceramic of the formula (4) was carried out at 200-1000° C. in a furnace which can be operated under air or nitrogen.
The layer thicknesses formed are 10-200 nm.
The measurements were carried out by means of ellipsometry.
p-type silicon: a 1 Ωcm FZ p-type silicon wafer having a thickness of 250 μm was used as substrate. After standard RCA cleaning, the various polysilazane layers were applied.
n-type silicon: the POCl emitter diffusion on an FZ silicon wafer having a thickness of 250 nm was carried out in a standard furnace and led to emitter layer resistances of 40 Ω/sqr.
The life was determined according to the principle of the decrease in photoconductivity (QSSPC) on a Sinton Apparatus model WCT-100. The greater the life, the better the passivation.
The layer composition was determined on a polysilazane powder which had in each case been ceramicized under the same conditions as on the silicon.
| Number | Date | Country | Kind |
|---|---|---|---|
| 10 2008 044 769.2 | Aug 2008 | DE | national |
| PCT/EP2009/006160 | Aug 2009 | EP | regional |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/EP09/06160 | 8/26/2009 | WO | 00 | 2/23/2011 |
