METHODS OF CLEANING CMP POLISHING PADS

Abstract

The present invention provides methods for cleaning the surface of CMP polishing pads comprising blowing a stream or curtain of forced air or gas from a source onto the surface of a CMP polishing pad substrate at a pressure of from 170 kPa (24.66 psig) to 600 kPa (87 psig), towards a vacuum source, the forced air or gas blowing at an angle of from 6 to 15? from a vertical plane which lies normal to the surface of the substrate, traverses the entire width of the surface of the substrate, and passes through the source of the forced air or gas, while, at the same time conveying along a horizontal plane the CMP polishing pad so that the entire surface of the CMP polishing pad surface is exposed to the forced air or gas at least one time; and, vacuuming the surface of the CMP polishing pad at a point on the surface which is downstream from a point at which the stream curtain of forced air or gas contacts the surface of the CMP polishing pad.

Claims

1. A method for cleaning a surface of CMP polishing pads comprising: blowing blowing a stream or curtain of forced air or gas from a source onto the surface of a CMP polishing pad substrate at a pressure of from 170 kPa (24.66 psig) to 600 kPa (87 psig), towards a vacuum source, the forced air or gas blowing at an angle of from 6 to 15? from a vertical plane which lies normal to the surface of the CMP polishing pad substrate, traverses the entire width of the surface of the CMP polishing pad substrate, and passes through the source of the forced air or gas while, at the same time conveying along a horizontal plane the CMP polishing pad substrate horizontally disposed on a flat platen so that the entire surface of the CMP polishing pad substrate is exposed to the forced air or gas at least one time; and, vacuuming the surface of the CMP polishing pad at a point on the surface which is downstream from a point at which the stream or curtain of forced air or gas contacts the surface of the CMP polishing pad substrate.

2. The method as claimed in claim 1, wherein the blowing forced air or gas comprises blowing at an angle of from 8 to 12.5? from a vertical plane which lies normal to the surface of the CMP polishing pad substrate, traverses the entire width of the surface of the CMP polishing pad substrate and passes through the source of the forced air or gas.

3. The method as claimed in claim 1, wherein the source of forced air or gas is located 20 mm or less, from the surface of the CMP polishing pad substrate as it is conveyed through the source of the forced air or gas, and wherein the stream or curtain of forced air or gas comprises a curtain that traverses the entire width of the surface of the CMP polishing pad substrate as the CMP polishing pad substrate is conveyed through the curtain or stream of forced air or gas.

4. The method as claimed in claim 1, wherein the conveying of the CMP polishing pad substrate along the horizontal plane comprises moving the CMP polishing pad substrate disposed on the flat platen along a track or conveyor so that the entire surface of the CMP polishing pad substrate is exposed to the forced air or gas at least twice, in a back and forth fashion, during the blowing of the stream or curtain of forced air or gas.

5. The method as claimed in claim 1, wherein the flat platen comprises a vacuum platen to hold the CMP polishing pad substrate in place.

6. The method as claimed in claim 1, wherein the vacuuming comprises applying vacuum from a vacuum source disposed parallel to the curtain of forced air or gas which traverses the entire width of the surface of the CMP polishing pad substrate and located less than 20 mm from the CMP polishing pad substrate surface as it is conveyed past the vacuum source.

7. The method as claimed in claim 1, wherein the vacuuming comprises applying vacuum continuously during the blowing of the stream of forced air or gas.

8. The method as claimed in claim 1, further comprising brushing the surface of the CMP polishing pad substrate at a point downstream of the point at which the CMP polishing pad substrate is vacuumed, while at the same time blowing the stream or curtain of forced air or gas onto the CMP polishing pad substrate and vacuuming, wherein the brushing comprises continuously contacting a brush element with the surface of the CMP polishing pad substrate during the conveying, the vacuuming and the blowing.

9. The method as claimed in claim 8, wherein the brush element traverses the entire width of the surface of the CMP polishing pad substrate, is disposed parallel to each of the curtain of forced air or gas and the vacuum source, and contacts the CMP polishing pad substrate downstream of the vacuum source.

10. The method as claimed in claim 1, wherein in the conveying along the horizontal plane, the CMP polishing pad substrate is disposed surface side up or surface side down, and, further wherein, when the CMP polishing pad substrate is disposed surface side down all of the blowing of forced air or gas, the vacuuming source, the brushing are directed up to the CMP polishing pad substrate so that the brush element contacts the surface of the CMP polishing pad substrate and each of the source of forced air or gas, and the vacuum source, is disposed a distance of less than 20 mm below the surface of the CMP polishing pad substrate.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0027] FIG. 1 depicts an apparatus for use in accordance with the methods of the present invention with the CMP polishing layer or pad and shows an example of a flat bed platen or flat platen, the track or conveyor and the forced air, vacuum, brush element and static dissipation bar useful in the methods of the present invention.

[0028] FIG. 2 depicts a cut away view of a forced air bar, vacuum hood, brush element and static dissipation bar useful in the methods of the present invention.

[0029] In accordance with the present invention, methods of cleaning CMP polishing layers or pads enable the removal of contaminants from the surface of the CMP polishing layers or pads after they are manufactured and before they are conditioned for polishing. The present inventors have surprisingly found that blowing a stream or curtain of forced air or gas at a high pressure of at least 276 kPa or, preferably, at least 360 kPa from a short distance of 30 mm or less onto the surface of a CMP polishing layer or pad and against a brush element that acts as a dam to trap debris and particles which have been freed from the pad surface will remove as much as eighty (80%) of such debris and particles found on such pads after manufacture. A vacuum source disposed between the stream or curtain of forced air or gas and the brush element effectively removes the trapped debris and particles. Further, treating the CMP polishing layer substrate with a static dissipation element prior to blowing the forced air or gas onto the substrate enables such a high debris and particles removal rate.

[0030] The methods of the present invention are scalable to fit CMP polishing layers of various sizes, as the size of any of the stream or curtain of forced air or gas, the vacuum source, the brush element and/or the static dissipation bar can be varied. In accordance with the methods of the present invention, the flat platen should be larger than the CMP polishing layer or, preferably, of a size having a radius that is equal to or within 10 cm longer than the radius of the CMP polishing layer. The methods thus are scalable to treat CMP polishing layers having a radius of from 100 mm to 610 mm.

[0031] The methods of the present invention are conducted in a dry environment and can be conducted in an air tight or climate controlled chamber wherein no additional contaminants are present aside from the debris and particles located in or on the surface of the CMP polishing layer.

[0032] The methods of the present invention enable the provision of CMP polishing layers or pads useful in back end CMP polishing. Suitable pads have a compressibility, as defined above, of from 10 to 30%.

[0033] Suitable CMP polishing layers for use in accordance with the methods of the present invention preferably comprise a porous polymer or filler containing porous polymer material such as a porous polyurethane. As used herein, the term porous polymer refers to polymers having pores within them; as used herein, the term poromeric is refers to a polymer matrix having pores within the polymer.

[0034] The methods of the present invention can be carried out on any pad, including those made from soft polymers, such as polyurethanes and find particular use in treating soft pads having a compressibility of from 10 to 30%. Pores can be provided by spaces in the pad polymer matrix.

[0035] The methods of the present invention can be performed on single layer or solo pads, as well as on stacked pads having a subpad layer.

[0036] As shown in FIG. 1, the methods of the present invention are carried out on the surface of a flat platen (10) with vacuum ports, not shown. In FIG. 1, the flat platen (10) carries the CMP polishing layer substrate under, moving from left to right, a static dissipation bar (12), a brush element (14), a vacuum hood (16) and an air bar (20). The various items are arranged so that air bar (20) blows forced air at an slight angle down onto a CMP polishing layer substrate, with the slight angle leaning toward vacuum hood (16) and brush element (14). In FIG. 1, as flat platen (10) carrying a CMP polishing layer substrate is conveyed along the track (18), the static dissipation bar acts on the substrate before it reaches any forced air or gas curtain.

[0037] As shown in FIG. 2, in the methods of the present invention the CMP polishing layer substrate is acted upon, in order, as it moves from left to right, a static dissipation bar (12), a brush element (14), a vacuum hood (16) and an air bar (20).

[0038] In the apparatus of the present invention, each of the source of forced air or gas, the vacuum source, the brush element and the static dissipation bar are mounted on the same bracket and can be raised and lowered in unison, such as via a mechanical actuator such as a ball screw, or an electric servo motor mechanically linked to a gear that raises and lowers the bracket. Preferably, the brush element has an additional finely threaded ball screw so that it can be independently raised and lowered at least a total distance of 30 mm.

[0039] Preferably, in the methods of the present invention the CMP polishing layer substrate is conveyed past all of the static dissipation bar, the brush element, the vacuum source and the source of forced air or gas once so that the whole surface is treated. In FIG. 1, this conveying consists of moving the CMP polishing layer substrate on a platen from left to right so that the whole substrate passes under the source of forced air or gas.

[0040] More preferably, in the methods of the present invention, the CMP polishing layer substrate is conveyed past all of the static dissipation bar, the brush element, the vacuum source and the source of forced air or gas twice so that each of two times the whole surface is treated. In FIG. 1, this conveying consists of moving the CMP polishing layer substrate on a platen from left to right all the way past the source of forced air or gas, and then moving it from right to left back to its starting point.

[0041] A suitable apparatus useful in the methods of the present invention is a Neutro-Vac? tool (Simco-Ion, Hatfield, Pa.), which can come in a customized width.

[0042] In the methods of the present invention, the composition of forced air or gas is not limited except that it must be inert. Suitable gases include air, carbon dioxide or helium.

[0043] The stream or curtain of forced air or gas in accordance with the present invention can comprise a curtain flowing from an air bar or other linear source of air having a plurality of forced air or gas outlet opens disposed all along its length. Preferably, the stream or curtain of forced air or gas flows from a source wherein at each point along the source forced air or gas travels a path that has one and the same length before reaching the substrate. Such a source of forced air or gas can be any that are disposed parallel to the surface of the CMP polishing layer substrate and that run at least the width of the CMP polishing layer or pad.

[0044] The stream or curtain of forced air or gas could fan out from a single point to form a fan as wide as the CMP polishing layer substrate; however, such a fan will provide less force in proportion to the distance of the substrate from the fan source. The flat bed platen in the apparatus of the present invention contains a plurality of small holes, for example, from 0.5 to 5 mm in diameter, through the platen which are connected to a vacuum. The holes can be arranged in any suitable manner to hold the CMP polishing layer substrate in place during grinding, such as, for example, along a series of spokes extending outward from the center point of the flat platen or in a series of concentric rings.

[0045] The vacuum source used in the methods of the present invention is connected to a vacuum pump, whereby debris and particles can be removed from the CMP polishing layer substrate.

[0046] The vacuum from the vacuum source can be provided at a pressure of from 0.01 bar (1 kPa) to 0.5 bar (50.5 kPa) or, preferably, from 0.03 bar (3 kPa) to 0.2 bar (20.2 kPa).

[0047] The vacuum provided by the flat platen can be provided at the same pressure as the vacuum from the vacuum source.

[0048] The brush element used in the methods of the present invention can be any inert plastic, for example, polyamide, hard rubber or natural, for example, horse hair brush material that effectively blocks the flow of debris and particles loosened by the stream or curtain of forced air or gas. In the methods of the present invention, the brush element is at least in contact with the surface of the CMP polishing layer substrate.

[0049] The static dissipation bar used in the methods of the present invention can comprise an electrically powered source of ionized particles or charges, such as tungsten emitters, directed at the CMP polishing layer substrate. The static dissipation bar is disposed a distance of less than 20 mm or, preferably, less than 10 mm from the surface of CMP polishing pad.

[0050] The static dissipation bar used in the methods of the present invention may touch the CMP polishing layer substrate surface in the methods of the present invention. In such a case, the static dissipation bar can comprise an antistatic material, such as, for example, conductive positively charged polymers like polyaniline or polyethylenimine; conductive materials, such as carbon black; antistatic material coated materials, such as indium tin oxide coated ceramics or inorganic oxide materials. The antistatic material can be in fibrous form, in a sheet form, or it can be a composite of particles molded in the form of a bar or strip.

EXAMPLES

[0051] In the following examples, unless otherwise stated, all units of pressure are standard pressure (?101 kPa) and all units of temperature are room temperature (21-23? C.).

[0052] The following test method was used in the Examples that follow:

[0053] Particle Count:

[0054] Particles were counted using monochromatic lighting in a 7.62 cm?7.62 cm (3?3) area of the given pad substrate. The area with the highest and lowest particle count was chosen and an average value was calculated before cleaning the pad and after cleaning the pad to determine % of particles removed.

Example 1

[0055] Experiments were conducted using 50.8 cm (20) diameter, and 1.524 mm (60 mil) thick Politex? porous polyurethane soft pad having a weight density of 0.286 g/cm.sup.3 and a compressibility of 15% (The Dow Chemical Co., Midland, Mich. (Dow)). In the methods of the Examples, a static bar was used to neutralize the charge of the material and aid with dislodging particles from the pad surface. An air knife was used to blow compressed air onto the surface of the CMP polishing layer substrate to dislodge particles. The air knife was set at an angle of about 6? to a vertical plane which lies normal to the surface of the substrate and passing through the source of the forced air in the air knife. For Comparative pads 1-4, 9-12 and 17-20, the (compressed) air pressure was set at 48.26 kPA (7 psi); and, for inventive pads 5-8, 13-16, and 21-26, the air pressure was set at 413.69 kPA (60 psi). No brush was used. The pads were conveyed so that they passed twice, once forth and once back under the forced air and vacuum sources at a rate of about 1.1 m/min.

[0056] A vacuum source was set to draw debris and particles from the indicated pad substrates at an average velocity of 19.8 m/s (3902 fpm). The vacuum source was set at a distance from the flat platen that varied from 0.508 to 1.016 cm (0.2 to 0.4). The results are shown in Table 1 below.

Example 1b

[0057] Experiments were conducted using 50.8 cm (20) diameter, and 1.524 mm (60 Mil) thick Politex? porous polyurethane soft pad having a weight density of 0.286 g/cm.sup.3 and a compressibility of 15% (The Dow Chemical Co., Midland, Mich. (Dow)). In the methods of the Examples, a static bar was used to neutralize the charge of the material and aid with dislodging particles from the pad surface. An air knife was used to blow compressed air at a force of, for pads 1 (comparative), 4 (comparative), 5, 8, 9, 10, 11, and 12 172.37 kPa (25 psig), and, for Comparative pads 2, 3, 6, and 7 34.37 kPa (5 psig) onto the surface of the CMP polishing layer substrate to dislodge particles.

[0058] The air knife was set at a given angle of from 5 to 30? from a vertical plane which lies normal to the surface of the substrate and passing through the source of the forced air in the air knife; for Comparative pads 1-2, the air knife was set at an angle of about 25? from the vertical plane; for Comparative pads 3-4, the air knife was set at an angle of about 20? from the vertical plane; for pads 5-6, the air knife was set at an angle of about 10? from the vertical plane; and, for pads 7-12 an angle of 6? from the vertical plane. Dislodged particles were captured using a vacuum source set to draw debris and particles from the indicated pad substrates at an average velocity of 19.8 m/s (3902 fpm). No brush was used. The pads were conveyed so that they passed twice, once forth and once back under the forced air and vacuum sources at a rate of about 1.1 m/min.

[0059] A vacuum source was set to draw debris and particles from the indicated pad substrates at an average velocity of 19.8 m/s (3902 fpm). The vacuum nozzle distance from the flat platen was 9.5 mm. The results are shown in Table 1 b, below.

TABLE-US-00001 TABLE 1 Vacuum Nozzle Distance Trial After After After Initial Initial Initial Vacuum Cleaning Cleaning Cleaning High Low Average Nozzle High Low Average Pad Particle Particle Particle Dist Particle Particle Particle Removal # Count Count Count (mm) Count Count Count % Vacuum Nozzle Distance Test *1 75 17 46 5.08 33 3 18 61% *2 43 5 24 5.08 38 4 21 13% *3 72 16 44 5.08 61 10 35.5 19% *4 69 12 40.5 5.08 53 6 29.5 27% 5 89 4 46.5 5.08 31 3 17 63% 6 60 11 35.5 5.08 9 1 5 86% 7 82 5 43.5 5.08 15 0 7.5 83% 8 119 11 65 5.08 15 1 8 88% Average 55% *9 104 15 59.5 6.35 82 7 44.5 25% *10 36 10 23 6.35 45 7 26 ?13% *11 72 18 45 6.35 49 9 29 36% *12 65 6 35.5 6.35 56 4 30 15% 13 95 10 52.5 6.35 16 2 9 83% 14 167 22 94.5 6.35 31 3 17 82% 15 38 11 24.5 6.35 8 1 4.5 82% 16 58 4 31 6.35 21 0 10.5 66% Average 52% *17 108 11 59.5 10.16 68 10 39 34% *18 148 21 84.5 10.16 104 19 61.5 27% *19 137 40 88.5 10.16 103 25 64 28% *20 242 34 138 10.16 149 34 91.5 34% 21 43 5 24 10.16 28 9 18.5 23% 22 123 18 70.5 10.16 17 3 10 86% 23 107 11 59 10.16 28 1 14.5 75% 24 42 12 27 10.16 11 3 7 74% 25 37 5 21 6.35 15 1 8 62% 26 92 9 50.5 6.35 14 2 8 84% Average 48% *Denotes Comparative Example.

[0060] As shown in Table 1, above the pads cleaned using a forced air pressure within the inventive range gave dramatically better particle removal. The only exception was in Example 21 where the pad itself had a very low number of particles or impurities to begin with. Further, the absence of a brush impaired control of the methods so the results varied more than with the brush. Compare Table 2, below.

TABLE-US-00002 TABLE 1b Air Knife Angle Testing After After After Initial Initial Initial Vacuum Cleaning Cleaning Cleaning High Low Average Slot High Low Average Pad Particle Particle Particle Distance Particle Particle Particle Removal # Count Count Count (mm) Count Count Count % *1 57 6 31.5 9.5 43 5 24 24% *2 109 10 59.5 9.5 145 7 76 ?28% *3 121 8 64.5 9.5 143 5 74 ?15% *4 175 8 91.5 9.5 82 1 41.5 55% 5 72 14 43 9.5 28 3 15.5 64% *6 77 27 52 9.5 39 8 23.5 55% *7 127 9 68 9.5 124 6 65 4% 8 38 5 21.5 9.5 13 0 6.5 70% 9 95 15 55 9.5 42 4 23 58% 10 51 10 30.5 9.5 56 2 29 5% 11 153 7 80 9.5 69 4 36.5 54% 12 99 19 59 9.5 62 8 35 41%

[0061] As shown in Table 1 b, above, in the absence of a brush element, the pad cleaning methods of the present invention are not nearly as effective as they are with the brush. Compare Table 2, below. This is surprising because the brush itself only traps particles for vacuum removal and does not itself remove the particles from the pad. Inventive Example 10 shows that the methods of the present invention lack preferred consistency without use of a brush; although the pad of Example 10 had a very low initial average count. Compare Examples 8, 9, 11 and 12.

Example 2

[0062] Example 1 was repeated except that a brush was installed adjacent to the vacuum nozzle downstream from the air knife which was used to blow compressed air at a pressure of 413.7 kPA (60 psig) onto the surface of the CMP polishing layer substrate to dislodge particles. The air knife was set at an angle of about 10? from a vertical plane which lies normal to the surface of the substrate and passing through the source of the forced air in the air knife. The brush bristles lightly contacted the pad. The pads were conveyed so that they passed twice, once forth and once back under the forced air and vacuum sources at a rate of about 1.1 m/min. The brush dislodged particles from the surface of the pad and directed them toward the vacuum nozzle.

TABLE-US-00003 TABLE 2 Brush Installed Testing In Table 2, below, pads 1-1 to 1-4 were all tested the same day and pads 1-2 to 10-2 were tested on the same day. After After After Initial Initial Initial Cleaning Cleaning Cleaning High Low Average Vacuum High Low Average Pad Particle Particle Particle Nozzle Particle Particle Particle Removal # Count Count Count Dist Count Count Count % 1-1 43 4 23.5 9.5 9 0 4.5 81% 2-1 137 13 75 9.5 21 0 10.5 86% 3-1 53 11 32 9.5 19 0 9.5 70% 4-1 77 12 44.5 9.5 14 0 7 84% 1-2 203 9 106 9.5 31 0 15.5 85% 2-2 174 15 94.5 9.5 32 2 17 82% 3-2 30 5 17.5 9.5 7 0 3.5 80% 4-2 88 16 52 9.5 12 0 6 88% 5-2 54 5 29.5 9.5 7 0 3.5 88% 6-2 207 17 112 9.5 56 0 28 75% 7-2 152 4 78 9.5 25 0 12.5 84% 8-2 201 11 106 9.5 53 0 26.5 75% 9-2 253 10 131.5 9.5 31 0 15.5 88% 10-2 39 4 21.5 9.5 6 0 3 86% Average 82%

[0063] As shown in Table 2, above, in methods of the present invention wherein the air knife was set at an inventive angle to a vertical plane which lies normal to the surface of the substrate and passing through the source of air, a static brush element was used in the inventive manner, and the forced air was blown at inventive pressures, the average amount of particles removed was 82%. This was a consistently excellent result.