SE456626B - SET TO MANUFACTURE SEMICONDUCTOR CELLS - Google Patents

Description

Any step following the hydrogen passivation should not cause the hydrogen introduced into the silicon to diffuse back out of the passivated substrate.

Thus, for example, it has been found that a hydrogen passivated cell, which has been subjected to a temperature of 600 ° C for half an hour in vacuum, loses substantially all of the bound hydrogen and returns to its previous level of passivation, as can be seen from the observation of an throne beam induced current activity. In this regard, it should be noted that the transition diffusion step in solar cell production typically involves temperatures of the order of 900 ° C.

It has also been found that hydrogen passivation normally heats the cell to a sufficiently high temperature to cause base metals, such as copper, to migrate through the junction, thereby causing a "soft" diode or short circuit. As shown, for example, by C. H.

Seager, DJ Sharp, JKG Panitz, and RV D'Aiello in Journal of Vacuum Science and Technology, vol 20, no. 3, pages 430 - 435 (March 1982), polycrystalline silicon passivation can be performed with a Kaufman-type ion source, which is used to generate a hydrogen ion beam in the kiloelectron voltage range. Relatively short exposure times (eg between 0.5 and 4 min) in an interval with high ion energy and large flux (eg 1 - 3 mA / cm 2) prove to be optimal. Such exposures generally result in the temperature of the substrate rising to at least about 275 ° C, if the substrate is carefully contacted with a suitable cooler. Otherwise, temperatures in excess of 400 ° C are easily achieved.

However, it is important that temperatures be limited to less than about 300 ° C to avoid rapid migration of base metals into the silicon matrix. However, manipulation of the substrate and coolant to achieve heat regulation during passivation easily becomes the rate limiting factor in highly productive treatment with such ion sources. Accordingly, it is desirable to avoid cooling in order to obtain a cheap, highly productive process. In addition, the surface irregularities of EFG-type silicon strips, which can be manufactured profitably, make cooling difficult.

In addition, hydrogen passivation is most effective when the silicon base surface is exposed. Thus, any plating mask, such as the silicon nitride layer used in prior art processes, must be removed before passivation. Consequently, the fronty electrodes must be deposited before passivation.

As described in application No. 563,061, the altered surface layer produced by hydrogen ion beam passivation can be used as a plating mask for subsequent metallization steps. More particularly, a preferred embodiment of the process described therein in detail, which is applied in the manufacture of silicon solar cells, includes the following steps: (1) forming a plating mask of a dielectric material on the front surface of a silicon strip with a shallow transition, in order to leaving these silicon areas exposed to be later covered by the front surface electrode, (2) depositing a thin layer of nickel (or similar material) on the exposed silicon, (3) removing the plating mask, (4) hydrogen passivating the transition side of the strip, (5 sintering the nickel to partially form a nickel silicide, (6) plating additional metal (s) on the metal-coated parts of the cell, and (7) anti-reflective coating of the exposed surface of the silicon. The silicon can then be further processed, for example by preparing it for connection to electrical circuits. The passivation changes the exposed surface of the transition side of the substrate so that it acts as a mask for the second plating step (6). in an alternative method, heating the sample during passivation provides at least some of the energy for the nickel sintering step.

This method is further simplified in application 456 626 4 with serial number 563 292 where it is taught that a negative plating mask (ie a mask covering only the areas of the front surface which are later to be covered by the front surface electrode) deposited on the front surface of the substrate can be used for to regulate the extent of the altered surface layer, which is formed by the ion beam, and thereby allow passivation before any front surface metallization, whereby the mask is removed after passivation and before metallization.

The two processes just described allow hydrogen passivation with the attendant improvement in cell performance of polycrystalline substrates, with a preferred temperature result. However, even the simplest of the two procedures just specified requires on the order of a dozen steps. In addition, both methods require that a surface layer on the substrate be formed in a temporary disposable mask, with an associated cost of both time and uncoverable material.

OBJECTS OF THE INVENTION Accordingly, it is an object of the present invention to eliminate the treatment steps in the manufacture of solar cells from EFG-type tapes or similar substrates.

A further object of the present invention is to provide a treatment sequence for the manufacture of solar cells, which treatment sequence comprises a hydrogen passivation step after the treatment steps at high temperature, but before any base metals are incorporated into the structure, which further minimizes the use of uncoverable material.

Brief Description of the Invention These and other objects are met by the present invention, where a removable and reusable mechanical mask is used to shield the ion beam used for hydrogen passivation on the front surface of the substrate. The altered surface layer, which is formed by the ion beam passing through the openings of the mask, forms a plating 456 626 mask, which delimits the areas of subsequent front surface metallization by immersion plating.

It will be appreciated that by using a removable mechanical mask, the areas of the substrate exposed to the ion beam are defined, thereby avoiding the chemical etching steps used to form and remove the plating masks according to the treatment sequences of the prior art. In addition, the mask of the present invention is reusable. As a result, a greatly simplified and more economical procedure is possible.

Other objects of the invention will be apparent in part and in part in the following. Accordingly, the invention includes the particular steps and the relationship between one or more of such steps with respect to each of the others, which steps are exemplified in the following detailed description, and in the scope of the claims set forth in the application.

Brief Description of the Drawing For a more complete understanding of the nature and objects of the present invention, reference should be made to the following detailed description, taken in conjunction with the accompanying drawings, which show a number of steps involved in the manufacture of solar cells in a preferred form. of the invention.

Throughout the drawing, the same reference numerals refer to a similar structure.

The drawings do not show the thicknesses and depths of the particular coatings and areas exactly in accordance with their relative proportions, in order to provide a more convenient illustration.

Detailed Description of the Invention With reference to the drawing, the preferred embodiment of the invention relates to the manufacture of solar cells of p-type EFG-cultured silicon bands. As a first process requirement, a side (hereinafter referred to as the "front side") of a pre-cleaned p-type type EFG silicon strip 2 is subjected to a phosphorus diffusion process which is calculated to give a relatively shallow transition 4. (ie a transition with a depth of between about 3000 and about 7000 Å), and a region 6 of the n-conducting type. As an example, a conductive type silicon strip, manufactured by the EFG method and having a resistivity of about 5 ohm-cm, is cleaned by etching in a solution of HNO 3 (70%): HF (49%) in a ratio of between about 4: 1 and 9: 1 for about 1 to 3 minutes at a temperature of about 25 ° C. Thereafter, the tape is subjected to a phosphorus diffusion process, which is well known in the art. A layer of phosphosilicate glass 8, formed as detailed in U.S. Patent No. 4,152,824, can be used, for example, as a source of the phosphorus dopant.

Then the layer of phosphosilicate glass 8 is etched away by immersing the substrate in a buffered HF solution.

Thus, (P 2 O 5) x (SiO 2) y, a phosphosilicate glass, can be removed from the substrate by immersing the latter in 10ONH 4 F (40%): 1HF at a temperature of between about 25 and about 40 ° C for a period of between about 15 s and 2 min.

Then the back of the substrate is coated with a layer of an aluminum paste. The aluminum paste used to form the layer 10 preferably comprises aluminum powders in a volatile organic carrier medium, such as terpineol, which can be removed by evaporation.

This substrate: E12.

H51! steps are then followed by an alloy step, _: _.-. h '- heated for about 0.25 - 2.0 min vi: is: temperature higher than about 575 ° C to deposit “= any volatile or pyrolysable organic components in the paste and to alloy the aluminum in the paste with the silicon substrate.

In the alloying step, the aluminum coating alloys with the back of the substrate to form a p + region 12 with a depth of from about 1 to about 5 microns. 456 626 7 The cell is then hydrogen passivated. A preferred way is to expose the front surface of the substrate 2 to the hydrogen ion beam from the Kaufman-type ion source (wide beam), which is located about 15 cm from the substrate. A mask 14 is placed between the ion source and the substrate. This ion source is preferably operated at a pressure of between about 20 and 50 millitorr (hydrogen), at a hydrogen flow rate of the order of about 25 - 40 cm 3 / minute, at a potential substrate, and 3 mA / cm of about 1700 volts DC between the source and and at a radiation flux of between about 1 2 on the substrate. An exposure time of between about 1 and about 4 minutes has been found to be suitable both to minimize the recombination loss of minority carriers, which typically occurs in EFG-type silicon cells, (a passivation zone with a depth of about 20 to 80 μm, or about 100 times as deep as transition 4 is achieved), while at the same time an altered surface layer 18 is achieved, with a depth of about 200 Å on the exposed parts of the substrate 2.

The exact nature of the altered surface layer 18 is not known. However, it is believed that there is a damaged zone, where the crystal structure has been somewhat interrupted, the silicon partially forming SiH or SiH2 with the hydrogen from the ion beam, and where the material is still amorphous. A small amount of carbon or one or more hydrocarbons proves to be necessary for the formation of the desired altered surface layer. As initially installed, the Kaufman ion source used was equipped with a graphite mounting device of about 13 cm (5 ") diameter on which the substrates, which were typically 5 cm to cm (2" to 4 ") on one side, were centrally located. In some cases, such as when a silicon mounting device replaced the graphite device, the altered layer did not appear as a plating mask, such as when the graphite device was used, and on this basis it has been assumed that carbon or hydrocarbon vapor formed by the action of hydrogen The ion beam on the graphite device, may form a dielectric layer on the surface of the substrate.In spite of its nature, it has been found that an altered surface layer 18, produced according to this method with acceleration voltages between about 1400 and about 1700 volts and exposure times as short as 1 min, are sufficient to prevent subsequent immersion plating metallization of the substrate on the altered layers 18.

The mask 14 is a metallic network with the desired multi-fingered electrode grid pattern, such as an electrode having the shape shown in U.S. Patent No. 3,686,036. The mask 14 is preferably made of molybdenum, although other metals such as invar, stainless steel, titanium, nickel or similar, or graphite and similar non-metallic materials at high temperature can also be used.

The mask 14 is relative to the front surface of the substrate 2 and the ion beam source is located so as to shield the desired electrode grid pattern on the front surface 20 when the ion beam source is activated. This means that the mask 14 obstructs the ion beam 16 in the desired electrode grid pattern, while allowing irradiation of the substrate on the areas between the electrodes.

Thereafter, metallization of the cell takes place. Both sides of the substrate are immersion plated with nickel, an adhesive deposit of nickel forming a nickel layer 22 on the back of the entire surface of the aluminum coating 10, while the adhesive deposit of nickel on the front forms a nickel layer 22 directly on the surface of the substrate 2 only in the areas free from altered surface layers 18. In this plating step, the altered surface layers 18 of silicon form a plating mask, to which nickel does not adhere. Immersion plating of the nickel layers can be done in different ways.

Preferably, they are performed according to a nickel immersion plating method which is the same as or similar to the method described in U.S. Patent No. 4,321,283.

As used herein, the term "immersion plating" 456 626 9 discloses a method in which an object is plated without the use of an externally applied electric field, by immersing it in a plating bath which does not contain a reducing agent, and the plating involves a displacement reaction. In contrast, the term "electroplating" refers to plating without the use of an externally applied electric field by immersing the object to be plated in a plating bath containing a reducing agent.

As a preliminary step, the cleaned silicon substrate surface is pre-activated by a suitable agent. This pre-activation process is desirable because the silicon surface 8 often does not itself favor the electrofree plating process, and any nickel plated on an untreated surface generally adheres only weakly thereto. Preferably, gold chloride is used as the activating agent, although platinum chloride, tin chloride-palladium chloride, or other well known activators may be used, as described, for example, in U.S. Patent No. 3,489,603.

Thereafter, both sides of the silicon belt are coated with a layer of nickel, preferably by immersing the belt in a water bath, as described in the aforementioned U.S. Patent No. 4,321,283, or in a water bath of nickel sulfamate and ammonium fluoride at a pH of about 2.9. and at approximately room temperature for a period of about 2 - 6 min.

After the nickel is applied, the substrate is heated in an inert atmosphere or in a mixture, such as nitrogen and hydrogen, to a temperature and for a time sufficient to sinter the nickel layers and cause the nickel layer 20 on the front side of the substrate to react with the adjacent silicon. to form an ohmic contact of nickel silicide. For this purpose, the substrate is preferably heated to a temperature of about 300 ° C for between about 15 and about 40 minutes. This produces a nickel silicide layer with a depth of about 300 Å at the interface between the nickel layer 20 and the substrate 2. The nickel layer 22 on the back forms an alloy with the aluminum layer 10. The temperature at this sintering step should not greatly exceed 300 ° C, as higher temperatures lead to a poor quality nickel layer, and, as previously noted, can cause some of the passivating hydrogen to diffuse back out of the substrate material.

Preferably, the deposition and sintering of nickel is controlled so that the nickel layer 20 on the front side of the substrate has a thickness of not more than about 750 Å.

Thereafter, the nickel in the layers and 22 is preferably subjected to etching, such as with nitric acid, and to further metallization, such as with a second layer of nickel and one or more copper layers. The additional nickel is applied by immersion plating, preferably in the manner previously described to form layers 20 and 22, since the added nickel is plated on layers 20 and 22, but not on the exposed areas of the altered layer 18.

Preferably, copper is applied by immersion plating and / or electroplating, by well known techniques.

No masking of the altered layer 18 is required, since copper formed by immersion plating or electroplating does not adhere to the altered surface 18.

After metallization, the cell edges (not shown) are trimmed and an anti-reflective coating 24 is applied to the front surface of the cell. This can be done by any of a number of known methods, such as by chemical vapor deposition or evaporation of, for example, TiO2. Alternatively, an anti-reflective coating 24 may be formed by plasma deposition of silicon nitride at a temperature of about 150 ° C, as is well known in the art.

The preferred method of applying the following invention includes, for example, performing the individual steps described above in the preferred manner described in detail for each step and in the described u :. l0 456 626 ll sequence.

It has been determined that solar cells manufactured according to the previous procedure of EFG-cultured tapes show between 10 and 20% increase in average efficiency.

In addition, the hydrogen passivation step of this material significantly reduces the distribution of cell efficiencies.

It will be appreciated that the process of the present invention greatly simplifies the production of solar cells. The method not only greatly reduces the number of steps required in the process but also eliminates the use of very uncoverable material. Thus, by eliminating the use of a photoresist to form a plating mask for the front surface electrodes, the present method eliminates the need to coat, expose, and develop such a resist, along with the required step of removing residual resist. Similar considerations apply to equivalent chemical etching procedures.

It will be appreciated that the method set forth in detail above may be modified without departing from the spirit of the invention. Thus, although the preferred embodiment of the method of the present invention utilizes an altered layer formed by hydrogen passivation to mask subsequent plating, in addition to previously plated nickel, the method may be used by those skilled in the art of the front metal other than nickel layer. As will be appreciated in the art, for example, the original surface electrodes of a shallow transition silicon device can be deposited by immersion plating some of a number of materials with low reactivity and capable of forming (preferably at low temperature) an ohmic contact and acting as a barrier to diffusion. of copper or any other base metal deposited at a later stage. Suitable metals for use with copper inbergiper palladium, platinum, cobalt and rhodium, as well as nickel. Although all of these materials form silicides, a silicide layer is not necessary. However, it is important that the original metal layer adheres properly, acts as an ohmic contact and acts as a barrier to the migration of any later deposited metal, as well as to insignificant migration to the transition itself.

Still other changes may be made without departing from the main principles of the invention, such as (a) forming the p + back region of the cell by using flame-sprayed aluminum instead of an aluminum paste, or (b) using different means of applying the second and the subsequent coatings of nickel or other low reactivity materials, such as palladium, platinum, cobalt and rhodium, or (c) formation of the transition by ion implantation. If no masking layer is deposited on the altered layer, the added nickel (or other low reactivity metals of the type described) must be applied by immersion plating.

Of course, the process accomplished by this invention is not limited to the production of solar cells from EFG substrates. Thus, for example, cast polycrystalline substrates, epitaxial silicon upon metallurgical type silicon or fine type polysilicon layer formed by chemical or physical vapor deposition can be used to form relatively high efficiency solar cells according to the present invention. In addition, the process is applicable to single crystalline silicon.

Then the method can also be applied with both n-type and p-type substrates.

In any of these cases, it will be appreciated that the transition can be formed by various methods, and not merely by phosphorus diffusion. Since these and other compounds can be manufactured by the above-mentioned processes without departing from the meaning of the invention contained herein, it is intended that all material included in the above description or shown in the accompanying drawings be construed as illustrative. and not in a limiting way.

Claims (7)

456 626 10 15 20 25 30 14 PATENT REQUIREMENTS A method of manufacturing semiconductor devices in a solid state, characterized in that the method in sequence comprises the steps of: (a) providing a silicon substrate having opposite first and second surfaces; (b) applying a mechanical mask with selected openings directly to the first surface; (c) exposing the first surface to a hydrogen ion beam through the selected apertures at an intensity and for a duration sufficient to reduce the losses of minority carriers in the substrate and to form on the first surface a surface layer corresponding to the apertures and to which metals only weakly adhere to, and to metallize the parts of the front surface which are masked with the mask. 2. A method according to claim 1, characterized in that the entire front surface is exposed to metal during metallization. A method according to claim 1, characterized in that it further comprises the step of forming a pn junction in the substrate adjacent the first surface before exposing the surface to the hydrogen jet. 4. A method according to claim 1, characterized in that the metallization is carried out using a metal selected from the group of metals including nickel, palladium, cobalt, platinum and rhodium. 5. A method according to claim 1, characterized in that it further comprises the step of immersion plating to metallize the parts of the front surface which were masked by the mask. 6. A method according to claim 5, characterized in that the metallization step is carried out in such a way that it also metallises the second surface of the substrate. 123 456 626 15 A method according to claim 1, characterized in that the metallization comprises the deposition of nickel from a bath containing a nickel salt and fluoride ions.