TW202235720A - Polishing cloth - Google Patents

Abstract

Provided is a polishing cloth including, as forming materials of the polishing cloth, an unwoven fabric and a resin that is impregnated in the unwoven fabric, wherein an Asker-C hardness is 80 or more, and an interquartile range of mean values of vertical abundance ratios of the forming materials in a width of 100 μm is 10.5 or less.

Description

abrasive cloth

The invention relates to a grinding cloth.

A polishing cloth is used for polishing a workpiece such as a silicon wafer (for example, Patent Document 1). The abrasive cloth includes a non-woven fabric and a resin impregnated into the non-woven fabric as a forming material for forming the abrasive cloth.

In such a polishing cloth, it is known that edge sagging occurs when the flexibility is too high. Previously, by increasing the amount of resin impregnated to harden the abrasive cloth, excessive contact to the end was prevented, thereby reducing edge sag. [Prior Art Literature] [Patent Document]

[Patent Document 1] Japanese Patent Laid-Open No. 2006-43811

[Problem to be solved by the invention]

However, if the resin content is increased, local unevenness tends to occur at intervals of approximately 100 to 200 μm in the entire polishing cloth. In particular, the non-uniformity caused by the longitudinal entanglement of the non-woven fabric will cause a slight deviation in hardness, resulting in local excessive contact and a decrease in the flatness of the object to be ground.

Therefore, in view of the above-mentioned problems, the subject of the present invention is to provide a polishing cloth capable of suppressing minute variations in hardness. [Technical means to solve the problem]

The abrasive cloth of the present invention has a non-woven fabric and a resin impregnated in the non-woven fabric as the forming material for forming the abrasive cloth, the Asker-C hardness is 80 or more, and the average value of the aspect ratio of the 100 μm width of the above-mentioned forming material is four The quantile range is below 10.5.

One embodiment of the present invention will be described below.

The polishing cloth of this embodiment includes a nonwoven fabric and a resin impregnated with the nonwoven fabric as a forming material for forming the polishing cloth. That is, the polishing cloth of the present embodiment includes a forming material including a nonwoven fabric and a resin impregnated into the nonwoven fabric.

The abrasive cloth of this embodiment has an Asker-C hardness of 80 or more, preferably 85 or more. On the other hand, the Asker-C hardness is preferably 95 or less from the point of view that defects (such as damage, etc.) are not likely to be generated on the object to be ground. In addition, the case where the Asker-C hardness satisfies the above-mentioned range refers to the case where two non-overlapping areas (N=2) are measured in one polishing cloth, and the above-mentioned range is satisfied in either area.

為5.08 mm、高度為2.54 mm，例如Asker-C硬度為90之情形時，測定範圍為

2.2 mm之圓。因此，僅就Asker-C硬度而言，無法反映以約100～200 μm之間隔產生之細微之硬度偏差。 Asker-C hardness is the hardness measured by the rubber hardness meter adopted in JIS K 7312, and it is usually used as an index of hardness in abrasive cloths. Asker-C hardness is measured on one surface of the abrasive cloth, that is, the abrasive surface. As its indenter shape, in

is 5.08 mm and the height is 2.54 mm, for example, when the Asker-C hardness is 90, the measurement range is

2.2 mm circle. Therefore, only the Asker-C hardness cannot reflect the subtle hardness deviations that occur at intervals of about 100-200 μm.

The inventors of the present invention have found through painstaking research that the average value of the longitudinal abundance ratio obtained through image analysis in the thickness direction of the abrasive cloth is directly proportional to the hardness. Furthermore, based on this finding, it was concluded that by narrowing the measurement range to a width of 100 μm and setting the aspect ratio to a specific range, it is possible to suppress slight variations in hardness.

That is, in the polishing cloth of this embodiment, the interquartile range of the average value of the vertical abundance ratio of the 100 μm width of the forming material is 10.5 or less. The interquartile range of the average value of the above-mentioned longitudinal abundance ratio is preferably at most 10.0, more preferably at most 9.0. In addition, in this polishing cloth, the interquartile range of the average value of the vertical abundance ratio of the 100 μm width of the forming material may be 7.2 or more. In addition, the case where the interquartile range of the average value of the above-mentioned longitudinal abundance ratio satisfies the above-mentioned range means that when two non-overlapping regions (N=2) are measured in one polishing cloth, any region satisfies the Situations within the above scope.

The interquartile range of the average value of the vertical abundance ratio of the 100 μm width of the forming material can be obtained as follows.

First, for a thin section (cross-section) perpendicular to the surface of the abrasive cloth, the measurement area is photographed with a field of view of 2000 μm×2000 μm, and multiple small areas of 100 μm×100 μm are extracted. For example, if it is a polishing cloth with a thickness of 1.3 mm, in the measurement area of 1200 μm (the direction parallel to the surface of the polishing cloth) × 1200 μm (the direction perpendicular to the surface of the polishing cloth, that is, the thickness direction of the polishing cloth), extract 144 A small area of 100 μm×100 μm. In this case, the measurement area in the thickness direction depends on the thickness of the polishing cloth.

And, the value obtained by arithmetically averaging the abundance ratios of the wale lines arranged from the front side to the back side of the polishing cloth was taken as "the average value of the longitudinal abundance ratios in a width of 100 μm" (amount of one line). Furthermore, in this measurement, 17 photographs were taken at 100 μm intervals of the cross-section of the polishing cloth in one measurement, and the average value of the vertical abundance ratio of 100 μm width in 12 lines×17 pictures=204 lines was obtained And calculate the interquartile range, and measure two non-overlapping regions (N=2) in one polishing cloth.

Furthermore, the abundance ratio of the forming material in each small region of 100 μm×100 μm refers to the presence of the forming material in each small region of 100 μm×100 μm when the area of each small region is taken as 100% as a whole The area ratio of the part.

In the above measurement, the polishing cloth was photographed by CT-scan. Specifically, for a thin section (cross-section) perpendicular to the surface of the abrasive cloth, two measurement areas were photographed with a field of view of 2000 μm × 2000 μm for each 100 μm (the two measurement areas do not overlap). In the measurement area, multiple small areas of 100 μm × 100 μm are extracted, and in the image of each small area of 100 μm × 100 μm, it is classified into voids and parts other than voids (parts with forming materials) Value processing is used to determine the abundance ratio (area ratio).

As the CT apparatus, a three-dimensional measurement X-ray CT (Computed Tomography, computer tomography) apparatus (TDM1000H-1) manufactured by Daiwa Scientific Co., Ltd. can be used. In addition, as CT image processing software, image processing software VGStudio Max 2.1 manufactured by Visual Science Volume Graphics Co., Ltd. of Japan can be used. Furthermore, Image J (Rasband, W.S., U.S. National Institutes of Health, Bethesda, Maryland, USA) can be used as image analysis software for calculating the abundance ratio (area ratio) of the forming materials.

For example, the abundance ratio (area ratio) of the above-mentioned forming materials is measured under the following conditions.

In the above measurement, the measurement area of the polishing cloth was continuously measured with the following field of view. The size of the field of view (vertical × horizontal × height): 2,000 μm × 2,000 μm × the entire area in the thickness direction

In addition, the above measurement conditions are as follows. Number of views per rotation: 1500 Frames/View: 10 X-ray tube voltage [KV]: 28.000 Expanding axis position [mm]: 7.416 Reconstructed pixel size X[mm]: 0.003880 Reconstructed pixel size Y[mm]: 0.003880 Reconstructed pixel size Z[mm]: 0.003880

In each measurement area, the binarization process of classification into voids and portions other than voids (portions where forming materials exist) is as follows.

In the binarization process, in order to classify into voids and parts other than voids (the part where the forming material exists), the contrast of the image of the measurement area is adjusted by the above-mentioned image processing software VGStudio Max. Contrast adjustment is performed in Ramp mode.

In the adjustment of the contrast, the difference between the void and the portion other than the void (the portion where the forming material exists) is clarified.

In VGStudio Max, the contrast adjustment is described as "opacity adjustment". Specifically, in the opacity adjustment screen of VGStudio Max, set the lower limit of the gray value as the peak, and then set the upper limit of the gray value in the range of "the peak value of the peak + 100±5". Furthermore, since the transmittance of light varies from material to material, the adjustment range of contrast is not limited thereto.

For the 2D image after the above-mentioned contrast adjustment, an image of the measurement area, that is, a thin cut section is obtained. Furthermore, since the viewing angle must be set larger for the measurement area, when saving the cross-section as an image, set the position so that the surface of the polishing cloth contacts the top of the image.

Next, with respect to the image of the measurement area obtained above, the abundance ratio of the forming material is measured by the above image processing software Image J.

Here, the measurement range in Image J is 1200 μm×thickness of abrasive cloth. For example, if it is a grinding cloth with a thickness of 1.0 mm, set the upper left of the image data as (x, y) = (0 pix, 0 pix) position, and extract 312 pix (= 1201.2 μm) and 234 pix (=900.9 μm) in the longitudinal direction as the measurement range. Then, convert the image type from RGB color to 8 bit, and binarize the image. Under these binarization conditions, the grayscale range of "129" to "255" corresponds to the portion where the forming material exists. The image size of this image data is reduced to 12 pix in width x 9 pix in length. 1 pix×1 pix of the reduced image becomes a surface of 100 μm×100 μm. This operation was performed for 17 images in the vertical direction of the polished surface for each 100 μm. In addition, in the binarization process in Image J, the part whose gradation range is "129" - "255" is a part except a void (the part where a formation material exists).

The compressibility of the abrasive cloth of this embodiment is preferably 5% or less, more preferably 3.5% or less. In addition, the case where the compressibility satisfies the above-mentioned range refers to the case where two non-overlapping regions (N=2) of one polishing cloth are measured, and any region satisfies the above-mentioned range.

The compressibility can be calculated|required by the following method. That is, it is possible to use the compression elasticity testing machine (indenter area: 50 mm ² ) described in JIS L1096:2010, apply a pressure of 300 gf/cm ² to the abrasive cloth in the thickness direction by the indenter, and hold it for 60 seconds. The thickness T1 of the abrasive cloth is measured. Next, the pressure of 1800 gf/ ^cm2 is applied to the abrasive cloth in the thickness direction by the pressure head and the thickness T2 of the abrasive cloth is measured after holding for 60 seconds. out the compression ratio. Compression ratio = (T1-T2) × 100/T1

The thickness of the polishing cloth of this embodiment is preferably not less than 0.8 mm and not more than 3.0 mm, more preferably not less than 1.0 mm and not more than 2.0 mm. By setting the thickness of the above-mentioned polishing cloth to 0.8 mm or more, there is an advantage that it is easy to alleviate the adverse effect on the polishing performance due to the state of the pressure plate of the polishing machine. In addition, there is also an advantage that, for example, the object to be polished can be easily flattened stably. On the other hand, by making the thickness of the above-mentioned polishing cloth 3.0 mm or less, the amount of deformation of the polishing cloth during polishing can be reduced, and as a result, there is an advantage that edge sagging of the object to be polished is less likely to occur. In addition, the case where the thickness satisfies the above-mentioned range refers to the case where the above-mentioned range is satisfied in any of the two non-overlapping regions (N=2) of one polishing cloth measured.

The polishing cloth of this embodiment preferably has a compressibility of 5% or less and a thickness of 0.8 mm to 3.0 mm. With this configuration, it is possible to further suppress minute variations in hardness.

In the polishing cloth of this embodiment, the apparent density of the forming material is preferably from 0.30 g/cm ³ to 0.50 g/cm ³ , more preferably from 0.40 g/cm ³ to 0.50 g/cm ³ . When the apparent density of the forming material is not less than 0.30 g/cm ³ and not more than 0.50 g/cm ³ , fine variations in hardness can be further suppressed. In addition, apparent density can be measured based on JISK7222:2005. In addition, the case where the apparent density satisfies the above-mentioned range refers to the case where the above-mentioned range is satisfied in any of the two non-overlapping regions (N=2) of one polishing cloth when measured.

As fiber which comprises the said nonwoven fabric, a polyester fiber, a nylon fiber, etc. are mentioned, for example.

The weight per unit area of the nonwoven fabric is preferably not less than 200 g/m ² and not more than 600 g/m ² . By setting the weight per unit area of the nonwoven fabric above 200 g/m ² , the hardness tends to be increased, and as a result, there is an advantage that edge sagging is less likely to occur on the object to be polished. In addition, since the weight per unit area of the above-mentioned nonwoven fabric is ²⁰⁰ g/m2 or more and 600 g/ ^m2 or less, it is easy to have voids at an appropriate ratio on the grinding surface. If the voids are clogged and the grinding performance changes.

As said resin, a polyurethane resin etc. are mentioned, for example.

A silicon wafer etc. can be mentioned as a to-be-polished object polished with the polishing cloth of this embodiment.

The polishing cloth of this embodiment is comprised as mentioned above, Next, the manufacturing method of the polishing cloth of this embodiment is demonstrated.

Hereinafter, regarding the manufacturing method of the abrasive cloth of this embodiment, a method of wet impregnating nonwoven fabric with polyurethane resin and then dry impregnating nonwoven fabric with polyurethane resin will be described as an example.

In wet impregnation, a polyurethane resin is dissolved in a water-soluble organic solvent to obtain a first impregnating liquid. Examples of the water-soluble organic solvent include dimethylformamide, dimethylsulfoxide, tetrahydrofuran, dimethylacetamide, and the like. Furthermore, the first immersion liquid may contain a filler. Carbon black etc. can be mentioned as this filler. In addition, the first immersion liquid may contain a dispersion stabilizer. Surfactants etc. are mentioned as this dispersion stabilizer.

Next, the nonwoven fabric is dipped in the first dipping liquid, and the nonwoven fabric dipped in the first dipping liquid is dipped in water. Thereby, the water-soluble organic solvent adhered to the first dipping liquid of the nonwoven fabric is replaced with water, the polyurethane resin is solidified, and the polyurethane resin adheres to the surface of the nonwoven fabric.

In dry impregnation, a prepolymer having an isocyanate group as a terminal group, a curing agent as an organic compound having active hydrogen, and an organic solvent are mixed to obtain a second impregnating liquid. As said organic solvent, methyl ethyl ketone, acetone, alcohol, ethyl acetate etc. can be mentioned.

Then, the nonwoven fabric after wet impregnation is immersed in the 2nd immersion liquid, and the nonwoven fabric after immersion in the 2nd immersion liquid is heated by a drying oven. Thereby, the organic solvent evaporates, and the prepolymer and the curing agent undergo a hardening reaction to form a polyurethane resin. As a result, the polyurethane resin further adheres to the surface of the nonwoven fabric.

Since the polishing cloth of this embodiment is comprised as mentioned above, it has the following advantages.

That is, the polishing cloth of the present embodiment is a polishing cloth having a nonwoven fabric and a resin impregnated into the nonwoven fabric as a forming material for forming the polishing cloth, the Asker-C hardness is 80 or more, and the aspect ratio of the 100 μm width of the forming material is The interquartile range of the mean is below 10.5.

Since the Asker-C hardness of the abrasive cloth is 80 or more, and the interquartile range of the average value of the longitudinal abundance ratio of the 100 μm width of the forming material is 10.5 or less, unevenness caused by longitudinal intertwining of the nonwoven fabric can be reduced. As a result, it is possible to suppress slight variations in hardness that cannot be reflected only by Asker-C hardness.

Furthermore, the polishing cloth of the present invention is not limited to the above-mentioned embodiments. In addition, the polishing cloth of the present invention is not limited to the above-mentioned effects. Various modifications can be made to the polishing cloth of the present invention without departing from the scope of the present invention. [Example]

Next, the present invention will be further described by way of examples and comparative examples.

The abrasive cloths of the examples whose physical properties are shown in Table 1 and 2 were produced. Moreover, the polishing cloth (commercially available) of the comparative example whose physical property is shown in Table 1, 2 was prepared. Furthermore, the Asker-C hardness, the compressibility, the apparent density of the forming material, the average value of the aspect ratio of the 100 μm width of the forming material, and the interquartile range of the average value were measured by the above method.

The flatness of the polishing cloth with a thickness of 1.2 to 1.3 mm and the polishing cloth with a thickness of 0.9 to 1.1 mm was evaluated by changing the conditions as follows.

Abrasive cloth with a thickness of 1.2-1.3 mm: Comparative Example 1, Examples 1-3 GBIR (Global Backsurface-referenced Ideal plane/Range, overall back surface-reference ideal plane/range) is measured according to the shape of the wafer (wafer) when the wafer is polished with a polishing cloth. Grinding is performed after setting the grinding cloth on the double-sided grinding machine and before grinding, and then performing grinding. Furthermore, the GBIR and ESFQR (Edge Site Front Surface-referenced least sQuares/Range) of the wafer adopt the value of GBIR/ESFQR that optimizes the gap between 0 and 3 μm. The results are shown in Table 1 and FIG. 1 .

In addition, the polishing conditions are as follows. Grinder: Speed FAM 20B Dresser: Speed FAM Genuine Dresser ♯100/♯120 Gap setting: 0～3 μm Grinding load: 1500 kg Number of revolutions: 10 rpm Wafer: 12 inch P-wafer Grinding liquid: a solution with a solid matter content (115°C) of 36.6%, an average particle size of 108 nm, and a pH of 11.3.

Polishing pads with a thickness of 0.9-1.1 mm: Comparative Example 2, Examples 4-5 Grinding is performed after the grinding is arranged on the single-sided machine and before grinding, and then the grinding is performed. However, in a single-sided machine, since it is greatly affected by the shape before grinding, the difference between the shape before grinding and the shape after grinding is analyzed, and the flatness is evaluated in the form of the difference GBIR. Also, the difference GBIR is a parameter that is very related to the polishing rate, so the shape comparison is performed in the form of difference GBIR/polishing rate. The results are shown in Table 2 and FIG. 2 .

In addition, the polishing conditions are as follows. Grinding Machine: Poli762 Dresser: Kinik Dresser♯150 Grinding Pressure: 300 gf/cm ² Head/Platen: 40/43 rpm Wafer: 8''(P-) Polishing Fluid: NP6610 (Nitta・Dupont Corporation system) Add 7.14% to DIW (pure water) Flow rate of used polishing liquid: 600 mL/min

[Table 1] Table 1 Comparative example 1 Example 1 Example 2 Example 3 N=1 N=2 N=1 N=2 N=1 N=2 N=1 N=2 physical properties Thickness (mm) 1.27 1.27 1.26 1.26 1.26 1.26 1.27 1.27 Asker-C hardness 88 88 85 85 86 86 88 88 Compression ratio(%) 2.0 2.0 2.0 2.3 2.3 2.0 2.1 1.9 Apparent density (g/cm ³ ) 0.45 0.46 0.44 0.44 0.45 0.45 0.45 0.45 Abundance ratio of 100 μm width average value(%) 36.1 38.0 35.5 35.5 36.9 36.7 36.5 36.1 Above interquartile range (%) 11.1 11.4 8.3 8.9 8.5 9.3 7.2 8.4 Evaluation GBIR: Improvement rate from Comparative Example 1 (%) 100 69 74 68 ESFQR: Improvement rate from Comparative Example 1 (%) 100 59 59 63

[Table 2] Table 2 Comparative example 2 Example 4 Example 5 N=1 N=2 N=1 N=2 N=1 N=2 physical properties Thickness (mm) 1.02 1.01 1.02 1.02 0.99 1.00 Asker-C hardness 84 83 81 82 83 84 Compression ratio(%) 3.1 3.4 3.5 3.3 3.3 3.0 Apparent density (g/cm ³ ) 0.46 0.46 0.44 0.46 0.45 0.45 Abundance ratio of 100 μm width average value(%) 35.2 36.3 37.7 34.7 36.4 36.9 Above interquartile range (%) 13.6 16.8 10.2 10.4 10.2 10.5 Evaluation Difference GBIR/grinding rate: improvement ratio (%) from Comparative Example 1 100 85 80

As shown in Tables 1 and 2 and FIGS. 1 and 2, the polishing cloths of the examples satisfying all the constituent requirements of the present invention can suppress slight variations in hardness, and as a result, the flatness of the wafer is good. [Cross-reference of related applications]

This case claims the priority of Japanese Patent Application No. 2020-215154, which is incorporated by reference into the description of this case.

1 is a graph showing the interquartile range of the average value of the vertical abundance ratio of the 100 μm width of the forming material and the flatness of the object to be polished in the polishing cloths of Comparative Example 1 and Examples 1 to 3. 2 is a graph showing the interquartile range of the average value of the vertical abundance ratio of the 100 μm width of the forming material and the flatness of the object to be polished in the polishing cloths of Comparative Example 2 and Examples 4 to 5.

Claims (3)

An abrasive cloth comprising a non-woven fabric and a resin impregnated into the non-woven fabric as a forming material for forming the abrasive cloth, Asker-C hardness is above 80, The interquartile range of the mean value of the vertical abundance ratio of the 100 μm width of the formation material is 10.5 or less. The abrasive cloth of claim 1 has a compressibility of 5% or less, and a thickness of 0.8 mm or more and 3.0 mm or less. The polishing cloth according to claim 1 or 2, wherein the apparent density of the above-mentioned forming material is not less than 0.30 g/cm ³ and not more than 0.50 g/cm ³ .

Images