JP4431712B2 - Manufacturing method of solar cell - Google Patents

Description

The present invention relates to a method for manufacturing a solar cell.

  Currently, high efficiency and low cost are simultaneously required for solar cells. Thus, recently, an OECO (Obliquely Evaporated Contact) method has attracted attention as a method for producing a high-efficiency and low-cost solar cell. The OECO method is a solar cell fabrication method devised by R. Hezel et al. Of Institut fur Solarenergie forschung Hameln / Emmerthal (ISFH) in Germany. For example, it is disclosed in Vol. (Hereinafter, a solar cell manufactured by the OECO method is also referred to as an OECO solar cell). A method for manufacturing the OECO solar cell will be briefly described with reference to FIG. First, the OECO solar cell includes a plurality of parallel grooves 2 ′ (rectangular shape) on a main surface to be a light receiving surface 1 ′ a of a silicon single crystal substrate 24 ′ (hereinafter also referred to as a semiconductor single crystal substrate or simply a substrate). And has a structure in which an electrode 6 'for output extraction is formed on the inner side surface 2'a on one side in the width direction of each groove 2'. As a groove forming method, a dicing saw is usually used. That is, the semiconductor single crystal substrate 24 ′ is placed on the surface of the processing table, and the dicing saw is horizontally moved in a state of being spaced apart from the surface of the processing table, thereby forming the groove forming surface of the semiconductor single crystal substrate 24 ′. (Light receiving surface) A groove 2 'is formed in 1'a. Formation of the electrode 6 'on the groove side surface 2'a is performed by vacuum vapor deposition from an oblique direction with respect to the main surface serving as the light receiving surface 1'a of the semiconductor single crystal substrate 24'. In this method, the electrode material can be selectively deposited on the side surface 2'a of the groove and the upper surface 2'b of the convex portion by the shielding effect of the deposited metal by the groove itself. Since the metal deposited in this manner has a different thickness between the convex portion upper surface 2′b and the side surface 2′a, an etching process is performed after the deposition step, and the thickness of the side surface 2′a and the upper surface 2′b. The metal is removed so that the electrode remains on the side surface by the difference of.

  By adopting the above-described structure, the shadowing loss of the solar cell is reduced to about 5% of the entire light receiving surface 1'a. For example, in the case of a solar cell in which electrodes are produced by a screen printing method, the shadowing loss generally reaches about 12%. Therefore, it can be said that the shadowing loss in an OECO solar cell is significantly small. As a result, high energy conversion is achieved. Efficiency can be achieved.

  The solar cells that are in practical use today can be classified into silicon (single crystal, polycrystal and amorphous), compound semiconductor, and others. The silicon single crystal is excellent in energy conversion efficiency and manufacturing cost. In general, solar cells using are widely manufactured. Usually, a single crystal obtained by a Czochralski method (hereinafter simply referred to as CZ method) or a floating zone melting method (hereinafter referred to simply as FZ method) is used for a silicon single crystal substrate for solar cells. A substrate obtained by slicing an ingot with a wire saw is used. A substrate sliced with a wire saw is used for as-slicing.

  However, when slicing with a wire saw, as the amount of cut into the ingot increases, the number of abrasive grains stagnating at the processing site increases and the cutting width increases. As a result, as schematically shown in FIG. 13, the thickness of the semiconductor single crystal substrate 24 ′ decreases from the cutting start side (left side of the drawing) toward the cutting end side (right side of the drawing). For this reason, the use of the semiconductor single crystal substrate 24 ′ in the as-slicing can greatly reduce the cost by shortening the process, while the thickness of the semiconductor single crystal substrate 24 ′ becomes non-uniform. is there. The variation in thickness is 20 to 30 μm in a substrate having a diameter of 4 inches, and the variation increases as the substrate has a larger diameter.

  When grooves are formed on such a solar cell substrate 24 ′ by a conventional method, the thickness of the substrate 24 ′ is not uniform, and therefore the depth of the groove 2 ′ from the light receiving surface 1′a is the light receiving surface. It is not constant over the entire surface of 1'a. That is, in the upper blade type dicing saw, the lower edge position of the teeth is constant, so that the depth of the cut becomes larger and the groove 2 'becomes deeper as the substrate portion has a larger thickness.

  In the substrate 24 ′ in which the depth of the groove 2 ′ is not uniform as described above, when the electrode 6 ′ is formed on the groove side surface 2′a by the above-described deposition method, the height of the formed electrode 6 ′ is designed. The value is insufficient or the metal is deposited on the groove bottom surface 2'c. For example, if the height of the electrode 6 'is less than the set value, the resistance loss at the electrode 6' increases. Further, if the metal is formed on the groove bottom surface 2'c, the shadowing loss increases, so that there is a problem that the energy conversion efficiency is deteriorated. If an etching process is performed to remove this, the electrode formation region becomes insufficient and the resistance loss increases. These increases in shadow loss and resistance loss naturally lead to a decrease in energy conversion efficiency of the solar cell.

  Further, not only in the case of the OECO solar cell as described above, if the depth of the groove formed in the substrate 24 ′ is not uniform, the characteristics of the manufactured solar cell may vary. For example, when the depth of the electrode contact groove formed on the back surface of the solar cell is deeper than a set value, the interface area between the electrode and the substrate increases, and the recombination rate at the interface increases. On the other hand, when it is shallower than the set value, the contact between the electrode and the substrate cannot be sufficiently obtained, and the contact resistance increases. Further, when grooves are formed on the light receiving surface of the solar cell, if the depth of the grooves is non-uniform, the recombination rate in the thickness direction and the vertical plane is non-uniform due to surface area variations. These variations in characteristics cause variations in the output voltage, and as a result, the output of the solar cell may be reduced.

  A first object of the present invention is to provide a solar cell in which electrodes are appropriately formed so that shadow loss and resistance loss can be reliably suppressed in the OECO method. A second object is to provide a solar cell manufacturing method that can easily make the groove depth uniform in a solar cell in which a groove such as an OECO solar cell is formed, and thus can achieve high efficiency of the solar cell at low cost. That is.

Means for Solving the Problems and Effects of the Invention

In order to achieve the first object, the solar cell according to the first aspect of the present invention is:
It has a light receiving surface in which a plurality of grooves substantially parallel to each other are formed on the first main surface of a semiconductor single crystal substrate, and an inner surface on one side in the width direction of each groove is used as an electrode forming region, and an electrode for output extraction is used here A solar cell provided with
In any cross section perpendicular to the longitudinal direction of each groove, for the groove having the maximum groove depth appearing in the cross section, the lower end of the formed electrode in the thickness direction of the semiconductor single crystal substrate and the non-electrode formation in the groove When an angle formed between a straight line connecting the upper ends of the side surfaces and a reference line orthogonal to the substrate thickness direction is θ, and a groove width defined as a distance between the opening edges of the grooves is W ₁ ,
The minimum groove depth h in each groove is
h ≧ W ₁ tan θ (1)
It is characterized by satisfying.

  As shown in FIG. 4, the lower end LE of the electrode 6 formed on one inner side surface 2 a of the groove 2 in the thickness direction of the semiconductor single crystal substrate 24 and the upper end TE of the non-electrode-formed inner side surface 2 c of the groove 2 are The angle θ formed by the connecting straight line and the reference line orthogonal to the substrate thickness direction corresponds to the incident angle (deposition angle) of the electrode material vapor when the electrode 6 is formed by oblique deposition. The minimum groove depth h in each groove 2 is the groove depth at the position where the groove depth is the smallest in the grooves 2 having a depth distribution in the length direction.

The present inventors have studied intensively, and the minimum groove depth h, the groove width W ₁ and the angle θ is, when satisfying the ▲ 1 ▼ expression conditions, or electrode is formed on the groove bottom surface, the electrode formation area It has been found that problems such as shortage are effectively suppressed. In such a solar cell, the groove depth is sufficient for the short groove formed in the thinnest part of the substrate having a non-uniform thickness, and the electrode is appropriately formed in the substrate thickness direction. The resistance loss does not increase beyond the desired value. At the same time, the effect of reducing the shadow loss, which is the original purpose of the OECO solar cell, is also good.

  In order to maintain the mechanical strength of the solar cell itself, for example, a silicon single crystal substrate has a substrate thickness of about 150 to 300 μm. Of course, since it is necessary to reduce the thickness of the substrate in order to reduce the manufacturing cost of the solar cell, the mechanical strength of the solar cell itself may not be maintained simply by forming the groove deeply. That is, it is preferable that a high level of mechanical strength is maintained while maintaining the groove depth defined in the present invention.

  In addition, if the groove is formed so that the thickness of the semiconductor single crystal substrate at the bottom position of each groove observed in the cross section is uniform, the in-plane distribution of the substrate strength is also uniform, so the specific position where the thickness is insufficient Therefore, there is no risk of the substrate cracking. Here, the uniform thickness means that the variation is within a range of ± 10 μm.

  The advantages of forming the groove in the above direction are as follows. That is, the electrode formed in each groove | channel 2 is connected by the below-mentioned current collection bus-bar electrode formed in a groove | channel arrangement direction as what is called a finger electrode. On the other hand, as shown in FIGS. 3A and 3B, the groove 2 formed in the above-described direction is P2 side with the thickest position of the semiconductor single crystal substrate being P2 and the thinnest position being P1. The depth distribution is such that the depth gradually decreases toward the position P1. Therefore, the groove inner surface area where the finger electrode can be formed per unit groove length decreases as the position approaches the position P1 side.

  On the other hand, as shown in FIGS. 14, 15, and 16, the solar cell 1 is formed in each groove on the first main surface of the semiconductor single crystal substrate 24 so as to extend over each groove 2. A current collecting bus bar electrode 30 for electrically connecting the electrodes 6 (hereinafter referred to as finger electrodes) 6 to each other is formed. The bus bar electrode 30 can be formed so as to follow the inner surface of the groove 2 as shown in FIG. 18, or can be formed so as to fill each groove 2 as shown in FIG.

  As shown in FIG. 14, FIG. 15 and FIG. 16, it is desirable to determine the formation position of the bus bar electrode 30 so that it is closer to the thickest position P2 than to the thinnest position P1. If the bus bar electrode 30 is formed on the side close to the position P <b> 2, a region where the groove inner surface area for forming the finger electrode 6 may be insufficient can be kept away from the bus bar electrode 30. Since the current density in the finger electrode decreases as the distance from the bus bar electrode 30 increases, the resistance loss actually caused by the voltage drop is small because the current density itself is small even if the cause of the resistance loss exists. Therefore, in the groove formed as described above, the position where the groove becomes shallow can be aligned with the groove tip side away from the bus bar electrode 30. Accordingly, in the vapor deposition process, even if the electrode formation region is slightly short of the designed value due to problems such as processing accuracy, such shortage occurs only at a position away from the bus bar electrode 30, so that resistance loss There is almost no increase. In any of the configurations of FIG. 14, FIG. 15 and FIG. 16, a straight line DL orthogonal to the line segment passes through the midpoint of the line segment connecting the thinnest position P1 and the thickest position P2. When the semiconductor single crystal substrate 24 is divided into two, the entire bus bar electrode 30 is within the region to which the position P2 where the thickness is the thickest belongs. If it does in this way, the effect which keeps the tip by the side of position P1 of the finger electrode which affects resistance loss increase from busbar electrode 30 can be heightened further.

  In this specification, it is determined as follows whether the bus bar electrode 30 is closer to the position P2 or the position P1. First, as shown in FIG. 17, base lines L2 and L1 in contact with the outline of the semiconductor single crystal substrate 24 are drawn at the positions P2 and P1, respectively. Also, a pair of area defining lines L3 and L4 that are orthogonal to the base line L2 and in contact with the outline of the semiconductor single crystal substrate 24, and a pair of areas that are orthogonal to the base line L1 and that are in contact with the outline of the semiconductor single crystal substrate 24 Draw area defining lines L5 and L6. When the area surrounded by the bus bar electrode 30, the base line L2, and the area defining lines L3, L4 is S2, and the area surrounded by the bus bar electrode 30, the base line L1, and the area defining lines L5, L6 is S1, S2> If it is S1, it will be considered that the bus-bar electrode 30 is close to the position P2 side.

  Next, as shown in FIG. 3A, the longitudinal direction of each groove 2 is the first main line connecting the thickest position P2 (thickness h2) and the thinnest position P1 (thickness h1) of the semiconductor single crystal substrate. The angle formed between the straight line L and the straight line along the surface is preferably within 45 °, and is preferably as parallel as possible (that is, the angle is as close to 0 ° as possible). This is because, when the straight line L and the direction of the groove 2 are parallel, the groove depth on the position P2 side of each groove 2 is maximized, and conversely, the groove depth on the position P1 side is minimized. This is because the effect of minimizing the occurrence of the resistance loss is most remarkable. When the angle formed by the straight line L and the groove direction exceeds 45 °, the groove depth of the groove 2 located near the position P1 and the area of the groove inner surface are insufficient, and resistance loss becomes remarkable. Not desirable.

  A silicon single crystal substrate is most suitable as the solar cell substrate material. For example, most silicon single crystals produced by the CZ method have a pulling axis direction of <100>, and thus the plane orientation of the substrate produced as a solar cell substrate material is {100}. When a large number of grooves are formed along a <110> direction on the main surface of a substrate having a plane orientation of approximately {100} (hereinafter also referred to simply as a {100} substrate), stress is likely to concentrate on the groove cross-sectional shape. In the case where a portion is formed or a lot of damage is left during the groove forming process, there is a problem that the substrate is easily cleaved along the groove and is broken only by a slight external force. By setting the direction of formation of each groove in a direction that does not coincide with the <110> direction, the mechanical strength of the obtained solar cell can be significantly improved.

Next, the manufacturing method of the solar cell according to the second aspect of the present invention is as follows.
Forming a plurality of grooves on at least one main surface of the semiconductor substrate;
From the substrate feed surface of the processing table having a flat substrate feed surface, the blade portion of the grooving blade protrudes by a certain height, and the grooving blade is rotated.
One first main surface of the semiconductor substrate is brought into close contact with the substrate feed surface, and in this state, the substrate is placed along the substrate feed surface at a right angle to the thickness direction of the grooving blade. The groove is formed on the one main surface by being relatively moved toward the surface.

  According to the method for manufacturing a solar cell of the present invention, one main surface on which the groove of the semiconductor substrate is to be formed is brought into close contact with the flat substrate feed surface of the processing table. As a matter of course, the surface and the substrate feed surface are parallel to each other. In addition, since the blade portion of the grooving blade protrudes from the substrate feed surface by a certain height, the semiconductor substrate can be relatively moved at right angles to the thickness direction of the rotating grooving blade. For example, a groove having a certain height is formed on one main surface of the semiconductor substrate maintained parallel to the substrate feed surface. Thereby, in the OECO solar cell, the shielding effect with respect to the adjacent grooves is uniform in each groove, and therefore, an equivalent electrode is deposited in each groove. As a result, each electrode can be formed as set, and resistance loss or shadowing loss can be suppressed. Also, in solar cells other than the OECO solar cell, the depths of the various grooves can be made uniform, so that variations in the characteristics of the solar cell can be suppressed, and consequently a decrease in output can be suppressed.

  In the present invention, the blade portions of a plurality of integrally grooving blades (hereinafter also referred to as grooving blade assemblies) coupled to the rotary shaft at regular intervals are equal to each other from the substrate feed surface of the processing table. Make it protrude by height. By adopting such a grooving blade assembly, a plurality of grooves corresponding to a plurality of grooving blades can be collectively formed on the main surface only by relatively moving the semiconductor substrate once. At this time, since the protruding heights of the respective grooving blades are equal to each other, the depths of the plurality of grooves formed collectively are equal to each other. In addition, since each grooving blade is coupled to the grooving blade assembly at regular intervals, a plurality of parallel grooves arranged at regular intervals are formed on the main surface of the semiconductor substrate. In this case, the plurality of grooves may be collectively formed several times to form grooves on the entire main surface, but the number of each grooving blade to be coupled to the grooving blade assembly is determined. It is more desirable from the viewpoint of productivity that the number of grooves to be formed in the semiconductor substrate is equal to or more than the number of grooves to be formed, and the grooves are formed on the entire main surface by one relative movement of the semiconductor substrate.

  Furthermore, as a processing table used in the manufacturing method of the present invention, a processing table in which a shaving powder discharge groove is formed on the substrate feed surface can be used. When a groove is formed on the main surface of the semiconductor substrate, a portion to be a groove of the semiconductor substrate is cut by the blade portion of the grooving blade, so that shavings are generated as the groove is formed. If the generated shaving powder is left as it is, the shaving powder will be interposed between the semiconductor substrate and the processing table, and the close contact state between the semiconductor substrate and the processing table cannot be maintained well. . Specifically, since the semiconductor substrate is lifted from the processing table by the shaving powder to the extent of the size of the shaving powder, the substrate feed surface of the processing table and the main surface where the grooves of the semiconductor substrate are formed are slightly shifted from parallel. Become. Thereby, the depth of the groove formed on the main surface is not kept constant over the entire main surface. For example, the depth of the groove formed after a large amount of shavings is generated tends to be relatively shallow compared to the depth of the groove that is initially shaved.

  However, if the cutting powder discharge groove is formed on the substrate feed surface of the processing table, even if the groove formation proceeds and a large amount of cutting powder is generated, the generated cutting powder is taken into the groove. It is possible to maintain a form in which the substrate feed surface and the main surface on which the grooves of the semiconductor substrate are formed are always in good contact, and therefore the depth of the grooves formed on the main surface is determined at the beginning and end of the groove formation. It can be kept constant. Thereby, as mentioned above, the resistance loss and shadowing loss of the solar cell are reduced, and the energy conversion efficiency is improved. Further, the formation of the shaving powder discharge groove reduces the contact area between the semiconductor substrate and the processing table, and also reduces the frictional resistance when the semiconductor substrate is relatively moved. Thereby, the processing efficiency of groove formation is improved and the productivity is also improved.

  In addition, the depth of the formed groove can be made uniform by the method for manufacturing a solar cell as described above, and the width (difference between the maximum value and the minimum value) of the groove depth distribution over the entire groove forming surface is A solar cell having a thickness distribution of the semiconductor substrate of 5% or less of the width (difference between the maximum thickness and the minimum thickness) can be manufactured.

  DESCRIPTION OF THE PREFERRED EMBODIMENTS The best mode for carrying out the embodiments of the present invention will be described below with reference to the drawings.

(Embodiment 1)
FIG. 1 is an enlarged cross-sectional view partially showing one embodiment of a solar cell according to the first aspect of the present invention. In the solar cell 1, on the first main surface 24a of the p-type silicon single crystal substrate 24 cut out from the silicon single crystal ingot, for example, a large number of grooves having a width of about several hundred μm and a depth of about several tens to 100 μm. 2 are formed in parallel to each other. These grooves 2 can be collectively engraved by hundreds to thousands of rotating blades that are coaxially coupled and integrally rotated, but may be engraved in several operations. .

  An emitter layer 4 is formed on the first main surface 24a of the grooved substrate 24 by thermally diffusing phosphorus, which is an n-type dopant, and a pn junction is formed. A thin silicon oxide film 5 that functions as a tunnel insulating film is formed on the emitter layer 4 by, for example, a thermal oxidation method.

  An electrode 6 is formed on the silicon oxide film 5 on the inner surface on one side in the width direction of the groove 2. The electrode 6 is formed by vapor-depositing an electrode material (for example, a metal such as aluminum) on the inner surface of the groove in the vapor deposition apparatus. At the time of vapor deposition, the inner surface 2a on one side in the groove width direction is formed. In order to preferentially deposit the electrode material, the substrate 24 is arranged to be inclined relative to the vapor deposition source at a predetermined angle or more. During the vapor deposition, extra electrode material is also deposited on the top surface of the ridge 23 formed between the grooves 2 and 2, but this is removed with an etching solution such as a hydrochloric acid solution. The entire first main surface 24a of the substrate 24 including the electrode 6 is covered with a silicon nitride film 7 that functions as a protective layer and an antireflection film.

  Each groove 2 has an outer shape in a cross section perpendicular to the longitudinal direction of the groove 2 as one of a rectangular shape shown in FIG. 2A, a semicircular shape shown in FIG. 2B, and a V shape shown in FIG. 2C. Therefore, it is suitable for the present invention.

  The grooves 2 intersect each other as shown in FIG. 2D or FIG. 2E when the outline shape in the cross section orthogonal to the longitudinal direction of the grooves 2 is, for example, a rectangle shown in FIG. 2A or a V shape shown in FIG. 2C. Two sides 2a and 2b appear. In the case of a rectangular groove, as shown in FIG. 2D, the side portions 2a and 2b correspond to the groove side wall and the groove bottom, respectively, and the crossing angle between them is approximately 90 °. On the other hand, in the case of a V-shaped groove, the side portions 2a and 2b intersect at an acute angle at the groove low. In any case, if the crossing positions of these side portions are formed sharply, stress concentration tends to be caused, leading to a decrease in the strength of the solar cell. Therefore, the mechanical strength of the solar cell can be further increased by setting the cross-sectional outline shape of the groove 2 to a shape in which the rounded R1 or R2 is applied to the intersecting position of the side portions 2a and 2b.

  The size of R 1 or R 2 is set within a range in which the effect of preventing stress concentration is sufficiently achieved and the effect of reducing the series resistance derived from the groove shape is not impaired, for example, about 2 to 20 μm. It is desirable. Also, such rounding can be easily formed by carrying out chemical etching after engraving a groove by cutting an outer peripheral blade or the like. This chemical etching may also be used for etching for removing damage that occurs during the groove engraving process. In this case, it is desirable that the etching thickness be in the range of about 5 to 20 μm in order to make the formed radius within the above desired range. As the chemical etching solution, an aqueous potassium hydroxide solution or the like can be used.

  By the way, a single crystal ingot obtained by the Czochralski method (hereinafter simply referred to as the CZ method) or the floating zone melting method (hereinafter simply referred to as the FZ method) is obtained by slicing with a wire saw. The thickness of the substrate to be obtained is non-uniform (see FIG. 3A), for example, 20 to 30 μm for a 4-inch substrate, and this variation increases as the substrate has a larger diameter. This is because in a wire saw, as the amount of cut into the ingot increases, abrasive grains stagnating at the machining location increase, and the cutting width gradually increases.

  When the grooves 2 are collectively provided with a dicing saw for the substrate 24 having a non-uniform thickness, the groove depth is also non-uniform as shown in the schematic cross-sectional view of FIG. In the OECO method, the extent to which the electrode is formed on one side of the inner surface of the groove is determined by the groove depth and the deposition direction (evap in FIG. 4). The deposition angle θ is such that the depth of the groove 2 appearing in the cross section is the maximum in an arbitrary cross section (that is, a cross section appearing on the paper surface) perpendicular to the longitudinal direction of the groove 2 (the direction perpendicular to the paper surface). For the groove 2u to be formed, a straight line connecting the lower end LE of the formed electrode 6 in the thickness direction T of the substrate 24 and the upper end TE of the inner surface (electrode non-formed inner side surface) 2c where the electrode of the groove 2u is not formed. , Defined as an angle formed by a reference line HL perpendicular to the substrate thickness direction T.

  When the height of the ridge 23 formed between the grooves 2 and 2 is insufficient to form an electrode in a desired region, not only the side 2a on the ridge 23 side but also the groove depending on the deposition angle θ. It is also deposited on the side 2b corresponding to the bottom. The area to be removed by etching increases, and the required electrode cannot be obtained.

In the present invention, even at the thinnest position of the substrate 24 (position P1 in FIG. 3A), as shown in FIG. 4, it is sufficient that the electrode 6 is not deposited on the side 2b corresponding to the groove bottom. The groove depth is secured. As shown in FIG. 5A, in an arbitrary cross section perpendicular to the longitudinal direction of the groove, assuming that the groove depth is h and the electrode height is he, the groove width W ₁ defined as the distance between the opening edges of the grooves, The height h in the groove depth direction of the electrode (the thickness direction of the substrate) from the previously defined vapor deposition angle θ is:
he = W ₁ tan θ
It is represented by Therefore, the groove depth h ′ at the thinnest position is
h ′ ≧ W ₁ tan θ (the above formula (1))
As long as it is sufficiently large so as to satisfy the above, the electrode 6 is not deposited on the side 2b corresponding to the groove bottom even at this position. Since the groove depth is the smallest at the above position, the groove depth h at other positions always satisfies the above (1), and as a result, any groove 2 has a side 2b corresponding to the bottom. A problem that the electrode 6 is deposited can be prevented.

  In the case of a V-shaped groove as shown in FIG. 5B, equation (1) is approximately established. If the groove of interest is the deepest groove formed in the substrate, an extra electrode is formed if the groove depth of the shallowest groove in the substrate is greater than or equal to h. There will be no shortage.

  Returning to FIG. 1, the back electrode layer 8 is formed on the other main surface 24 b (hereinafter referred to as a second main surface) side of the solar cell 1 with a silicon nitride film 10 interposed therebetween, for example. The silicon nitride film 10 is formed as a protective film, and can be formed by, for example, a vapor deposition method (CVD: Chemical Vapor Deposition) method or the like. Further, the back electrode layer 8 covers substantially the entire surface of the second main surface 24b, and is configured as an aluminum vapor deposition layer, for example. The back electrode layer 8 is electrically connected to the underlying silicon single crystal substrate 24 through a contact through portion 10a penetrating the silicon nitride film 10 in the film thickness direction. The contact penetrating portion 10a may be formed by photolithography, but in the present embodiment, it is a groove by machining or a hole by laser processing.

  Moreover, in order to maintain the mechanical strength of the solar cell itself, the thickness of the silicon single crystal substrate 24 is set to about 150 to 300 μm. Of course, since it is necessary to reduce the thickness of the substrate in order to reduce the manufacturing cost of the solar cell, the mechanical strength of the solar cell itself may not be maintained simply by forming the groove deeply. That is, while maintaining the groove depth defined in the present invention, the depth is preferably suppressed to, for example, 100 μm or less.

  Further, in FIG. 1, each groove 2 of solar cell 1 is formed in a direction that does not coincide with the <110> direction on first main surface 24 a. Thereby, the mechanical strength of the solar cell 1 is improved. Note that in this specification, even if the crystal principal axis of the single crystal substrate used is tilted from <100> to about 6 ° due to off-angle provision, the substrate is regarded as having a {100} plane orientation.

Hereinafter, an example of a method for manufacturing the solar cell 1 will be described.
First, a silicon single crystal ingot in which a group III element such as boron or gallium is added to high-purity silicon is prepared, and a p-type silicon single crystal substrate having a plane orientation {100} is cut out therefrom. The specific resistance of the p-type silicon single crystal substrate is, for example, 0.5 to 5 Ω · cm. As shown in step (a) of FIG. 6, on the first main surface 24a of the p-type {100} substrate, a high-speed rotary blade DS (FIG. 3B) causes a direction different from <110>, for example, <100> direction. In addition, a plurality of parallel grooves 2 having a depth of 20 to 100 μm are formed. The silicon single crystal substrate may be produced by any of the CZ method and the FZ method, but is preferably produced by the CZ method from the viewpoint of the mechanical strength of the obtained substrate. Further, although the substrate thickness can maintain sufficient mechanical strength even at 40 μm, it is desirable to set it to 150 to 300 μm in consideration of the groove formation depth for the convenience of slicing.

  As the high-speed rotary blade, for example, a blade having a rectangular cross-section, a blade having a semicircular cross-section, and a blade having a chevron-shaped cross-section are appropriately selected and used according to the form of the groove to be formed. Using a high-speed rotary blade, the main surface of the substrate 24 is cut at a speed of, for example, about 0.1 to 4 cm per second while spraying the cutting fluid, and the grooves 2 are formed. The high-speed rotary blade can be replaced with a dicer or a wire saw.

As already described with reference to FIG. 3A, the longitudinal direction of each groove 2 is the first connecting the thickest position P2 (thickness h2) and the thinnest position P1 (thickness h1) of the semiconductor single crystal substrate. In order to reduce resistance loss of an electrode (finger electrode) formed on the inner surface of the groove, it is desirable that the angle formed between the straight line L and the straight line along the main surface is within 45 °. Further, when the direction of the straight line L and the groove 2 is set in parallel, the groove satisfying the expression (1) can be formed as follows. That is, the positions P1 and P2 are obtained in advance by measuring the thickness distribution of the substrate. Next, in FIG. 3B, the substrate 24 is set on the table DT so that the cutting direction of the high-speed rotary blade DS coincides with the direction connecting P1 and P2. Then, the value of W ₁ tan θ is calculated using the groove width W ₁ and the deposition angle θ determined as design values, and the groove depth h ′ becomes larger than the W ₁ tan θ at the position P1. In addition, the cutting depth of the high-speed rotary blade DS is set so that the substrate thickness T ′ (for example, 40 μm or more) that can sufficiently ensure the strength remains at each groove bottom position. At this time, the substrate thickness T ′ remaining on the bottom surface of each groove 2 can be represented by h1-h ′.

  Next, the substrate damage after the groove formation is removed by the chemical etching described above. In the case of the rectangle shown in FIG. 2A or the V-shaped groove shown in FIG. 2C, it is desirable to set the etching for removing the damage so as to enable the rounding as shown in FIG. 2D or 2E. When the damage removal etching is completed, a texture structure is formed on the substrate by a known technique such as anisotropic etching as a surface roughening treatment of the main surface for reducing reflection loss. After forming the texture, washing is carried out in an acidic aqueous solution of hydrochloric acid, sulfuric acid, nitric acid, hydrofluoric acid or the like, or a mixture thereof, but washing in hydrochloric acid is preferred from the viewpoint of economy and efficiency.

  Next, as shown in step (b) of FIG. 6, the emitter layer 4 is formed on the cleaned substrate surface. As the method for forming the emitter layer, any method such as a coating diffusion method using diphosphorus pentoxide or an ion implantation method in which phosphorus ions are directly implanted can be used. However, from an economical viewpoint, phosphorus oxychloride is used. It is preferable to employ a gas phase diffusion method. For example, the n-type emitter layer 4 can be formed on the surface by heat-treating the substrate at around 850 ° C. in a phosphorus oxychloride atmosphere. The thickness of the emitter layer 4 to be formed is about 0.5 μm, for example, and the sheet resistance can be 40 to 100Ω / □. Note that the phosphorus glass formed on the substrate surface by this treatment is removed in a hydrofluoric acid solution.

  Next, an electrode is formed on the second main surface 24b side of the substrate. First, as shown in step (c) of FIG. 6, a silicon nitride layer 8 as a passivation film is formed on the second main surface 24b. The silicon nitride layer 8 can be formed by a CVD (Chemical Vapor Deposition) method. In this case, any method such as atmospheric pressure CVD method, reduced pressure CVD method and photo CVD method is possible, but when adopting the remote plasma CVD method, it is a low temperature process of about 350 to 400 ° C., In addition, it can be said that it is suitable for the present invention in that the surface recombination rate of the obtained silicon nitride layer 8 can be reduced. The direct thermal nitriding method is not preferable because a sufficient film thickness cannot be obtained.

  Then, as shown in step (d) of FIG. 6, the formed silicon nitride layer 8 is used for conducting an electrode that reaches the underlying p-type silicon single crystal substrate 24 by using a high-speed rotary blade similar to that used for groove formation. A groove 8a is formed. The shape of the blade is, for example, one of a rectangle, a semicircle, and a mountain according to the groove cross-sectional shape. After the formation of the groove 8a, the groove 8a is covered with the electrode 9 together with the surrounding silicon nitride layer 8 as shown in step (e) of FIG. Silver or copper can be used as the electrode material, but aluminum (including an alloy) is most preferable from the viewpoints of economy and workability. The aluminum can be deposited by either sputtering or vacuum evaporation. The electrode forming process on the second main surface 24b side is thus completed.

  Next, as shown in step (f) of FIG. 6, a silicon oxide film 5 is formed on the first main surface 24a by a thermal oxidation method. This layer functions as a tunnel insulating layer between the electrode 6 on the first main surface 24a and the substrate 24. In order to optimize the tunnel effect while preventing a short circuit, the layer thickness is 5 to 30 mm. To do. The silicon oxide film 5 can be formed by various known methods such as dry oxidation, wet oxidation, steam oxidation, pyrogenic oxidation and hydrochloric acid oxidation. However, a high quality and easy dry film thickness control method can be used. It is preferable to adopt.

On the substrate 24 on which the silicon oxide film 5 is formed, the electrode 6 is deposited on the inner surface on one side in the width direction of the groove 2 by, for example, an oblique vapor deposition method. The electrode material is preferably aluminum (including an alloy), but is not limited thereto, and other metals such as silver and copper are also possible. Specifically, on the basis of the state in which the first main surface 24a faces the vapor deposition source so that the extending direction of the groove 2 is perpendicular to the vapor deposition source, the main axis of the substrate 24 is 70 ° with respect to the vapor deposition source. The substrate 24 is arranged in the vapor deposition apparatus in a form inclined by ˜89 °. With such an arrangement, the electrode material can be preferentially deposited on the inner surface on one side in the width direction of the groove 2. In addition, it is desirable to perform the vapor deposition after the degree of vacuum in the apparatus reaches a level of 2 × 10 ⁻⁵ Pa or less, and the vapor deposition rate is, for example, 10 to 15 毎 per second (however, it is not limited thereto). ). As shown in step (g) of FIG. 6, the substrate 24 on which the electrode 6 is deposited is dipped in an acidic aqueous solution of hydrochloric acid, sulfuric acid, nitric acid, hydrofluoric acid, phosphoric acid or a mixture thereof, thereby forming grooves. Unnecessary electrode material deposited on the top of the ridge 23 generated between the two is removed.

  The substrate 24 having been subjected to the above steps is formed with a bus bar electrode 30 as shown in FIGS. 14 to 16 by a known method, and further, as a surface passivation and antireflection film, for example, by a remote plasma CVD method. The final solar cell 1 is obtained by depositing the silicon nitride layer 7 uniformly on the main surface 24a, for example, 600 to 700 mm (step (h) in FIG. 6).

  FIG. 14 shows an example in which the bus bar electrode 30 is formed in an arc shape along the outer peripheral edge of the semiconductor single crystal substrate 24. FIG. 16 shows an example in which the semiconductor single crystal substrate 24 is formed in a polygonal line along the outer peripheral edge.

  Further, in FIG. 15, the bus bar electrode 30 has the thickest finger electrode 2 in each groove 2 with respect to a part 40 of the array including the maximum length groove 2 ′ in the array of grooves 2. It arrange | positions in the form divided | segmented into the 1st part 2a by the side of the position P2, and the 2nd part 2b longer than the 1st part 2a by the side of the position P1 used as the thinnest. And the position of the bus-bar electrode 30 is adjusted so that the length of the 1st part 2a becomes large as the groove | channel 1 with a long full length. Specifically, the bus bar electrode 30 is arranged in parallel with the straight line DL at the position of the groove array 40, and the length of the first portion 2a is longer as the groove 2 has a longer overall length in the groove array 40. . In the groove 2 having a long overall length, the length of the finger electrode 2 tends to increase and the resistance loss tends to increase. However, if the position of the bus bar electrode 30 is devised as described above and the length of the first portion 2a is increased, Since the length of the two portions 2b is reduced correspondingly, it is more effective in reducing resistance loss.

  14 to 16, the bus bar electrode 30 is illustrated as having a uniform width. However, a lead wire for output output is connected to the bus bar electrode 30, and current collection is performed at a portion closer to the lead wire. Since a large current flows forward, it is desirable that the width be increased toward the lead wire connection position. For example, when a lead wire is connected to the first end portion of the bus bar electrode 30, the width may be increased from the second end portion toward the first end portion. Further, when the lead wire is connected to the intermediate position in the longitudinal direction of the bus bar electrode 30, the width may be formed so as to approach the intermediate position to which the lead wire is connected from both ends of the bus bar electrode 30.

  The bus bar electrode 30 is formed by, for example, a method of forming an electrode pattern by electrolytic plating or electroless plating of a metal such as Ni (which may be a combination of both, such as electroplating after electroless plating). , A method of forming an electrode pattern by vapor deposition of a metal such as Al, a screen printing of a metal paste such as Ni, a method of forming a paste print pattern, and firing it into an electrode, a metal thin film for an electrode such as aluminum Examples thereof include a method of attaching a band to a semiconductor single crystal substrate by pressure bonding or the like (heat setting can be performed if necessary). In addition, when using a metal ribbon attachment, it becomes difficult to make a ribbon contact | adhere to the groove | channel inner surface in which the finger electrode is formed, and a problem may arise in ensuring conduction | electrical_connection. Therefore, when a metal paste layer is formed on one surface of the metal ribbon and pasted on the metal paste layer side, the metal paste is filled into the groove inner surface, and it is easy to ensure conduction with the finger electrode. . In this case, it is desirable to perform firing after attaching the metal ribbon from the viewpoint of securing the adhesion of the electrode.

(Embodiment 2)
Hereinafter, an embodiment of a method for manufacturing a solar cell according to the second aspect of the present invention will be described. The target solar cell is also the OECO cell shown in FIG. 1 in the second embodiment, similar to the first embodiment. In the first aspect, an upper blade type high-speed rotary blade is used, and the distribution of the depth of the groove 2 according to the unevenness of the substrate thickness is admitted, and the groove depth h at the minimum depth position is accepted. By adding a restriction like the formula (1) to ', shadowing loss and electrode resistance loss were reduced. On the other hand, the second aspect aims to make the groove depth as uniform as possible in one substrate.

  In this embodiment, a p-type silicon single crystal ingot is prepared by adding a group III element such as boron or gallium to high-purity silicon by the CZ method or the FZ method, and the p-type silicon single crystal substrate 24 is cut out therefrom. The preferable specific resistance and substrate thickness range of the p-type silicon single crystal substrate 24 are the same as those in the embodiment. In addition, in the manufacturing method of this embodiment, a polycrystalline silicon can also be used and what was made by the crystal manufacturing techniques known until now, such as a HEM method, a casting method, and an EFG method, can be used.

  Next, as shown in FIG. 9, a plurality of grooves 2 parallel to each other are formed on the first main surface 24 a of the substrate 24 by the grooving blade 13. Specifically, the processing table 12 is fixedly arranged so that the substrate feed surface 12a for placing the semiconductor single crystal substrate 24 faces upward, and the grooving blades 13 are coaxially coupled at a constant interval. A grooving blade assembly 15 in which a plurality of grooving blades 13 rotate integrally is projected from an opening portion 14 formed in the substrate feed surface 12a of the processing table 12 so that each of the grooving blades 13 protrudes. Protruding so that the lengths are equal to each other. Then, the grooving blade assembly 15 is integrally rotated about the axis 16, and the first main surface 24 a on which the groove 2 is to be formed is brought into close contact with the substrate feed surface 12 a of the processing table 12. In this state, the semiconductor single crystal substrate 24 is moved perpendicularly to the thickness direction of the grooving blade 13 and toward the grooving blade assembly 15 (as indicated by the feed direction 9 in FIG. 9). . At this time, the second main surface 24b of the semiconductor single crystal substrate 24 is moved while being held by the chuck. 9 (the same applies to FIG. 13 described above and FIGS. 10 and 11 described later), the thickness of the semiconductor single crystal substrate 24 is exaggerated, and the actual thickness of the semiconductor single crystal substrate 24 is shown. Is approximately 1/500 of the diameter of the semiconductor single crystal substrate 24.

  Accordingly, even in the case of the semiconductor single crystal substrate 24 having a non-uniform thickness as shown in FIG. 10, the protruding height h of the grooving blade assembly 15 from the substrate feed surface 12a, the thickness of the grooving blade It is possible to form a plurality of grooves 2 having a groove depth H, a groove width W1, and a groove interval W2, each having the same groove depth H, corresponding to the length w1 and the interval w2 between the grooving blades 13, respectively. . The configuration and arrangement of these grooving blade assemblies 15 can be changed as appropriate according to the desired groove 2 to be formed in the semiconductor single crystal substrate 24. For example, the number of grooving blades 13 coupled to the grooving blade assembly 15 is 100 to 200, the protruding height h of the grooving blade 13 is several tens to 100 μm, and the thickness w1 of the blade is several tens to ten. The width w2 between the blades and the blades is set to about 100 μm and several hundreds μm. In addition, as a kind of blade of the grooving blade 13 used, for example, a diamond blade (for example, diamond particles having a particle diameter of 5 μm to 10 μm uniformly adhered to the blade surface) can be employed. . In addition, by making the cross-sectional shape in the cross section including the rotation axis of the grooving blade 13 into a rectangular shape, a semicircular shape, a V shape or a U shape, on the first main surface 24a of the semiconductor single crystal substrate 24, It is also possible to form a groove whose cross-sectional shape in a cross section orthogonal to its longitudinal direction corresponds to the cross-sectional shape of the grooving blade. Using such a grooving blade 13 (and a grooving blade assembly 15 formed by combining them), for example, a speed of about 0.1 to 4 cm per second while ejecting a cutting fluid to the processing portion. Then, the first main surface 24a of the substrate 24 is cut.

  Further, the groove formation on the first main surface of the solar cell can be performed using a processing table 12 'as shown in FIG. On the substrate feed surface 12'a of the processing table 12 ', a shaving powder discharge groove 17 is formed. The cutting powder discharge groove 17 is, for example, a strip-shaped one having a width of 0.5 to 5 mm, a length of 10 to 30 cm, and a depth of 1 to 10 mm, and the semiconductor single crystal substrate 24 is expected to contact at an interval of 1 to 10 mm. It is uniformly formed on the substrate feed surface 12'a. By forming the shaving powder discharge groove 17 as described above, the contact area between the semiconductor single crystal substrate 24 and the substrate feed surface 12'a is reduced, and the frictional resistance during the movement of the substrate 24 is reduced. Will be improved. Further, as the groove formation progresses, the cutting powder generated during the grinding of the groove is taken into the cutting powder discharge groove 17, so that the gap between the first main surface 24a of the semiconductor single crystal substrate 24 and the substrate feed surface 12'a is increased. The shavings are less likely to intervene. As a result, the adhesion between the first main surface 24a of the substrate 24 and the substrate feed surface 12'a can be maintained satisfactorily. As a result, the height of the groove 2 to be formed is kept constant on the first main surface. Can keep.

  It is more desirable that the direction in which the shaving powder discharge groove 17 formed on the substrate feed surface 12'a is not coincident with the substrate feed direction. When such a cutting powder discharge groove 17 is formed, the cutting powder generated by the groove formation is interposed on the substrate feed surface 12 ′ a of the processing table 12 ′, and further the semiconductor single crystal substrate 24 in the feed direction 9. Along with the movement, it moves in the direction of the feeding direction 9. Therefore, the cutting powder is finally taken into the cutting powder discharge groove by making the formation direction of the cutting powder discharge groove 17 not coincide with the feeding direction 9 of the substrate 24.

  Steps after forming the groove 2 are the same as steps (b) to (h) shown in FIG. 6 of the first embodiment, and a solar cell 1 having the same configuration as that shown in FIG. 1 is obtained. It is done. According to the method of the present invention, the groove 2 having a certain height is formed on one main surface 24 a of the substrate 24. Thereby, in the OECO solar cell, the shielding effect with respect to the adjacent grooves 2 becomes uniform in each groove 2, so that the equivalent electrode 6 is deposited in each groove 2. Each electrode 6 can be formed as set, and resistance loss or shadowing loss can be suppressed.

  As described above, in the groove forming step according to the second embodiment, the shape of the grooving blade as described above, the shape and structure of the processing table, and the arrangement thereof are adopted, but the present invention is limited to this. Instead, these may be changed as appropriate based on the technical knowledge of those skilled in the art within the scope of the present invention.

  Needless to say, the present invention is not limited to this embodiment, and can be implemented in various modes without departing from the spirit of the present invention. For example, in both the first aspect and the second aspect, the solar cell produced by the OECO method has been described as an example, but the present invention is not limited to this. And even if it is other than the OECO solar cell, it can be applied to any solar cell as long as it is in the form of a solar cell that requires groove formation on at least one of the main surfaces, and variations in the characteristics of the solar cell Is effective in suppressing the output decrease.

  Hereinafter, the present invention will be described more specifically with reference to examples.

Example 1
Prepare silicon wafer 1 for p-type single crystal solar cells (group 4 element diameter, substrate minimum thickness 270 μm, substrate maximum thickness 300 μm, resistivity 0.5 Ωcm) using Group III element gallium as an impurity element. The thickness distribution was examined using a dial gauge. Based on the result, as shown in FIG. 3, a square (rectangular) cross-sectional groove 2 is formed using a dicing saw (high-speed rotating blade) so as to be parallel to a straight line connecting the thickest point and the thinnest point of the wafer. Were formed in parallel. The shape of the grinding wheel blade was 450 μm in thickness, 50 mm in diameter, and 50 μm between the grinding wheel blades. Here, since the thickness of the grindstone blade was 450 μm and the vapor deposition angle described later was 5 °, the minimum depth h of the groove was 50 μm. The processing conditions were a grinding wheel blade rotation speed of 50 rpm and a wafer feed speed of 1 mm / s, and cooling water was supplied to the processing site. The wafer was chucked on the surface opposite to the processed surface by a vacuum chuck. In addition, for comparison, a wafer having a uniform thickness was fabricated by grooving under the same conditions (this is referred to as comparative product 1).

Next, the grooved wafer was etched with an aqueous potassium hydroxide solution to remove the damaged layer, and a silicon nitride film 8 was formed on the back surface using a plasma CVD apparatus. Subsequently, an n ⁺ region 4 in which phosphorus of a group V element was used as an impurity by thermal diffusion was prepared on the light receiving surface side so that the sheet resistance was 100Ω / □. Further, a groove 8a for electrode conduction was formed on the back surface by a rotary blade, and an electrode 9 having a thickness of 2 μm was formed by vacuum deposition of aluminum. Next, a tunnel oxide film 5 having a thickness of 2 nm is formed on the light-receiving surface by thermal oxidation, and subsequently aluminum that serves as an electrode from a direction perpendicular to the parallel grooves of the light-receiving surface and at an angle of 5 ° to the wafer surface. The electrode 6 having a thickness of 7 μm was formed only on one side of the groove. Here, the metal deposited on the upper surface of the ridge formed between the grooves was removed by etching with an aqueous phosphoric acid solution. A thick bus bar electrode is formed so as to be connected to each electrode, and a silicon nitride film 7 having a thickness of 70 nm is formed on the light receiving surface by plasma CVD to prevent sunlight reflection and protect the surface. A battery (invention product 1) and a solar cell for comparison (comparative product 1) were obtained.

Next, the output characteristic of the produced solar cell was measured using a solar simulator (light intensity: 1 kW / m ² , spectrum: AM1.5 global). The obtained output characteristics are shown in Table 1.

  In this measurement result, there is no significant difference in the output obtained between the product 1 of the present invention and the comparative product 1. That is, even in the solar cell of the present invention using a wafer with a non-uniform thickness, a high output equivalent to that when using a wafer with a uniform thickness can be obtained.

  Next, in the cross section perpendicular to the longitudinal direction of the groove, the electrode height (length in the thickness direction of the substrate) of the groove side surface of the product 1 of the present invention was measured using an SEM. The measurement results are shown in FIG. As shown in FIG. 7, the measurement points are the center point of the wafer and the electrodes in the vicinity of a total of nine points including points a to i rotated by 45 ° from the groove forming direction on the circle having a radius of 3.5 cm from the center. did. An arrow SL shown in FIG. 7 indicates the cutting direction of the silicon crystal rod, and the substrate is gradually thinned from position 12 to position 13.

  In this measurement result, the standard deviation σ of the electrode height was 1.7, which was about 10% at maximum when converted to the difference in electrode cross-sectional area. Usually, the resistance loss due to the electrode of the solar cell is about 5% of the output, so that the influence of the current variation on the characteristics is estimated to be about 0.5% of the output. Therefore, this variation in electrode height is not a problem because it does not greatly affect the energy conversion efficiency of the solar cell.

(Example 2)
On the light-receiving surface side of a silicon wafer 1 for p-type single crystal solar cells (group 4 element diameter boron, substrate minimum thickness 270 μm, substrate maximum thickness 300 μm, resistivity 2.0 Ωcm) using boron, a group III element as an impurity element, A plurality of square (rectangular) cross-sectional grooves were formed in parallel by a dicing saw. The groove was formed to be parallel to a straight line connecting the thickest point and the thinnest point of the wafer. The shape of the grindstone blade was set to 450 μm in thickness, 50 mm in diameter, and 50 μm between grindstone blades as in Example 1. Here, since the thickness of the grindstone blade was 450 μm and the vapor deposition angle was 5 °, the minimum depth h of the groove was 50 μm. The processing conditions were a grinding wheel blade rotation speed of 60 rpm and a wafer feed speed of 1.5 mm / s, and cooling water was supplied to the processing site. The wafer was chucked on the surface opposite to the processed surface by a vacuum chuck. For comparison, a wafer is also produced in which the longitudinal direction of the groove is perpendicular to the straight line connecting the thickest point and the thinnest point of the wafer (perpendicular to the groove direction shown in FIG. 3). (This is referred to as comparative product 2). Furthermore, the same solar cell production process as Example 1 was performed about each, and the solar cell of this invention (this invention product 2) and the solar cell for a comparison (comparative product 2) were obtained.

Next, the output characteristic of the produced solar cell was measured using a solar simulator (light intensity: 1 kW / m ² , spectrum: AM1.5 global). Table 2 shows the obtained output characteristics.

  The product 2 of the present invention showed a fill factor that was about 1.9% larger than that of the comparative product 2, and showed a high output. That is, higher output can be obtained by applying the present invention.

  Furthermore, the current-voltage characteristics of the solar cell were measured by a four-terminal method in a completely dark state, and the series resistance was estimated. Table 3 shows the measurement results.

In this measurement result, a difference of 0.55 Ω · cm ² occurred in the series resistance between the product 2 of the present invention and the comparative product 2. It is considered that the difference in fill factor is caused by the difference in series resistance. In the OECO solar cell, by forming the groove so as to be parallel to the straight line connecting the thickest point and the thinnest point of the wafer, the resistance loss in the finger electrode is reduced and high output can be obtained.

(Example 3)
On the light-receiving surface side of a p-type silicon single crystal substrate (10 cm square, substrate thickness 300 μm, resistivity 0.5 Ωcm, maximum thickness variation 25 μm) for solar cells using a group III element gallium as a dopant element, the present invention In accordance with this manufacturing method, a plurality of square (rectangular) cross-sectional grooves are formed in parallel. The shape of the grooving blade was 450 μm in thickness and 50 mm in diameter, and these grooving blades were bonded together at an interval of 50 μm to form a grooving blade assembly. The protrusion height h of the grooving blade assembly from the processing table was 50 μm. Further, the processing conditions were a grooving blade assembly rotational speed of 50 rpm and a substrate feed speed of 1 mm / s. The substrate was chucked on the surface opposite to the first main surface (second main surface) where the groove was to be formed by a vacuum chuck, and the first main surface was brought into close contact with the substrate feed surface of the processing table. Then, as described with reference to FIG. 9, the semiconductor single crystal substrate is moved in a direction perpendicular to the thickness direction of the grooving blade while supplying cooling water to the processing site, and a plurality of semiconductor single crystal substrates are formed on the semiconductor single crystal substrate. Grooves were formed. The semiconductor single crystal substrate obtained by this step is referred to as product 3 of the present invention.

  In addition, for comparison, a conventional method in which a substrate is fixed to a processing table, a grooving blade is cut into the upper surface of the substrate, and the grooving blade is moved in parallel with the substrate feed surface of the processing table in that state. A groove was formed in the semiconductor single crystal substrate by the method. The semiconductor single crystal substrate obtained in this step is referred to as Comparative Example Product 3.

  Next, each grooved substrate was etched with an aqueous potassium hydroxide solution to remove the damaged layer, and a silicon nitride film was formed on the back surface of the substrate using a plasma CVD apparatus. Subsequently, an n-type emitter layer having phosphorus as a group V element as an impurity was produced on the light-receiving surface side by thermal diffusion so that the sheet resistance was 100Ω / □. Further, an electrode having a thickness of 2 μm was formed on the back surface by vacuum deposition of aluminum. Next, a tunnel oxide film having a thickness of 2 nm is formed on the light-receiving surface by thermal oxidation, and subsequently aluminum is vacuumed from a direction perpendicular to the parallel grooves of the light-receiving surface and at an angle of 5 degrees with respect to the first main surface. Evaporation was performed to form an electrode having a thickness of 5 μm only on one side of the groove. Here, the metal deposited on the upper surface of the convex portion between the grooves was removed by etching with an aqueous phosphoric acid solution. Finally, a silicon nitride film having a film thickness of 70 nm was formed on the light receiving surface by plasma CVD to prevent sunlight reflection and protect the surface, thereby producing a solar cell.

Using a solar simulator (light intensity: 1 kW / m ² , spectrum: AM1.5 global), the current-voltage characteristics of the produced solar cell were measured. Table 4 shows various characteristics of the solar cell obtained from the current-voltage characteristics.

  In this measurement result, the product 3 of the present invention showed a fill factor (curve factor of the current-voltage curve) which was about 5.7% larger than that of the comparative product 3, and showed a high maximum output. Next, the electrode shape on the surface of the solar cell was observed using SEM. As a result, it was found that the comparative example product 3 has a shortage of the electrode height in the groove on the entire light receiving surface of about 15 μm at the maximum as compared with the product 3 of the present invention. The output difference in Table 1 is mainly due to the difference in fill factor. This difference in fill factor is thought to be caused by the increase in series resistance due to the shortage of the electrode cross-sectional area found from SEM observation.

  As described above, by applying the present invention, the height of the electrode formed on the side surface of the groove can be made substantially constant for all the finger electrodes, so that variation in characteristics over the entire light receiving surface is reduced. And it becomes possible to form the electrode which can suppress the increase in series resistance, and it becomes possible to obtain a high output solar cell.

Example 4
Groove machining according to the present invention was performed using a machining table in which shaving powder discharge grooves parallel to the substrate feed direction were formed on the surface with a width of 2 mm, a depth of 5 mm, and a groove interval of 2 mm as shown in FIG. The substrate used is a p-type silicon single crystal substrate for solar cells (group 10 cm square, substrate thickness 250 μm, resistivity 1.0 Ωcm, maximum thickness variation 20 μm) using Group III element boron as an impurity element. As the grooving blade, square grooving blades having a thickness of 450 μm and a diameter of 50 mm, which were arranged at intervals of 50 μm, were combined, and this was used as a grooving blade assembly. The protrusion height h from the substrate feed surface of the processing table was 50 μm. The processing conditions were a grooving blade rotation speed of 60 rpm and a substrate feed speed of 1.25 mm / s, and the substrate was moved in the same manner as in Example 3 to form grooves on the first main surface.

  Moreover, the test which groove-grows on the same conditions using the process table which has not formed the shaving powder discharge groove | channel was also done. As a result of processing 50 samples of each sample, when the chip discharging groove was not formed on the processing table, there were no problems such as cracks, chipping or deviation from the set value of the groove depth due to the groove processing. When the product was a non-defective product, the non-defective product rate was 83%. On the other hand, when the shaving powder discharge groove was formed, the yield rate was 98%, and the yield rate was improved by 15%. Thereby, it turns out that productivity improved more.

The expanded sectional view which shows one embodiment of the solar cell of this invention partially. The schematic diagram which shows the 1st example of groove shape. The schematic diagram which shows the 2nd example of groove shape. The schematic diagram which shows the 3rd example of groove shape. The schematic diagram which shows the 1st example which forms R at the corner | angular part in a groove | channel. The schematic diagram which shows the 2nd example which forms a round at the corner | angular part in a groove | channel. The schematic diagram which shows the single crystal substrate with nonuniform thickness. FIG. 3B is a schematic diagram illustrating a groove forming method for the substrate of FIG. 3A. The cross-sectional schematic diagram of the solar cell of this invention with sufficient groove depth. Explanatory drawing which illustrates typically the prescription | regulation method of the groove depth of a rectangular cross-section groove | channel. Explanatory drawing which illustrates typically the prescription | regulation method of the groove depth of a V-shaped cross-sectional groove | channel. Manufacturing process explanatory drawing of a solar cell. The figure which shows the measurement location of the electrode formation length to the board | substrate thickness direction in a solar cell employ | adopted in Example 1. FIG. 3 is a graph showing measurement results of electrode height in Example 1. The schematic diagram explaining an example of the groove | channel formation method in the manufacturing method of the solar cell of this invention. Sectional drawing which shows an example of the groove | channel formation method in the manufacturing method of the solar cell of this invention. The schematic diagram explaining the modification of the groove | channel formation method in the manufacturing method of the solar cell of this invention. The figure which shows the electrode vapor deposition shape in the manufacturing method of an OECO solar cell. The figure which shows the electrode vapor deposition shape in case there exists dispersion | variation in the thickness of a board | substrate. The top view which shows 1st embodiment of the formation form of a bus-bar electrode. The top view which shows 2nd embodiment of the formation form of a bus-bar electrode. The top view which shows 3rd embodiment of the formation form of a bus-bar electrode. The figure explaining the definition for discriminating which of the positions P2 and P1 a bus-bar electrode is near. Sectional drawing which shows the 1st example of the longitudinal direction cross-sectional shape of a bus-bar electrode. Sectional drawing which shows the 2nd example of the longitudinal direction cross-sectional shape of a bus-bar electrode.

Claims (4)

Forming a plurality of grooves on at least one main surface of the semiconductor substrate;
From the substrate feed surface of the processing table having a flat substrate feed surface, the blade portion of the grooving blade protrudes by a certain height, and the grooving blade is rotated.
One main surface of the semiconductor substrate is brought into close contact with the substrate feeding surface, and in this state, the substrate is moved along the substrate feeding surface and perpendicular to the thickness direction of the grooving blade toward the grooving blade. The method of manufacturing a solar cell, wherein the groove is formed on the one main surface by relative movement. Projecting the blade portions of the plurality of grooving blades, which are integrally rotated coaxially at regular intervals, respectively, by the same height,
The method for manufacturing a solar cell according to claim 1, wherein a plurality of parallel grooves arranged at regular intervals are collectively formed on the one main surface of the semiconductor substrate by the grooving blades.   The method for manufacturing a solar cell according to claim 1 or 2, wherein the processing table uses a chip discharge groove formed on the substrate feed surface. The processing table is fixedly arranged so that the substrate feeding surface faces upward, and the other main surface side of the substrate is chucked and held,
4. The groove according to claim 1, wherein the groove is formed by moving the substrate held by the chuck toward the grooving blade fixedly arranged. 5. A method for manufacturing a solar cell.

Images