JP2006121011A - Method for processing transparent layer and thin-film photoelectric converting device using same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method of efficiently insulating and separating a transparent electrode layer without damaging an insulating primary layer, and to improve output characteristics and reliability of an integrated thin-film photoelectric converting device. <P>SOLUTION: The integrated thin-film photoelectric converting device includes the transparent electrode layer 42 comprising an insulating primary layer 42a and a conductive oxide film layer 42b which are deposited in order on a glass substrate 41, a photoelectric converting layer 44, a reverse-surface electrode layer 46, a sealing resin layer 48, and an organic protective layer 49. Sunlight to photoelectrically be converted is made incident on the photoelectric converting device from the side of the glass substrate 41. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention provides a method for patterning a transparent electrode layer and a manufacturing method for an integrated thin film photoelectric conversion device.

  Solar cells that convert solar energy into electrical energy are attracting attention as a new energy source with low environmental impact. In particular, a thin-film silicon type solar cell formed by a CVD process is attracting attention as a low-cost solar cell because it has a smaller film thickness than a crystalline solar cell and uses less raw materials. As a method for manufacturing a thin film silicon solar cell, a predetermined device structure is manufactured by combining a film forming process by a CVD method or the like and a processing process for forming a solar cell. For example, an integrated structure in which a plurality of unit cells are connected in series or in parallel on a single substrate is employed to form a solar cell module.

  FIG. 1 shows the structure of such an integrated thin film silicon solar cell. On the transparent substrate 1, a transparent electrode layer 2, a photoelectric conversion layer 3, and a back electrode layer 4 are sequentially laminated. The transparent electrode layer 2 has a separation groove 5 for insulating and dividing, the photoelectric conversion layer has a connection groove 6 for electrically connecting the transparent electrode layer and the back electrode layer, and the thin film silicon layer and the back electrode layer are divided. The unit cells 8 are connected in series by the separating grooves 7 that are connected.

As the transparent conductive film used for the transparent electrode layer, indium tin oxide (hereinafter referred to as ITO), tin oxide (hereinafter referred to as SnO ₂ ), zinc oxide (hereinafter referred to as ZnO) or the like is usually used. ITO is widely used because of its high electrical conductivity. However, if In is a rare metal and the production volume is small, there is a problem in stable supply when the demand for transparent conductive films increases. SnO ₂ is cheaper than ITO and has a low free electron concentration, so that a film with high transmittance can be obtained. However, it has drawbacks in that it has low conductivity and low plasma resistance. In contrast, zinc is abundant and inexpensive as a resource. In addition, ZnO is suitable as a transparent conductive film for solar cells because it has characteristics such as high plasma resistance and high mobility because of its high mobility.

  The photoelectric conversion layer is composed of a thin film silicon layer, which is composed of a pin junction, and an amorphous silicon layer or a crystalline silicon layer is used as a power generation layer. The amorphous silicon layer has sensitivity characteristics mainly in the visible region of the solar spectrum, and the crystalline silicon layer mainly has sensitivity properties in the near infrared region. And a tandem structure in which crystalline silicon is laminated.

  Since the back electrode layer also has the purpose of reflecting light into the cell, a metal film such as silver (Ag), aluminum (Al), or chromium (Cr) is used. In the process of manufacturing a unit cell, there are a method using chemical etching such as a photofab and a method of mechanical scribing as a method of processing these thin films to form separation grooves and connection grooves. From the viewpoint of cost, a method of scribing with a laser beam is advantageous.

  When scribing such a thin-film silicon solar cell, a YAG laser is often used because of cost and stability, and the wavelength is appropriately selected according to the film to be processed. In particular, scribe to the transparent electrode layer is a problem in scribe. That is, the purpose of this scribe is to electrically separate adjacent electrodes, but if insulation is insufficient, a leak path will be generated between cells, leading to deterioration of photoelectric conversion characteristics due to leakage current. In addition, when the laser beam processing power is increased to ensure sufficient insulation, the doped doping material diffuses to lower the resistance value of the transparent electrode layer such as Al, Ga, B, etc. due to the increase in the film temperature. However, this also causes a leak path.

  In addition, when glass is used as the transparent substrate, an insulating base layer is inserted as an alkali barrier layer in order to prevent alkali metals such as Na in the glass from diffusing into the transparent electrode layer and the power generation layer. By increasing the processing power, scribing is performed at the same time, and an alkali metal penetrates into the transparent electrode layer, resulting in a decrease in reliability.

  Similarly, when glass is used, by increasing the processing power, minute cracks are generated in the glass itself, and the reliability with respect to strength is lowered. Moreover, it goes without saying that the width of the separation groove is increased and the power generation efficiency of the solar cell is reduced.

In particular, when ZnO is used as the transparent electrode layer, the above problem becomes significant. This is because when ZnO is sublimated during sublimation, the resistivity is one or more orders of magnitude smaller than that of SnO ₂ , so that an insulation failure between the separation grooves is likely to occur, and the power is increased to prevent this. The reason is that the alkali barrier layer is scribed.

In addition, it is desirable to form fine irregularities on the ZnO surface in order to scatter light and obtain a large current. However, when only a chemical vapor phase method such as CVD is used as a forming method, a film is formed to form the irregularities. The thickness must be at least 1 μm. Therefore, the film thickness becomes thicker than that of SnO ₂ and scribing becomes more difficult.

  In the method disclosed in Patent Document 1 as a method for solving such a problem, after scribing with a fundamental wave of a YAG laser (wavelength 1064 nm) on a transparent electrode layer made only of ZnO formed on glass, an acid is added. Alternatively, the zinc oxide residue is removed by washing with an alkaline solution. However, in this method, not only the residue but also the film itself may be etched and the physical properties of the film may change. Further, no alkali barrier layer is inserted between the glass and ZnO, and no mention is made of any preventive measures for the diffusion of alkali components from the glass.

The method described in Patent Document 2 discloses a structure in which an insulating base layer made of insulating fine particles is inserted between glass and a transparent electrode layer mainly composed of zinc oxide. Is not mentioned at all, and if the method according to Patent Document 1 is used, a change in physical properties of the film becomes a problem, and the known method does not provide an optimum method.
JP 2000-114555 A JP2003-243676

  In this way, when the transparent electrode layer is scribed with a laser, if the laser power is weakened, insulation failure occurs and the characteristics of the solar cell deteriorate. Conversely, when the power is increased, damage to the insulating base layer and the glass as the alkali barrier layer increases, leading to a decrease in reliability. Moreover, since the width of the separation groove is increased, the power generation area is reduced. The present invention solves these drawbacks of the prior art, and in the scribing to the transparent electrode layer of the integrated photoelectric conversion device, the scribing method having excellent reliability and the high conversion efficiency created by this method It is an object to provide a thin film photoelectric conversion device having high reliability.

  In order to solve the above-mentioned problems, a transparent electrode layer processing method according to the present invention uses a laser beam to deepen a transparent electrode layer in which at least one insulating underlayer and a zinc oxide layer are sequentially laminated on a glass substrate. In the step of forming the separation groove in the vertical direction, the laser beam has a uniform output intensity distribution in the beam cross section and a flat front end of the beam, and the front end of the separation groove is formed under the insulating property. It is characterized by not penetrating the formation.

  The zinc oxide layer is preferably formed by chemical vapor deposition.

  Moreover, it is preferable that the film thickness of an insulating base layer is 0.1-1.0 micrometer, and the film thickness of a conductive oxide film layer is 1.0 micrometer or more.

  And it is preferable that an insulating base layer contains the microparticles | fine-particles which consist at least of silicon oxide, and the particle size of this microparticles is 0.1-1.0 micrometer.

  In addition, a thin film photoelectric conversion device according to the present invention is an integrated photoelectric conversion device in which a transparent electrode layer, a photoelectric conversion layer, and a back electrode layer are laminated in this order on a glass substrate, and includes a transparent electrode layer processed according to the present invention. Yes.

  The photoelectric conversion layer may be composed of at least one amorphous silicon-based photoelectric conversion unit and at least one crystalline silicon-based photoelectric conversion unit.

  According to the present invention, it is possible to insulate and separate the transparent electrode layer with high yield without damaging the insulating base layer and the glass. Further, the output characteristics and reliability of the integrated thin film photoelectric conversion device can be improved.

  FIG. 4 is a schematic cross-sectional view of an integrated thin film photoelectric conversion device according to an embodiment of the present invention. This integrated thin film photoelectric conversion device includes an insulating underlayer 42a and a transparent electrode layer 42 comprising a conductive oxide film layer 42b sequentially deposited on a glass substrate 41, a photoelectric conversion layer 44, a back electrode layer 46, a sealing resin. A layer 48 and an organic protective layer 49 are included. For this photoelectric conversion device, sunlight to be subjected to photoelectric conversion is incident from the glass substrate 41 side.

As the glass substrate, it is desirable to use a soda lime plate glass having a large surface area that can be obtained at low cost, having high transparency and insulating properties, and having a smooth main surface of both SiO ₂ , Na ₂ O and CaO. .

The insulating underlayer 42a formed in the present invention preferably contains fine particles made of at least silicon oxide (SiO ₂ ). This is because SiO ₂ has a refractive index lower than that of the transparent electrode layer and has a value close to that of a glass substrate. Further, since SiO ₂ has high transparency, it is suitable as a material used on the light incident side. Further, for the purpose of adjusting the refractive index of the insulating underlayer 106, in addition to SiO _2, titanium oxide (TiO _2), aluminum oxide (Al ₂ O _3), indium tin oxide (ITO), zirconium oxide (ZrO ₂₎ Or fine particles such as magnesium fluoride (MgF ₂ ).

A method for forming the insulating base layer 42a containing fine particles on the surface of the glass substrate 41 is not particularly limited, but a method of applying together with a binder forming material containing a solvent is desirable. The adhesive layer that plays a role in improving the adhesion strength between the fine particles and between the fine particles and the glass substrate 41 is preferably an inorganic material in consideration of long-term reliability and durability against photoelectric conversion layer formation conditions (particularly temperature). Specific examples include metal oxides such as silicon oxide, aluminum oxide, titanium oxide, zirconium oxide, and tantalum oxide. In particular, when SiO ₂ fine particles are adhered to a glass substrate, if silicon oxide mainly composed of the same silicon is used for the adhesive layer, the adhesion is strong due to the formation of silicide bonds, the transparency is good, and the refractive index is also high. It is preferable because it is close to the substrate and fine particles.

  Examples of the method for applying the coating solution to the surface of the glass substrate 41 include a dipping method, a spin coating method, a bar coating method, a spray method, a die coating method, a roll coating method, and a flow coating method. A roll coating method is preferably used for uniformly forming the underlayer. When the coating operation is completed, the coated thin film is immediately dried by heating. Since the film thus formed contains fine particles, the shape of the convex portion is a curved surface, and the height of the concave and convex portions is relatively uniform. Therefore, an electrical short circuit to the conductive oxide film layer 42b and the photoelectric conversion layer 44 to be formed later is not caused.

  Further, in order to enhance the effect as an alkali barrier film for preventing alkali metal from glass from entering the conductive oxide film layer 42b and the photoelectric conversion layer 44, and when the thickness of the conductive oxide film layer is not constant In order to prevent color unevenness due to light interference, a combination of a plurality of thin films having different refractive indexes may be interposed between the insulating base layer 42a and the conductive oxide film layer 42b.

  As a material of the conductive oxide film layer 42b, it is preferable to use a transparent conductive oxide film containing at least ZnO formed on the surface in contact with the photoelectric conversion layer 44 by a chemical vapor deposition method. This is because according to this method, ZnO is suitable for a photoelectric conversion device because it can form a fine concavo-convex structure having a light confinement effect even at a low temperature of 200 ° C. or lower and is a material having high plasma resistance. For example, ZnO of the substrate for a photoelectric conversion device of the present invention is formed by a CVD method under a reduced pressure condition using a dopant composed of organic zinc, an oxidizing agent, and a Group 3 element as a raw material at a substrate temperature of 200 ° C. or lower. Further, in this case, a thin film in which the interval between the uneven peaks and valleys is approximately 50 to 500 nm and the uneven process difference is approximately 20 to 200 nm is preferable in terms of obtaining the light confinement effect of the photoelectric conversion device. The substrate temperature here means the temperature of the surface where the substrate is in contact with the heating unit of the film forming apparatus.

  When the conductive oxide film layer 42b is composed only of a thin film mainly composed of ZnO, the average thickness of the ZnO film is preferably 0.5 to 5 μm, and more preferably 1 to 3 μm. This is because if the ZnO film is too thin, it will be difficult to sufficiently provide unevenness that effectively contributes to the light confinement effect, and it will be difficult to obtain the necessary conductivity as a transparent electrode, and if it is too thick, the light absorption by the ZnO film itself will be difficult. This is because the amount of light that passes through ZnO and reaches the photoelectric conversion layer is reduced, so that the conversion efficiency is lowered. Furthermore, when it is too thick, the film forming cost increases due to an increase in the film forming time.

In order to separate the transparent electrode layer 42 into a plurality of regions corresponding to a plurality of integrated photoelectric conversion cells, a separation groove 43 is formed by laser scribing. In this case, the second harmonic of the Nd-YVO4 laser is used in the present invention. This is because, compared with the Nd-YAG laser, the beam quality is high quality and the oscillation is repeated in a high frequency region, so that scribing is possible at high speed and the productivity is excellent. In addition, the second harmonic (wavelength of 532 nm) was used because the purpose is to reduce the cost of the apparatus by processing at the same wavelength including the scribe after formation of the photoelectric conversion layer and the back electrode layer. In addition, damage to the insulating underlayer or glass is reduced, and conversion efficiency and reliability are improved. Of course, there is no problem with scribing by chemical etching or mechanical processing.
The second harmonic of the Nd-YVO4 laser preferably uses a beam whose output intensity distribution in the beam cross section is uniform and whose beam tip is flat. This is because, in the case of Gaussian-type output intensity distribution, the output intensity at the center of the beam is particularly increased by increasing the laser power when the scribe is defective, and the separation groove 43 is formed through the insulating base layer 42a as an alkali barrier layer. This is because there is a risk that the function of the system is impaired. Of course, you may use the fundamental wave (wavelength 1064nm) of Nd-YVO4 laser or Nd-YAG laser. However, even in such a case, it is preferable to use the one in which the output intensity distribution of the beam cross section is made uniform and the beam tip is flat.

  A photoelectric conversion layer 44 is formed on the transparent electrode layer 42 on which the separation groove 43 is formed. The photoelectric conversion layer 44 may be a single unit as illustrated, or a plurality of units may be stacked. The photoelectric conversion layer 44 preferably has absorption in the main wavelength region (400 to 1200 nm) of sunlight, for example, a unit using a crystalline silicon-based thin film. In addition to silicon, “silicon-based” materials include semiconductor materials containing 50% or more of silicon, such as silicon carbide and silicon germanium.

  The photoelectric conversion layer 44 thus laminated is divided into a plurality of strip-shaped semiconductor regions by connection grooves 45 formed by laser scribing as in the case of the transparent electrode layer 42. In the present invention, the second harmonic of an Nd-YVO4 laser is used. This is because the transparent conductive layer has less absorption compared to the fundamental wave, so that the damage given is reduced, and since the absorption to the photoelectric conversion layer is large, the photoelectric conversion layer can be efficiently scribed. .

  Also in this case, it is preferable to use a beam whose output intensity distribution in the beam cross section is uniform and whose beam tip is flat.

A back electrode layer 46 is formed on the laser-patterned photoelectric conversion layer 3. As the back electrode layer 46, it is preferable to form at least one metal layer made of at least one material selected from Al, Ag, Au, Cu, Pt and Cr by sputtering or vapor deposition. Further, a layer made of a conductive oxide such as ITO, SnO ₂ , or ZnO may be formed between the photoelectric conversion layer 44 (not shown). This conductive oxide layer has a function of improving the adhesion between the photoelectric conversion layer 44 and the back electrode layer 46, increasing the light reflectance of the back electrode layer 46, and preventing the chemical change of the photoelectric conversion layer 44. .

  The back electrode layer 46 is scribed in the same manner as the photoelectric conversion layer 44. In this case, a beam is incident from the glass side and the photoelectric conversion layer 44 is thermally sublimated to scribe with the back electrode layer 46 to form a plurality of separation grooves 47.

  In this case as well, it is preferable to use a beam whose output intensity distribution in the beam cross section is uniform and whose beam tip is flat. It is also possible to scribe from the back electrode layer side using the third harmonic and the fourth harmonic. There is no problem with scribing by chemical etching or mechanical processing.

  Further, as an example of the thin film photoelectric conversion device of the present invention, a tandem thin film solar cell in which an amorphous silicon photoelectric conversion unit layer and a crystalline silicon photoelectric conversion unit layer are sequentially stacked on the transparent electrode layer 42 may be used (see FIG. Not shown). The amorphous silicon photoelectric conversion layer is sensitive to light of about 360 to 800 nm, and the crystalline silicon photoelectric conversion layer can photoelectrically convert light up to about 1200 nm longer than that. A photoelectric conversion device arranged in this order from the amorphous silicon photoelectric conversion unit layer and the crystalline silicon photoelectric conversion unit layer is a photoelectric conversion device that can effectively use incident light in a wider range.

  Further, the back side of the photoelectric conversion device is sealed with an organic protective layer 49 via a sealing resin layer 48. As the sealing resin layer 48, a resin capable of bonding the organic protective layer 49 to the photoelectric conversion device is used. Examples of such a resin include EVA (ethylene / vinyl acetate copolymer), PVB (polyvinyl butyral), PIB (polyisobutylene), and a silicone resin. As the organic protective layer 49, an insulating film having excellent moisture resistance and water resistance such as a fluororesin film such as a polyvinyl fluoride film (for example, Tedlar film (registered trademark)) or a PET film is used. It is done. The organic protective layer 49 may have a single layer structure or a stacked structure in which these layers are stacked. Furthermore, the organic protective layer 49 may have a structure in which a metal foil made of aluminum or the like is sandwiched between these films. Since a metal foil such as an aluminum foil has a function of improving moisture resistance and water resistance, the organic protective layer 49 having such a structure can effectively protect the photoelectric conversion device from moisture. These sealing resin layer 48 / organic protective layer 49 can be simultaneously attached to the back side of the photoelectric conversion device by a vacuum laminating method.

Embodiments of the present invention will be described below.
As a first embodiment, scribing of the transparent electrode layer was performed. As shown in FIG. 5, first, a white plate glass having a thickness of 4 mm and a size of 910 × 455 mm was used as the glass substrate 51, and a SiO 2 layer 52 was deposited to a thickness of 100 nm on one side thereof by a roll coating method. Next, a ZnO layer 53 was deposited on the SiO2 layer by CVD. The glass substrate was carried into the film forming chamber and the substrate temperature was adjusted to 150 ° C. Subsequently, 1000 sccm of hydrogen, 500 sccm of diborane diluted to 5000 ppm with hydrogen, 100 sccm of water, and 50 sccm of diethylzinc were introduced, and the pressure in the deposition chamber was set to 1 Torr. Under this condition, a ZnO layer was deposited by 1.5 μm. About the produced ZnO layer, the sheet resistance, the haze rate, and the transmittance | permeability were measured using the resistance measuring device, the haze meter, and the spectrophotometer, respectively. As a result, the resistivity was 9Ω / □, the haze ratio was 20%, and the transmittance at a wavelength of 1000 nm was 81%.

  Next, scribing was performed on the transparent electrode layer by the following method. Since the beam shape immediately after being emitted from the laser oscillator is Gaussian, the configuration shown in FIG. That is, the beam 21 emitted from the laser oscillator is passed through the expander 22 to thicken the beam, and the focused beam through the condenser lens 23 is passed through the imaging slit 24. By adjusting the center of the beam to the center of the opening of the imaging slit 24 and adjusting the width of the slit opening, a beam having a flat tip can be obtained. By passing this beam further through the imaging lens 25, a beam having a predetermined width is imaged on the processing surface. As a result, the beam output intensity distribution immediately after emission from the laser oscillator is Gaussian as shown in FIG. 2A, but the output intensity distribution as shown in FIG.

  The second harmonic of the Nd-YVO4 laser is obtained by converting the fundamental wave (wavelength 1064 nm) output from the Nd-YVO4 rod excited by the laser diode through a nonlinear optical crystal. The processing conditions were a Q switch frequency of 20 kHz, a processing speed of 400 mm / sec, a processing point power of 1.6 W, and a beam diameter of 30 μm. The substrate was placed on an XY table, a laser beam was incident from the glass surface on which the transparent electrode layer was not formed, and the stage was moved to perform scribing perpendicular to the short side direction of the substrate. 100 separation grooves were formed at a distance of 8.9 mm between the separation grooves. The separation groove width at this time was 30 μm. The insulation resistance between the adjacent electrodes was measured with a tester without cleaning after scribing. When measured at an applied voltage of 250 V, the insulation resistance was 1 MΩ or more in all stages. When the depth of the separation groove was measured with a laser step microscope, it was 1.55 μm, and it was revealed that the scribing was only halfway through the SiO 2 layer.

  Next, a first comparison form will be described. First, the SiO2 layer 52 and the ZnO layer 53 are formed in this order on the glass substrate 51 as in the first embodiment. Next, this transparent electrode layer was scribed with a processing optical system shown in FIG. 3 using a fundamental wave (wavelength 1064 nm) output from an Nd-YAG rod excited by a krypton arc lamp. That is, the surface to be processed is irradiated with the beam emitted from the laser oscillator through the attenuator 31 and the condenser lens 32 to perform scribing. In this case, the energy output intensity distribution of the fundamental wave of the Nd-YAG laser is Gaussian. As specific processing conditions, the Q switch frequency was 10 kHz, the processing speed was 200 mm / sec, the processing point power was 5.2 W, and the beam diameter was 60 μm. The scribe pattern was the same as in the first example. As a result, the separation groove width was 60 μm. When the insulation resistance between adjacent electrodes was measured by the same method as in the first embodiment, the insulation resistance of 21 stages out of 100 was 1 MΩ or less. Further, ultrasonic waves were applied and cleaning was performed with pure water in a state where a rotating brush was applied to the scribe surface (hereinafter referred to as ultrasonic cleaning), and the ZnO residue was removed. 1 MΩ or less. When the depth of the separation groove was measured, it was 1.55 μm.

  Next, a second comparative embodiment will be described. The second comparative example differs from the first comparative example only in that the laser power is different in the first comparative example. That is, the laser beam was scanned with a laser power of 6.5 W and a beam diameter of 75 μm. As a result, the width of the separation groove was 75 μm, and the insulation resistance before ultrasonic cleaning was 1 MΩ or less at 5 out of 100 stages, and the insulation resistance was 1 MΩ or more at all stages after ultrasonic cleaning. However, the depth of the separation groove was 1.6 μm and reached the glass surface.

  From the above results, the output intensity distribution of the beam cross section is uniformed, and when a beam with a flat tip is used, the power is evenly incident on the scribe surface, so that the separation groove does not penetrate through the insulating base layer. It is thought that can be formed. On the other hand, when the output intensity distribution is a Gaussian type, the tip of the beam is sharp, so the incident power density is different at the center and the periphery of the beam, and the processed shape is non-uniform. In other words, if insulation failure occurs, the minimum energy required to increase the laser output and thermally sublimate ZnO to the periphery is incident on the center, and more energy is incident on the center, and the insulating underlayer also scribes. It is thought that it was done.

  Next, a second embodiment will be described. As in the first embodiment, as shown in FIG. 5, first, a white plate glass having a thickness of 4 mm and a size of 910 × 455 mm is used as the glass substrate 51, and the SiO 2 layer 52 is formed to a thickness of 100 nm by roll coating on one side thereof. Deposited on. Next, a ZnO 3 layer 53 was deposited on the SiO 2 layer 52 by CVD. Next, separation grooves 54 were formed by laser scribe on the transparent conductive layer by the method according to the present invention. The second harmonic of the Nd-YVO4 laser has a flat tip at the configuration shown in FIG.

  The processing conditions were a Q switch frequency of 20 kHz, a processing speed of 400 mm / sec, a processing point power of 1.6 W, a beam diameter of 30 μm, and 100 scribing was performed in the short side direction of the substrate. Next, after ultrasonic cleaning was performed on the substrate, an amorphous silicon photoelectric conversion unit 55 having a film thickness of about 300 nm was formed on the ZnO layer 53 by a plasma CVD method. The amorphous silicon photoelectric conversion unit 55 includes a p-type microcrystalline layer 551, a p-type amorphous silicon carbide layer 552, an i-type amorphous silicon layer 553, and an n-type layer 554.

The p-type microcrystalline layer 551 was formed by introducing silane, diborane, and hydrogen, applying a pressure of 350 Pa, and high frequency power for plasma excitation at a density of 150 mW / cm ² , and the film thickness was set to 15 nm. The reason why such a microcrystalline layer is formed on the ZnO layer is that the ohmic characteristics of the p-type amorphous silicon carbide layer 552 and the ZnO layer are not excellent. F. This is because of a decrease. Therefore, in order to further improve the ohmic characteristics, a microcrystalline layer formed by adding methane to silane, diborane, or hydrogen may be inserted between the microcrystalline layer and ZnO.

In order to clean the surface of ZnO and improve bonding, plasma treatment may be performed with hydrogen, argon, nitrogen, or the like. Subsequently, an amorphous silicon carbide layer 552 is formed by introducing silane, diborane, hydrogen, and methane into the chamber and applying a pressure of 133 Pa and high frequency power for plasma excitation at a density of 170 mW / cm ² . At this time, the film thickness was set to 10 nm.

Next, silane and hydrogen are introduced as a film forming gas, and an i-type amorphous silicon layer 553 is formed to a thickness of 300 nm by applying a pressure of 50 Pa and a plasma excitation high frequency power of 120 mW / cm ^2. The Further, silane, phosphine, and hydrogen were introduced into the chamber as a film forming gas, the pressure was set to about 350 Pa, and the n-type layer 554 was formed to a thickness of about 10 nm. Next, a crystalline silicon photoelectric conversion unit 56 was produced by a plasma CVD method.

The crystalline silicon photoelectric conversion unit 56 includes a p-type layer 561, an i-type crystalline silicon layer 562, and an n-type layer 563. First, the p-type layer 561 was formed by introducing silane, diborane, and hydrogen into the chamber and applying a pressure of 200 to 600 Pa and high frequency power for plasma excitation at a density of 150 mW / cm ² . The film thickness at this time was set to 15 nm.

Next, silane and hydrogen are introduced as a film forming gas, and an i-type crystalline silicon layer 562 is formed to a thickness of 1.4 μm by applying a pressure of 850 Pa and high frequency power for plasma excitation of 150 mW / cm ^2. It was done. Further, silane, phosphine, and hydrogen were introduced into the chamber as a film forming gas, the pressure was set to about 350 Pa, and the n-type layer 563 was formed to a thickness of about 15 nm. This substrate was placed in a sputtering chamber, and argon was introduced as a sputtering gas, and ZnO doped with aluminum (Al) was formed by RF discharge to a film thickness of 30 nm (not shown).

  Next, the crystalline photoelectric conversion unit 56 and the amorphous photoelectric conversion unit 55 are separated so that the distance between the midpoint of the separation groove 54 and the midpoint of the connection groove 57 of the connection groove 57 reaching the upper surface of the transparent conductive layer is 55 μm. Formed in parallel with each other. In this case, the second harmonic (wavelength 532 nm) of Nd—YVO 4 laser was used. The processing conditions were a Q switch frequency of 20 kHz, a processing speed of 400 mm / sec, a processing point power of 0.09 W, and a beam diameter of 40 μm. As a result, the line width of the connection groove was 40 μm.

  Subsequently, this substrate was placed in a sputtering chamber, and a back electrode layer 58 was formed. The back electrode layer 58 was composed of a ZnO layer 581 and an Ag layer 582, and the ZnO layer 581 was formed by a sputtering method. Argon was introduced as a sputtering gas, and ZnO doped with aluminum (Al) was deposited by RF discharge to a thickness of 60 nm. On the ZnO layer 581, an Ag layer 582 having a thickness of 200 nm was formed by a sputtering method. Next, the separation groove 59 that separates the back electrode layer 58, the crystalline photoelectric conversion unit 56, and the amorphous photoelectric conversion unit 55 and reaches the upper surface of the transparent electrode layer is set to a distance of 105 μm between the separation groove 54 and the center of the separation groove 59. They were formed in parallel so that In this case, the second harmonic (wavelength 532 nm) of Nd—YVO 4 laser was used. The processing conditions were a Q switch frequency of 20 kHz, a processing speed of 400 mm / sec, a processing point power of 0.09 W, and a beam diameter of 40 μm. As a result, the separation groove width was 40 μm.

  Further, an EVA sheet is placed as the sealing resin layer 60 on the back side of the tandem photoelectric conversion device, and a black fluororesin-based sheet (trade name: Tedlar) is placed thereon as the organic protective layer 61 and laminated by a vacuum laminating method. The tandem photoelectric conversion device was sealed.

The thus obtained 100-stage integrated thin film photoelectric conversion device was irradiated with AM1.5 light at a light quantity of 100 mW / cm ² , and the output characteristics were measured. The short-circuit current density was 12.4 mA / cm ² , the fill factor was 72.0%, and the conversion efficiency was 11.9%.

Next, in order to carry out a reliability test of the integrated thin film photoelectric conversion device, it was left open in an oven maintained at an atmospheric pressure and an air atmosphere at a temperature of 85 ° C. and a humidity of 85% RH according to JIS standards. The output characteristics were evaluated after standing for a period of time. As a result, the open circuit voltage (Voc) was 1.33 V, the short circuit current density (Jsc) was 12.4 mA / cm ² , the fill factor (FF) was 71.9%, and the conversion efficiency (Eff) was 11. It was 9% and the characteristics did not change.

  On the other hand, a third comparative embodiment will be described. As shown in FIG. 5, the second comparative example is used until the substrate in which the SiO 2 layer 52 and the ZnO layer 53 are formed as transparent electrode layers is scribed with 100 separation grooves 54 in the short side direction of the substrate. The method used was the same. Next, after ultrasonic cleaning was performed on the substrate, an amorphous silicon photoelectric conversion unit 55 having a film thickness of about 300 nm was formed on the ZnO layer 53 by a plasma CVD method. The amorphous silicon photoelectric conversion unit 55 includes a p-type microcrystalline layer 551, a p-type amorphous silicon carbide layer 552, an i-type amorphous silicon layer 553, and an n-type layer 554. The film forming conditions and film thicknesses of these layers were the same as in the second embodiment. Next, a crystalline silicon photoelectric conversion unit 56 was manufactured under the same film forming conditions and film thickness as in the third example. This substrate was put into a sputtering chamber, and argon was introduced as a sputtering gas, and ZnO doped with aluminum (Al) was formed by RF discharge to a thickness of 30 nm (not shown).

  Next, the crystalline photoelectric conversion unit 56 and the amorphous photoelectric conversion unit 55 are separated, and a connection groove 57 reaching the upper surface of the transparent conductive layer is formed in parallel so that the distance between the center of the separation groove 54 and the connection groove 57 is 95 μm. did. In this case, scribing was performed with the processing optical system shown in FIG. 3 using the second harmonic (wavelength 532 nm) output from the Nd-YAG rod excited by the arc lamp. As a result, the connection groove width was 75 μm.

  Subsequently, this substrate was placed in a sputtering chamber, and a back electrode layer 58 was formed under the same film forming conditions and film thickness as in the second embodiment. Next, the back electrode layer 58, the crystalline photoelectric conversion unit 56, and the amorphous photoelectric conversion unit 55 are separated, and the separation groove 59 reaching the upper surface of the transparent conductive layer is separated. The distance between the separation groove 54 and the center of the separation groove 59 is 195 μm. They were formed in parallel so that In this case, scribing was performed with the processing optical system shown in FIG. 3 using the second harmonic (wavelength 532 nm) output from the Nd-YAG rod excited by the arc lamp. As a result, the separation groove width was 75 μm.

Further, an EVA sheet as the sealing resin layer 60 is placed on the back side of the integrated thin film photoelectric conversion device, and a black fluororesin-based sheet (trade name: Tedlar) is placed thereon as the organic protective layer 61, and laminated by a vacuum laminating method. The integrated thin film photoelectric conversion device was sealed.
The integrated thin-film photoelectric conversion device thus obtained was irradiated with AM 1.5 light at a light amount of 100 mW / cm ² and measured for output characteristics. As a result, the open circuit voltage per stage was 1.33 V, and the short-circuit current density. Was 12.2 mA / cm ² , the fill factor was 71.5%, and the conversion efficiency was 11.6%.

Next, in order to carry out a reliability test of the integrated thin film photoelectric conversion device, it was left open in an oven maintained at an atmospheric pressure and an air atmosphere at a temperature of 85 ° C. and a humidity of 85% RH according to JIS standards. The output characteristics were evaluated after standing for a period of time. As a result, the open circuit voltage (Voc) is 1.33 V, the short circuit current density (Jsc) is 12.0 mA / cm ² , the fill factor (FF) is 70.0%, and the conversion efficiency (Eff) is 11. 2%. F. The characteristics deteriorated due to the decrease in the thickness.
From the above results, in the second embodiment in which the scribing to the transparent electrode layer is performed without penetrating the insulating base layer, the initial characteristics are maintained in the reliability test. On the other hand, in the third comparative embodiment in which scribing was made through the insulating base layer, F.R. F. As a result, the initial characteristics deteriorated.

It is typical sectional drawing which shows the structure of an integrated thin film silicon solar cell. An explanatory view of a laser beam optical system concerning an embodiment of the invention, and a figure showing distribution of energy density of a laser beam. It is explanatory drawing of the laser beam optical system concerning the comparison form of this invention. It is typical sectional drawing which shows the laminated structure of the integrated thin film photoelectric conversion apparatus sealed. It is typical sectional drawing which shows the laminated structure of the integrated thin film photoelectric conversion apparatus sealed.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Transparent substrate 2 Transparent electrode layer 3 Photoelectric conversion layer 4 Back surface electrode layer 5 Separation groove 6 Connection groove 7 Separation groove 21 Beam 22 Condensing lens 23 Slit 24 Imaging lens 31 Attenuator 32 Condensing lens 41 Glass substrate 42 Transparent electrode Layer 42a insulating base layer 42b conductive oxide film layer 43 separation groove 44 photoelectric conversion layer 45 connection groove 46 back electrode layer 47 separation groove 48 sealing resin layer 49 organic protective layer 51 glass substrate 52 SiO2 layer 53 ZnO layer 54 separation groove 55 Amorphous silicon photoelectric conversion unit 56 Crystalline silicon photoelectric conversion unit 57 Connection groove 58 Back electrode layer 59 Separation groove 60 Sealing resin layer 61 Organic protective layer 551 p-type microcrystalline layer 552 p-type amorphous silicon carbide layer 553 i-type amorphous silicon layer 554 n-type layer 561 p-type layer 562 i-type crystalline silicon Layer 563 n-type layer 581 ZnO layer 582 Ag layer

Claims (6)

  In the step of forming a separation groove in the depth direction using a laser beam on a transparent electrode layer in which at least one insulating underlayer and a zinc oxide layer are sequentially laminated on a glass substrate, the laser beam has a beam cross section. A method for processing a transparent electrode layer, characterized in that the output intensity distribution is uniform, the beam tip is flat, and the tip of the separation groove to be formed does not penetrate the insulating base layer.   The method for processing a transparent electrode layer according to claim 1, wherein the zinc oxide layer is formed by a chemical vapor deposition method.   2. The transparent electrode layer according to claim 1, wherein the insulating base layer has a thickness of 0.1 to 1.0 [mu] m, and the conductive oxide film layer has a thickness of 1.0 [mu] m or more. Method.   2. The method for processing a transparent electrode layer according to claim 1, wherein the insulating base layer includes at least fine particles made of silicon oxide, and the fine particles have a particle size of 0.1 to 1.0 [mu] m.   An integrated photoelectric conversion device in which a transparent electrode layer, a photoelectric conversion layer, and a back electrode layer are laminated in this order on a glass substrate, comprising a transparent electrode layer processed by the method according to claim 1, Conversion device. The thin film photoelectric conversion device according to claim 6, wherein the photoelectric conversion layer includes at least one amorphous silicon photoelectric conversion unit and at least one crystalline silicon photoelectric conversion unit.

Images