Sheet for cleaning, method for manufacturing semiconductor device, and method for manufacturing sheet for cleaning

Abstract

To provide a sheet for cleaning which is low in cost, and is excellent in rigidity, dust emission property, water repellency or the like while being composed of recycled paper and virgin pulp paper, a method for manufacturing a semiconductor device, and a method for manufacturing a sheet for cleaning.SOLUTION: A sheet for cleaning, which is interposed between molding surfaces for resin molding and cleans the molding surfaces for resin molding by a cleaning resin, is composed of a recycled paper material containing cellulose as a main component, wherein a mounting hole for mounting the mold for resin molding is provided on at least one end of outer periphery, 11 to 19 (prepreg ratio%) of a phenol resin is impregnated into a rear face layer from a surface layer of the recycled paper material, and the phenol resin is cured by heating at 200 to 230°C for a predetermined time.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a cleaning sheet, a method for manufacturing a semiconductor device, and a method for manufacturing a cleaning sheet. [Background technology]

[0002] In the resin molding process of a resin molded semiconductor device, molding is repeated many times. As a result, contaminants such as resin burrs, oxide films, oil, and dust accumulate inside the mold dies where the molding resin is filled, i.e., in the cavities, runners, air vents, and around the cull block of the upper and lower molds that form a pair of mold dies.

[0003] Such stains have a negative effect on molding quality, and stains on the mold in particular are likely to result in poor demoldability. If this happens, when the semiconductor product is released from the pair of upper and lower molds, considerable stress will be placed on the semiconductor chip inside the molded product, which may result in cracks in the chip, stains due to surface contamination of the molded product, a decrease in mark strength, etc. Therefore, it is necessary to clean the mold die after every certain number of shots.

[0004] In response to such demands, for example, a method has been proposed in which a lead frame not mounted with a semiconductor chip (hereinafter referred to as a dummy lead frame) is clamped between the main surfaces (mating surfaces) of a molding die, and then a cleaning resin made of a melamine resin or the like is injected into the molding die and hardened onto the dummy lead frame, and dirt is then adhered to the surface of the cleaning resin, which is then removed together with the cleaning resin to perform cleaning (see, for example, Patent Document 1).

[0005] Also, for example, a method for cleaning a molding die has been proposed, which comprises the steps of clamping a sheet-like member made of cotton cloth (nonwoven fabric) that can be impregnated with and permeated with a cleaning resin between the open mold halves, and filling the cavity of the closed molding die with molten cleaning resin (see, for example, Patent Document 2).

[0006] Furthermore, a method for cleaning a mold has been proposed, which comprises the steps of placing a mold cleaning sheet made of paper, which allows resin and cleaning resin to not penetrate or penetrate from the front to the back of the paper, on the mold and clamping it between the first and second molds, and supplying cleaning resin from a pot to the mold cleaning sheet and filling the cleaning resin into the cavity block through an opening in the mold cleaning sheet (see, for example, Patent Document 3). [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Japanese Patent Application Publication No. 1-95010 [Patent Document 2] Japanese Patent Application Publication No. 6-254866 [Patent Document 3] JP 2007-208158 A Summary of the Invention [Problem to be solved by the invention]

[0008] However, the technique described in Patent Document 1 has a problem in that an expensive dummy frame is used as the cleaning sheet, resulting in high costs.

[0009] Furthermore, the techniques described in Patent Documents 2 and 3 disclose cleaning sheets using unbleached paper made from virgin pulp and having heat resistance, or cleaning sheets using nonwoven fabric. Although cleaning sheets using the nonwoven fabric described in Patent Document 2 were used for a time and have been attempted, the problems of dust generation and peeling of the paper surface and release surface could not be solved, and they have not been generally adopted.

[0010] Furthermore, Patent Document 3 was made by Mr. Tsuchida, who is also the inventor of the present application, and discloses the requirements for a cleaning sheet made of paper, such as that the mold cleaning sheet is unbleached paper made of virgin pulp and has heat resistance, that the cleaning sheet is impregnated with a release agent (such as a thermosetting resin) beforehand and then subjected to a cleaning shot, thereby improving the release properties and preventing the cleaning sheet and cleaning resin from adhering to the mold, and that when considering the release properties between the cleaning sheet and the mold, it is preferable that the material of the cleaning sheet has a high hardness, and for example, if the material of the cleaning sheet is paper, it is possible to prevent adhesion and generation of dust by applying heat and pressure to the cleaning sheet itself to make it high density, but does not disclose any specific content for satisfying all of these requirements, and there is no disclosure in other patent documents other than those in which the applicant previously obtained a patent decision.

[0011] Therefore, the present invention has been made in consideration of the above-mentioned problems, and aims to provide a cleaning sheet that is made of recycled paper, is low cost, and has excellent rigidity, dust generation, water repellency, etc., as well as a method for manufacturing a semiconductor device and a method for manufacturing a cleaning sheet. [Means for solving the problem]

[0012] Mode 1: One or more embodiments of the present invention propose a cleaning sheet which is interposed between surfaces of a resin molding die and cleans the surfaces of the resin molding die with a cleaning resin, the cleaning sheet being made of recycled paper material whose main component is cellulose, and at least one end of the outer periphery is provided with an attachment hole for attachment to the resin molding die, and the recycled paper material is impregnated with 11 to 19% (prepreg ratio) of phenolic resin from the front layer to the back layer and cured by heating at 200°C to 230°C for a predetermined time.

[0013] Mode 2: One or more embodiments of the present invention propose a cleaning sheet to be placed between a pair of first and second dies of a molding die having a cavity block which is a mounting area for a lead frame or a substrate frame.

[0014] Mode 3: One or more embodiments of the present invention propose a method for manufacturing a semiconductor device, comprising the step of cleaning the mating surfaces of a molding die consisting of a pair of a first die and a second die having a cavity block which is a mounting area for a lead frame, using the cleaning sheet according to mode 1, the method comprising the steps of: preparing at least the cleaning sheet when placed between the first die and the second die; automatically placing the cleaning sheet on the molding die and clamping it between the first die and the second die; supplying a cleaning resin from a pot and filling the cleaning resin into the cleaning sheet; and releasing the cleaning resin and the cleaning sheet from the molding die after hardening the cleaning resin.

[0015] Mode 4: One or more embodiments of the present invention propose a method for manufacturing a semiconductor device, wherein the semiconductor device includes at least a DIL-P, a QFP, a diode, and a transistor.

[0016] Mode 5: One or more embodiments of the present invention propose a method for producing a cleaning sheet, comprising the steps of: impregnating recycled paper, the main component of which is cellulose, with phenolic resin at 11 to 19% (prepreg ratio); and curing the phenolic resin by subjecting the paper material impregnated with the phenolic resin to a heat treatment including a treatment of heating the paper material at least at 200°C to 230°C for a predetermined period of time. Effect of the Invention

[0017] According to one or more embodiments of the present invention, it is possible to provide a cleaning sheet made of recycled paper, which is low cost and has excellent rigidity, dust generation properties, water repellency, etc., as well as a method for manufacturing a semiconductor device and a method for manufacturing a cleaning sheet. [Brief description of the drawings]

[0018] [Figure 1] FIG. 1 is a perspective view of a cleaning sheet according to a first embodiment of the present invention. [Diagram 2] 3A to 3C are diagrams illustrating an impregnation step of the cleaning sheet according to the first embodiment of the present invention. [Diagram 3] FIG. 4 is a diagram illustrating an example of an experimental result in the impregnation step of the cleaning sheet according to the first embodiment of the present invention. [Figure 4] 3A to 3C are diagrams illustrating a thermal curing process of the cleaning sheet according to the first embodiment of the present invention. [Diagram 5] 3A to 3C are diagrams illustrating a die-cutting process for the cleaning sheet according to the first embodiment of the present invention. [Figure 6] 3A to 3C are diagrams illustrating a thermal compression treatment process for the cleaning sheet according to the first embodiment of the present invention. [Figure 7] FIG. 1 is a diagram showing the configurations of Sample 1, Sample 2, and an inventive product in relation to a cleaning sheet according to a first embodiment of the present invention. [Figure 8] FIG. 4 is a diagram showing the results of a heat resistance test of Sample 1 of the cleaning sheet according to the first embodiment of the present invention. [Figure 9] FIG. 4 is a diagram showing the results of a heat resistance test of Sample 2 of the cleaning sheet according to the first embodiment of the present invention. [Figure 10] FIG. 2 is a diagram showing the results of a heat resistance test of the inventive cleaning sheet according to the first embodiment of the present invention. [Figure 11] FIG. 2 is a diagram showing the results of tensile strength tests of Sample 1, Sample 2, and the invention product, regarding the cleaning sheet according to the first embodiment of the present invention. [Figure 12] FIG. 2 is a diagram showing the results of a dust generation test for Sample 1, Sample 2, and the inventive product, with regard to the cleaning sheet according to the first embodiment of the present invention. [Figure 13] FIG. 2 is a diagram showing the results of a suction property test of sample 1, sample 2, and the invention product, with regard to the cleaning sheet according to the first embodiment of the present invention. [Figure 14] FIG. 2 is a diagram showing the results of tests on the cleaning sheet according to the first embodiment of the present invention, for Sample 1, Sample 2, and the invention product. [Figure 15] FIG. 1 is a perspective view showing an example of the structure of a transfer molding device used in the present invention. [Figure 16] 13 is a partial cross-sectional view for explaining the structure of a molding die in the transfer molding apparatus of FIG. 12. [Figure 17] FIG. 2 is a plan view showing a lead frame on which a semiconductor chip is mounted. [Figure 18] 1 is a perspective view showing an example of a structure of a semiconductor device, partially in cross section; [Figure 19] 1 is a cross-sectional view showing an example of a structure of a semiconductor device. [Figure 20] 1A to 1C are cross-sectional process flow diagrams showing an example of a manufacturing process for a semiconductor device. [Figure 21] 1 is a partial cross-sectional view of a molding die showing a state in which a lead frame to which a semiconductor chip is bonded is placed in the molding die. [Figure 22]1 is a partial cross-sectional view of a molding die showing a state in which a lead frame to which a semiconductor chip is bonded is placed in the molding die, the die is closed, and molding resin is filled between the die. FIG. [Diagram 23] FIG. 2 is a partial cross-sectional view of a molding die showing a state in which the die is opened after filling with molding resin. [Figure 24] FIG. 2 is a partial cross-sectional view of the mold showing a state in which a mold cleaning sheet is placed inside the mold. [Diagram 25] FIG. 11 is a partial cross-sectional view of a molding die showing a state in which a molding die cleaning sheet is placed inside the molding die, the die is closed, and cleaning resin is filled between the die members. [Figure 26] 6A to 6C are diagrams illustrating an impregnation step of a cleaning sheet according to a second embodiment of the present invention. [Figure 27] FIG. 11 is a diagram illustrating an example of an experimental result in an impregnation step of a cleaning sheet according to a second embodiment of the present invention. [Figure 28] 6A to 6C are diagrams illustrating a thermal curing process of the cleaning sheet according to the second embodiment of the present invention. [Figure 29] 6A to 6C are diagrams illustrating a die-cutting process for a cleaning sheet according to a second embodiment of the present invention. [Diagram 30] 6A to 6C are diagrams illustrating a thermal compression treatment process for a cleaning sheet according to a second embodiment of the present invention. [Diagram 31] FIG. 11 is a diagram showing configurations of Sample 1, Sample 2, and Sample 3 of a cleaning sheet according to a second embodiment of the present invention. [Diagram 32] FIG. 11 is a diagram showing the results of a heat resistance test of Sample 1 of the cleaning sheet according to the second embodiment of the present invention. [Diagram 33] FIG. 11 is a diagram showing the results of a heat resistance test of Sample 2 of the cleaning sheet according to the second embodiment of the present invention. [Diagram 34] FIG. 11 is a diagram showing the results of a heat resistance test of Sample 3 of the cleaning sheet according to the second embodiment of the present invention. [Diagram 35]FIG. 11 is a diagram showing the results of a tensile strength test of Samples 1, 2 and 3 regarding the cleaning sheet according to the second embodiment of the present invention. [Diagram 36] FIG. 11 is a diagram showing the results of a dust generation test for samples 1, 2 and 3 of the cleaning sheet according to the second embodiment of the present invention. [Figure 37] FIG. 11 is a diagram showing the results of a suction property test of Samples 1, 2 and 3 of the cleaning sheet according to the second embodiment of the present invention. [Figure 38] FIG. 11 is a diagram showing all the test results of Sample 1, Sample 2, and Sample 3 for the cleaning sheet according to the second embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0019] <First embodiment> Hereinafter, an embodiment of the present invention will be described with reference to FIGS.

[0020] <Configuration of cleaning sheet 100> The cleaning sheet 100 is made by impregnating recycled paper, the main component of which is cellulose and has a thickness of about 0.40 mm, with phenolic resin, which is unlikely to leak during molding of the cleaning resin, and then by subjecting the paper material impregnated with the phenolic resin to a heat treatment that includes heating the paper material at a temperature of at least 200°C to 230°C for a predetermined period of time, so that the cleaning resin does not penetrate or pass through it, as it does with nonwoven fabric. Specifically, the cleaning sheet 100 is made of recycled paper, and is impregnated with a thermosetting resin from the front layer 101 to the back layer 102 of the paper material, and is then thermally cured.

[0021] As shown in FIG. 1, the cleaning sheet 100 is rectangular and has a plurality of circular positioning holes 103 formed on the outer periphery. The cleaning sheet 100 is picked up by an automatic conveyor or a robot in an area other than the area where the round positioning holes are provided, and is loaded into the lower die of a mold of a semiconductor manufacturing device.

[0022] The press-punched cross section of the cleaning sheet 100 is impregnated with and cured with phenolic resin, and its thickness is, for example, about 0.50 mm.

[0023] <Manufacturing method of cleaning sheet 100> The method for producing the cleaning sheet according to this embodiment will be described with reference to FIGS.

[0024] First, as shown in Figure 2, 2 A roll of recycled paper material, mainly composed of cellulose and approximately 0.40 mm thick, is unwound and placed at a specified speed (time) into a resin impregnation tank filled with phenolic resin diluted with methanol at a prepreg ratio of approximately 19%, and while being wound up on a winding machine, the surface layer and back layer are dried in a dryer (impregnation process).

[0025] The results are shown in Figure 3. In FIG. 3, "M" indicates the right end of the rolled paper in the width direction perpendicular to the transport direction of the rolled paper, and "G" indicates the left end of the rolled paper in the width direction perpendicular to the transport direction of the rolled paper. Also, "VC" is an abbreviation for volatile content, which indicates the amount of remaining volatile matter, and in this experiment, indicates the remaining amount of methanol used as a diluent.

[0026] As a result of the experiment, as shown in Figure 3, no significant variation was observed in the amount of resin at the right and left ends of the roll width direction, perpendicular to the transport direction of the rolled paper material, at the start, middle, and end points of the roll, and the amount of remaining volatile matter was also at an acceptable level.

[0027] Next, as shown in FIG. 4, in the impregnation step, the rolled recycled paper material mainly composed of cellulose impregnated with the phenolic resin is unwound and subjected to a heat curing treatment (heat curing treatment step). As shown in FIG. 4, this heat curing process is roughly divided into primary curing and secondary curing. In the primary curing, a thermal curing process is performed on the rolled recycled paper material that has undergone the impregnation process by using an upper heater and a lower heater that are provided perpendicular to the transport direction of the rolled recycled paper material. At this time, the heating temperatures of the upper and lower heaters are both 200°C.

[0028] The secondary curing is a thermal curing process using far-infrared rays, and similar to the primary curing, is performed by an upper far-infrared heater and a lower far-infrared heater arranged perpendicular to the transport direction of the roll-shaped recycled paper material. In the secondary curing, the above-mentioned upper far-infrared heaters and lower far-infrared heaters are arranged in four pairs at a predetermined interval, and the far-infrared output of each of the upper far-infrared heaters and lower far-infrared heaters in each pair ((0), (1), (2), (3)) is the same. In order from the conveying direction, the outputs of the groups ((0), (1), (2), (3)) are 150(V), 170(V), 170(V), and 210(V). The heat-hardened roll of recycled paper is then wound up on a winding machine.

[0029] As shown in FIG. 5, the rolled recycled paper material that has been heat-cured in the heat-curing process is cut into rectangular recycled paper materials of a predetermined size. A rectangular piece of recycled paper material of a predetermined size is subjected to a heat compression process. In the thermal compression process, as shown in FIG. 6, a cleaning sheet (substrate) is sandwiched between protective plates and heating plates on the top and bottom, and the thermal compression molding is performed by applying pressure to the heating plates from above and below. Then, the cut and heat-compressed rectangular recycled paper material of a specified size is placed in a press die and punched (pressing process) to form the cleaning sheet 100 as shown in FIG. 1. After an inspection process, the sheet is packaged.

[0030] <Evaluation of physical properties of cleaning sheet 100> Hereinafter, the results of the evaluation of the physical properties of the cleaning sheet 100 will be described with reference to FIGS.

[0031] [Test sample] The physical properties of the cleaning sheets were evaluated using three recycled paper materials, including the inventive product, as shown in Figure 7. The invention is a phenolic resin molded sheet NL-1 manufactured by Oji Materials Co., Ltd., which contains 85-95% (wt%) cellulose and is composed of a hardened formaldehyde-phenol polycondensation compound. The base paper is impregnated with 19% phenolic resin (prepreg ratio), and has a thickness of 0.38 mm and a basis weight of 410 g / m 2 It was. Sample 1 is a phenolic resin molded sheet NS Bowl F manufactured by Nippon Tokai Industrial Paper Supply Co., Ltd., and is composed mainly of pulp, as well as paper strength agents and aluminum sulfate. The base paper is impregnated with 18% phenolic resin (prepreg ratio), and has a thickness of 0.37 mm and a basis weight of 387 g / m 2 It was. Sample 2 is a phenolic resin molded sheet NLS manufactured by Oji Materials Co., Ltd., and is composed mainly of cellulose, as well as other ingredients such as a fixing agent, a paper strength agent, and a sizing agent. The base paper is impregnated with 18% phenolic resin (prepreg ratio), and has a thickness of 0.39 mm and a basis weight of 379 g / m 2 It was.

[0032] [Heat resistance test] We commissioned the Yamagata Prefectural Industrial Technology Center to conduct heat resistance tests on the invention, sample 1, and sample 2.

[0033] <Evaluation method> Analysis samples were prepared by stacking multiple disks punched out from each material using a φ4 mm punch. The measurements were performed using a TG8120 manufactured by Rigaku Corporation. The differential thermal change was measured when the temperature was raised from room temperature to 800°C at 10°C per minute in an air atmosphere at 200 mL / min. Differential heat indicates the heat of reaction of endothermic and exothermic reactions occurring at each temperature. Measurement charts for each sample are shown in FIGS.

[0034] <Sample 1> The measurement chart of Sample 1 is shown in FIG. The weight change rate can be seen to be divided into four stages, that is, between room temperature and 100°C, at 200°C, between 200°C and 360°C, between 360°C and 500°C, and between 500°C and 700°C. No significant change in the differential heat is observed between room temperature and 200°C. Between 200°C and 500°C, an exothermic reaction occurs after 220°C, and two large exothermic peaks appear to overlap at 366°C and 456°C. The weight loss between these temperatures is believed to be due to this highly exothermic reaction. This heat generation caused a gradual rise in temperature, and it is believed that combustion was occurring. After that, between 500°C and 700°C, no clear reaction heat is visible on the chart, but there is a gradual exothermic reaction. The two exothermic peaks between 200° C. and 500° C. tend to be similar to those of the invention product and sample 2, but the peaks of sample 1 are sharper.

[0035] <Sample 2> The measurement chart of Sample 2 is shown in FIG. The weight change rate can be seen to be divided into four stages, that is, between room temperature and 100°C, at 200°C, between 200°C and 360°C, between 360°C and 500°C, and between 500°C and 700°C. No significant change in the differential heat is observed between room temperature and 200°C. Between 200°C and 500°C, an exothermic reaction occurs after 220°C, and two large exothermic peaks appear to overlap at 363°C and 425°C. The weight loss between these temperatures is believed to be due to this highly exothermic reaction. This heat generation caused a gradual rise in temperature, and it is believed that combustion was occurring. After that, between 500°C and 700°C, no clear reaction heat is visible on the chart, but there is a gradual exothermic reaction.

[0036] <Inventions> The measurement chart of the invention is shown in Figure 10. The weight change rate can be seen to be divided into four stages, that is, between room temperature and 100°C, at 200°C, between 200°C and 360°C, between 360°C and 500°C, and between 500°C and 700°C. No significant change in the differential heat is observed between room temperature and 200°C. Between 200°C and 500°C, an exothermic reaction occurs after 220°C, and two large exothermic peaks appear to overlap at 360°C and 436°C. The weight loss between these temperatures is believed to be due to this highly exothermic reaction. This heat generation caused a gradual rise in temperature, and it is believed that combustion was occurring. After that, between 500°C and 700°C, no clear reaction heat is visible on the chart, but there is a gradual exothermic reaction.

[0037] <Consideration> Looking at the weight change rate of each sample at 220°C and 360°C, at 220°C, the invention product, sample 1, and sample 2 were 3% to 4%. The difference between the invention, sample 1, and sample 2 is less than 1%. At 360°C, the invention product, sample 1, and sample 2 were 55% to 59%. The weight loss rate at 360°C mainly indicates the combustion reaction of the fibers. Taking this into consideration, we believe that the fiber and resin content has an effect, just as at 220°C.

[0038] <Appearance Observation> The appearance of the sample was observed after it was left in an oven heated to 200°C for 1 hour. Each sample was tested in triplicate. There was no difference in color among the three samples. Each sample showed a color change to a different extent. To quantify the color change, a Konica Minolta color difference meter was used for evaluation. As a result, the invention was classified as "slightly different." Sample 2 was classified as "very slightly different." The cause of the color change is believed to be a change in the color of the resin. Therefore, it is considered that the change between the invention product and sample 2, which originally turned brown after sample processing, was small.

[0039] <Consideration> From the evaluation of the thermal properties described above, it is believed that up to around 220°C, moisture adsorbed to the fibers is released, from 220°C to around 360°C, the fibers are burned, and beyond that, the resin is thermally decomposed. It is believed that the degree of weight loss during heating differs depending on the fiber and resin content. In terms of heat resistance, although there is a possibility of a change in color and a change in dimensions due to moisture release, there was no decomposition of the fibers or resin in any of the samples up to 220°C, and it is believed that there will be no significant effect on short-term performance. Regarding the change in hue, the invention product and sample 2 are only slightly different and are not considered to have any effect.

[0040] [Tensile strength test] As shown in FIG. 11, the evaluation was carried out using the invention product, sample 1, and sample 2 as test samples.

[0041] <Evaluation method> The tensile strength test was carried out in accordance with the tensile test method for paper and paperboard specified in JIS P8113. The test machine used was an Autograph AG-100kNX manufactured by Shimadzu Corporation. The test was performed 10 times for each material. The tensile strength, specific tensile strength, tensile energy absorption, tensile modulus and their average values ​​are shown in FIG. Regarding the direction, the same direction as the winding direction is defined as the vertical direction, and the perpendicular direction is defined as the horizontal direction. The average longitudinal and transverse values ​​were added to the table. The correlation values ​​between the average values ​​of tensile strength and tensile modulus and the hardness value were 0.9 or more, showing a strong positive correlation. Since the tensile index is the tensile strength value divided by the basis weight, the correlation value is about 0.6, which is a degree of correlation that can be recognized. For this reason, the tensile strength and tensile modulus are considered using the average values ​​in the machine direction and the cross direction.

[0042] <Consideration> Regarding tensile strength, the invention product had a tensile strength of 25 kN / m, while sample 2 and sample 1 had tensile strengths of 22 kN / m and 21 kN / m, respectively. In terms of specific tensile strength, calculated by dividing tensile strength by basis weight, sample 2 showed a high value of 67 Nm / g. The tensile modulus of elasticity of sample 2 was as high as 5458 MPa. The invention and sample 2 had values ​​of 5048 MPa and 5111 MPa, respectively, which were lower than sample 2. In either case, the results showed that there was no problem with the cleaning sheet.

[0043] [Dust generation evaluation test] As shown in FIG. 12, evaluation was performed using sample 1, sample 2, and the invention product as test samples.

[0044] <Evaluation method> The evaluation was performed by the tumbling method in accordance with JIS B 9923. The evaluation was carried out using one piece of each material, measuring 15 cm x 15 cm and not having been washed.

[0045] The evaluation test was performed using a tumbling dust generation tester CW-HDT-102 (manufactured by Akado Seisakusho) with a drum rotation speed of 30 rpm and a flow rate of 0.0102 m 3The flow rate of the old beverage was 1 cubic foot per minute, the particle counter was a Met One A2400B (Hach Ultra Analytics), and the flow rate of the old beverage was 1 cubic foot per minute.

[0046] The evaluation test is 1) The process of running a tumbling-type dust generation tester installed in a clean room (cleanliness: ISO class 5 (class 100)) without any contamination to confirm that the inside of the tester is dust-free. 2) The process of placing the sample into the drum of the testing machine and starting the testing machine. 3) Starting 30 seconds after the start of operation of the tester, measurements are taken continuously for 1 minute at a speed of 1 cubic foot per minute for a total of 5 times, the maximum and minimum values ​​are excluded, and the average of the remaining measurements is taken as the number of particles generated. The procedure was as follows.

[0047] <Evaluation Results> The evaluation results are shown in Figure 12. The number of particles with a diameter of 0.3 μm or more and less than 0.5 μm was 107 particles / sec (296 particles / ft) for sample 1. 3 ), and 42.7 particles / sec (118 particles / ft) for sample 2. 3 ), 31.8 pieces / sec (88 pieces / ft 3 ) was. The number of particles with a diameter of 0.5 μm or more and less than 1 μm was 116 particles / sec (321 particles / ft) for sample 1. 3 ), and 61.8 particles / sec (171 particles / ft) for sample 2. 3 ), 46.6 pieces / sec (129 pieces / ft 3 ) was. The number of particles with a diameter of 1 μm or more and less than 5 μm was 76.9 particles / sec (213 particles / ft) for sample 1. 3 ), and 70.0 particles / sec (194 particles / ft) for sample 2. 3 ), 44.6 pieces / sec (124 pieces / ft 3 ) was. The number of particles with a diameter of 5 μm or more and less than 10 μm was 10.7 particles / sec (30 particles / ft) for sample 1. 3 ), and 12.9 particles / sec (36 particles / ft) for sample 2. 3 ), 9.3 / sec (26 / ft3 ) was. The number of particles with a diameter of 10 μm or more and less than 25 μm was 0.5 particles / sec (1 particle / ft) for sample 1. 3 ), and 1.2 particles / sec (3 particles / ft) for sample 2. 3 ), 0.8 pieces / sec (2 pieces / ft 3 ) was. The number of particles with a diameter of 25 μm or more was 0 particles / sec (0 particles / ft) for sample 1. 3 ), and 0 pieces / sec (0 pieces / ft 3 ), 0 pieces / sec (0 pieces / ft 3 ) was.

[0048] <Consideration> The above evaluation shows that for particles with a particle size of 10 μm or more, there is no significant difference between Sample 1, Sample 2, and the invention product, but the smaller the particle size, the more significantly the dust generation of the invention product is superior to Sample 1 and Sample 2.

[0049] [Suction characteristic test] As shown in FIG. 13, evaluation was performed using sample 1, sample 2, and the invention product as test samples.

[0050] <Evaluation method> A suction system was attached to a Shimadzu Corporation material testing machine Autograph AG-X1000kN, and the load was measured when the sample was moved from the suction pad in the direction in which it was being pulled away from the suction pad. An oil rotary pump was used for suction, and a regulator was installed between the pump and the suction pad to adjust the suction pressure. A suction pad made of nitrile rubber and having a diameter of 20 mm was used. The test machine and the sample were fixed with a commercially available double-sided tape (Nicetack), and the sample did not peel off from the double-sided tape during the test. Suction was performed at the lowest setting of -0.001MPa (regulator reading). The conditions were set taking into account the suction settings at the production site. Five test samples were prepared for each evaluation, and each sample was tested three times, for a total of 15 tests.

[0051] <Evaluation Results> The evaluation results of the maximum load when peeling off the suction pad are shown in FIG. In FIG. 13, the number before the test number indicates the test sample prepared for each evaluation subject, and the number after the test number indicates the repetition. The unit is N (Newton). Looking at the average values, Sample 1, Sample 2, and the invention show high values ​​of over 10N. The average values ​​of Sample 1, Sample 2, and the invention are similar, ranging from 10.6N to 12.6N. In terms of the standard deviation, sample 2 had a slightly larger variation than the other samples, at 1.3 N.

[0052] <Summary of evaluation results> FIG. 14 shows a summary of the evaluation results of the above evaluation tests. There were no significant differences that were of concern between Sample 1, Sample 2, and the invention. However, as shown in FIG. 14, the product of the present invention showed good results in all evaluation tests. In this respect, the invention can be said to have an overall advantage over Sample 1 and Sample 2. In particular, this is an extremely effective cleaning sheet that is made from recycled paper and yet is effective in terms of vacuum suction, dust generation, rigidity, and peeling of the front and back of the sheet, and has performance and structure similar to that of metal frames and resin substrate frames. In conventional cleaning of molding dies, a cleaning sheet was manually set on the second die, indicated by the symbol 4 in FIG. 15. However, with the cleaning sheet of this embodiment, as shown in the results of the adhesion test, the frame can be automatically set on the second lower die, indicated by the symbol 4, from the loader, indicated by the symbol 1 in FIG. 15, using the claws and suction pads, and then the cleaning work can be automated by transporting it to the unloader after molding. Therefore, automated operation improves operator efficiency, significantly reducing manpower requirements and improving cleaning accuracy and quality.

[0053] <Method of Manufacturing Semiconductor Device> Hereinafter, the method for manufacturing the semiconductor device will be described with reference to FIGS. In the following, a transfer molding device will be described as an example of a device that performs molding.

[0054] <Configuration of semiconductor manufacturing equipment> The transfer molding apparatus shown in Figure 15 has a first mold 3 which is an upper mold, a second mold 4 which is a pair with the first mold, a molding die 5 which has the first mold 3 and the second mold 4, a loader 1 which loads a work (here, for example, a lead frame which has completed die bonding and wire bonding) into the molding die 5, and an unloader 2 which removes the work from the molding die 5.

[0055] In the transfer molding apparatus, the lead frame 201 to which the semiconductor chip 24 (see FIG. 17) is bonded is carried from the loader 1 shown in FIG. 15 into the molding die 5, where the semiconductor chip 24 and the like are resin-molded. A QFP (Quad Flat Package) 19, which is a semiconductor device after the above molding is carried out to an unloader 2 and accommodated therein.

[0056] The molding die 5 shown in FIG. 16 is provided with, for example, a cavity 6 having a shape corresponding to the molded portion 22 of the QFP 19 shown in FIG. 18, a cull 7, a runner 8, a pot 9, a plunger 10, ejector plates 11 and 15, ejector pins 12 and 16, a gate 13, and an air vent 14.

[0057] As shown in Figure 20, cavities 6 of a predetermined shape, which are first recesses in which semiconductor chips 24 are placed, are formed in multiple locations on the mating surface 26 of the second mold 4 of the mold forming section 28 (see Figure 16) (note that cavities 6 are also formed on the mating surface 26 of the first mold 3 in the same manner as the second mold 4). Molding resin such as tablets 212 is set in a predetermined position in the second mold 4, and multiple cylindrical pots 9 having second recesses are formed through the mold, and culls 7 are provided in each portion of the first mold 3 corresponding to the pots 9 as shown in Figure 21.

[0058] A number of runners 8, which are connected to a number of cavities 6, branch out from the cull 7, and when the first mold 3 and the second mold 4 are in close contact with each other, the pot 9 is connected to the number of cavities 6 via the cull 7 and the runners 8. An air vent 14 is formed on the outside of the cavity 6 to allow air in the cavity 6 to escape to the outside and complete filling of the resin.

[0059] The cleaning sheet 100 is used to clean the inside of the molded portion 28 by being placed between the first mold 3 and the second mold 4 of the molded portion 28 after molding of the periphery of a semiconductor chip 24, such as the semiconductor chip 24 and gold wire 21 as shown in FIG. 19 mounted on a lead frame 201 as shown in FIG. 17, has been performed any number of times.

[0060] During cleaning, only the cleaning sheet 100 is clamped between the first mold 3 and the second mold 4, and in this state, the cleaning resin 25 is supplied into the cavity 6 as shown in FIG. As a result, the cleaning resin 25 fills the cavity 6, and the cleaning sheet 100 is filled into every corner of the cavity 6 without the cleaning sheet 100 moving up and down (lifting) due to the effect of the resin flow speed.

[0061] The cleaning sheet 100 is formed in a size such that it can be guided by the positioning wedges 18 provided on each outer periphery of the mating surface 26 of the second mold 4 for positioning the upper and lower molds. Therefore, if the cleaning sheet 100 is placed in accordance with the positioning wedges 18 on each side, when the cleaning sheet 100 is placed on the second mold 4, it is not necessary to perform highly accurate positioning between the cleaning sheet 100 and the second mold 4. When positioning the sheet in the mold without providing positioning notches, the sheet dimensions may be adjusted to match the inner dimensions of each positioning wedge 18.

[0062] A QFP 19 shown in FIG. 18 is an example of a semiconductor device that is molded and assembled by the transfer molding apparatus shown in FIG. Here, the structure of the semiconductor device manufactured by the method for manufacturing a semiconductor device according to this embodiment will be described with reference to FIGS. In this embodiment, a QFP (Quad Flat Package) shown in FIG. 18 will be described as an example of a resin molded surface-mounted semiconductor device using a lead frame.

[0063] The QFP shown in FIG. 18 and FIG. 19 is used as, for example, a microcomputer. The structure of the QFP consists of a tab 200 of a small tab structure on which a semiconductor chip 24 having a semiconductor integrated circuit formed thereon is mounted, inner leads 20 and outer leads 23 formed by cutting from a lead frame 201 (see Figure 17), gold wires 21 (which may be copper wires other than gold wires) that electrically connect the bonding pads 203 of the semiconductor chip 24 to the inner leads 20, and a resin molded part 22 formed by resin molding the semiconductor chip 24 and the gold wires 21, etc. (see Figure 19).

[0064] In addition, in the QFP 19, the planar shape of the resin molded portion 22 is approximately square. Moreover, the semiconductor chip 24 is fixed to the tab 200 by a bonding material 202 such as silver paste. A plurality of outer leads 23 protruding from the four sides of the resin molded portion 22 are bent into a gull-wing shape.

[0065] <Method of Manufacturing Semiconductor Device> Next, a method for manufacturing a semiconductor device will be described with reference to FIG. The method for manufacturing a semiconductor device is characterized in that it includes a molding (resin molding) process of a semiconductor chip 24 using a transfer molding apparatus shown in FIG. 14, and a cleaning process of the inside of the mold forming section 28 of the transfer molding apparatus using a cleaning sheet 100 shown in FIG.

[0066] The details of the manufacturing method including the molding and cleaning steps are as follows. First, as shown in FIG. 20(a), a lead frame 201 is prepared. FIG. 16 shows a plan view of the lead frame 201. For convenience of illustration and explanation, the inner lead 20 and the outer lead 23 are only partially shown.

[0067] As shown in FIG. 20(b), a die bonding process (also called a die attachment process) is performed in which a semiconductor chip 24 is mounted on a tab 200 of a lead frame 201. In the die bonding process, as shown in FIG. 20(b), a bonding material 202, for example, a silver-containing epoxy adhesive, that is, a silver paste, is applied onto a tab 200 of a lead frame 201. Next, the semiconductor chip 24 is mounted on the tab 200 to which the bonding material 202 has been applied, using a collet (not shown).

[0068] As shown in FIG. 20(c), a wire bonding process is performed in which bonding pads 203 formed on electrodes of the semiconductor chip 24 mounted on the tab 200 are bonded and electrically connected to the corresponding inner leads 20 by gold wires 21. Thereafter, as shown in FIG. 20(d), a molding process is performed in which the semiconductor chip 24 mounted on the lead frame 201 and the peripheral parts of the semiconductor chip 24, such as the gold wires 21, are molded using resin to protect them from the external atmosphere, such as dust and humidity, and from mechanical shocks.

[0069] As shown in FIG. 20(e), using a cutting / forming die (not shown), the residual resin on the front and back surfaces of the frame that has adhered and formed near the gate 13 is separated from the lead frame 201, thereby separating the unit frame portion having the resin molded portion 22 from the lead frame 201, and bending the outer lead 23 into a specified shape (in this embodiment, a gull-wing shape).

[0070] Next, the resin molding step in the above molding process will be described with reference to FIGS. A solid tablet 212 (molding resin) heated by a preheater is placed on top of the plunger 10 shown in FIG. 21, and then the lead frame 201 having the semiconductor chip 24 and the inner lead 20 wire-bonded thereto is transported from the loader 1 shown in FIG. 15 to the molding die 5.

[0071] In this state, the second mold 4 is moved toward the first mold 3, thereby forming a space including the cavity 6 between the first mold 3 and the second mold 4, which form the molded portion 28. 22, when the molten molding resin is pushed out into the cull 7 by the plunger 10, the molding resin flows through the runner 8 and the gate 13 into the cavity 6. Furthermore, the molding resin filled in the cavity 6 is thermally cured by heat and curing, and then the second mold 4 is moved downward to open the mold.

[0072] Subsequently, as shown in FIG. 23, the ejector plate 15 is caused to move downward and return, and the ejector plate 11 is caused to move upward and return. As a result, the ejector pins 12 and 16 project, completing the mold opening, and the resin-molded semiconductor device 213 with the outer frame 214 is removed.

[0073] <Action and effect> As described above, the cleaning sheet 100 according to this embodiment is a cleaning sheet that is placed between the surfaces of a resin molding die and cleans the resin molding die surfaces with a cleaning resin, and is made of recycled paper whose main component is cellulose, and is recycled. In other words, since the cleaning sheet 100 is made from recycled paper, it is extremely inexpensive and provides the greatest economical benefits compared to conventional dummy frames.

[0074] Moreover, the cleaning sheet 100 according to this embodiment is impregnated with 19% (prepreg ratio) of phenol resin from the front layer to the back layer of the recycled paper material, and cured by heating at 200°C to 230°C for a predetermined time. In other words, considering that the temperature of semiconductor molding dies is 170°C to 180°C, this cleaning sheet has sufficient heat resistance and is ideal for a wide variety of molding dies.

[0075] Moreover, the cleaning sheet 100 according to this embodiment is impregnated with 19% (prepreg ratio) of phenol resin from the front layer to the back layer of the recycled paper material, and is thermally cured. This cleaning sheet 100 was rated better than the samples in the tensile strength, tensile strength ratio, tensile energy absorption, tensile modulus, etc. in the tensile strength test, and was at a level that would not cause any problems in practical use as a cleaning sheet.

[0076] Furthermore, the cleaning sheet 100 according to this embodiment exhibited a higher hardness than the sample product in a hardness test. As a result, it is possible to reduce warping on the mold compared to conventional products, making it an ideal cleaning sheet for a wide variety of molding dies.

[0077] In addition, in the heat resistance characteristic test, the cleaning sheet 100 according to this embodiment showed that the heat generation peak was lower than that of the sample product, the peak shifted to the higher temperature range, and the weight decay curve also showed a decay point in the higher temperature range and a gentle decay gradient. This makes it an ideal cleaning sheet for a wide variety of mold dies.

[0078] Furthermore, the cleaning sheet 100 according to this embodiment exhibited results in a tear and tensile strength test that showed higher rigidity and specific strength than the sample product. Therefore, it is possible to reduce warping on the mold compared to the sample product, making it an ideal cleaning sheet for a wide variety of molds.

[0079] Furthermore, in a dust generation evaluation test, the cleaning sheet 100 according to this embodiment showed results showing that it scattered fewer particles than the sample product. Therefore, it is expected that the generation of dust during use, which was an issue with cleaning sheets using the sample product as a base material, can be significantly reduced.

[0080] Generally, cleaning sheets are used for semiconductor devices such as DIP, QFP, and TSOP, but as described above, the cleaning sheet 100 according to this embodiment is resistant to deformation or warping and has high rigidity, and therefore can also be used as a dummy frame for MAP, BGA, CSP, etc. Therefore, by replacing the currently used expensive substrate frames with dummy frames having similar physical properties to the cleaning sheet 100 according to this embodiment, a significant cost reduction can be expected. The cleaning sheet 100 in this embodiment may be provided with openings as appropriate depending on the semiconductor device used.

[0081] Furthermore, the cleaning sheet 100 according to this embodiment is basically made of recycled paper and resin, so that it can be easily disposed of after use and has no impact on the environment. Furthermore, even as prices of metals (eg, Cu) and resin substrates continue to rise, the product is less susceptible to this effect.

[0082] The cleaning sheet according to this embodiment is made of recycled paper, and as shown by the results of the above-mentioned tests, it meets all of the requirements for a cleaning sheet in terms of vacuum suction, dust generation, heat resistance, rigidity and strength, and peeling of the sheet from the front to the back. In addition, the cleaning sheet made from paper (made from recycled paper and virgin pulp) that is suitable for use in the mold chamber clean rooms used in semiconductor manufacturing is a first for the global semiconductor industry.

[0083] Although an embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to this embodiment, and designs that do not deviate from the gist of the present invention are also included.

[0084] Regarding the cleaning sheet 100 according to this embodiment, without any sales, marketing, or advertising activities, the product outline has spread by word of mouth, and inquiries have been received from well-known semiconductor manufacturers both in Japan and overseas, and trading companies that handle the product. Starting with a delivery to one of Japan's leading semiconductor manufacturers in the fall of this year, discussions are currently underway for delivery to semiconductor manufacturers and trading companies both in Japan and overseas.

[0085] In other words, the cleaning sheet 100 according to the embodiment has achieved commercial success, which is a determining factor for positively inferring an inventive step.

[0086] In addition, the company that has contacted us and is scheduled to start making deliveries this fall is a leading company (applicant of Patent Document 3) that has been conducting research into replacing metal dummy frames with cleaning sheets. The fact that such a company is adopting a cleaning sheet related to the cleaning sheet 100 according to the embodiment is clear evidence that the cleaning sheet 100 according to the embodiment is not an invention that a person skilled in the art could easily conceive of. <Second embodiment>

[0087] This embodiment will be described below with reference to FIGS. The cleaning sheet 1A of this embodiment is an improvement on the cleaning sheet A of the first embodiment, which satisfies the requirements as a cleaning sheet and has no problems in practical use, in order to further improve the quality of the cleaning sheet. Specifically, measures were taken to address the issue of the ink contained in recycled paper transferring to a mold, which became evident during implementation testing of cleaning sheet A according to the first embodiment, and consideration was given to changing the base paper.

[0088] <Configuration of cleaning sheet 100A> The cleaning sheet 100A is made by impregnating recycled paper, the main component of which is cellulose and has a thickness of about 0.60 mm, with phenolic resin, which is unlikely to leak during molding of the cleaning resin, and then by subjecting the paper material impregnated with the phenolic resin to a heat treatment that includes heating the paper material at a temperature of at least 200°C to 230°C for a predetermined period of time, so that the cleaning resin does not penetrate or pass through it, as it does through nonwoven fabric.

[0089] The cleaning sheet 100A according to this embodiment has a rectangular shape like the cleaning sheet 100, and is provided with a plurality of circular positioning holes 103 on the outer periphery. The cleaning sheet 100 is picked up by an automatic conveyor or a robot in an area other than the area where the round positioning holes are provided, and is loaded into the lower die of a mold of a semiconductor manufacturing device.

[0090] The press-punched cross section of the cleaning sheet 100A is impregnated with and cured with phenolic resin, and the thickness of the sheet is, for example, about 0.60 mm.

[0091] <Manufacturing method for cleaning sheets> A method for producing the cleaning sheet A according to this embodiment will be described with reference to FIGS.

[0092] First, as shown in FIG. 26, 2 A roll of recycled paper material, mainly composed of cellulose and approximately 0.60 mm thick, is unwound and placed at a specified speed (time) into a resin impregnation tank filled with phenolic resin diluted with methanol at a prepreg ratio of approximately 11% to 19%, and while being wound up by a winder, the surface layer and back layer are dried in a dryer (impregnation process).

[0093] The results are shown in FIG. In Figure 27, "M" indicates the right end of the rolled paper in the width direction perpendicular to the transport direction of the rolled paper, and "G" indicates the left end of the rolled paper in the width direction perpendicular to the transport direction of the rolled paper. Also, "VC" is an abbreviation for volatile content, which indicates the amount of remaining volatile matter, and in this experiment, indicates the remaining amount of methanol used as a diluent.

[0094] As a result of the experiment, as shown in Figure 27, no significant variation was observed in the amount of resin at the right and left ends of the rolled paper in the width direction of the paper material, which is perpendicular to the transport direction, before cutting, after cutting, and after thermal curing, and the amount of remaining volatile matter was also at an acceptable level.

[0095] Next, as shown in FIG. 28, in the impregnation step, the rolled recycled paper material mainly composed of cellulose impregnated with the phenolic resin is unwound and subjected to a heat curing treatment (heat curing treatment step). As shown in FIG. 28, this heat curing process is roughly divided into a primary curing process and a secondary curing process. In the primary curing, a thermal curing process is performed on the rolled recycled paper material that has undergone the impregnation process by using an upper heater and a lower heater that are provided perpendicular to the transport direction of the rolled recycled paper material. At this time, the heating temperatures of the upper and lower heaters are both 200°C.

[0096] The secondary curing is a thermal curing process using far-infrared rays, and similar to the primary curing, is performed by an upper far-infrared heater and a lower far-infrared heater arranged perpendicular to the transport direction of the roll-shaped recycled paper material. In the secondary curing, the above-mentioned upper far-infrared heaters and lower far-infrared heaters are arranged in four pairs at a predetermined interval, and the far-infrared output of each of the upper far-infrared heaters and lower far-infrared heaters in each pair ((0), (1), (2), (3)) is the same. In order from the conveying direction, the outputs of the groups ((0), (1), (2), (3)) are 150(V), 170(V), 170(V), and 210(V). The heat-hardened roll of recycled paper is then wound up on a winding machine.

[0097] As shown in FIG. 29, the rolled recycled paper material that has been heat-cured in the heat-curing process is cut into rectangular recycled paper materials of a predetermined size. A rectangular piece of recycled paper material of a predetermined size is subjected to a heat compression process. In the thermal compression process, as shown in FIG. 30, a cleaning sheet (substrate) is sandwiched between protective plates and heating plates on the top and bottom, and the thermal compression molding is performed by applying pressure to the heating plates from above and below. Then, the cut and heat-compressed rectangular recycled paper material of a specified size is placed in a press die and punched (press process) to form, for example, a cleaning sheet 100A of the shape shown in FIG. 1, which is then inspected and packaged.

[0098] <Evaluation of the physical properties of cleaning sheets> Hereinafter, the results of the evaluation of the physical properties of the cleaning sheet 100A will be described with reference to FIGS.

[0099] [Test sample] The physical properties of cleaning sheets were evaluated using three recycled paper materials, as shown in FIG. Samples 1, 2, and 3 were all made from phenolic resin molded sheet CPA manufactured by Oji Materials Co., Ltd. Phenolic resin molded sheet CPA manufactured by Oji Materials Co., Ltd. contains 85-95% (wt%) cellulose and is also composed of a hardened formaldehyde-phenol polycondensation compound. As shown in FIG. 31, sample 1 was obtained by impregnating the base paper with phenol resin at 11% relative to the prepreg and subjecting it to a heat curing treatment. As shown in FIG. 31, sample 2 is obtained by impregnating the base paper with phenol resin at 15% relative to the prepreg and subjecting it to a heat curing treatment. As shown in FIG. 31, sample 3 is obtained by impregnating the base paper with phenol resin at 19% relative to the prepreg and subjecting it to a heat curing treatment.

[0100] [Heat resistance test] The Yamagata Prefectural Industrial Technology Center was commissioned to conduct heat resistance property tests on Samples 1, 2, and 3.

[0101] <Evaluation method> Analysis samples were prepared by stacking multiple disks punched out from each material using a φ4 mm punch. The measurements were performed using a TG8120 manufactured by Rigaku Corporation. The differential thermal change was measured when the temperature was raised from room temperature to 800°C at 10°C per minute in an air atmosphere at 200 mL / min. Differential heat indicates the heat of reaction of endothermic and exothermic reactions occurring at each temperature. Measurement charts for each sample are shown in FIGS.

[0102] <Sample 1> The measurement chart of Sample 1 is shown in FIG. The weight change rate can be seen to be divided into four stages, that is, between room temperature and 100°C, at 200°C, between 200°C and 360°C, between 360°C and 500°C, and between 500°C and 700°C. No significant change in the differential heat is observed between room temperature and 200°C. Between 200°C and 500°C, an exothermic reaction occurs after 220°C, with two large exothermic peaks overlapping at 375.7°C and 459.3°C. The weight loss between these temperatures is believed to be due to this highly exothermic reaction. This heat generation caused a gradual rise in temperature, and it is believed that combustion was occurring. After that, between 500°C and 700°C, no clear reaction heat is visible on the chart, but there is a gradual exothermic reaction. The two exothermic peaks between 200° C. and 500° C. tend to be similar to those of Samples 2 and 3, but the peaks of Sample 1 are sharper.

[0103] <Sample 2> The measurement chart of sample 2 is shown in FIG. The weight change rate can be seen to be divided into four stages, that is, between room temperature and 100°C, at 200°C, between 200°C and 360°C, between 360°C and 500°C, and between 500°C and 700°C. No significant change in the differential heat is observed between room temperature and 200°C. Between 200°C and 500°C, an exothermic reaction occurs after 220°C, with two large exothermic peaks overlapping at 371.3°C and 461.7°C. The weight loss between these temperatures is believed to be due to this highly exothermic reaction. This heat generation caused a gradual rise in temperature, and it is believed that combustion was occurring. After that, between 500°C and 700°C, no clear reaction heat is visible on the chart, but there is a gradual exothermic reaction.

[0104] <Sample 3> The measurement chart of Sample 3 is shown in FIG. The weight change rate can be seen to be divided into four stages, that is, between room temperature and 100°C, at 200°C, between 200°C and 360°C, between 360°C and 500°C, and between 500°C and 700°C. No significant change in the differential heat is observed between room temperature and 200°C. Between 200°C and 500°C, an exothermic reaction occurs after 220°C, with two large exothermic peaks overlapping at 469.1°C and 492.7°C. The weight loss between these temperatures is believed to be due to this highly exothermic reaction. This heat generation caused a gradual rise in temperature, and it is believed that combustion was occurring. After that, between 500°C and 700°C, no clear reaction heat is visible on the chart, but there is a gradual exothermic reaction.

[0105] <Consideration> Looking at the weight change rate of each sample at 220°C and 360°C, around 250°C, Samples 1, 2, and 3 were 3.8% to 4.8%. The difference between sample 1, sample 2 and sample 3 is less than 1%. At around 360°C, Samples 1, 2 and 3 were between 42% and 53%. The weight loss rate at 360°C mainly indicates the combustion reaction of the fibers. Taking this into consideration, we believe that the fiber and resin content has an effect, just as at 250°C.

[0106] <Appearance Observation> The appearance of the sample was observed after it was left in an oven heated to 200°C for 1 hour. Each sample was tested in triplicate. There was no difference in color among the three samples. Each sample showed a color change to a different extent. To quantify the color change, a Konica Minolta color difference meter was used for evaluation. As a result, sample 3 was classified as "slightly different." Sample 2 was classified as "very slightly different." The cause of the color change is believed to be a change in the color of the resin. Therefore, it is considered that the change between the invention product and sample 2, which originally turned brown after sample processing, was small.

[0107] <Consideration> From the evaluation of the thermal properties described above, it is believed that up to around 250°C, the moisture adsorbed to the fiber is released, from 250°C to around 360°C, the fiber is burned, and beyond that, the resin is thermally decomposed. It is believed that the degree of weight loss during heating differs depending on the fiber and resin content. In terms of heat resistance, although there is a possibility of a change in color and a change in dimensions due to moisture release, there was no decomposition of the fibers or resin in any of the samples up to 250°C, and it is believed that there will be no significant effect on short-term performance. Regarding the change in hue, there is only a slight difference between Sample 2 and Sample 3, and it is considered that there is no effect.

[0108] [Tensile strength test] As shown in FIG. 31, evaluation was performed using Sample 1, Sample 2, and Sample 3 as test samples.

[0109] <Evaluation method> The tensile strength test was carried out in accordance with the tensile test method for paper and paperboard specified in JIS P8113. The test machine used was an Autograph AG-100kNX manufactured by Shimadzu Corporation. The test was performed 10 times for each material. The tensile strength, tensile strength index, tensile energy absorption amount, tensile modulus, and their average values ​​are shown in FIG. Regarding the direction, the same direction as the winding direction is defined as the longitudinal direction, and the perpendicular direction is defined as the transverse direction. The average values ​​of the longitudinal and transverse directions are added to the table. The correlation values ​​between the average values ​​of tensile strength and tensile modulus and the hardness value were 0.9 or more, showing a strong positive correlation. Since the tensile index is the tensile strength value divided by the basis weight, the correlation value is about 0.6, which is a degree of correlation that can be recognized. For this reason, the tensile strength and tensile modulus are considered using the average values ​​in the machine direction and the cross direction.

[0110] <Consideration> In terms of tensile strength, Sample 1 had a value of 27.4 kN / m, Sample 2 had a value of 31.8 kN / m, and Sample 3 had a value of 38.6 kN / m. In terms of specific tensile strength, calculated by dividing tensile strength by basis weight, sample 1 showed high values ​​of 53 Nm / g, sample 2 showed high values ​​of 60.7 Nm / g, and sample 3 showed high values ​​of 68.9 Nm / g. The tensile energy absorption rate of sample 1 was 446 J / m 2 , Sample 2 is 395.7 J / m 2 , Sample 3 is 424.8 J / m 2 , and showed high values. The tensile modulus of elasticity was high, at 6,324.5 MPa for sample 1, 5,957.4 MPa for sample 2, and 6,872 MPa for sample 3. In either case, the results showed that there was no problem with the cleaning sheet.

[0111] [Dust generation evaluation test] As shown in FIG. 31, evaluation was performed using Sample 1, Sample 2, and Sample 3 as test samples.

[0112] <Evaluation method> The evaluation was performed by the tumbling method in accordance with JIS B 9923. The evaluation was performed using one piece of each material measuring 15 cm x 15 cm that had not been washed.

[0113] The evaluation test was performed using a tumbling dust generation tester CW-HDT-102 (manufactured by Akado Seisakusho) with a drum rotation speed of 30 rpm and a flow rate of 0.0102 m 3 The flow rate of the old beverage was 1 cubic foot per minute, the particle counter was a Met One A2400B (Hach Ultra Analytics), and the flow rate of the old beverage was 1 cubic foot per minute.

[0114] The evaluation test is 1) The process of running a tumbling-type dust generation tester installed in a clean room (cleanliness: ISO class 5 (class 100)) without any contamination to confirm that the inside of the tester is dust-free. 2) The process of placing the sample into the drum of the testing machine and starting the testing machine. 3) Starting 30 seconds after the start of operation of the tester, measurements are taken continuously for 1 minute at a speed of 1 cubic foot per minute for a total of 5 times, the maximum and minimum values ​​are excluded, and the average of the remaining measurements is taken as the number of particles generated. The procedure was as follows.

[0115] <Evaluation Results> The evaluation results are shown in Figure 36. The number of particles with a diameter of 0.3 μm or more and less than 0.5 μm was 286 particles / sec (793 particles / ft) for sample 1. 3 ), and 97.4 particles / sec (270 particles / ft) for sample 2. 3 ), and 5.6 particles / sec (16 particles / ft) for sample 3. 3 ) was. The number of particles with a diameter of 0.5 μm or more and less than 1 μm was 332 particles / sec (894 particles / ft) for sample 1. 3 ), and 152 particles / sec (421 particles / ft) for sample 2. 3 ), and 11.2 particles / sec (31 particles / ft) for sample 3. 3 ) was. The number of particles with a diameter of 1 μm or more and less than 5 μm was 232 particles / sec (643 particles / ft) for sample 1. 3 ), and 150 pieces / sec (417 pieces / ft) for sample 2 3 ), and 14.1 particles / sec (39 particles / ft) for sample 3. 3 ) was. The number of particles with a diameter of 5 μm or more and less than 10 μm was 34 particles / sec (94 particles / ft) for sample 1. 3 ), and 38.9 particles / sec (108 particles / ft) for sample 2. 3 ), and 3.0 pieces / sec (8 pieces / ft 3 ) was. The number of particles with a diameter of 10 μm or more and less than 25 μm was 1.7 particles / sec (5 particles / ft) for sample 1. 3 ), and 4.0 pieces / sec (11 pieces / ft) for sample 2 3 ), and 0 pieces / sec (0 pieces / ft 3 ) was. The number of particles with a diameter of 25 μm or more was 0 particles / sec (0 particles / ft) for sample 1.3 ), and 0 pieces / sec (0 pieces / ft 3 ), 0 pieces / sec (0 pieces / ft 3 ) was.

[0116] <Consideration> From the above evaluation, there is no significant difference between Sample 1, Sample 2, and Sample 3 for particles with a particle size of 10 μm or more.

[0117] [Suction characteristic test] As shown in FIG. 31, evaluation was performed using Sample 1, Sample 2, and Sample 3 as test samples.

[0118] <Evaluation method> A suction system was attached to a Shimadzu Corporation material testing machine Autograph AG-X1000kN, and the load was measured when the sample was moved from the suction pad in the direction in which it was being pulled away from the suction pad. An oil rotary pump was used for suction, and a regulator was installed between the pump and the suction pad to adjust the suction pressure. A suction pad made of nitrile rubber and having a diameter of 20 mm was used. The test machine and the sample were fixed with a commercially available double-sided tape (Nicetack), and the sample did not peel off from the double-sided tape during the test. Suction was performed at the lowest setting of -0.001MPa (regulator reading). The conditions were set taking into account the suction settings at the production site. Three test samples were prepared for each evaluation, and each sample was tested three times, for a total of nine tests.

[0119] <Evaluation Results> The evaluation results of the maximum load when peeling off the suction pad are shown in FIG. In FIG. 37, the number before the test number indicates the test sample prepared for each evaluation subject, and the number after the test number indicates the repetition. The unit is N (Newton). Looking at the average values, samples 1, 2, and 3 show high values ​​of 10N or more. The average values ​​of samples 1, 2, and 3 are similar, ranging from 12.6N to 16.5N.

[0120] <Transfer confirmation test> As shown in FIG. 31, evaluation was performed using Sample 1, Sample 2, and Sample 3 as test samples.

[0121] <Evaluation method> In the actual machine state, Samples 1, 2, and 3 were visually inspected for the presence or absence of transfer onto the mold.

[0122] <Evaluation Results> For Sample 1, Sample 2, and Sample 3, no stains or transfer of deposits to the mold were observed.

[0123] <Summary of evaluation results> FIG. 38 shows a summary of the evaluation results of the above evaluation tests. Although there were differences in the data between Sample 1, Sample 2, and Sample 3, there were no significant differences that should be taken into consideration. In other words, even though the raw material is recycled paper, it is effective in terms of vacuum suction, dust generation, rigidity, and peeling of the front and back of the sheet, overcoming the issues unique to recycled paper, making it an extremely effective cleaning sheet with performance and structure similar to that of metal frames and resin substrate frames. In conventional cleaning of molding dies, a cleaning sheet was manually set on the second die indicated by symbol 4 in FIG. 15. However, with cleaning sheet A of this embodiment, as shown in the results of the adhesion test, the frame can be automatically set on the second lower die indicated by symbol 4 from the loader indicated by symbol 1 in FIG. 15 using claws and suction pads, making it possible to automate the cleaning work by transporting it to the unloader after molding. Therefore, automated operation improves operator efficiency, significantly reducing manpower requirements and improving cleaning accuracy and quality.

[0124] <Action and effect> As described above, the cleaning sheet 100A of this embodiment is a cleaning sheet that is placed between the surfaces of a resin molding die and cleans the resin molding die surfaces with a cleaning resin, and is made of recycled paper whose main component is cellulose. In other words, the cleaning sheet 100A is made of recycled paper, so it is extremely inexpensive and provides the greatest economical benefits compared to conventional metal dummy frames.

[0125] The cleaning sheet 100A according to this embodiment is impregnated with 11 to 19% (prepreg ratio) of phenol resin from the front layer to the back layer of the recycled paper material, and cured by heating at 200°C to 230°C for a predetermined time. In other words, considering that the temperature of semiconductor molding dies is 170°C to 180°C, this cleaning sheet has sufficient heat resistance and is ideal for a wide variety of molding dies.

[0126] Moreover, the cleaning sheet 100A according to this embodiment is impregnated with 11 to 19% (prepreg ratio) of phenol resin from the front layer to the back layer of the recycled paper material, and is thermally cured. This cleaning sheet 100 received excellent evaluations in the tensile strength test, tensile strength index, tensile energy absorption amount, tensile modulus, transfer confirmation test, etc., and was at a level that was problem-free for practical use as a cleaning sheet.

[0127] Furthermore, the cleaning sheet 100A according to this embodiment exhibited high hardness in a hardness test. As a result, it is possible to reduce warping on the mold compared to conventional products, making it an ideal cleaning sheet for a wide variety of molding dies.

[0128] In addition, in a heat resistance characteristic test, the cleaning sheet 100A according to this embodiment showed a low heat generation peak that shifted to a higher temperature range, and the weight decay curve also showed a decay point in the higher temperature range and a gentle decay gradient. This makes it an ideal cleaning sheet for a wide variety of mold dies.

[0129] Furthermore, the cleaning sheet 100A according to this embodiment exhibited high rigidity and specific strength in tear and tensile strength tests. Therefore, it is possible to reduce warping on the mold compared to the sample product, making it an ideal cleaning sheet for a wide variety of molds.

[0130] Furthermore, the cleaning sheet 100A according to this embodiment showed results in a dust generation evaluation test in which little particles were scattered. Therefore, it is expected that the generation of dust during use, which was an issue with cleaning sheets using the sample product as a base material, can be significantly reduced.

[0131] Generally, cleaning sheets are used for semiconductor devices such as DIP, QFP, and TSOP, but as described above, the cleaning sheet 100A according to this embodiment is resistant to deformation or warping and has high rigidity, and therefore can also be used as a dummy frame for MAP, BGA, CSP, etc. Therefore, by replacing the currently used expensive substrate frames with dummy frames having similar physical properties to the cleaning sheet 100A according to this embodiment, a significant cost reduction can be expected. The cleaning sheet 100 in this embodiment may be provided with openings as appropriate depending on the semiconductor device used.

[0132] Furthermore, the cleaning sheet 100A according to this embodiment is basically made of recycled paper and resin, so that it can be easily disposed of after use and does not have an adverse effect on the environment. Furthermore, even as prices of metals (eg, Cu) and resin substrates continue to rise, the product is less susceptible to this effect.

[0133] Cleaning sheet A in this embodiment is made of paper, and as shown by the results of the above-mentioned test evaluations, it meets all of the requirements for a cleaning sheet in terms of vacuum suction properties, dust generation, heat resistance, rigidity strength, and peeling of the sheet from the front to the back. In addition, the cleaning sheet made from paper (made from recycled paper and virgin pulp) suitable for use in the mold chamber clean rooms used in semiconductor manufacturing is a first for the global semiconductor industry.

[0134] Although an embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to this embodiment, and designs that do not deviate from the gist of the present invention are also included.

[0135] Regarding the cleaning sheet 100 according to this embodiment, without any sales, marketing, or advertising activities, the product outline has spread by word of mouth, and inquiries have been received from well-known semiconductor manufacturers both in Japan and overseas, and trading companies that handle the product. Starting with a delivery to one of Japan's leading semiconductor manufacturers in the fall of this year, discussions are currently underway for delivery to semiconductor manufacturers and trading companies both in Japan and overseas.

[0136] In other words, the cleaning sheet 100 according to the embodiment has achieved commercial success, which is a determining factor for positively inferring an inventive step.

[0137] In addition, the company that has contacted us and is scheduled to start making deliveries this fall is a leading company that has been conducting research into replacing metal dummy frames with cleaning sheets, and the fact that such a company is adopting a cleaning sheet related to the cleaning sheet 100 of the embodiment is clear evidence that the cleaning sheet 100 of the embodiment is not an invention that a person skilled in the art could easily conceive of. [Explanation of symbols]

[0138] 1; Loader 2; Unloader 3. First mold 4; Second mold 5;Molding tool 6;Cavity 7; Cal 8. Runner 9;Pot 10; plunger 11; Ejector plate 12;Ejector pin 13. Gate 14;Air vent 15;Ejector plate 16;Ejector pin 18: Positioning watch 19; QFP (semiconductor device) 20; Inner lead 21; Bonding wire 22: Molded part 23;Outer lead 24: Semiconductor chips 25;Cleaning resin 26;Mating surface 28;Mold forming section 100; cleaning sheets 101; surface layer 102; Back layer 103;Mounting hole 200;tab 201; Lead frame 202;Jointing material 203; Bonding pad 213: Semiconductor device with resin-molded outer frame 214; Outer frame

Claims

1. A cleaning sheet which is interposed between surfaces of a resin molding die and cleans the resin molding die surface with a cleaning resin, The cleaning sheet is made of recycled paper material whose main component is cellulose, has an attachment hole at least at one end of its outer periphery for attachment to the resin molding die, is impregnated with 11 to 19% (prepreg ratio) of phenolic resin from the front layer to the back layer of the recycled paper material, and is hardened by heating at 200°C to 230°C for a predetermined period of time.

2. A cleaning sheet is disposed between a pair of first and second dies of a molding die having a cavity block which is a mounting area for a lead frame.

3. A method for manufacturing a semiconductor device, comprising the step of cleaning a mating surface of a pair of a first mold and a second mold, the pair having a cavity block which is a mounting area for a lead frame and a substrate frame, using the cleaning sheet according to claim 1, providing at least a cleaning sheet when disposed between the first mold and the second mold; automatically placing the cleaning sheet into the mold and clamping it between the first mold and the second mold; supplying a cleaning resin from a pot and filling the cleaning sheet with the cleaning resin; releasing the cleaning resin and the cleaning sheet from the mold after the cleaning resin is cured; 1. A method for manufacturing a semiconductor device comprising the steps of:

4. 4. The method for manufacturing a semiconductor device according to claim 3, wherein the semiconductor device includes at least a DIL-P, a QFP, a diode, and a transistor.

5. A process of impregnating recycled paper mainly composed of cellulose with phenolic resin at 11 to 19% (prepreg ratio); a step of curing the phenolic resin by subjecting the paper material impregnated with the phenolic resin to a heat treatment including a treatment of heating the paper material at a temperature of at least 200° C. to 230° C. for a predetermined period of time; A method for producing a cleaning sheet, comprising the steps of:

Images