JP2005167291A - Solar cell manufacturing method and semiconductor device manufacturing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To obtain a manufacturing method of a solar cell with a structure in which electric separation of pn junction can be realized by simple method and manufacturing cost and productivity can be improved remarkably, and a semiconductor device manufacturing method having a new electrode forming method when using metal paste material in a structure body having a silicon nitride film or a titanium oxide film on a surface of a substrate. <P>SOLUTION: By preparing and sintering glass paste 104, which is made of glass which has property to dissolve silicon, as the main component on an n-type diffusion layer 101 of pn junction, the n-type diffusion layer 101 is eroded by the glass paste 104. A p-type inversion layer 105 is formed by diffusing aluminum to an n-type diffusion layer 101 under a P electrode 103 which is made of aluminum silver paste and inverting to p-type. Thereby, electric separation of pn junction is realized. And moreover, by preparing and sintering metal paste material on an insulating film, the insulating film is penetrated so as to electrically contact with a semiconductor substrate. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a method for manufacturing a solar cell and a method for manufacturing a semiconductor device. In particular, the electrical isolation of a pn junction necessary for forming an electrode of a silicon solar cell is realized by a simple and mass-productive method. Semiconductor that forms an electrode capable of greatly reducing contact resistance with a semiconductor substrate by applying a physical phenomenon similar to a method of manufacturing a solar cell and an electrical isolation method of a pn junction in forming an electrode of a solar cell The present invention relates to a device manufacturing method.

Currently, the mainstream of power solar cells used on the earth is silicon solar cells, but it is common to attempt to reduce the manufacturing cost by simplifying the process flow at the mass production level as much as possible.
Hereinafter, an example of the manufacturing method will be described with reference to FIG.
In the p-type Si substrate 1 shown in FIG. 17A, the n-type diffusion layer 2 in which the conductivity type is inverted is formed by thermally diffusing phosphorus (P), for example, in FIG. Usually, phosphorus oxychloride (POCl 3) is often used as a phosphorus diffusion source. Further, the n-type diffusion layer 2 is formed on the entire surface of the p-type Si substrate 1 unless otherwise specified. The n-type diffusion layer 2 has a sheet resistance of about several tens of ohms / square and a depth of about 0.3 to 0.5 μm.

  Further, although not described in detail, the n-type diffusion layer 2 is formed on only one main surface (incident light surface) as shown in FIG. The resist is removed using an organic solvent or the like. Thereafter, the aluminum paste 3 is printed on the opposite surface of the n-type diffusion layer 2 in FIG. 2C by, for example, screen printing, and then baked in a near infrared furnace at 700 to 900 ° C. for several minutes to several tens of minutes. As a result, aluminum is diffused as impurities from the aluminum paste into the p-type Si substrate 1, and a BSF (Back Surface Field) layer 4 made of a p + layer containing high concentration impurities is formed as shown in FIG. The This BSF (Back Surface Field) layer 4 contributes to the improvement of the energy conversion efficiency of the solar cell.

  Subsequently, as shown in FIG. 5 (e), on the n-type diffusion layer 2 formed only on one main surface, as an n electrode 5, a comb-shaped silver paste electrode pattern is usually formed by screen printing or the like. A solar cell is completed by baking at 700-800 degreeC. Here, for the sake of simplicity, the description with reference to the drawings is omitted, but an antireflection film such as TiO 2, SiN, or SiO 2 is formed on the n-type diffusion layer 2 for the purpose of further improving the conversion efficiency. There are many.

  In general, a solar cell is manufactured by the above-described process. In FIG. 17F, the n-type diffusion layer 2 on the side surface of the p-type Si substrate 1 is formed on the n-type diffusion layer 2 shown in FIG. The solar cell structure when p and n electrodes are formed while remaining is shown. In this case, the portion indicated by reference numeral 6 in the figure is a p + / n junction, but since the impurity (phosphorus) concentration in the vicinity of the surface of the n-type diffusion layer 2 is as high as 1020 to 1021 cm −3, In particular, the structure is likely to cause leakage, and the conversion efficiency of the solar cell is greatly reduced. That is, as shown in FIG. 17C, a solar cell having good characteristics cannot be produced unless the n-type diffusion layer 2 is formed only on one surface (incident light surface). In the example shown here, as described above, the steps of surface protection of the diffusion layer by resist, etching, and resist removal are required, but such a process needs to be simplified as much as possible in order to reduce the manufacturing cost. is there.

  Therefore, the present invention has been made in view of the above-described points, and electrical separation of the pn junction necessary for the configuration of the solar cell can be realized by a very simple method, and the manufacturing cost and productivity are greatly improved. It is an object of the present invention to obtain a method for manufacturing a solar cell having a structure that can be used.

  Also, the manufacture of a semiconductor device capable of forming an electrode capable of significantly reducing contact resistance with a semiconductor substrate by applying a physical phenomenon similar to a method of electrically separating pn junctions when forming an electrode of a solar cell. The purpose is to obtain a method.

  The method for manufacturing a solar cell according to the present invention comprises a silicon layer serving as a power generation layer and a thin film layer having a conductivity type opposite to the silicon layer formed on the entire surface including the front surface, both surfaces of the front surface and the back surface of the silicon layer, or the edges. In the method for manufacturing a silicon solar cell having a pn junction, a step of electrically separating the pn junction by providing a material mainly composed of glass having a property of melting silicon on the pn junction and firing the material. And after the step of firing the material containing glass as a main component, the p-electrode and n-electrode required for collecting solar cells are formed on the surface of the pn junction or both the front and back surfaces, and the glass is the main component. Forming and firing outside the region on the pn junction where the material to be formed is formed, and as the p electrode and the n electrode, silver, aluminum silver, aluminum paste, etc. Characterized by using a combination of fee.

  According to another aspect of the present invention, there is provided a method for manufacturing a semiconductor device, comprising: forming a semiconductor substrate having a junction, and forming an insulating film on both the front and back surfaces of the semiconductor substrate; Providing a metal paste material containing glass having a melting property and firing it to form an electrode penetrating the insulating film and making electrical contact with the semiconductor substrate. In the case of a single layer, a silicon nitride film is used, and in the case of a plurality of layers, a silicon nitride film is used as at least one of the layers.

  According to the present invention, electrical separation of the pn junction necessary for the configuration of the solar cell can be realized by a very simple method, and the manufacturing cost and productivity can be greatly improved.

  In addition, an electrode that can greatly reduce the contact resistance with the semiconductor substrate can be formed by applying a physical phenomenon similar to the electrical separation method of the pn junction in forming the electrode of the solar cell.

  For example, lead boron glass is added to a silver paste used for an electrode material used for a normal silicon solar cell. This glass is frit-like, for example, composed of lead (Pb) 5-30%, boron (B) 5-10%, silicon (Si) 5-15%, oxygen (0) 30-60%, Furthermore, zinc (Zn), cadmium (Cd), etc. may be mixed about several percent.

  Such lead boron glass has a property of melting by heating at several hundred degrees C. (for example, about 700.degree. C.) depending on the composition and eroding silicon at that time. For details of such a phenomenon, for example, GCCheek et al. (IEEE Transactions on Electron Devices, vo1. ED31, No. 5, 1984, pp.602-609) and R. Mertens et al. (Conference Record of the 17th IEEE Photovoltaic Specialists Conference, 1984, pp.1347-1351).

In general, electrical contact between silicon and silver paste is obtained by utilizing such characteristics of glass frit. Therefore, as a point to keep in mind, for example, the glass frit amount and the firing temperature are devised so that the glass frit in the silver paste erodes the n-type diffusion diffusion layer 2 shown in FIG. Yes.
The point of this invention is that the electrical separation of the pn junction can be easily performed by actively utilizing the properties of the glass frit.

The present invention will be described below with reference to the drawings.
In the present invention, a suitable amount of diethylene glycol monobutyl ether and ethylene glycol monomethyl ether are mixed in the lead boron glass having the above-described composition, adjusted to a predetermined viscosity capable of screen printing, and pasty. (Hereinafter referred to as glass paste).

Embodiment 1 FIG.
FIG. 1 is an example of a flowchart of the most basic solar cell manufacturing process according to the present invention.
That is, in the present invention, first, a prepared concavo-convex structure called a texture is formed on the surface by etching silicon with an alkaline aqueous solution such as sodium hydroxide having a concentration of 1 to several percent on a prepared p-type silicon substrate, for example. By causing multiple reflection of light on the surface of the solar cell, the reflectance is effectively reduced and the conversion efficiency is improved (steps S1 and S2). Note that the present invention can be applied to any case without this texture structure without any problem.

  Next, an n-type diffusion layer is formed by thermal diffusion of phosphorus with POCl3 (step S3), and an antireflection film is formed on the surface of the n-type diffusion layer (step S4). Then, after passing through the printing / drying step (step S5) of the light receiving surface silver paste electrode, the printing / drying step (step S6) of the back surface aluminum paste electrode, and the printing / drying step (step S7) of the back surface glass paste, firing is performed. Thus, the solar cell is completed (steps S8 and S9). Note that the present invention can be applied without any problem even when the antireflection film is not formed on the light receiving surface side. Here, there is no problem even if the printing order of the front electrode and the back electrode is changed.

It will be described with reference to FIG. 2 that a solar cell can be manufactured by this process flow.
First, FIG. 2 (a) shows that silicon can be eroded by glass paste.
That is, a linear glass paste pattern having a width of 400 μm is formed on the surface of a p-type single crystal silicon substrate by screen printing, dried for 10 minutes at 100 ° C. in dry air, and then several times at 650 ° C. in dry air. The amount of silicon erosion by the glass paste after baking for a minute was measured with a surface roughness meter. The evaluation of the amount of erosion is carried out after removing the fired glass paste with hydrofluoric acid. Etching removability of glass paste is somewhat poor and can be easily removed by applying external force, for example, applying ultrasonic waves.

  As a result, as shown in FIG. 2A, it can be seen that silicon is eroded by about 0.5 μm. As shown below, for example, if the depth of the diffusion layer in which phosphorus is thermally diffused using POCl3 as a raw material is set to 0.3 μm, it can be said that the condition is sufficient for erosion and removal of this diffusion layer. This junction depth can be controlled with good reproducibility by adjusting the diffusion temperature and time.

FIG. 2 (b 1) and FIG. 2 (b 1) are a top view and a cross-sectional view of a sample for evaluating whether or not pn junction can be electrically separated by this method.
In the figure, 100 is a p-type silicon substrate having a resistivity of about 2 Ωcm, 101 is an n-type diffusion layer in which phosphorus is thermally diffused by POCl 3, and has a thickness of about 0.3 μm by SIMS (Secondary Ion Mass Spectroscopy) evaluation. Confirm that there is. Further, 102 is an n electrode made of silver paste, 103 is a p electrode made of aluminum silver paste containing several percent of aluminum, and 104 is a glass paste. Here, an aluminum silver paste is used as the p-electrode 103, but an aluminum paste containing no silver may be used, and it has been confirmed that the effects described below can be obtained similarly.

  FIG. 2D shows the current of the electrode between the n electrode 102 made of silver paste and the p electrode 103 made of aluminum silver paste after baking the sample shown in FIG. 2B 12 for several minutes at 700 ° C. (Vertical axis) One voltage (horizontal axis) characteristic (denoted with Glass Paste in the figure) and only the n electrode 102 made of silver paste and the p electrode 103 made of aluminum silver paste were patterned without forming a glass paste pattern. The current-voltage characteristics (indicated as “without Glass Paste” in the figure) are shown. As is clear from this result, clear diode characteristics (rectification) are obtained by the presence of the glass paste 104, and it can be seen that this method is extremely effective for electrical separation of pn junctions. Of course, the glass paste 104 after firing is an insulator, and even if the glass paste 104 is removed, the diode characteristics shown in FIG. 2D do not change.

  Although the figure numbers are mixed, it is clear from this result that the sample after firing the sample of FIG. 2 (b-12) has a shape as shown in FIG. 2 (c). As shown in FIG. 2 (c), the glass paste 104 erodes the n-type diffusion layer 101, and aluminum diffuses into the n-type diffusion layer 101 under the p-electrode 103 made of aluminum silver paste. A p-type inversion layer 105 in which the layer is inverted to p-type is formed.

  In the present embodiment, the n-type diffusion layer 101 is formed by thermally diffusing phosphorus with POCl 3, but other methods such as phosphorus or arsenic ion implantation or silane-based gas such as monosilane (SiH 4), A mixture of disilane (Si2H6), dichlorosilane (SiCl2H2), trichlorosilane (SiCl3H), tetrachlorosilane (SiCl4) and phosphine (PH3) gas as an impurity source is decomposed by heat or plasma or other external energy. The same effect can be obtained even if an n-type diffusion layer formed in this way is used.

  Furthermore, in order to improve the characteristics by suppressing the surface recombination of the solar cell, the window effect, etc. by using the pn junction as a heterojunction, for example, silicon carbide (SiC) formed from a mixed gas such as monosilane and methane or ethane is used. Also good. In addition, the thin film formed by such a method can be made amorphous, microcrystalline, polycrystalline, or single crystal by changing the forming conditions such as the film forming temperature. If the thickness is several hundred angstroms to several μm, the same effect can be obtained by devising the glass paste composition and firing conditions.

  In this embodiment, as an example of means for realizing electrical isolation of the pn junction, as shown in FIG. 2 (b-12), a method of firing after patterning all of the electrode / glass paste is performed. As shown, electrode formation after patterning / firing of glass paste, electrode formation after patterning / firing / etching removal of glass paste, glass paste patterning / firing after patterning / firing of electrode, etc. It is possible to achieve electrical isolation of the pn junction as described above by various methods. Moreover, as specific firing temperature conditions, it has been confirmed that the above-described pn junction separation effect can be obtained within a range of 600 to 950 ° C. for several minutes to several tens of minutes.

FIG. 3 shows a process diagram for actually producing a solar cell based on the above basic experimental facts, and corresponds to the flowchart shown in FIG.
The structure of the solar cell in FIG. 3 is the same as that described in the conventional example shown in FIG. Here, as the p-type silicon substrate 100 shown in FIG. 3A, a substrate having a thickness of 600 μm, a size of 10 cm × 10 cm, a specific resistance of 20 Ωcm, and a plane orientation (100) was used. The n-type diffusion layer 101 shown in FIG. 3B is formed by thermal diffusion of phosphorus with the above-described POCl 3. As diffusion conditions, for example, a diffusion layer having a diffusion layer sheet resistance of about 50Ω / □ and a junction depth of 0.3 μm was formed by treatment at 860 ° C. for 10 minutes.

  Here, the point is that the glass paste 104 is formed at the positions shown in FIGS. 3C and 3D, and the pattern width of the glass paste 104 is, for example, 0.5 to 2 mm. The shape of the n-electrode 107 made of silver paste formed on the light-receiving surface (incident light surface) was a comb pattern as shown in FIG. 3E (electrode pitch is 2.5 mm, for example). Here, the n electrode 107 and the glass paste 104 were formed by baking for 7 minutes at 700 ° C. in dry air.

  The characteristics of the solar cell produced by this method were inferior to those of the conventional method shown in FIGS. 17 (a) to 17 (e). This is because the BSF layer 4 and the n-type diffusion layer 101 are electrically separated from each other by the glass paste 104, as indicated by the edge portion 115 in FIG. Although not shown in FIG. 3, as shown in the flowchart of FIG. 1, after providing an antireflection film on the surface of the n-type diffusion layer 2, an n electrode 107 made of silver paste formed on the light receiving surface, Even when the p electrode 103 made of aluminum paste and the glass paste 104 formed on the back surface are formed, the solar cell can be manufactured in exactly the same manner.

  In this case, the electrical contact with the n-type diffusion layer 101 is realized by the phenomenon that the glass component contained in the silver paste of the n-electrode 107 breaks through the antireflection film during firing (fire-through). Here, as the antireflection film, for example, a silicon nitride film (SiNx), titanium oxide (TiOx), silicon oxide film (SiOx), or the like formed by using a CVD method, a sputtering method, or the like has a thickness of several hundreds. To about 1000 angstroms.

Although not shown in FIG. 3, as shown in the flowchart of FIG. 1, generally, after providing an uneven structure called texture (see Embodiment 6 described later) on the substrate surface, the above-described process method is performed. Even if applied, there is no hindrance to this invention.
Further, as a method for forming the antireflection film, for example, when a silicon nitride film is formed by a low pressure CVD method, a silicon nitride film is also formed on the back surface side of the substrate.
However, in the present invention, it was also confirmed that a solar cell can be manufactured with the above-described manufacturing flow as it is without removing the silicon nitride film formed on the back surface as in this example. This is because the glass component 104 formed on the back surface and the glass component contained in the silver paste of the n-electrode 107 fire through the silicon nitride film on the back surface, thereby electrically isolating the pn junction and the electrode semiconductor. This is because contact can be realized. Of course, the same applies not only to the silicon nitride film but also to other antireflection film materials.

As described above, according to this embodiment, the unnecessary step of removing the diffusion layer formed on the back surface of the substrate as described in the conventional example is completely unnecessary, and the process can be greatly simplified. .
It goes without saying that the various procedures described above can be selected as the process flow for forming the glass paste 104 and the electrode paste. In this embodiment, even when the conductivity type of the substrate is reversed, the present invention can be applied based on the same idea as that of the fourth embodiment described later.
Furthermore, in order to form the electrode and the glass paste pattern in the present invention, a roll coater type printing method may be used in addition to the screen printing described above.
The various process methods described above can be similarly applied to the realization of some embodiments described below, and can be selected depending on the structure of the solar cell and the crystal quality of silicon.

Embodiment 2. FIG.
FIG. 4 explains a solar cell manufacturing process and structure for explaining the second embodiment of the present invention. Hereinafter, the manufacturing process will be described in order with reference to the drawings.
FIG. 4A shows a state in which the pattern of the glass paste 104 is formed on the p-type silicon substrate 100 on which the n-type diffusion layer 101 is formed, as in the first embodiment. FIG. 4B shows a state where the glass paste 104 is removed by etching after firing the sample shown in FIG. Here, the n-type diffusion layer 101 is removed by the mechanism shown in the first embodiment, and the p-type silicon substrate surface 106 is exposed as shown in FIG. 4B.

FIG. 4C shows a state in which an n electrode 102 made of silver paste is formed on an n type diffusion layer 101 on which the glass paste 104 has not been patterned, and a p electrode 103 made of aluminum silver paste is formed on the p type silicon substrate surface 106. It is. These electrode pattern pitches are about 1 to 3 mm.
FIG. 4D is an n-type formed on the surface opposite to the surface on which the process is performed in FIGS. 4A to 4C (which becomes the main incident light surface (light receiving surface) when functioning as a solar cell). The state which formed the electrode to the diffused layer 101 is shown. Here, 107 is an n electrode made of silver paste formed on the main incident light surface side, and the pattern pitch is about 2 to 3 mm.

  Here, the electrodes 102, 103, and 107 may be fired separately. However, in order to simplify the process, the electrodes 102, 103, and 107 may be continuously printed and then fired at the same time. Good. The n-type diffusion layer 101 to be formed on the main incident light surface side can be formed by various methods described in the first embodiment, and the order of formation is also the n-type diffusion layer 101 on the back surface side. Or after formation of the n-type diffusion layer 101 on the back surface side.

The schematic diagram of the electrode pattern of the solar cell completed by the above is shown to FIG.4 (e) and (f).
The solar cell having this structure has a structure having an n-electrode 107 on the main incident light surface side and an n-electrode 102 and a p-electrode 103 on the back surface (opposite surface), which is effective for improving the conversion efficiency of the solar cell. Although known as one (for example, T. Warabisakoand K. Matsukuma, Technical Digest of the 7th lnternational Photovoltaic Science and Engineering Conference, Nagoya, Japan, 1993, pp. 57), According to the present invention, it can be realized by a simple process.

Embodiment 3 FIG.
FIG. 5 is a process diagram illustrating a manufacturing method according to Embodiment 3 for realizing the structure of the solar cell of Embodiment 2 described above.
In the above-described second embodiment, the method of removing the glass paste 104 has been described in order to electrically isolate the pn junction. However, as already described in the first embodiment, it is not always necessary to remove the glass paste 104. Absent. Here, an embodiment of a solar cell manufacturing process when the glass paste 104 is not removed will be described.

5A, in the same manner as in the first embodiment, the silver paste pattern of the n electrode 102 and the aluminum silver paste pattern of the p electrode 103 are printed on the p type silicon substrate 100 on which the n type diffusion layer 101 is formed. The state after firing is shown, and 105 is a p-type inversion layer equivalent to that already described.
FIG. 5B shows a state in which a glass paste 104 is printed for electrical separation between the n-electrode 102 and the p-electrode 103, and FIG. 5C shows that the pn junction is electrically separated by firing this. FIG. 5 (d) shows the state after the n electrode 107 pattern made of silver paste is formed on the main incident light surface side. As shown here, the electrodes 102, 103 and 107 and the glass paste 104 may be fired separately, but for the sake of simplification of the process, the electrodes 102, 103 and 107 are fired simultaneously after printing. May be performed.
As described above, according to the present embodiment, it is possible to manufacture a solar cell by a process further simplified than that of the second embodiment.

Embodiment 4 FIG.
FIG. 6 is a process diagram for explaining the manufacturing method according to the fourth embodiment for realizing the solar cell structure of the second embodiment. In the above-described third embodiment, as shown in FIGS. 4A and 4B, the p electrode, the n electrode, and the glass paste pattern are separately formed. However, in order to further simplify the manufacturing process. The following methods are effective for this.

  6A shows an n electrode 102 pattern made of silver paste, a p electrode 103 pattern made of aluminum silver paste, and a glass paste 104 pattern on a p type silicon substrate 100 on which an n type diffusion layer 101 is formed. FIG. 6B shows a state after printing all, and FIG. 6B shows a state where it is fired at once. Reference numeral 105 denotes a p-type inversion layer equivalent to that already described.

  Further, FIG. 6C shows a state in which the pattern of the n-electrode 107 made of silver paste is formed on the main incident light surface side. As shown here, the electrodes 102, 103 and 107 and the glass paste 104 may be baked separately, but the electrodes 102, 103 and 107 and the glass paste 104 are printed to simplify the process. Thereafter, all may be fired simultaneously.

As described above, according to the present embodiment, it is possible to manufacture a solar cell by a process further simplified than that of the third embodiment.
In Embodiments 1 to 4, the optimum conditions are not specified, but the selection of the actual solar cell manufacturing method procedure, that is, the electrode material, the printing order of the glass paste, the firing temperature and the number of firings, firing The selection of the order and the like largely depends on the quality of the silicon crystal used, so it is not uniquely determined but should be determined in consideration of the thermal influence. In any case, according to the present invention, it is apparent that a solar cell can be manufactured by a simplified process.

Embodiment 5 FIG.
FIG. 7 is a process diagram for explaining a method for manufacturing a solar cell according to still another embodiment 5 of the present invention.
In the first to fourth embodiments described above, the case where the conductivity type of the substrate is p-type has been described. However, the present invention can also be applied to the case where the substrate is n-type.

  In FIG. 7A, 108 is an n-type silicon substrate having a resistivity of about 2 Ωcm, for example, and 109 is a p-type diffusion layer in which boron is thermally diffused, for example. The p-type diffusion layer 109 is formed by decomposing ion implantation of impurities such as boron or a mixture of silane-based gas and diborane (B2H6) gas by heat or external energy such as plasma. A type silicon layer may be used, and if the thickness is several hundred angstroms to several μm, the following effects can be obtained for the same reason as described in the first embodiment.

  In the present embodiment, first, glass paste 104 is patterned on this p-type silicon layer 109 by screen printing. The pattern shape, pitch, and the like may be the same as those already described in the second embodiment. FIG. 7B shows a state after the glass paste 104 is removed by etching with hydrofluoric acid after baking FIG. 7A. Due to the nature of the glass paste 104, the target silicon conductivity type is eroded by firing regardless of n and p, and as a result, a shape as shown in FIG. 7B can be obtained. Here, reference numeral 110 denotes an n-type silicon substrate surface which is eroded and exposed by the glass paste 104.

  The state after printing and baking the silver paste pattern to be the n-type electrode 102 on the n-type silicon substrate surface 110 and the aluminum silver paste pattern to be the p-electrode 103 on the partially left p-type diffusion layer 109 It is FIG.7 (c). Furthermore, on the p-type diffusion layer 109 formed on the surface on which the process is performed in FIGS. 7A to 7C and the opposing surface (which becomes a main incident light surface (light receiving surface) when functioning as a solar cell). FIG. 7D shows a state in which an aluminum silver paste electrode to be the p electrode 111 is formed.

Here, the electrodes 102, 103, and 111 may be fired separately. However, in order to simplify the process, the electrodes 102, 103, and 111 may be fired simultaneously after printing. The p-type diffusion layer 109 to be formed on the main incident light surface side can be formed by the various methods described above, and the order of formation is the same as that of the p-type layer on the back surface side or the p-type layer on the back surface side. It may be after forming the mold layer.
The schematic diagram of the electrode pattern of the solar cell completed by the above is shown to FIG.7 (e) and (f). Moreover, it is also possible to use the aluminum paste which does not contain silver for the p electrode 111 instead of an aluminum silver paste.

Embodiment 6 FIG.
FIG. 8 shows the structure of a solar cell for explaining Embodiment 6 according to the present invention. FIG. 8 (a) is a sectional structural view thereof, and FIG. 8 (b) is an external view thereof. Unlike the structures shown in the second to fifth embodiments, the structure shown here is characterized by using a p-type silicon substrate 112 provided with a via hole 113 on the substrate. Also, by providing the n-type diffusion layer 101, the generated current is collected by the n electrode 102 and the p electrode 103 arranged on the back surface of the solar cell.
The reason for adopting such a structure is that when the p-type silicon substrate 112 that contributes to power generation is thinned, a decrease in the generated current is inevitable. Therefore, by arranging all the electrodes on the back surface, the light receiving area is increased, and the photocurrent is increased. This is because it aims to increase the number of people.

The basic process for realizing this structure is shown by, for example, M. Deguchi (Conference Record of IEEE First World Conference on Photovoltaic Energy Conversion, Hawaii, voI.II, pp.1287 (1994)). Then, detailed explanation of the process method is omitted.
Only the final structure when this invention is applied to such a solar cell structure is shown in FIGS. 8 (a), (b) and (c).

In FIG. 8A, after the n-type diffusion layer 101 is provided on the entire surface of the p-type silicon substrate 112, electrode formation and pn junction are performed in the same process as shown in the second, third, or fourth embodiment. In this state, electrical separation can be achieved. Here, the n-type diffusion layer 101 can be formed by various methods as described in Embodiment Mode 1.
FIG. 8C is a top view of the electrode arrangement surface. By realizing the separation of the pn junction and the arrangement of the electrode pattern with the pattern shown here, the p-type silicon substrate 112 having the via hole 113 is used. The present invention is also applicable to the solar cell.

  In FIG. 8, 114 is an uneven structure of the surface generally called texture. When a silicon substrate having a (100) plane orientation is used, silicon is etched with an alkaline aqueous solution such as sodium hydroxide having a concentration of 1 to several percent. However, by providing this, multiple reflection of light occurs on the surface of the solar cell, effectively reducing the reflectance and improving the conversion efficiency.

In the first to fifth embodiments, the case where this texture structure is not provided has been described in order to simplify the description. Of course, the present invention can be applied without any problem even if this texture structure is provided.
Furthermore, for the sake of simplification, the structure in which the antireflection film is not formed on the light receiving surface side has been described in the second to sixth embodiments. However, the antireflection film is formed by the method described in the first embodiment. Needless to say, the present invention can be applied without any problem even in the case of providing the above.

Embodiment 7 FIG.
Embodiments 1 to 6 described above describe a manufacturing method that easily realizes electrical isolation of pn junctions in the formation of solar cell electrodes. In the seventh embodiment, the formation of solar cell electrodes is described. A method of manufacturing a semiconductor device capable of forming an electrode capable of significantly reducing contact resistance with a semiconductor substrate by applying a physical phenomenon similar to the electrical isolation method of the pn junction at the time will be described in detail.
That is, the method for manufacturing a semiconductor device according to the seventh embodiment or later to be described later mainly relates to a silicon semiconductor device, particularly when a metal paste material is used in a structure having a silicon nitride film or a titanium oxide film on the surface of a silicon substrate. Provided is a novel semiconductor device manufacturing method capable of simplifying an electrode forming method and improving productivity and reducing a contact resistance between an electrode made of a metal paste material and silicon.

There are a wide variety of types of semiconductor devices to which such a method for manufacturing a semiconductor device can be applied. Among them, the method is particularly effective for light receiving elements such as photodiodes and solar cells. Here, taking a solar cell as a specific example, the background of the invention will be described.
Currently, silicon solar cells are the mainstream of power solar cells used on the earth, but the process flow at the mass production level is generally as simple as possible to reduce the manufacturing cost. Regarding the formation, a method of forming a metal paste by screen printing or the like is employed.

In the solar cell manufacturing flow that is generally performed, the steps up to the formation of the antireflection film follow the manufacturing flow shown in FIGS. 17 (a) to 17 (d), and then on the surface of the n-type diffusion layer 2. An antireflection film is formed.
FIG. 9 and FIG. 10 are flowcharts showing an example of the surface electrode forming method after forming the antireflection film.
FIG. 9 shows a method disclosed in, for example, Japanese Patent Publication No. 5-72114, and FIG. 10 shows a method disclosed in, for example, Japanese Patent Publication No. 3-46985.

  The case of FIG. 9 is an electrode formation method when the antireflection film 7A is made of a silicon nitride film formed by plasma CVD after the processes shown in FIGS. In this case, first, the antireflection film 7A made of a silicon nitride film is selectively etched away only at the portion where the surface electrode is formed (FIG. 9A → FIG. 9B). As a result, the n-type diffusion layer 2 is exposed at the portion where the silicon nitride film, which is an insulator, is removed, and the silver paste electrode 5A is directly printed on this, and the surface electrode is formed through a drying and firing process. Completion is complete (FIG. 9C).

  The case of FIG. 10 is an electrode formation method in the case where the antireflection film 7B is made of a titanium oxide film formed by a CVD method or the like after the processes shown in FIGS. In this case, unlike FIG. 9, the silver paste electrode 5B which becomes a surface electrode is printed directly on the titanium oxide film, and a drying and baking process is performed (FIG. 10 (a)-> (b)). As a result, the antireflection film 7B made of a titanium oxide film is melted and the silver paste electrode 5B reaches and contacts the n-type diffusion layer 2 so that electrical contact is achieved (FIG. 10C).

  Such a method is generally called fire-through and is a phenomenon that has been known for a long time. In particular, as in this conventional example, it is known that fire-through occurs relatively easily in a titanium oxide film or a silicon oxide film. However, when the silver paste electrode 5 is formed on the n-type shallow diffusion layer 2 in a pn junction type solar cell as shown in FIGS. And there is a problem that pn junction breakdown occurs. Regarding this point, a method for suppressing the destruction of the n-type diffusion layer 2 by mixing the Group 5 element of the periodic table into the silver paste and diffusing the Group 5 element into the n-type diffusion layer 2 during firing is also proposed. Has been.

  Such an example is disclosed, for example, in Japanese Patent Publication No. 4-67347 in which phosphorus, arsenic, antimony, or the like is mixed. Further, in the example of Japanese Patent Publication No. 3-46985 shown in FIG. 10, an element of Group 5 of the periodic table, specifically, phosphorus reacts with an antireflection film made of a titanium oxide film or a silicon oxide film, It has been shown to be easier to promote.

  As described above, various contrivances have been made for the electrode formation method of the silicon solar cell. In particular, in the solar cell using the silicon nitride film as the antireflection film, the fire-through has not been performed. I needed a process as shown. Alternatively, as shown in FIG. 10, a method of promoting the fire-through by mixing elements of Group 5 of the periodic table such as phosphorus into the silver paste electrode has been proposed. The possibility of forming electrodes by fire-through is not disclosed.

  In view of this, the seventh embodiment provides a method for manufacturing a semiconductor device that enables fire-through even when a silicon nitride film, which has been difficult in the past, is used, and that can greatly reduce the contact resistance with the semiconductor substrate. Also, a semiconductor device manufacturing method in which the performance of a solar cell is improved as compared with the conventional method and the contact resistance can be greatly reduced in a semiconductor device in which electrodes are formed using a fire-through of a titanium oxide film. To do.

  In the seventh embodiment, the metal paste material has a property of melting and penetrating an insulating film formed on the semiconductor substrate. In order to realize this, a frit-like material mixed in the metal paste material is used. The glass component is the point. According to the experiments by the present inventors, for example, as a main component of glass, lead (Pb) 5-30%, boron (B) 5-10%, silicon (Si) 5-15%, oxygen (O) 30-60. It has been clarified that the effects shown in the following embodiments described later can be easily achieved by using the composition of% as a basis. In the present embodiment, in addition to the glass having these compositions, diethylene glycol monobutyl ether (Diethyleneglycol Monobtylether), ethylene glycol monomethyl ether (Ethyleneglycol Monobtylether) and the like are mixed in an appropriate amount, and adjusted to a predetermined viscosity capable of screen printing. Using. It was also confirmed that there was no need to use a metal paste material mixed with elements of Group 5 of the periodic table as introduced in the prior art. As long as there is no notice in particular, the metal paste material used by the following specific embodiment concerning a semiconductor device uses such a metal paste material.

FIG. 11 is a flowchart relating to the semiconductor device manufacturing method according to the seventh embodiment and a ninth embodiment to be described later.
A specific embodiment will be described below based on this flowchart. The process flow is not limited to one semiconductor device, but a solar cell will be described below as a specific example.

  In the seventh embodiment, in the flowchart shown in FIG. 11, step S11-step S12- (step S13) -step S14-step S15a- (step S16) -step S17- (step S18) -step S19-step S20. The solar cell manufactured by the flow will be described with reference to a characteristic cross-sectional structure shown in FIG. In FIG. 11, the steps in parentheses, such as (Step S3), are not essentially required in the manufacturing process of the solar cell, and the steps or steps that can further improve the characteristics by adding this step. It is shown that it may be omitted for simplification.

  First, step S11 in FIG. 11 is a step of cleaning a single crystal manufactured by, for example, a pulling method or a polycrystalline silicon substrate manufactured by a casting method as a semiconductor substrate, but in the case of a solar cell, it is sliced from an ingot. In many cases, the substrate is used as it is. In this case, in order to remove substrate surface damage and contamination of the wafer slicing process due to the wire saw used for slicing, an alkaline aqueous solution such as potassium hydroxide or sodium hydroxide aqueous solution or a mixed solution of hydrofluoric acid and nitric acid is used. The substrate surface is etched by about 10 to 20 μm. Further, a step of washing with a mixed solution of hydrochloric acid and hydrogen peroxide may be added to remove heavy metals such as iron adhering to the substrate surface. In FIG. 12A, the silicon substrate obtained by this process is denoted by reference numeral 200.

  Subsequently, in step S12, if the substrate to be used is a p-type silicon substrate 200, for example, an n-type diffusion layer 201 is formed to form a pn junction. As a method for forming this n-type diffusion layer, phosphorus diffusion using phosphorus oxychloride (POCl 3) as described with reference to FIG. 17B is used. Other methods include, for example, phosphorus or arsenic ion implantation or silane-based gases such as monosilane (SiH4), disilane (Si2H6), dichlorosilane (SiCl2H2), trichlorosilane (SiCl3H), tetrachlorosilane (SiCl4) The same effect can be obtained by using an n-type silicon layer formed by decomposing a mixture of phosphine (PH3) gas with external energy such as heat or plasma. Furthermore, in order to improve the characteristics by suppressing the surface recombination of the solar cell, the window effect, etc. by using the pn junction as a heterojunction, for example, silicon carbide (SiC) formed from a mixed gas such as monosilane and methane or ethane may be used. Good.

  That is, the junction formed on the silicon substrate is configured by using a semiconductor layer made of Group 4 using an element belonging to Group 3 or Group 5 of the periodic table as an impurity. A pn junction can be formed, and a semiconductor layer made of a Group 4 element is directly formed on the surface of a semiconductor layer formed by impurity diffusion or ion implantation into a silicon substrate or on the surface of the silicon substrate by heat of source gas or plasma. By using a semiconductor thin film layer composed of a single element or a plurality of elements deposited and formed by a decomposition reaction, the impurity concentration, thickness, etc. of the semiconductor layer can be controlled with high accuracy. In addition, the thin film formed by such a method can be made amorphous, microcrystalline, polycrystalline, or single crystal by changing the forming conditions, for example, the film forming temperature. Can also be controlled. The thickness may be arbitrarily selected in accordance with the purpose within the range of several hundred angstroms to several μm. The n-type diffusion layer formed as described above is denoted by reference numeral 201 in FIG.

  Next, the pn junction separation of the front and back surfaces of the substrate in step S13 can be performed by a flow as shown in FIG. When the n-type layer formation process by ion implantation or CVD method as described in step S12 is selected, since the n-type layer is basically formed only on one side, FIG. The flow shown in Fig. 1 is not necessarily required. Furthermore, even when using a diffusion method in which a liquid coating material containing phosphorus, such as PSG (Phospho-Slicate-Glass), is spin-coated on only one surface of the substrate and annealed under appropriate conditions, The flow as shown in FIG. 17C is not necessarily required. Of course, when there is a possibility that the n-type layer is formed up to the back surface of the substrate, the use of this process can improve the integrity.

  In step S14, the substrate on which the n-type diffusion layer 201 is formed as described above is thermally oxidized (treated in oxygen at a temperature of 800 ° C. to 1000 ° C. for several minutes) by, for example, a thermal oxidation method. This is a step of forming a silicon oxide film having an appropriate thickness (for example, about 100 to 200 angstroms) on the surface of the mold diffusion layer. By this step, an oxide film is formed not only on the surface of the n-type diffusion layer but also on the surface of the p-type silicon substrate 200 on the opposite surface. By forming this oxide film, the surface of the n-type diffusion layer and the p-type on the opposite surface are formed. The solar cell characteristics are improved by reducing the defect density (surface state density) on the surface of the silicon substrate and suppressing surface reconstitution. This method is widely known as one of surface passivation techniques. The silicon oxide film may be formed by a plasma CVD method or a thermal CVD method using a mixed gas of SiH4 and 02 as a raw material. FIG. 12A shows a structure in which the silicon oxide film 202 obtained by the above method is formed particularly on the surface of the n-type diffusion layer 201.

  Step S15a is a step of forming the silicon nitride film 203 (see FIG. 12A). When step S14 is performed, the structure is formed on the silicon oxide film 202. However, as described above, the silicon oxide film 202 is essentially not necessary in terms of the configuration of the solar cell, and thus is directly n. It may be formed on the mold diffusion layer 201. Since the silicon nitride film 203 functions as an antireflection film, it is possible to reduce the surface reflectance of the solar cell with respect to the incident light, and thus it is possible to greatly increase the generated current. Although the thickness of the silicon nitride film 203 depends on its refractive index, for example, about 700 to 800 angstroms is appropriate when the refractive index is about 1.9 to 2.0.

  The silicon nitride film 203 is formed by using a low pressure thermal CVD method or a plasma CVD method. In the case of the thermal CVD method, dichlorosilane (SiCl2H2) and ammonia (NH3) are often used as raw materials. For example, the gas flow rate ratio is NH3 / SiCl2H2 = 10 to 20, the pressure in the reaction chamber is 0.2 to 0.5 Torr, and the temperature is 760. Film formation is performed under the condition of ° C. In this method, the silicon nitride film 203 contains almost no hydrogen due to high-temperature pyrolysis, the composition ratio of Si and N becomes Si3N4, which is almost stoichiometric, and the refractive index is also about 1.96. The range is 1.98. Therefore, such a film has a feature that the film quality (film thickness, refractive index) does not change even if heat treatment is applied in a later step, and the film is extremely dense.

  In addition, a mixed gas of SiH4 and NH3 is generally used as a source gas in the case of forming by plasma CVD. As the film forming conditions, for example, the gas flow ratio NH3 / SiH4 = 0.5 to 1.5, the pressure in the reaction chamber is 1 to 2 Torr, the temperature is 300 to 550 ° C., and the frequency of the high frequency power source necessary for plasma discharge is several hundreds. kHz or more is appropriate. In the case of this plasma CVD method, the silicon nitride film 203 contains hydrogen because it is formed at a lower temperature than the thermal CVD method, and the composition ratio of Si and N is greatly changed because gas decomposition is caused by plasma. It has the feature that it can be. Specifically, by changing conditions such as gas flow ratio, pressure, and temperature, the composition ratio of Si, N, and hydrogen changes, and the refractive index is about 1.8 to 2.5. A silicon nitride film can be formed. In the case of such film quality, when a heat treatment is applied in a later process, the refractive index and film thickness may change from the values immediately after the film formation due to a phenomenon such as hydrogen desorption in the electrode firing process, for example. . In this case, it is possible to obtain a silicon nitride film necessary for a solar cell by taking into account the change in film quality due to heat treatment in a later step and determining the film formation conditions.

  Since details of step S16 will be described later, the subsequent steps will be described first. Step S17 is a step of printing and drying an electrode 204 (see FIG. 12A) made of silver aluminum or aluminum paste as a metal paste material according to the present invention having a predetermined pattern by screen printing on the back surface of the wafer, Step S18 is a step of firing this paste. Specifically, by baking in dry air, for example, at a temperature of 700 to 800 ° C. for several tens of seconds to several minutes, ohmic contact with the p-type substrate is obtained, and aluminum diffuses into the silicon substrate. A diffusion layer 205 (see FIG. 12B) is formed.

  In step S19, the silver paste electrode 206 according to the present invention is directly printed and dried on the silicon nitride film 203 formed on the n-type diffusion layer 201 side. In the subsequent step S20, the silver paste 206 is baked. However, as long as the silver paste containing the glass component according to the present invention, silver aluminum, or aluminum paste is used, step S18 is not necessary. The electrode materials printed and dried in S19 can be baked together in step S20. In the present invention, the fire-through for the silicon nitride film 203, which has not been possible until now in step S20, is easily realized. As shown in FIG. 12B, the silver paste electrode 206 after the fire-through is made of silicon nitride. The film 203 and the silicon oxide film 202 can be melted and penetrated to make electrical contact with the n-type diffusion layer 201.

This point will be described in detail below based on experimental facts.
As described above, the silicon nitride film formed by the thermal CVD method is usually denser than the silicon nitride film formed by the plasma CVD method, and it is easily predicted that it is difficult to realize further fire-through. Therefore, ohmic contact on the light-receiving surface electrode side can be easily obtained regardless of the film forming method.
FIG. 13 shows the relationship with the fill factor (FF), which is a measure of whether or not a solar cell is actually manufactured with the structure shown in FIG. It is the result of having investigated experimentally.

  In this experiment, polycrystalline silicon by casting is used as a semiconductor substrate, phosphorus diffusion by phosphorus oxychloride (POCl 3) is used for the n-type diffusion layer 201, thermal oxidation is used for the silicon oxide film 202, and the silicon nitride film 203 is formed. Regarding the film method, both the plasma CVD method and the low pressure thermal CVD method have been studied. The electrodes were fired at the same time as the electrodes 204 and 206 in FIG. 12A and were performed in dry air. The firing time was 45 seconds at a firing temperature (peak value) of 675 and 700 ° C, 22 seconds at 720 ° C to 750 ° C, and 10 seconds at 775 ° C to 800 ° C. In addition, all the solar cell sizes are 10 cm × 10 cm. As shown in FIG. 13, a practically sized solar cell of 10 cm × 10 cm has a fill factor of 0.76 or more obtained under almost all firing conditions, confirming the effectiveness of the present invention. did it.

  The refractive index of the silicon nitride film 203 by the plasma CVD method shown in FIG. 12 is 2.1 and the film thickness is 750 angstroms, but other refractive indexes are 1.9, 2.0, 2.2, 2 It was confirmed that similar results were obtained for the films of .3, 2.4. In addition, although description regarding step S16 was abbreviate | omitted in said description, this process assumes the following situations. For example, when the silicon nitride film 203 is formed by the low pressure CVD method, a method of fixing only a part of the periphery of the substrate is often used. In this case, if the silicon nitride film 203 is only on one side as shown in FIG. However, it is also formed on the opposite surface. In that case, the silicon nitride film formed on the opposite surface may interfere with the formation of the back electrode 204. That is, when the fire-through cannot be performed, electrical contact with the substrate 200 is hindered.

  Therefore, in order to avoid such a situation, it is possible to make electrical contact with the substrate 200 by, for example, performing electrode formation after removing part or all of the silicon nitride film partially formed on the opposing surface. This is the process in step S16. However, in the case of using silver aluminum or aluminum paste according to the present invention, it has been confirmed that the silicon nitride film 203 can be fired through under the above baking conditions, and this step is not necessarily required. It has also been confirmed that even if a silicon nitride film formed by the plasma CVD method exists on the back side, it can be fire-through in exactly the same way. In other words, in the present invention, even when a silicon nitride film is formed on both surfaces of the substrate positively or by force majeure, a solar cell can be manufactured without requiring any patterning process.

Embodiment 8 FIG.
In the eighth embodiment, of the steps shown in FIG. 11, the invention according to steps S21 and S22 after the seventh embodiment will be described. In particular, step S22 is a characteristic process according to the present invention, in which the solar cell that has undergone step S20 is immersed in an aqueous hydrofluoric acid (HF) solution. After the immersion, the substrate is thoroughly washed with pure water and then dried.

  FIG. 14 shows the actual experimental results. In addition, what was shown here is the result about the solar cell when the silicon nitride film 203 is formed by the low pressure thermal CVD method. The firing conditions are those when firing is performed longer than in Embodiment 7, for example, when firing is performed at 720 ° C. for 5 minutes. The fill factor value immediately after firing was indicated by white circles in FIG. 14, but was a very low value of about 0.65. On the other hand, as a result of immersing the solar cell in this state in an aqueous hydrofluoric acid solution under various conditions shown in FIG. 14, it was found that there was a remarkable improvement effect (black circle in FIG. 14).

  As a result of conducting another experiment to understand this phenomenon, the specific resistance of the electrode 206 (see FIG. 12B) before and after the hydrofluoric acid aqueous solution immersion treatment was 3 × 10 −3 Ωcm, which was not changed at all. On the other hand, the contact resistivity was 2.68 × 10 −1 Ωcm 2 before the treatment, but 2.60 × 10 -3 Ωcm 2 after the treatment, which is 2 digits lower. That is, the fill factor is improved when the silicon nitride film 203 is fired through, depending on the firing conditions, the glass component that increases the contact resistance is segregated at the interface between the electrode 206 (see FIG. 12B) and the n-type diffusion layer 201. It is assumed that this is removed by the hydrofluoric acid immersion treatment and the characteristics are improved.

  In other words, this method is an epoch-making product that enables solar cell production under a wide range of electrode firing conditions and greatly contributes to yield improvement in the case of forming electrodes by firing through the silicon nitride film 203. It can be said that it is a thing. Further, in this step, since the completed solar cell is immersed in the hydrofluoric acid aqueous solution, for example, there is a concern that a silicon nitride film as an antireflection film is etched, but at least within the range shown in FIG. It was not recognized that the silicon nitride film formed by the CVD and plasma CVD methods was etched and the reflectance of the solar cell was changed in an inappropriate direction.

  In addition, it has been confirmed that the adhesion force of the electrode is not greatly reduced in this step, and that no problem occurs with respect to manufacturing the solar cell module by a normal method. Note that the hydrogen annealing in step S21 has the effect of contributing to the reduction of the interface state density between the silicon nitride film 203 or the silicon oxide film 202 and the n-type diffusion layer 201 in the eighth embodiment. As specific conditions, for example, annealing is performed at 400 to 450 ° C. for 30 minutes with a mixed gas of hydrogen gas: nitrogen gas = 1: 9 for 30 minutes. As a result of the experiment, for example, an effect was confirmed in which the open circuit voltage was 587 mV before annealing and became 596 mV after the treatment.

  It has also been confirmed that even if the order of step S21 and step S22 is changed, the final solar cell characteristics are not affected. In the above description, the case where hydrofluoric acid is used has been described. However, the present invention is not limited to this as long as the etching solution has a property of dissolving the glass component. It has been confirmed that there is an effect.

Embodiment 9 FIG.
In the ninth embodiment, in the flow chart shown in FIG. 11, the flow of step S11-step S12-step S13-step S14-step S15b- (step S16) -step S17- (step S18) -step S19-step S20. The solar cell produced in will be described.
FIG. 12 shows a characteristic cross-sectional structure of a solar cell manufactured by the process of the present embodiment. Except for step S15b, it is exactly the same as that described in the seventh embodiment and will not be described. However, the ninth embodiment is characterized in that step S13 is always performed.

  Step S15b is a step of forming the titanium oxide film 207 shown in FIG. The titanium oxide film 207 can be formed by applying and baking an organic titanate (an organic liquid material containing titanium) typified by TPT (tetra-i-propyl titanate) on a substrate. When applying by spin coating, the thickness is controlled by the amount of TPT, the number of rotations and the time. After firing, the film thickness decreases due to volume reduction, but the thickness after firing is adjusted to be about 700 to 800 angstroms. The firing temperature and time are, for example, 300 to 400 ° C. and several minutes to 30 minutes. As another film formation method, the titanium oxide film 207 can also be formed by performing thermal decomposition (thermal CVD method) at 250 to 300 ° C. in a state where TPT vapor and water vapor are mixed. With the above method, a refractive index of about 2.0 to 2.3 is obtained.

  Here, the reason why the silicon oxide film 202 is provided in step S13 is to reduce the defect density on the wafer surface as already described, but in particular, the sun using the titanium oxide film 207 as an antireflection film. In a battery, it is effective to have a double structure of silicon oxide film / titanium oxide film. The reason is that when the titanium oxide film 207 is directly formed on the semiconductor by the method as described above, the defect density at the interface between the titanium oxide film 207 and the semiconductor substrate (in this case, the n-type diffusion layer 201) is high. This is because the conversion efficiency of the battery is adversely affected. According to this embodiment, as shown in FIG. 15, the electrode 206 (see FIG. 15B) can be easily fired even on the insulating film having a double structure of the silicon oxide film 202 / titanium oxide film 207. Although it was possible to pass through, the experiment showed that the conditions were very characteristic (see FIG. 16).

  In general, the metal paste material is fired in dry air. However, as shown in FIG. 16, the oxygen concentration (mixed gas of oxygen and nitrogen) is an important point for the structure of the present invention. You can see that For example, when firing at a high temperature of 800 ° C., a high fill factor (FF) can be obtained even in normal dry air, but when the temperature is lowered, a high fill factor FF cannot be obtained with dry air. On the other hand, by setting the oxygen concentration to 50%, for example, a high fill factor FF can be obtained. From other systematic experimental results, it was confirmed that by setting the oxygen concentration to 30% or higher, a high fill factor FF can be obtained in a firing range of about 700 to 800 ° C.

  Although details are not shown, it was also confirmed that the hydrofluoric acid immersion treatment in step S22 (FIG. 11) shown in the eighth embodiment is effective for improving the fill factor FF even in this structure. For example, the fill factor FF of 0.68 was improved to 0.75 by immersing it in a solution of HF: H20 = 1: 50 for 30 seconds until the step S20.

As described above, according to the method for manufacturing a solar cell according to the present invention, the electrical separation of the pn junction necessary for the configuration of the solar cell can be realized by a very simple method. Cost and productivity can be greatly improved.
In particular, a method in which a material mainly composed of glass having a property of melting silicon is provided on the pn junction and baked without performing a conventional chemical etching process for electrical separation of the pn junction. Therefore, the solar cell manufacturing process can be greatly shortened, and the manufacturing cost can be reduced.

  The thin film layer has a thickness of several hundreds of angstroms to several microns and is decomposed by thermal impurity diffusion, ion implantation, heat of source gas or application of external energy such as plasma. Therefore, there is an effect that pn junction formation can be realized with good controllability and high reproducibility.

  Further, the material containing glass as a main component is in a paste form, and since this is applied and patterned, there is an effect that a predetermined pattern can be easily formed.

  In addition, since the material mainly composed of glass is electrically separated from the pn junction and is left in the formation portion without being removed, it is not necessary to remove the material mainly composed of glass. The process is simplified and has the effect of reducing contamination.

  Moreover, after electrically separating the pn junction by firing the material containing glass as a main component, the method further includes a step of removing the material containing glass as a main component. This has the effect of improving the patterning accuracy in electrode formation.

  The method further includes forming a p-electrode and an n-electrode necessary for collecting solar cells outside the region on the pn junction where the material containing the glass as a main component is formed. Since the electrode is formed outside the region of the material, there is an effect that it is possible to avoid stress or impurity contamination of the material mainly composed of glass on the electrode.

  In addition, the step of forming the p electrode and the n electrode is performed after the formation of the material mainly composed of the glass, thereby independently setting the electrode formation conditions and the material formation conditions mainly composed of the glass. There is an effect that it is easy to optimize the forming conditions.

  Further, the step of forming the p electrode and the n electrode is performed before the formation of the material containing the glass as a main component, thereby independently setting the electrode forming conditions and the material forming conditions containing the glass as a main component. That is, there is an effect that the formation conditions can be easily optimized.

  In addition, by using a combination of materials such as silver, aluminum silver, and aluminum paste as the p electrode and the n electrode, it is possible to use together with a material mainly composed of glass, and there is an effect that a manufacturing method with high productivity can be obtained. is there.

  In addition, the step of firing the glass-based material is produced by forming the p-electrode and n-electrode materials on the front surface or both the front and back surfaces of the pn junction, and firing the electrodes at the same time. There is an effect of making a highly producible manufacturing method.

  In addition, after the step of firing the glass-based material, the method further includes the step of forming and firing the p electrode and the n electrode on the front surface or both the front and back surfaces of the pn junction, There is an effect that the material forming conditions mainly composed of glass are set independently, that is, the forming conditions can be easily optimized.

  In addition, the step of firing the electrode is performed after removing the material containing the fired glass as a main component, thereby independently setting the electrode forming conditions and the material forming conditions containing the glass as a main component, that is, forming It is easy to optimize the conditions.

  In addition, before the step of firing the material containing the glass as a main component, the method further includes the step of forming and firing the p electrode and the n electrode on the front surface or both the front and back surfaces of the pn junction layer. In addition, there is an effect that the material forming conditions mainly including glass are set independently, that is, the forming conditions can be easily optimized.

  In addition, the material mainly composed of glass is patterned by screen printing or roll coater printing so as to be formed on the pn junction surface, so that pattern formation can be performed in a short time, and productivity can be achieved. There is an effect to increase.

  In addition, since the electrode material is patterned by screen printing or roll coater printing and formed on the pn junction, pattern formation can be performed in a short time, and the productivity can be improved.

  According to another aspect of the present invention, there is provided a pn junction comprising a substrate and a thin film layer having a conductivity type opposite to the substrate formed on the entire surface including the front surface or both surfaces of the substrate and the back surface of the substrate. In a solar cell having a p-type electrode and an n-type electrode formed on the pn junction, the main component is glass that electrically separates the pn junction between the p-type electrode and the n-type electrode. By providing the material, it is possible to achieve electrical separation between the pn electrodes with a simple structure, and there is an effect that a solar cell structure having characteristics comparable to those of the conventional structure can be obtained.

  According to another aspect of the invention, there is provided a method of manufacturing a semiconductor device, wherein an insulating film is formed on a surface of a semiconductor substrate having a junction or both surfaces of a front surface and a back surface. The method includes forming a metal paste material containing glass having a property of melting the insulating film and firing to form an electrode penetrating the insulating film and making electrical contact with the semiconductor substrate. Therefore, it is possible to form an electrode that can greatly reduce the contact resistance with the semiconductor substrate by applying the same physical phenomenon as the method of easily performing electrical separation of the pn junction when forming the electrode of the solar cell. effective.

  The insulating film is composed of a single layer or a plurality of layers. In the case of a single layer, a silicon nitride film is used. In the case of a plurality of layers, a fire-through is performed even when a silicon nitride film is used as at least one of the layers. Furthermore, it is possible to realize a structure of a semiconductor device and a method for manufacturing the same that can significantly reduce the contact resistance with the semiconductor substrate. In particular, in a structure in which an insulating film is formed on one main surface of a semiconductor substrate having a junction or both surfaces including the opposing surface, the main constituent material of the insulating film is a silicon nitride film, and silicon nitride is formed on the film. A metal paste material containing glass having a property of melting the glass is formed in a predetermined shape at a predetermined position and baked, so that the electrode melts and penetrates the silicon nitride film and makes electrical contact with the semiconductor substrate. Therefore, it is possible to omit the patterning / etching process of the silicon nitride film according to the electrode pattern, and to greatly reduce the manufacturing cost.

  Further, when the insulating layer is composed of a plurality of layers, by using a silicon oxide film in addition to the silicon nitride film, the insulating film has a two-layer structure, in particular, a silicon nitride film and a silicon oxide film. There is an effect that the effect of suppressing the surface recombination of carriers on the surface of the semiconductor device can be remarkably improved.

  Further, since the silicon nitride film is formed by a thermal CVD method or a plasma CVD method, it is easy to control the film thickness, refractive index, etc., and it is easy to realize a predetermined structure of the semiconductor device. .

  Further, when the insulating film is composed of a plurality of layers and a titanium oxide film is used as at least one of the layers, a metal paste containing glass having a property of melting the insulating film on the titanium oxide film By forming the material in a predetermined shape at a predetermined position and firing it, the electrode melts and penetrates the insulating film and makes electrical contact with the semiconductor substrate, so all patterning of the insulating film according to the electrode pattern The etching process is omitted, and the manufacturing cost can be greatly reduced.

  Further, when the insulating film is formed using a silicon oxide film in addition to the titanium oxide film, the insulating film has a two-layer structure of a titanium oxide film and a silicon oxide film, so that This has the effect of significantly improving the effect of suppressing the surface reconstitution of the carrier.

  In addition, the titanium oxide film is formed by applying an organic liquid material containing titanium onto the semiconductor substrate and baking it, or is formed by a thermal CVD method. There is.

  Further, in the semiconductor device having an insulating film having a two-layer structure of the titanium oxide film and the silicon oxide film, the metal paste material is fired in an atmosphere of a mixed gas of oxygen and nitrogen, and the oxygen concentration is 30% or more. Since there is, there is an effect that the contact resistance between the electrode and the semiconductor can be remarkably reduced.

  Further, since the silicon oxide film is formed by a thermal oxidation method, a thermal CVD method or a plasma CVD method, there is an effect that a high quality silicon oxide film can be obtained.

  In addition, the glass is composed mainly of a combination of lead, boron, silicon, and oxygen, so that the various insulating films can be easily melted when the metal paste material is baked to make electrical contact with the semiconductor. There is an effect that it is possible to make a plan, that is, fire-through can be easily performed.

  In addition, the metal paste material is silver, aluminum containing aluminum, or aluminum paste. Since one or more of these are used in combination, all electrodes can be formed through fire-through. There is.

  In addition, since single crystal or polycrystalline silicon is used for the semiconductor substrate, mass production is possible and the cost of the semiconductor device can be reduced.

  In addition, since the junction formed on the silicon substrate is configured by using a semiconductor layer of Group 4 using an element belonging to Group 3 or Group 5 of the periodic table as an impurity, it is a simple method. This has the effect of forming a pn junction.

  In addition, a semiconductor layer composed of the Group 4 element is deposited and formed on the semiconductor layer formed by impurity diffusion or ion implantation into the silicon substrate or on the surface of the silicon substrate by a decomposition reaction by heat of source gas or plasma. Since a semiconductor thin film layer made of a single element or a plurality of elements is used, there is an effect that the impurity concentration and thickness of the semiconductor layer can be controlled with high accuracy.

  And a step of immersing in an aqueous solution containing hydrofluoric acid or ammonium fluoride after a step of forming an electrode that is in electrical contact with the semiconductor substrate through the insulating film by firing the metal paste material As a result, the contact resistance of the electrode can be greatly reduced.

  Further, by adding a step of performing heat treatment in an atmosphere containing hydrogen before or after the step of immersing in the aqueous solution containing hydrofluoric acid or ammonium fluoride, the effect of further suppressing surface reconstitution in the semiconductor device There is.

It is a flowchart which shows the process explaining the manufacturing method of the solar cell using the glass paste which concerns on this invention. It is explanatory drawing explaining the electrical isolation method of the pn junction using the glass paste which concerns on this invention. It is explanatory drawing which shows the manufacturing method of the solar cell using the glass paste which concerns on Embodiment 1 of this invention. It is explanatory drawing which shows the manufacturing method of the solar cell using the glass paste which concerns on Embodiment 2 of this invention. It is explanatory drawing which shows the manufacturing method of the solar cell using the glass paste which concerns on Embodiment 3 of this invention. It is explanatory drawing which shows the manufacturing method of the solar cell using the glass paste which concerns on Embodiment 4 of this invention. It is explanatory drawing which shows the manufacturing method of the solar cell using the glass paste which concerns on Embodiment 5 of this invention. It is explanatory drawing which shows the manufacturing method of the solar cell using the glass paste which concerns on Embodiment 6 of this invention. It is for providing explanation of the manufacturing method of the semiconductor device which concerns on Embodiment 7 of this invention, and is a flowchart for the manufacturing flow of the conventional solar cell in case it has a silicon nitride film as an antireflection film. It is for providing explanation of the manufacturing method of the semiconductor device concerning Embodiment 7 of this invention, and is a process drawing of the manufacturing flow of the conventional solar cell when it has a titanium oxide film as an antireflection film. It is a flowchart which shows the process explaining the manufacturing method of the semiconductor device which concerns on Embodiment 7 and 9 of this invention. It is for demonstrating the manufacturing method of the semiconductor device which concerns on Embodiment 7 of this invention, and is sectional drawing which shows the structure of the solar cell manufactured by carrying out a fire-through of a silicon nitride film / silicon oxide film structure. It is for demonstrating the manufacturing method of the semiconductor device which concerns on Embodiment 7 of this invention, and is explanatory drawing which shows the relationship between the conditions through which a silicon nitride film is fired through, and the fill factor of a solar cell. It is for demonstrating Embodiment 8 of this invention, and is explanatory drawing which shows the relationship between hydrofluoric acid immersion process conditions and the fill factor of a solar cell. It is for demonstrating the manufacturing method of the semiconductor device which concerns on Embodiment 9 of this invention, and is sectional drawing which shows the structure of the solar cell manufactured by carrying out a fire-through of a titanium oxide film / silicon oxide film structure. FIG. 10 is an explanatory diagram showing a relationship between a firing condition and a fill factor of a solar cell when a titanium oxide film / silicon oxide film structure is fired through, for explaining a method of manufacturing a semiconductor device according to a ninth embodiment of the present invention. It is. It is process drawing explaining the manufacturing method of the solar cell which concerns on a prior art example.

Explanation of symbols

  100 p-type silicon substrate, 101 n-type diffusion layer, 102 n-electrode made of silver paste, 103 p-electrode made of aluminum silver paste, 104 glass paste, 105 p-type inversion layer, 106 p-type silicon exposed by erosion of glass paste Substrate surface, 107 n-electrode made of silver paste formed on the main incident light surface side, 108 n-type silicon substrate, 109 p-type diffusion layer, 110 n-type silicon substrate surface exposed by erosion of glass paste, 111 main incident light P-electrode made of aluminum silver paste formed on the surface side, p-type silicon substrate having 112 via holes, 113 via holes, 114 texture structure, 115 BSF / n-type silicon junction electrically separated by glass paste, 200 p-type silicon substrate, 201 n-type diffusion layer, 202 Silicon oxide film, 203 silicon nitride film, 204 aluminum paste electrode, 205 aluminum diffusion layer, 206 silver paste electrode, 207 titanium oxide film.

Claims (13)

Manufacturing method of silicon solar cell having pn junction consisting of silicon layer as power generation layer and thin film layer of opposite conductivity type to the silicon layer formed on the surface of the silicon layer or both surfaces of the front surface and the back surface or the entire surface including the edge In
A step of electrically separating the pn junction by providing and baking a material mainly composed of glass having a property of melting silicon on the pn junction;
After the step of firing the glass-based material, the p-electrode and n-electrode necessary for collecting solar cells are formed on the surface of the pn junction or both the front and back surfaces of the glass-based material. Forming and firing outside the region on the pn junction in which is formed,
A method for manufacturing a solar cell, wherein a combination of materials such as silver, aluminum silver, and aluminum paste is used as the p electrode and the n electrode.   The method for producing a solar cell according to claim 1, wherein the step of firing the electrode is performed after removing a material mainly containing glass after firing. In a method for manufacturing a semiconductor device in which an insulating film is formed on both the front surface and the front and back surfaces of a semiconductor substrate having bonding,
Forming a metal paste material containing glass having a property of melting the insulating film on the insulating film and baking it to form an electrode penetrating the insulating film and making electrical contact with the semiconductor substrate Including
The insulating film is composed of a single layer or a plurality of layers. In the case of a single layer, a silicon nitride film is used, and in the case of a plurality of layers, a silicon nitride film is used as at least one of the layers. Device manufacturing method.   4. The method of manufacturing a semiconductor device according to claim 3, wherein when the insulating layer is formed of a plurality of layers, a silicon oxide film is used in addition to the silicon nitride film.   5. The method of manufacturing a semiconductor device according to claim 3, wherein the silicon nitride film is formed by a thermal CVD method or a plasma CVD method.   4. The method of manufacturing a semiconductor device according to claim 3, wherein the insulating film is composed of a plurality of layers, and a titanium oxide film is used as at least one of the insulating films.   7. The method of manufacturing a semiconductor device according to claim 6, wherein the insulating film is made of a silicon oxide film in addition to the titanium oxide film.   8. The titanium oxide film according to claim 6 or 7, wherein the titanium oxide film is formed by applying an organic liquid material containing titanium onto the semiconductor substrate and baking it, or is formed by a thermal CVD method. A method for manufacturing a semiconductor device.   9. The method of manufacturing a semiconductor device according to claim 6, wherein the metal paste material is fired in an atmosphere of a mixed gas of oxygen and nitrogen, and the oxygen concentration thereof is 30% or more.   8. The method of manufacturing a semiconductor device according to claim 4, wherein the silicon oxide film is formed by a thermal oxidation method, a thermal CVD method or a plasma CVD method.   11. The method for manufacturing a semiconductor device according to claim 3, wherein the glass is a glass composed mainly of a combination of lead, boron, silicon, and oxygen.   After the step of forming an electrode that makes electrical contact with the semiconductor substrate through the insulating film by firing the metal paste material, a step of immersing in an aqueous solution containing hydrofluoric acid or ammonium fluoride is added. 12. The method of manufacturing a semiconductor device according to claim 3, wherein the method is a semiconductor device manufacturing method. 13. The method of manufacturing a semiconductor device according to claim 12, further comprising a step of performing a heat treatment in an atmosphere containing hydrogen before or after the step of immersing in the aqueous solution containing hydrofluoric acid or ammonium fluoride.

Images