JP2011222774A - Manufacturing device and manufacturing method of solar battery - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a highly efficient solar battery without using expensive and slow in throughput process such as photolithography technology, avoiding weakening of a silicon substrate caused by partial removal processing.SOLUTION: The manufacturing method includes a process for preparing substrates (step S101), a process for providing a flexible transfer layer on the substrate surface (step S121), and a process for providing recess/protrusion portions on the substrate surface to reduce light reflectivity by pressing a mold with recess/protrusion shape to the transfer layer to transfer the recess/protrusion shape (step S122). Recess/protrusion portions for reducing light reflectivity on the transfer layer formed on the substrate can thereby be formed without processing the substrate itself.

Description

  The present invention relates to a method and an apparatus for manufacturing a solar cell that increases the power generation efficiency by processing the surface of a substrate of a crystalline solar cell such as a single crystal solar cell or a polycrystalline solar cell.

  In general, in a crystalline solar cell using a polycrystalline silicon substrate, a single crystal silicon substrate, etc., forming a concavo-convex structure on the surface and efficiently taking incident light from the surface into the interior produces a highly efficient solar cell. In order to do so, various methods have been adopted.

  In solar cells using single crystal silicon, a method is known in which anisotropic etching using an alkaline aqueous solution using potassium hydroxide or sodium hydroxide is used, and a large number of pyramidal irregularities are formed by etching from the (100) plane. ing. This is a technique using the difference in etching rate depending on the crystal plane of silicon, and the effectiveness is lowered in the case of a polycrystalline silicon substrate whose crystal orientation is not constant.

  As described in Japanese Patent Application Laid-Open No. 2004-133620, as a method of manufacturing irregularities that can be handled even when a polycrystalline silicon substrate is used, a number of grooves are formed on the substrate surface by machining, and then hydrofluoric acid and nitric acid are used as components. It has been known to obtain desired irregularities by removing a damaged layer with an isotropic wet etching solution. However, in this method, since the period of the unevenness is restricted by the blade feed pitch of the dicing machine, there are problems that the unevenness becomes rough and the processing speed is slow.

  Apart from this, by adding phosphoric acid or a carboxylic acid having 3 to 6 carbon atoms to an etching solution mainly composed of hydrofluoric acid and nitric acid, impurities that cause the etching rate to deteriorate are unevenly distributed on the etching surface, and random unevenness is caused. Has been proposed (see Patent Document 2). Although this method does not involve machining, the generation of irregularities is affected by chance, so that it is difficult to obtain desired irregularities and there is a limit that only relatively flat irregularities can be obtained.

  Recently, a method of performing isotropic wet etching using a photoresist mask by laser patterning using a hologram lens to perform mask peeling has been proposed (see Non-Patent Document 1). However, this method also has good controllability of the mask because it uses photolithography technology, but requires a series of semiconductor processes such as coating, exposure, etching, stripping, and cleaning, which results in high manufacturing costs and limited throughput. There was an inconvenience.

  Furthermore, since these methods are methods of locally removing the silicon substrate, if the silicon substrate is processed with steep irregularities in order to increase the efficiency of taking in sunlight, the silicon substrate is likely to cause stress concentration and is fragile and cracked. There was a side effect of making it an easy substrate.

Japanese Patent Laid-Open No. 3-71677 JP-A-10-303443

J. et al. Plasma Fusion Res. 12 (2009) 829-832, PP. .

  The object of the present invention is to avoid the weakening of the substrate caused by performing the local removal processing of the silicon substrate in view of the above-described problems, without using an expensive and slow process such as a photolithography technique. The object is to provide a highly efficient solar cell.

  In order to solve the above problems and achieve the object of the present invention, the method for manufacturing a solar cell of the present invention includes a step of preparing a substrate and a step of forming a flexible transfer layer on the surface of the substrate. Further, the method has a step of pressing a mold having a concavo-convex shape formed on the transfer layer, transferring the concavo-convex shape to the transfer layer, and forming a concavo-convex portion on the substrate surface for reducing the light reflectance.

  In the method for producing a solar cell of the present invention, a transfer layer is formed on a substrate, and fine uneven portions are formed on the transfer layer by imprint processing. That is, there is no step of forming the uneven portion on the substrate itself, and there is no step of weakening the substrate. And in the manufacturing method of the solar cell of this invention, since an uneven | corrugated | grooved part is formed in the transfer layer formed on the board | substrate, a solar cell with sufficient incident efficiency can be formed.

  The apparatus for manufacturing a solar cell according to the present invention transfers a concavo-convex portion to a substrate placement portion on which a substrate having a flexible transfer layer formed on the surface is placed, and a flexible transfer layer formed on the surface of the substrate. And a mold having a concavo-convex shape formed thereon.

  In the solar cell manufacturing apparatus of the present invention, when the mold is pressed against the transfer layer formed on the substrate surface, the uneven shape of the mold is transferred to the transfer layer, and the uneven portion is formed on the transfer layer. it can. Then, by using the solar cell manufacturing apparatus of the present invention, it is possible to form uneven portions on the transfer layer formed on the substrate, so there is no need to form uneven portions on the substrate itself, and the substrate is weakened. Can be prevented.

  According to the present invention, since there is no local removal processing that weakens the substrate itself, the risk of the substrate breaking during the subsequent mounting process can be greatly reduced. Furthermore, since the low-speed dicing process and the expensive and low-throughput photolithography process are not used, the manufacturing cost can be greatly reduced. In addition, no powerful chemicals such as hydrofluoric acid or nitric acid are used, and no semiconductor cleaning process is required to neutralize and clean these chemical residues, greatly simplifying the production line and reducing environmental impact. it can.

It is a process flow of the cell process of the manufacturing method of the solar cell which concerns on the 1st Embodiment of this invention. It is a schematic block diagram of the sheet | seat nanoimprint apparatus used for the manufacturing method of the solar cell which concerns on the 1st Embodiment of this invention. It is a fine structure figure of the belt-like nanomold surface of the sheet nanoimprint apparatus used in a 1st embodiment of the present invention. In the 1st Embodiment of this invention, it is the figure which showed the fine shape of the nanoimprint layer immediately after imprinting by the belt-shaped nanomold. It is a process flow of the cell process of the manufacturing method of the solar cell which concerns on the 2nd Embodiment of this invention. In the 2nd Embodiment of this invention, it is the figure which showed the fine shape of the nanoimprint layer in the stage which the surface Ag baking process was complete | finished. It is a schematic block diagram of the pattern transfer apparatus used in the 2nd Embodiment of this invention. It is a fine structure figure of the nano mold surface of the pattern transfer device used in the 2nd embodiment of the present invention. It is a fine structure figure of the surface of a mold concerning other examples applicable in the present invention. It is a process flow of the cell process of the manufacturing method of the solar cell which concerns on the 3rd Embodiment of this invention. In the 3rd Embodiment of this invention, it is the figure which showed the fine shape of the nanoimprint layer immediately after imprinting by the belt-shaped nanomold.

  Hereinafter, an example of a solar cell manufacturing method and a solar cell manufacturing apparatus according to an embodiment of the present invention will be described with reference to FIGS. In addition, this invention is not limited to the following examples.

<1. First Embodiment>
FIG. 1 is a process flow of a cell process in the method for manufacturing a solar cell according to the first embodiment of the present invention. The cell process is centered on a process called a previous process in a general semiconductor process.

  First, a P-type ingot of P-type polycrystalline silicon is prepared (step S101), and a P-type polycrystalline substrate (hereinafter referred to as a wafer) is formed from the P-type ingot with a wire slicer to a thickness of about 0.1 to 0.2 mm. Wafer slicing is performed on the thin plate (step S102). In this wafer slicing step, the ingot is cut into a thin plate by sliding it against the side of the ingot while applying a suspension of abrasive grains to the wires arranged in multiple rows. By this wafer slicing step, a polycrystalline wafer in the form of a square thin plate is prepared from a rectangular parallelepiped polycrystalline silicon casting.

  Next, the damaged layer of the cut wafer is etched (step S103). In the wafer slicing process in step S102, fine irregularities are generated on the wafer surface, and further, a damaged layer is generated in which the crystal state is deteriorated due to the bite of abrasive grains or stress during processing. For this reason, in the damaged layer etching step, the surface layer is etched for the purpose of cleaning the surface layer portion of the wafer and removing the damaged layer. Wet treatment is performed in the order of oxidation cleaning with a mixed solution of hydrochloric acid and hydrogen peroxide, rinse cleaning with pure water, removal of an oxide film with dilute hydrofluoric acid, rinse with pure water, and drying.

  The process so far was the same as the known solar cell manufacturing process using polycrystalline silicon, but the following process is a manufacturing process unique to the present invention. In a general silicon crystal solar cell manufacturing process, roughening is often performed by roughening the surface layer in order to prevent reflection in order to reduce sunlight reflection and improve power generation efficiency. In the example, the surface is not roughened, and the process proceeds while the wafer surface is smooth.

  In this embodiment, the dope source is applied after the damaged layer etching process in step S103 (step S104). In this dope source coating step, a coating solution containing a phosphorus compound is used to diffuse N-type impurities into a P-type polycrystalline wafer cut out from an ingot cast by adding slightly P-type impurities. Spray uniformly on the surface and dry. The film made of this coating liquid becomes a dope source, and phosphorus atoms diffuse into the wafer in the next thermal process.

  Further, a diffusion layer (N type) is formed (step S105). This diffusion layer (N-type) forming step diffuses phosphorus in the wafer by heating the wafer coated on the surface with the above-mentioned phosphorus compound, which is an N-type impurity, in a heating furnace to about several hundred to 1000 degrees, The surface layer of the wafer is changed from the original P-type to N-type. At this time, a residual film containing phosphorus as a residue remains on the surface of the wafer.

  Next, the remaining film is removed (step S106). The nature of the remaining film varies depending on the heating method used in the previous step and the nature of the dope source, but here a cheap and simple heating process by heating in the atmosphere is performed. In this case, since the remaining film is mainly composed of ash-like fine powder, lift-off using slant etching of the surface oxide film with an ammonium fluoride solution is performed. Even in this embodiment, since the wafer surface is smooth even at this stage, the remaining film removal process in step S106 can be easily performed compared to other methods in which the surface is roughened and impurities enter the recesses. The etching amount is very small, and the processing time and the consumption of the etching solution are slight.

  Next, a silicon oxynitride film is formed on the wafer surface to form a silicon oxynitride film serving both as a protective film and an antireflection film (step S107). In this embodiment, this silicon oxynitride film also serves as a passivation film. In this silicon oxynitride film forming process, it is possible to oxidize the surface silicon only by heat treatment in an oxidation furnace and use this as a protective film, but the moisture blocking performance is low only with silicon oxide. Therefore, in the silicon oxynitride film forming step in step S107, sputtering film formation is performed using a sputtering source in which silicon oxide and silicon nitride are mixed.

  Next, a nanoimprint process is performed (step S120). In this nanoimprint process, first, SOG coating that is spin coating of water glass is performed on the wafer surface (step S121), and a flexible transfer layer (hereinafter referred to as nanoimprint layer) is formed. SOG is an abbreviation for Spin On Glass, and a glass material is formed by depositing a liquid glass material (water glass) on a wafer by spin coating and firing it. In this embodiment, in the SOG coating process, HSQ manufactured by Tokyo Ohka Kogyo Co., Ltd. that is commercially available as SOG for inorganic room temperature nanoimprint is used. Since HSQ has a high viscosity, in order to obtain a desired viscosity, it is diluted with a solvent to obtain a water glass diluted chemical solution for SOG coating. A water glass layer having a high viscosity is formed on the surface of the wafer 1 by spin-coating this water glass diluted chemical solution. The coating thickness is determined by the viscosity of the coating solution and the spin speed. Drying progresses naturally even during spin coating, and the coating liquid becomes highly viscous.Therefore, the film thickness varies depending on the discharge amount and discharge time, so that the required film thickness can be stably and uniformly applied on the surface. It is preferable to confirm the combination in advance by experiment. Here, the thickness of the coating film was set to 0.3 μm in the state before entering the next sheet nanoimprint process (step S122) after spin coating. And the nanoimprint layer 2 of the flexible viscosity suitable for nanoimprint is obtained.

As described above, in this embodiment, the sheet nanoimprint process (step S122) is performed after the SOG coating process in step S121. FIG. 2 is a schematic configuration diagram of a sheet nanoimprint apparatus 200 used in the sheet nanoimprint process of the present embodiment. In the present embodiment, as shown in FIG. 2, a sheet nanoimprint apparatus 200 using a belt-shaped nanomold 201 having an endless track shape is used.
Below, the sheet | seat nanoimprint apparatus 200 which is a manufacturing apparatus of the solar cell used by this embodiment is demonstrated.

  As shown in FIG. 2, the sheet nanoimprint apparatus 200 includes a belt-shaped nanomold 201, two upper and lower belts constituting a conveyance belt 202, a wafer supply unit 210, and a wafer take-out unit 220.

  The upper belt-shaped nanomold 201 is formed of a loop-shaped band, and has a fine uneven shape on the surface. The belt-shaped nanomold 201 is held by four rolls 203 to 206. Each roll in turn is a preheating roll 203 that preheats the belt-shaped nanomold 201 before the belt-shaped nanomold 201 contacts the nanoimprint layer 2 on the wafer 1, and heating that presses the nanoimprint layer 2 while heating the belt-shaped nanomold 201. The pressure roll 204, the peeling roll 205 that peels the belt-shaped nanomold 201 from the nanoimprint layer 2, and the return roll 206 that returns the belt-shaped nanomold 201 to the preheating roll 203 while maintaining the tension of the belt-shaped nanomold 201. These rolls 203 to 206 rotate in the same direction as indicated by an arrow a in FIG. 2, whereby the belt-shaped nanomold 201 rotates counterclockwise.

  The lower conveyance belt 202 constitutes a substrate placement portion for placing the wafer 1 on the surface, and is constituted by a loop-like belt-like body. The conveyor belt 202 is held at a position facing the upper belt-shaped nanomold 201 by six rolls 203 to 208 and is driven at a constant speed with the belt-shaped nanomold 201. Each roll performs the same function as the upper roll, the preheating roll 203, the heating and pressing roll 204, the peeling roll 205, the return roll 206, the input roll 207 for placing the wafer 1, and the take-out roll for taking out the wafer 1 after nanoimprinting. It consists of 208. Of the conveyor belt 202, a charging roll 207, a preheating roll 203, a heating and pressing roll 204, a peeling roll 205, a return roll 206, and a take-out roll 208, which are rollers for holding a portion on which the wafer 1 is carried, are transported. The upper surface of 202 is arranged so as to form a flat surface, and the wafer 1 can be transported horizontally at a constant speed. These rolls 203 to 208 rotate in the same direction as indicated by an arrow b in FIG. 2, whereby the transport belt 202 rotates clockwise and the wafer 1 is transported.

  Here, the length of the portion of the transport belt 202 on which the wafer 1 is placed and transported is configured to be longer than the length of the belt-shaped nanomold 201 at the portion facing it.

  Further, the preheating roll 203, the heating and pressing roll 204, and the peeling roll 205 of the lower conveyance belt 202 are not necessarily placed at the same positions as the corresponding rolls on the upper side, but the belt-shaped nanomold 201 is formed on the nanoimprint layer 2. If different temperatures and different pressure distributions are generated on the upper and lower surfaces of the wafer 1 in contact with each other, there is a risk that the microstructure formed on the nanoimprint layer 2 is deformed. For this reason, the preheating roll 203, the heating and pressing roll 204, and the peeling roll 205 provided on each of the upper belt-shaped nanomold 201 and the lower conveyance belt 202 are placed at the same position in the conveyance direction. However, the preheating roll of the upper belt-shaped nanomold 201 is different from the preheating roll 203, the heating and pressurizing roll 204, and the peeling roll 205 of the lower conveying belt 202, while the upper surface has the same height. 203, the heating and pressing roll 204, and the peeling roll 205 are arranged to have different heights.

The wafer supply unit 210 supplies the wafer 1 onto the transport belt 202, and includes a suction head 212 and an arm 211. The wafer 1 that is stably held while being vacuum-sucked by the suction head 212 is carried onto the transport belt 202 by the arm 211.
The wafer take-out unit 220 takes out the wafer 1 on which nanoimprinting has been completed on the transfer belt 202 from the transfer belt 202, and includes a suction head 222 and an arm 221. The wafer 1 that has undergone nanoimprinting is taken out of the conveyor belt 202 by the arm 221, stably held and moved while being vacuum-sucked by the suction head 222.

  The material of the nanoimprint layer 2 used in this embodiment is generally used as a room temperature imprint material. Here, imprinting is started at a temperature of about 50 degrees and peeling is performed at a temperature of about 40 degrees. Since the viscosity of the nanoimprint layer 2 decreases as the temperature is increased, by starting imprinting at a temperature of about 50 degrees, pressing becomes easier and the molding performance is improved. Further, when the temperature is slightly lowered at the time of peeling, the material of the nanoimprint layer 2 having a larger coefficient of thermal expansion than the mold shrinks, and the mold peeling becomes easier. On the other hand, since the belt-shaped nanomold 201 is pressed against the nanoimprint layer 2 to reversely transfer the surface shape of the belt-shaped nanomold 201, the HSQ constituting the nanoimprint layer 2 needs to flow. When the flow of HSQ is insufficient, the shape accuracy to be formed may be deteriorated. In view of these problems, in this embodiment, nanoimprinting is performed under the above temperature conditions.

  Here, details of the belt-shaped nanomold 201 and the fine structure of the surface will be described. In the belt-shaped nanomold 201, a small piece mold made of nickel produced by an electroforming method having a thickness of 0.07 mm and 160 × 160 mm is bonded to a loop-shaped stainless steel belt. The electroforming method applied to the formation of the mold of the present embodiment is such that palladium serving as a catalyst is attached to the surface of a matrix, a nickel conductive film is formed by an electroless plating method, and electroplating is performed using this nickel conductive film as a seed. This is one of the methods for producing a metal film that is formed by forming a film on the surface of the mother mold by thickening the film by peeling the plating film from the mother mold.

  The belt-shaped nanomold 201 can also be obtained by a manufacturing method other than the above. In the method of forming the belt-shaped nanomold 201 other than the above, first, the photosensitive resist is subjected to fine processing by electron beam drawing to obtain a primary master mold. Next, a primary stamper is formed by electroplating the primary matrix. Next, the primary stamper is repeatedly heat-imprinted on a long resin film to obtain a secondary master mold, and nickel is plated on the secondary master mold by an electroforming method to obtain a long secondary stamper. A desired belt-shaped nanomold 201 can be obtained by loop processing the obtained long secondary stamper.

  FIG. 3 illustrates the fine structure of the surface of the belt-shaped nanomold 201. The belt-shaped nanomold 201 has fine irregularities 201a arranged on the surface thereof in a grid-like arrangement. The mutual arrangement and schematic shape of the fine uneven shape 201a at this time are as follows. In other words, it is a regular lattice arrangement with an arrangement pitch P of 0.3 μm, the protrusion shape is a cylinder with a tip diameter of 0.1 μm and a height of 0.2 μm, and the root portion is slightly expanded and deformed into a truncated cone shape. ing. Although this uneven shape 201a can be a pure cylinder, in order to avoid wear and deformation of the mold, the tip portion is slightly rounded and the root portion is slightly expanded. Is preferred.

  Here, in order to form a fine shape on the upper surface of the wafer optically for reducing the surface reflection and increasing the light absorption rate into the solar cell, in this embodiment, it is formed on the belt-shaped nanomold 201. The fine concavo-convex shape to be formed is constituted by a protrusion 201a as shown in FIG. However, in order to form a structure capable of optically reducing surface reflection on the upper surface of the wafer, a mold having a recess having the same shape as the protrusion 201a shown in FIG. 3 may be used. Also, considering the air adhering to the uneven shape before the start of nanoimprinting and the lack of fluidity of the nanoimprint layer 2 that occurs during nanoimprint processing, the uneven shape formed on the belt-shaped nanomold 201 is easy to ensure fluidity. It is preferable to use a protrusion shape, that is, a protrusion shape that is not rectangular.

  In the sheet nanoimprint process (step S122 in FIG. 1) of the present embodiment, the uneven shape 201a formed on the belt-shaped nanomold 201 is transferred onto the nanoimprint layer 2 using the above sheet nanoimprint apparatus 200, and the uneven portion (Nanoimprint shape 2a) is formed. FIG. 4 shows a fine shape of the nanoimprint layer 2 immediately after imprinting by the belt-shaped nanomold 201. The nanoimprint layer 2 is applied to the surface layer of the wafer 1 by about 0.3 μm. However, the material of the portion where the fine uneven shape 201a is in contact with the belt-shaped nanomold 201 is pressed by the sheet nanoimprint process in step S122. Pushed away. Thereby, the nanoimprint layer 2 is deformed into a film having a large number of nanoimprint shapes 2a on the surface. At this time, the material of the nanoimprint layer 2 slightly remains on the bottom surface 2b of the nanoimprint shape 2a corresponding to the tip of the uneven shape 201a. This is due to the fact that the material of the nanoimprint layer 2 is used in a high viscosity state, so that the material cannot be completely pushed away.

  In practice, this residual film corresponds to the lower layer, and is not a problem because it is interdiffused with silicon oxynitride, which is the surface passivation film of the wafer 1, by a subsequent heat treatment process.

Returning to FIG. 1 again, the description of the method for manufacturing the solar cell will be continued.
After the sheet nanoimprint process in step S122, an N ₂ cure process (step S123) is performed. The nanoimprint layer 2 in a spin-coated state has an unevaporated volatile component inside, but contains a volatile component such as water in a state in which the nanoimprint layer 2 is taken into a molecule by a Si—H bond. For this reason, in subsequent high-temperature heat processing, in order to maintain the nanoimprint shape 2a, it heats in nitrogen atmosphere. In order to promote drying, the temperature is kept at 120 ° C. for a short time, and then heated at 400 ° C. for about 30 minutes to break the Si—H bond. By the N ₂ curing process of step S123, the nanoimprint layer 2 is slightly reduced in thickness, but the overall shape is kept substantially uniform, most of the material is changed to SiO ₂ , and heat resistance is improved.

  Then, a back surface Al coating process (step S108) and a back surface Al baking process (step S109) heated to about 800 degrees are performed, and a back surface Al electrode is formed on the back surface of the wafer 1. The back surface Al electrode has an effect of preventing carrier recombination on the back surface of the wafer 1 by diffusing Al atoms as a dopant in the vicinity of the back surface of the solar cell to form a high concentration P layer. The back surface Al electrode is formed by applying an aluminum paste to the entire back surface by silk screen printing.

The aluminum paste used as the back Al electrode is in a state of being attached to the wafer surface with organic components mainly composed of ethyl cellulose and glycol ether solvent immediately after printing, and is heated in the atmosphere to a temperature of about 700 to 800 degrees. For this reason, the organic component is burned and lost, and at the same time, the low melting point glass frit contained in the aluminum paste at about 3 w% is melted, and the silicon substrate and the aluminum layer are adhered.
The back surface Al firing process of step S109 is the first process of a series of electrode paste firing, and firing and alloy growth proceed in the subsequent firing process, so the firing time may be short.

  Furthermore, a back surface Ag application process (step S110) and a back surface Ag baking process (step S111) are implemented, and a back surface Ag electrode is formed on the back surface Al electrode. The back surface Ag electrode is formed directly on the back surface Al electrode formed in the back surface Al baking step, which is the previous process, for the purpose of collecting current from the back surface, and is formed into a striped electrode having a width of about 2 mm. In the back surface Ag coating process in step S110, silver paste is screen-printed and applied onto the back surface Al electrode, and then fired by heating again to a temperature of about 700 to 800 degrees in the back surface Ag firing process in step S111.

The Si—H bonds remaining in the nanoimprint layer 2 formed on the surface of the wafer 1 by the thermal processes of step S109 and step S111 are almost completely replaced with SiO _2, and the nanoimprint shape 2a of the nanoimprint layer 2 is almost destroyed. However, the film is strong as a whole, and a film having a refractive index of 1.45 to 1.46 is obtained.

  Thereafter, a surface Ag coating process (step S112) and a surface Ag baking process (step S113) are performed to form a surface silver electrode on the surface side of the wafer 1. The surface silver electrode includes a striped finger electrode having a width of about 0.1 mm and a pitch of about 2 mm, and a bus electrode having a width of about 2 mm and orthogonal thereto. In this case as well, the surface Ag coating process in step S112 is performed by screen printing, and it is necessary to determine in advance stable printing conditions for the narrow finger portion and the wide bus portion. Here, the film having the fine nanoimprint shape 2a described above is formed on the entire surface of the wafer 1, but it is not necessary to position the nanoimprint shape 2a and the surface silver electrode relative to each other. This is because the nanoimprint shape 2a is sufficiently fine and the pitch P is narrow, so that it is a uniform and flat base film from the viewpoint of the screen printing process. On the contrary, in the case of having fine irregularities on the order of sub-μm, the apparent contact angle with respect to the solvent in the silver paste is increased, so that bleeding of the printed pattern is reduced.

In the surface Ag firing process of step S113, the nanoimprint layer 2 is a SiO ₂ layer having a thickness of slightly more than 0.3 μm, and is eroded by the melting of the glass frit contained in the firing of the surface silver electrode together with the underlying silicon oxynitride layer. It does not prevent electrical bonding between the surface silver electrode and the phosphorous diffusion layer on the surface of the wafer 1.

  In this way, the wafer 1 on which electrode formation has been completed proceeds to the next mounting process (step S114), and wiring bonding between cells, resin sealing, and the like are performed. The nanoimprint shape 2 a of this embodiment example is arranged on the front and back of the wafer 1. Then, EVA (Ethylene Vinyl Acetate copolymer) which is melted and crystallized in the sealing process and becomes a transparent resin is formed on the nanoimprint layer 2 where the nanoimprint shape 2a is formed. Accordingly, an antireflection effect is obtained by the difference in refractive index between EVA, which is a transparent resin, and silicon oxide of the nanoimprint layer 2, and the fineness of the pitch of the fine nanoimprint shape 2a formed on the nanoimprint layer 2.

  Further, in the solar cell manufacturing method of the present embodiment, the wiring process and the resin sealing process itself do not require special consideration, and the conventional automatic stringer, layup device, and sealing device are used for the mounting process. Can be realized. Furthermore, in the solar cell manufacturing method according to the present embodiment, stress concentration is less likely to occur compared to the conventional solar cell in which the surface of the wafer 1 itself has been subjected to unevenness processing. There is an advantage that can be reduced.

  That is, according to the present embodiment example, since there is no local removal processing that weakens the wafer itself, the risk of the wafer breaking in the subsequent mounting process can be greatly reduced. Furthermore, since the low-speed dicing process and the expensive and low-throughput photolithography process are not used, the manufacturing cost can be greatly reduced. In addition, no powerful chemicals such as hydrofluoric acid or nitric acid are used, and no semiconductor cleaning process is required to neutralize and clean these chemical residues, greatly simplifying the production line and reducing environmental impact. it can.

  Although the first embodiment of the present invention has been described above, the present invention can also be realized by the following embodiment.

<2. Second Embodiment>
Hereinafter, the manufacturing method of the solar cell which concerns on the 2nd Embodiment of this invention, and the manufacturing apparatus used therewith are demonstrated using FIGS.

  FIG. 5 is a process flow showing a method for manufacturing a solar cell according to the second embodiment of the present invention. The present embodiment is different from the first embodiment in that the nanoimprint process (step S120) is performed after the surface Ag firing. Since it is the same up to the silicon oxynitride film forming step (step S107), the duplicate description is omitted. Further, in the present embodiment example, after the silicon oxynitride film forming process in step S107, the back surface Al coating process (step S108) to the surface Ag firing process (step S113) in the first embodiment are performed.

  FIG. 6 shows a schematic configuration diagram of the surface shape of the wafer 1 at the stage where the surface Ag baking process of step S113 is completed. As shown in FIG. 6, the surface Ag electrode 3 is formed on the surface of the wafer 1 by printing and baking. The front surface Ag electrode 3 is formed by two bus bar electrodes 3a that vertically cut the wafer 1 and a number of finger electrodes 3b orthogonal to the bus bar electrodes 3a. In FIG. 6, the shape generally called 2 buses is illustrated, but the number of bus bar electrodes 3a may be increased. For example, there is no problem even when the bus bar electrodes 3a are increased to three, four,... For the purpose of preventing resistance loss due to current flowing through the thin finger electrodes 3b. Further, the bus bar electrode 3a is formed on the back surface of the wafer 1, and the bus bar electrode 3a is made conductive by using a via hole penetrating the wafer 1 from the finger electrode 3b on the front surface, so that there is no bus bar electrode 3a on the surface of the wafer 1. Even in the case of a solar cell of the type, this process can be executed in the same manner.

  From this state, the nanoimprint process (step S120) is performed next. As shown in FIG. 6, the surface Ag electrode 3 is formed on the surface of the wafer 1, and the surface Ag electrode 3 surface is about several μm. Swelled from the surface of the wafer 1. Furthermore, the surface has irregularities of about 1 to 2 μm due to the sintered Ag particles, and is not suitable for imprinting. However, since the surface Ag electrode 3 originally does not transmit light, only the exposed portion 1a of the wafer 1 needs to be subjected to the antireflection treatment.

  In the nanoimprint process in step S120, first, an organic thin film is applied by spin coating (step S124) to form a nanoimprint layer. In this organic thin film coating step, a material having a UV curable acrylate-based material as a main component and a metal oxide nanoparticle increasing the refractive index as an additive is used as a material for the nanoimprint layer. Since the thin film is applied by spin coating, it is also effective to dilute with a reactive volatile component (for example, butyl acrylate).

  Since the surface Ag electrode 3 is formed on the surface of the wafer 1 as described above and this may inhibit the uniform wetting of the coating liquid, the rotation speed is about several tens of revolutions / minute for the first few seconds of the spin coating. In this state, the coating material is blended, and then the liquid supply is continued while increasing the number of revolutions to make the coating film uniform. Thereafter, when the liquid supply is stopped and a part of the reactive volatile component is dried by spin coating, the nanoimprint layer 2 having a viscosity suitable for nanoimprinting is obtained.

  Next, UV imprinting is performed (step S125). In the UV imprint process, a pattern transfer apparatus 300 shown in FIG. 7 is used.

  A pattern transfer apparatus 300 used as a solar cell manufacturing apparatus according to this embodiment includes an adsorption stage 320 that adsorbs and holds the wafer 1 and an upper mold that holds the nanomold 301 on which the concavo-convex shape to be transferred to the wafer 1 is formed. It comprises a pressure plate 302 and an ultraviolet irradiation device 360 as a light source.

  The suction stage 320 constitutes a substrate placement unit, and is disposed on the XYZθ table 321 so that the X direction, the Y direction, the Z direction, and the rotation angle θ can be positioned. The wafer 1 on which the nanoimprint layer 2 is formed is placed on the suction stage 320 and vacuum-sucked. A claw 322 driven by a claw driving mechanism 323 is installed on the side surface of the suction stage 320. When the wafer 1 is peeled from the nanomold 301, the wafer 1 on which the nanoimprint layer 2 is formed and the nanomold 301 are separated. The nail | claw 322 is inserted in an interface and the trigger of peeling is made.

  FIG. 8 is a microstructure diagram of the surface of the nano mold of the pattern transfer apparatus 300 used in this embodiment. As shown in FIG. 8, the nano mold 301 has grooves 301 a for avoiding the surface Ag electrode 3 formed on the wafer 1, and a fine surface portion 301 b in contact with the nano imprint layer 2 on the surface of the wafer 1 is fine. An uneven shape 301c is provided. Actually, since the surface Ag electrode 3 is on the order of mm, the depth of the groove 301a is about 10 μm, and the fine uneven shape 301c is about sub μm, it is difficult to show in actual scale, but in FIG. The shape schematic of the mold 301 is shown. Since the nano mold 301 is made of quartz having excellent ultraviolet transmittance and the surface of the groove 301a is also smooth, the surface Ag electrode 3 can be observed by the camera 350 through the groove 301a when the wafer 1 is brought close to the nano mold 301. It is.

  In addition to quartz, the nanomold 301 may be made of a resin material having ultraviolet transparency and having a fine pattern formed thereon. The nano mold 301 is held by suction on an upper pressure plate 302 made of quartz.

  As shown in FIG. 7, the upper pressure plate 302 has an outer periphery held by an outer frame 303 and is attached to an annular attachment member 304. The upper pressure plate 302 sucks and holds the nanomold 301 so that the concavo-convex shape of the nanomold 301 faces the nanoimprint layer 2 of the wafer 1 held on the suction stage 320. The nano mold 301 is vacuum-sucked by a suction passage 302 a provided on the outer periphery of the upper pressure plate 302 via an annular mounting member 304 on the back surface of the upper pressure plate 302. The vacuum supply is performed by a vacuum pump (not shown) by connecting the negative pressure tube 306 to the negative pressure tube attachment portion 305, and when the nanomold 301 is not attached or detached, the suction passage 302a is always kept at a negative pressure.

  The ultraviolet irradiation device 360 is disposed above the upper pressure plate 302 that holds the nanomold 301. When the nano mold 301 is pressed against the nano imprint layer 2 of the wafer 1 on the adsorption stage 320, the ultraviolet irradiation device 360 transmits the nano mold 301 and irradiates the nano imprint layer 2 with ultraviolet light.

  In the pattern transfer apparatus 300 of this embodiment, when the wafer 1 is placed on the suction stage 320 and fixed by suction, the XYZθ table 321 brings the wafer 1 close to the nanomold 301. Here, the camera 350 as an imaging device moves onto the nano mold 301, passes through the upper pressure plate 302 and the nano mold 301, observes the surface Ag electrode 3 of the wafer 1, and is positioned in the groove 301 a of the nano mold 301. To do. This positioning is performed by relatively moving the nanomold 301 and the wafer 1 using an image acquired by the camera 350.

  Thereafter, the outer frame 303 holding the outer periphery of the upper pressure plate 302 is lowered, and the nanomold 301 is pressed against the wafer 1. The outer frame 303 has three push-down mechanisms (not shown) at intervals of 120 degrees, and these push-down mechanisms are provided at rotationally symmetrical positions on the outer periphery of the nanomold 301. Then, each of the three push-down mechanisms generates a uniform push-down force, thereby pressing the nanomold 301 evenly against the wafer 1. The applied pressure is calculated to be 0.2 MPa, for example, by calculating the area of the flat portion 301b.

  Here, the camera 350 is retracted, and ultraviolet irradiation is performed by the ultraviolet irradiation device 360, and the nanoimprint layer 2 is cured by ultraviolet rays. Since the nanoimprint layer 2 shrinks in volume by about 1% due to ultraviolet curing, the adhesion between the nanomold 301 and the wafer 1 is reduced, and adhesion by the entire vacuum adsorption force is the main.

  In this state, when the outer frame 303 is pulled upward while pressing the nail 322 against the interface between the nanomold 301 and the wafer 1, the wafer 1 is peeled off from the nanomold 301 without difficulty while being vacuum-sucked by the suction stage 320. .

  As described above, a portion of the surface of the wafer 1 that is not covered with the surface Ag electrode 3 has a shape in which the fine nanoimprint shape 2a by nanoimprint covers the entire surface. Moreover, since the nanoimprint shape 2a is made of a material having a higher refractive index than EVA that seals the wafer 1 in the subsequent mounting process, the surface reflection of the solar cell can be reduced. In addition, since there are no dents or grooves that reduce the strength of the wafer 1 due to stress concentration on the surface of the wafer 1 itself, there is no possibility of damage to the wafer 1 due to stress due to heating and pressurization during sealing. In the second embodiment of the present invention, the same effect as that of the first embodiment can be obtained.

  In the first and second embodiments described above, it is assumed that the fine uneven shape 201a (FIG. 3) and the uneven shape 301c (FIG. 8) by nanoimprint are arranged at the regular lattice points with the pitch P. This is because the formation of the nanomold is made by electron beam drawing to facilitate positional accuracy, but in order to increase the density without reducing the size of the concavo-convex shape, as shown by the concavo-convex shape 201c in FIG. In addition, it is preferable to arrange the grids so that the lattice points come to the vertices of equilateral triangles (illustrated by broken lines). As described above, when the lattice points are arranged at the vertices of the equilateral triangle, the uneven shape can be increased when the same arrangement density is required, so that there is an advantage that damage to the nanomold can be reduced.

  Moreover, when the pitch P has periodicity at a pitch of 1 / integer of the wavelength to be absorbed by the solar cell, the nanoimprint layer 2 may work as a reflective film. On the other hand, a method in which the pitch is sufficiently narrowed to prevent interference is preferable, but the size of the fine shape is reduced accordingly. Since the wavelength of sunlight is broad, it is considered that unnecessary interference cannot be avoided with a size of about a fraction of the wavelength. In such a case, the adverse effect can be reduced by using a random arrangement so that the pitch P between adjacent fine shapes is not constant, instead of arranging the fine shapes in a lattice arrangement with periodicity. .

  Actually, the complete random arrangement is disadvantageous in terms of manufacturing cost in the process of manufacturing the nanomold. Therefore, pseudo-random arrangement is performed in the range of several μm × several μm, and the nanomold is manufactured using this as a master stamper. Thereby, unnecessary interference of light can be sufficiently avoided. This is because sunlight is poorly coherent, and interference is sufficiently avoided when several wavelengths are separated. The first and second exemplary embodiments are suitable when a sufficiently fine nanomold is not selected and an arrangement interval of a fraction of a wavelength is adopted as the pitch P. In addition, according to the 1st and 2nd embodiment, there exists an advantage which can extend the lifetime of a nanomold.

  In addition, as described above, the fine shape of the nanomold has been mainly explained with respect to the fine protrusions from the viewpoint of the shrinkage of the nanoimprint layer 2 and the ease of detachment of the gas component from the inside, but the present invention is not limited to this. . It is also possible to make the imprint shape into a projection with the mold side having a fine concave surface. In this case, since the mold manufacturing method becomes easy, there is an advantage that the nanomold 301 used particularly when the pattern transfer apparatus 300 of the second embodiment is used can be made inexpensive.

  Furthermore, in the first and second embodiments described above, the nanoimprint layer 2 provided on the wafer 1 is thinner than the wavelength of light, and the pitch P of the nanoimprint shape 2a is also less than a fraction of the wavelength of light. The explanation was given on the assumption. However, the present invention is not limited to this, and the manufacturing method of the present invention is effective even when the nanoimprint layer is larger than half the wavelength of light and several times as large as the wavelength of light. The manufacturing method of the solar cell in that case is demonstrated below.

<3. Third Embodiment>
In FIG. 10, the process flow of the manufacturing method of the solar cell which concerns on the 3rd Embodiment of this invention is shown. 10, parts corresponding to those in FIG. 1 are denoted by the same reference numerals, and redundant description is omitted.

  This embodiment is different from the first and second embodiments in that the nanoimprint process in step S120 is performed after the residual film removal process in step S106.

  In this case, the nanoimprint layer 2 formed in the SOG coating process (step S121) has a thickness of about 1 μm. Further, the sheet imprint process (step S122a) is performed using the sheet nanoimprint apparatus 200 shown in FIG. The fine uneven shape 201a (see FIG. 3) of the belt-shaped nanomold 201 of the sheet nanoimprint apparatus 200 used in the present embodiment is a substantially parabolic projection, and the imprint shape 2c is a substantially parabolic concave mirror shape. Become.

  FIG. 11 shows an imprint shape 2c formed on the belt-shaped nanomold 201 of the sheet nanoimprint apparatus 200 of the present embodiment. The imprint shape 2c is a substantially paraboloid that is cut off at the tip portion to become a flat portion 2d, and the flat portion 2d at the tip is a portion that hits the focal point of the concave mirror. Due to this shape, the sunlight reflected by the inclined surface can be efficiently collected on the flat portion 2 d, and the sunlight absorbed by the inclined surface is captured inside the nanoimprint layer 2 and efficiently delivered to the wafer 1. After forming the imprint shape, an antireflection film formed of a laminated film of silicon oxynitride and silicon oxide is formed (step S115).

  Thereafter, similarly to the first embodiment, the back surface Al coating process (step S108) is performed to the front surface Ag baking process (step S113), and the process proceeds to the mounting process (step S114).

  In this embodiment, the material of the nanoimprint layer 2 is SOG having excellent heat resistance, and the pattern transfer uses the sheet nanoimprint apparatus 200, thereby facilitating the formation of the antireflection film. In this embodiment, there is an advantage that the manufacturing process can be simplified as compared with the method of forming a large number of fine concave structures on the surface, which has been performed by the conventional exposure process and isotropic wet etching. Also in this embodiment, the same effect as that of the first embodiment can be obtained.

  DESCRIPTION OF SYMBOLS 1 ... Wafer, 2 ... Nanoimprint layer, 2a ... Nanoimprint shape, 3 ... Surface Ag electrode, 3a ... Busbar electrode, 3b ... Finger electrode, 200 ... Sheet nanoimprint apparatus, 201 ... belt-shaped nanomold, 201a ... uneven shape, 202 ... transport belt, 203 ... preheating roll, 204 ... heating and pressing roll, 205 ... peeling roll, 206 ... Return roll, 207... Input roll, 208... Take-out roll, 210... Wafer supply unit, 211... Arm, 212. Nano mold 302 ... Upper pressure plate 303 ... Outer frame 306 ... Negative pressure tube 320 ... Adsorption stage 321 ... XYZθ Table, 350 ... Camera, 360 ... Ultraviolet irradiation device

Claims (13)

Preparing a substrate;
Forming a flexible transfer layer on the substrate surface;
A step of pressing a mold having a concavo-convex shape formed on the transfer layer, transferring the concavo-convex shape to the transfer layer, and forming a concavo-convex portion for reducing light reflectance on the substrate surface. Production method. Including a step of forming a surface electrode after the step of forming the uneven portion,
The manufacturing method of the solar cell of Claim 1. The transfer layer having flexibility is made of a material containing water glass, and includes a step of heating the transfer layer to change to silicon oxide after the step of forming the concavo-convex portion. A method for manufacturing a solar cell. The mold is formed of a band-shaped body formed in a loop shape, and the mold is pressed against the transfer layer by rotating a roll disposed on an inner peripheral portion of the band-shaped mold, and an uneven portion is formed on the transfer layer. The manufacturing method of the solar cell as described in any one of Claims 1-3. Including a step of forming a surface electrode before the step of forming the uneven portion,
The manufacturing method of the solar cell of Claim 1. The transfer layer is made of a material containing a photosensitive organic material. After pressing the mold against the transfer layer, the transfer layer is cured by irradiating light to transfer the uneven shape of the mold. The method for manufacturing a solar cell according to claim 5, wherein uneven portions are formed in the layer. The solar cell manufacturing method according to claim 5 or 6, wherein positioning of the mold and the substrate is performed by observing a surface electrode formed on the substrate surface through the mold made of a light transmissive material. Method. The solar cell according to claim 1, wherein the uneven portion is a concave surface that forms a parabolic surface, and further includes a step of forming an antireflection film after the formation of the uneven portion that forms the parabolic surface. Manufacturing method. A substrate mounting portion for mounting a substrate having a flexible transfer layer formed on the surface;
A mold having a concavo-convex shape for transferring the concavo-convex portion to a flexible transfer layer formed on the surface of the substrate;
An apparatus for manufacturing a solar cell. The apparatus for manufacturing a solar cell according to claim 9, wherein the mold is formed of a belt-like body formed in a loop shape, and a fine uneven shape is formed on a surface thereof. The substrate mounting portion is composed of a band-shaped body formed in a loop shape, and is disposed at a position facing the mold formed in the loop shape,
The length of the part facing the said mold of the said board | substrate mounting part is longer than the part facing the said board | substrate mounting part of the said mold. The manufacturing apparatus of the solar cell of Claim 10. The mold is made of a light transmissive material,
The solar cell manufacturing apparatus according to claim 9, further comprising a light source for irradiating the transfer layer with light when the mold is pressed against the transfer layer to cure the transfer layer. An imaging device for obtaining an image of the substrate through the mold;
The solar cell manufacturing apparatus according to claim 12, wherein positioning of the mold and the substrate is performed by relatively moving the mold and the substrate using an image acquired by the imaging device.

Images