Offset Pore Microcellular Polishing Pad

A porous polyurethane polishing pad with interconnected pores and spring arm sections addresses high defect rates and selectivity issues, enhancing polishing efficiency and reducing contamination in semiconductor manufacturing.

JP7715529B2Active Publication Date: 2025-07-30DUPONT ELECTRONIC MATERIALS HLDG INC
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
JP2021069491
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Priority Date
2020-04-18
Filing Date
2021-04-16
Publication Date
2025-07-30
Estimated Expiration
2041-04-16

AI Technical Summary

Technical Problem

Existing microporous polishing pads used in semiconductor manufacturing face challenges with high defect rates and insufficient selectivity in removing TEOS relative to copper, leading to contamination and degraded functional performance.

Method used

A porous polyurethane polishing pad with interconnected large, medium, and small pores, featuring spring arm sections that enhance compressibility and contact area, is designed to improve polishing efficiency and reduce defects.

Benefits of technology

The pad achieves a lower defect rate and increased polishing rate, with improved TEOS:Cu selectivity, reducing contamination and enhancing wafer yield.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007715529000018
    Figure 0007715529000018
  • Figure 0007715529000019
    Figure 0007715529000019
  • Figure 0007715529000020
    Figure 0007715529000020
Patent Text Reader

Abstract

To provide a microporous polishing pad capable of further reducing defects caused by a polishing pad and improving a polishing rate.SOLUTION: The invention provides a porous polyurethane polishing pad that includes a porous substrate having large pores that extend upward from a base surface and open to an upper surface. The large pores extend to a top polishing surface and have a lower section and an upper section having a vertical orientation. The lower and upper sections are offset in a horizontal direction. Middle-sized pores having a columnar shape and a vertical orientation occur adjacent middle sections connecting the lower and upper sections, and small pores having a columnar shape and a vertical orientation occur between the middle-sized pores. The pores are combined so as to increase compressibility of the polishing pad and contact area of the top polishing surface during polishing.SELECTED DRAWING: Figure 4
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to a chemical mechanical polishing pad and a method of forming a polishing pad. More specifically, the present invention relates to a microporous chemical mechanical polishing pad and a method of forming a microporous polishing pad.

Background Art

[0002] In the manufacture of integrated circuits and other electronic devices, multiple layers of conductive, semiconductor, and dielectric materials are deposited on and removed from the surface of a semiconductor wafer. Thin layers of conductive, semiconductor, and dielectric materials can be deposited using several deposition techniques. Common deposition techniques in current wafer processing include, among others, sputtering, chemical vapor deposition (CVD), plasma enhanced chemical vapor deposition (PECVD), and physical vapor deposition (PVD) also known as electrochemical plating. Common removal techniques include, among others, wet and dry isotropic and anisotropic etching.

[0003] When layers of material are sequentially deposited and removed, the top surface of the wafer becomes non-planar. In subsequent semiconductor processing (such as photolithography), the surface of the wafer needs to be flat, so the wafer needs to be planarized. Planarization helps to remove undesirable surface topography and surface defects (such as rough surfaces, aggregated materials, damage to the crystal lattice, scratches, contaminated layers or materials).

[0004] Chemical mechanical planarization, or chemical mechanical polishing (CMP), is a common technique used to planarize or polish a workpiece (e.g., a semiconductor wafer). In conventional CMP, a wafer carrier or polishing head was attached to a carrier assembly. The polishing head holds the wafer and positions the wafer to contact the polishing layer of a polishing pad attached to a table or platen within the CMP apparatus. The carrier assembly provides a controllable pressure between the wafer and the polishing pad. At the same time, a polishing medium (e.g., slurry) is dispensed onto the polishing pad and drawn into the gap between the wafer and the polishing layer. To perform the polishing, the polishing pad and the wafer typically rotate relative to each other. When the polishing pad rotates under the wafer, the wafer typically sweeps an annular polishing track or polishing region. Here, the surface of the wafer faces directly the polishing layer. The wafer surface is polished and planarized by the chemical and mechanical action of the polishing layer and polishing medium on the surface.

[0005] The CMP process typically occurs in two or three steps on a single polishing tool. The first step planarizes the wafer and removes most of the excess material. After planarization, the next step removes the scratches or chattering marks introduced during the planarization step. The polishing pads used for these applications must be flexible and isotropic so as to polish the substrate without scratching it. Further, these polishing pads and slurries for these steps often require selective removal of material (e.g., high TEOS to metal removal rate). For the purposes of this specification, TEOS is the decomposition product of tetraethyl orthosilicate. Since TEOS is a harder material than metals such as copper, this has been a difficult problem that manufacturers have been working on for years.

[0006] Over the last several years, semiconductor manufacturers have increasingly shifted to microporous polishing pads (e.g., Politex™ and Optivision™ polyurethane pads for finishing or final polishing operations where low defect rates are a more critical requirement) (Politex and Optivision are trademarks of DuPont de Nemours, Inc. or one or more of its affiliates). For the purposes of this specification, the term microporous refers to a porous polyurethane polishing pad produced by coagulation from an aqueous solution, a non-aqueous solution, or a combination of aqueous and non-aqueous solutions. The advantages of these polishing pads are that they have few defects and can be removed efficiently. This reduction in defects can potentially lead to a dramatic increase in wafer yield.

[0007] A particularly important polishing application is copper barrier polishing, which requires a low defect rate in combination with the ability to simultaneously remove both copper and TEOS dielectrics. As a result, the TEOS removal rate is higher than the copper removal rate to meet advanced wafer integration designs. Commercially available pads (e.g., Politex polishing pads) do not provide a low enough defect rate for future designs, and the TEOS:Cu selectivity is not high enough either. Another commercially available pad contains a surfactant that produces an excessive amount of foam during polishing, which leaches out and interferes with the polishing. Furthermore, the surfactant may contain alkali metals that can contaminate the dielectric and degrade the functional performance of the semiconductor.

[0008] Despite the low TEOS removal rate associated with microporous polishing pads, some advanced polishing applications are transitioning to all-microporous pad CMP polishing operations because it may be possible to achieve a lower defect rate with microporous pads compared to another pad type (e.g., IC1000™ polishing pads). These operations have few defects, but the challenges of further reducing pad-induced defects and increasing the polishing rate remain. SUMMARY OF THE INVENTION MEANS FOR SOLVING THE PROBLEM

[0009] One aspect of the present invention provides a porous polyurethane polishing pad comprising: a porous substrate having large pores extending upward from a base surface and opening to an upper surface, the large pores being interconnected with tertiary pores, a portion of the large pores opening to a top polishing surface, at least a portion of the large pores extending to the top polishing surface, having a lower section and an upper section with a vertical orientation, where vertical is orthogonal to the base surface Orthogonal to the polishing surface Towards in an orthogonal direction, and an intermediate section connecting the lower and upper sections, the lower and upper sections being horizontally offset, medium pores having a columnar shape and a vertical orientation being generated adjacent to the intermediate section, and small pores having a columnar shape and a vertical orientation being generated between the medium pores, wherein the large pores, medium pores, and small pores having horizontally offset upper and lower sections are combined to increase the compressibility of the polishing pad during polishing and the contact area of the top polishing surface.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 3A

Figure 3B

Figure 4

Figure 4A

Figure 5

DETAILED DESCRIPTION OF THE INVENTION

[0011] The polishing pad of the present invention is useful for polishing at least one of a magnetic substrate, an optical substrate, and a semiconductor substrate. In particular, a polyurethane pad is useful for polishing a semiconductor wafer. In particular, for this pad, a very low defect rate is important compared to the planarization ability, and for advanced applications (e.g., copper-barrier applications) where it is necessary to simultaneously remove multiple materials such as, for example, copper, barrier metals, and dielectric materials (including but not limited to TEOS, low-k and ultra-low-k dielectrics). For the purposes of this specification, "polyurethane" is a product derived from bifunctional or polyfunctional isocyanates (e.g., polyether ureas, polyisocyanurates, polyurethanes, polyureas, polyurethane ureas, their copolymers, and their mixtures).

[0012] The porous polyurethane polishing pad includes a porous substrate having large pores extending upward from a base surface and opening to an upper surface, i.e., a polishing surface. The large pores interconnect with the tertiary pores. It is possible for all pores to open at the top surface, but usually only a portion of the large pores open to the top polishing surface. At least a portion of the large pores extends to the top polishing surface and has a lower section and an upper section with a vertical orientation. As used herein, vertical refers to a direction perpendicular to the base surface and toward the upper surface. Usually, the average diameter of the lower section of the large pores is larger than the average diameter of the upper section of the large pores.

[0013] The spring arm sections connect the lower section and the upper section. All of the spring arm sections extend in the same horizontal direction as measured from the vertical direction. Although it is possible to bend the spring arms in multiple directions, usually when the web is pulled by a shearing force, spring arm sections that all extend in the same direction are created. As a result, the intermediate sections, i.e., the spring arm sections, usually have an average diameter smaller than the average diameter of the lower section of the large pores. In the case of long intermediate arm sections, i.e., spring arm sections, they usually have an average diameter smaller than the average diameter of the lower section of the large pores and the average diameter of the upper section.

[0014] These spring arm sections are combined to increase the compressibility of the polishing pad and the contact area of the top polishing surface during polishing. Advantageously, the spring arm sections create a horizontal overlap between most of the lower section and the upper section of the large pores. This shift of the large pores facilitates the compression of the entire polishing pad. Most advantageously, the spring arm sections create a horizontal separation gap between most of the lower section and the upper section of the large pores. The longer the spring arms, the greater this effect and the higher the compressibility of the polishing pad. Increasing the compression ratio can enhance the conformity of the polishing pad on the wafer, increase the contact area, and increase the polishing rate. Advantageously, the spring arm sections have an angle between 15 and 90 degrees as measured from the upward vertical direction.

[0015] In addition to the large pores, medium pores start adjacent to the spring arm sections of the large pores, and these medium pores have a vertical orientation. The medium pores usually start from horizontal and upper adjacent positions of the spring arm sections. Similarly, small pores start between the medium pores and interconnect with them. As suggested, the large pores are the largest and usually have a vertical height approximately twice the vertical height of the medium pores. The large pores with spring arms or connection sections advantageously account for less than 50 percent of the total of the large pores plus the medium and small pores. The combination of large, medium, and small pores improves the compressibility of the polishing pad.

[0016] The polishing pad advantageously also has a compression ratio measured by a uniaxial compression tester having a Keyence laser thickness measuring instrument configured as follows. [Table 1] The deflection instrument operates by applying a first weight 1 to a rod that presses a solid metal probe with a diameter of 5 mm against a flat specimen and measuring the thickness (T1) after 60 seconds. Next, after waiting for another 60 seconds and adding a second weight to the rod to increase the weight, the probe is further pressed into the specimen. Next, the measured value after another 60 seconds represents the final thickness (T2) used to calculate the compression ratio by the above formula. For this application, particularly for the purpose of the example, all compression ratio data and ranges represent values measured by the above test method.

[0017] The polishing pad advantageously has a compression ratio of at least 5% in the above test. Most advantageously, the polishing pad has a compression ratio of 5 - 10% in the above test.

[0018] Advantageously, the polishing pad has an embossed surface that forms grooves extending to the periphery of the polishing pad. Usually, the embossing is in an X - Y square grid pattern. However, the embossing can be in any known pattern (e.g., circular, or circular and radial).

[0019] Referring to FIG. 1, a polyurethane - water - dimethylformamide ( "DMF") coating mixture 10 coats the felt roll 12 by controlling the backing blade 14 and the knife or doctor blade 16. The porous polishing layer is fixed to a polymer film substrate, woven, or formed on a woven or non - woven substrate so as to form a polishing pad. When depositing the porous polishing layer on a polymer substrate (e.g., a non - porous poly(ethylene terephthalate) film or sheet), it is often advantageous to use a binder (e.g., a proprietary urethane or an acrylic adhesive) to enhance adhesion to the film or sheet. These films or sheets may be porous, but preferably, these films or sheets are non - porous. The advantages of non - porous films or sheets are to promote uniform thickness or flatness, enhance the overall rigidity of the polishing pad, reduce the overall compressibility, and eliminate the slurry suction effect during polishing.

[0020] The doctor blade 16, which includes the felt roll 12, the backing blade 14, and side walls (not shown), together with the backing blade 14 forms a trough 18 that holds the coating mixture 10. The backing blade 14 presses the felt roll 12 against the backup roll 20 to prevent the coating mixture 10 from flowing out from the rear of the trough 18. The backup roll 20 rotates clockwise during the operation of the coating line.

[0021] The width of the gap 22 is determined by moving the back blade 14 closer to or farther from the backup roll 20. The smaller the gap 22, the greater the reverse tension of the felt roll 12. The dotted arrow 22A indicates the change in the width of the gap 22. This is achieved by moving the back blade 14 towards the backup roll 20 to decrease (-) the gap and increase the tension, or away from the backup roll 20 to increase (+) the gap and decrease the tension. The tension vector A indicates the direction of the reverse tension of the felt roll 12. The height of the doctor blade 16 determines the thickness of the coating 24 on the felt roll 12. Since the doctor blade 16 controls the thickness of the liquid coating mixture 10, it gives the felt roll 12 a reverse tension close to or equal to zero.

[0022] A tension roller (not shown) pulls the felt roll 12 having the coating 24 into the water bath 26. The tension vector B indicates the direction of the tension pulling both the felt roll 12 and the coating 24 through the water bath 26. As soon as it is immersed in the water bath 26, the DMF diffuses from the coating mixture 10 and is replaced by water with a lower concentration of DMF. This rapid diffusion creates pores within the coating 24. Moving the touch roll 28 up and down facilitates the adjustment of the tension and compression of the felt roll 12 having the coating 24. Since the coating mixture 10 is a liquid-solid mixture, there is no reverse tension in the coating 24 between the doctor blade 16 and the touch roll 28. Tension in the coating 24 exists only after the coating 24 has passed through the touch roll 28. The touch roll 28 rotates counterclockwise during the operation of the coating line. When the large pores 30 move and engage with the touch roll 28, the tension and compression forces combine to deform the pores 30. Increasing the line speed shortens the time for the substrate surrounding the pores to align and harden. The substrate should have sufficient strength to maintain its shape but insufficient strength to elastically deform and recover. This partial hardening before curing in the oven facilitates the formation of the deformed pores 30.

[0023] Referring to FIG. 2, the combination of the reverse tension on the felt roll 12 and the tensile tension of the coating portion 24 after the felt roll 12 and the touch roll 28 acts to create a shear zone illustrated by a dotted line with respect to the lower shear zone boundary 32 and the upper shear zone boundary 34. Arrow C indicates the direction of rotation of the touch roll 28. In the shear zone 33, between the dotted lines 32 and 34, the large pores 30 change from vertical pores to large vertical pores 40 having a curved portion of the spring arm section 42 (FIG. 3A). Arrow D gives the direction of the felt roll 12 at the touch roll 28. At the touch roll 28, the tension vectors A and B pull in the opposite direction through the lower shear zone boundary 32 to the upper shear zone boundary 34 or the upper end of the shear zone 33. The shear zone 33 defined between the lower boundary 32 and the upper boundary 34 gradually deforms the large pores 30. The pore 30A shows the first bend in the intermediate section. The pore 30B has a more distinct bend in the central portion. The pore 30C has a clearly defined bend with a moderately narrowed central portion. The pore 30D has a substantially final bend with a central portion having a substantially final narrowness. The pore 40 represents the final large pore including the spring arm section. These spring arm sections promote the high compressibility and compatibility of the final polishing pad.

[0024] Referring to FIGS. 3, 3A, and 3B, the large pore 30 includes a main section 50 having a teardrop shape, an intermediate section 52 having a tapered neck shape, and an upper section 54 having a vertical orientation and a slight taper. Arrow sections 50A, 52A, and 54A define the heights of the main section 50, the intermediate section 52, and the upper section 54, respectively. Typically, the shear zone boundaries 32 and 34 extend from the upper part of the main section 50 through the intermediate section 52 to the lower part of the upper section 54. During deformation, the upper part of the main section 50 deforms in the direction of being pulled. The intermediate section 52 deforms in multiple directions and sides. The pore is stretched and narrowed for the first bend from the vertical direction to the partially horizontal - partially vertical direction, and then bent upward to return from the partially horizontal - partially vertical direction to the vertical direction. When the pore is stretched and narrowed, the cross-sectional area or the average diameter decreases. This narrow region that extends at least partially in the horizontal direction is known as the spring arm section 60. Arrow 60A defines the height and length of the spring arm section 60. Arrow 60B extends from the vertical bisector of the main section 50 to the vertical bisector of the upper section 54 and defines the offset of the upper section. Advantageously, the spring arm section 60 has an angle of 15 degrees to 90 degrees from the vertical. Most advantageously, the spring arm section 60 has an angle of 25 degrees to 80 degrees from the vertical.

[0025] Referring to FIG. 3A, when the shear zone 33 is large, the upper section 54 shifts horizontally by a distance sufficient to create a horizontal gap 60B for the spring arm section 60 that extends beyond the lower section 50 of the large pore. Referring to FIG. 3B, when the shear zone 33 is small, the upper section 54 shifts horizontally by a distance insufficient to create a horizontal gap 60B for the spring arm section 60 that extends beyond the lower section 50 of the large pore 40. In this case, there is a horizontal overlap between the upper section of the spring arm section 60 and the outermost portion of the lower section 50 of the large pore 40. The force of the shear zone 33 combined with the polymer matrix-controlled yield strength controls the final length 60A of the spring arm section.

[0026] Referring to FIG. 4, the coated felt substrate 12 includes a plurality of large pores 40 that include spring arm sections 60. The plurality of spring arm sections are combined to increase compressibility and contact area during polishing. A series of large secondary pores (Medium-sized micropores) 70 start from a position adjacent to the spring arm section 60. Similarly, a set of upper secondary pores (Small-sized micropores) 72 start at approximately the middle position of the secondary pores 70. Usually, the large pores 40 have a maximum size. The secondary pores 70 are smaller than the large pores 40 but tend to be larger than the upper secondary pores 72. The large pores 40, secondary pores 70, and upper secondary pores 72 all extend to the skin layer 76 at the top surface. A large number of micropores 78 are present beneath the surface immediately below the skin layer 76.

[0027] After DMF removal, oven drying cures the thermoplastic polyurethane. Optionally, the high-pressure washing and drying steps further clean the substrate. After drying, referring to FIG. 4A, the buffing process removes the skin layer 76 and the fine pores 78 and opens the large pores 40, secondary pores 70, and upper secondary pores 72 to a controlled depth. This enables a consistent pore count and open pore area at the top surface. It is advantageous to use a stable abrasive that does not fall off and does not penetrate the porous substrate during buffing. Typically, diamond abrasives produce the most consistent texture and have the lowest likelihood of breaking during buffing. After buffing, the substrate has a normal "tooth" (i.e., pore) height of 10 - 30 mils (0.25 - 0.76 mm) and an overall thickness of 30 - 60 mils (0.76 - 1.52 mm). The average large pore diameter ranges from 5 - 85 μm. Typical density values are from 0.2 to 0.5 g / cm 3 3. The cross-sectional pore area is typically 10 - 30 percent, the surface roughness Ra is less than 14, and Rp is less than 40. The hardness of the buffing pad is preferably 40 - 74 on the Asker C scale.

[0028] In an alternative embodiment, the non-porous film serves as the base substrate. The most prominent drawback of the film is the possibility of air bubbles being trapped between the polishing pad and the platen of the polishing tool when the non-porous film or porous substrate is used as the base substrate in combination with the adhesive film. These air bubbles distort the polishing pad and cause defects during polishing. The patterned release liner facilitates the removal of air and removes air bubbles under these circumstances. This results in a major problem of increased polishing non-uniformity, increased defects, increased pad wear, and shortened pad life. When felt is used as the base substrate, these problems are eliminated because air permeates the felt and air bubbles cannot be trapped. Second, when the polishing layer is applied to the film, the adhesion of the polishing layer to the film depends on the strength of the adhesive bond. Under some aggressive polishing conditions, this bond can fail, resulting in catastrophic failure. When felt is used, the polishing layer actually penetrates to a specific depth within the felt, forming a strong, mechanically interlocked boundary layer. A woven structure is acceptable, but a non-woven structure can provide additional surface area for strong bonding to the porous polymer substrate. An excellent example of a suitable non-woven structure is a polyester felt impregnated with polyurethane to hold the fibers together. The thickness of a typical polyester felt roll is 0.5 to 1.5 mm.

[0029] The polishing pad of the present invention is suitable for polishing or planarizing at least one of a semiconductor substrate, an optical substrate, and a magnetic substrate using a polishing liquid and relative movement between the polishing pad and at least one of the semiconductor substrate, the optical substrate, and the magnetic substrate. The polishing layer has a continuous bubble polymer matrix. At least a part of the continuous bubble structure is open to the polishing surface. The large pores extend to the polishing surface having a vertical orientation. These large pores contained in the solidified polymer matrix form a "beard" layer up to a specific "beard" height. The height of the vertical pores is the same as the height of the "beard" layer. The orientation of the vertical pores is formed during the solidification process. For the purposes of this patent application, the vertical or up-down direction is perpendicular to the polishing surface. The average diameter of the vertical pores increases with the distance from the polishing surface or the distance below the polishing surface. The polishing layer typically has a thickness of 20 to 200 mils (0.5 to 5 mm), preferably 30 to 80 mils (0.76 to 2.0 mm). A continuous bubble polymer matrix having vertical pores and open channels interconnecting the vertical pores. Preferably, the continuous bubble polymer matrix has interconnected pores having a diameter sufficient to allow fluid transport. These interconnected pores have an average diameter much smaller than the average diameter of the vertical pores. The pore morphology is characterized by open primary pores at the top with a size of about 40 μm and interconnected micropores with a size of about 2 μm in the polyurethane layer.

[0030] The plurality of grooves in the polishing layer facilitate the distribution of slurry and the removal of polishing debris. Preferably, the plurality of grooves form an orthogonal grid pattern. Usually, these grooves form an X-Y coordinate grid pattern on the polishing layer. These grooves have an average width measured adjacent to the polishing surface. The plurality of grooves have a debris removal residence time such that a point on at least one of a semiconductor substrate, an optical substrate, and a magnetic substrate rotating at a constant speed passes through the width of the plurality of grooves. The plurality of protruding land regions within the plurality of grooves are preferably supported by a tapered support structure that extends outwardly and downwardly from the top or plane of the polishing surface of the plurality of protruding land regions. Preferably, at an inclination of 30 to 60 degrees measured from the plane of the polishing surface, and most preferably, the plurality of land regions have a frustum or non-pointed top that forms the polishing surface from a polymer matrix containing vertical pores. Usually, the protruding land regions have a shape selected from hemispherical, truncated pyramid-shaped, truncated trapezoidal, and combinations thereof, and are provided with a plurality of grooves extending linearly between the protruding land regions. The plurality of grooves have an average depth greater than the average height of the vertical pores. Further, the vertical pores have an average diameter that increases the depth of at least one below the polishing surface.

[0031] Melting and solidifying the thermoplastic polyurethane at the bottom of the inclined sidewalls closes most of the large and small pores and forms groove channels. Preferably, the plastic deformation of the sidewalls and the melting and solidifying process form a grid of interconnected grooves. The bottom surface of the groove channels has few or no open pores. This facilitates the smooth removal of debris and secures the microporous polishing pad within the open pore tapered pillow structure. Preferably, the grooves form a series of pillow structures formed from a porous matrix containing large and small pores. Preferably, the small pores have a diameter sufficient to allow the flow of deionized water between the vertical pores.

[0032] The base layer is important for forming a proper foundation. The base layer can be a polymer film or sheet. However, woven or non-woven fibers provide an optimal base material for the micro-porous polishing pad. For the purposes of this specification, micro-porous refers to breathable synthetic leather formed from the aqueous replacement of an organic solvent. Non-woven felt provides an excellent base material for most applications. Usually, these base materials are occupied by polyester fibers (e.g., polyethylene terephthalate fibers, or other polymer fibers formed by mixing, carding, and needle punching).

[0033] For consistent properties, it is important for the felt to have a consistent thickness, density, and compressibility. Forming the felt from consistent fibers with consistent physical properties results in a base substrate with consistent compressibility. To enhance consistency, shrinkable fibers and non-shrinkable fibers can be mixed, and the density of the felt can be controlled by passing the felt through a warm water bath. This has the advantage that the final felt density can be finely adjusted using the bath temperature and residence time. After forming the felt, it is sent to a polymer impregnation bath (e.g., an aqueous polyurethane solution), and the fibers are coated. After coating the fibers, curing the felt in an oven increases rigidity and elasticity.

[0034] The post-coating curing and subsequent buffing process control the thickness of the felt. To finely adjust the thickness, the buff can first be applied with coarse abrasive grains and then the felt can be finished with fine abrasive grains. After buffing the felt, it is preferable to wash and dry the felt to remove the sand and debris picked up during the buffing process. After drying, dimethylformamide (DMF) is then trapped on the back side to prepare the felt for the waterproofing process. For example, perfluorocarboxylic acid and its precursors (e.g., the AGC Chemicals textile repellent AG-E092) can waterproof the upper surface of the felt. After waterproofing, the felt needs to be dried, and an optional burning process can remove the ends of the fibers protruding from the top layer of the felt. Next, the waterproof felt is prepared for coating and coagulation.

[0035] A mixture of anionic and nonionic surfactants preferably forms pores during coagulation and contributes to improved hard-segment / soft-segment formation and optimal physical properties. With respect to anionic surfactants, the surface-active portion of the molecule bears a negative charge. Examples of anionic surfactants include, but are not limited to, carboxylates, sulfonates, sulfate esters, phosphates and polyphosphate esters, and fluorinated anions. More specific examples include, but are not limited to, dioctyl sodium sulfosuccinate, sodium alkylbenzene sulfonate, and salts of polyoxyethylenated fatty alcohol carboxylates. In the case of nonionic surfactants, the surface-active portion does not have an apparent ionic charge. Examples of nonionic surfactants include, but are not limited to, polyoxyethylene (POE) alkylphenol, POE linear alcohol, POE polyoxypropylene glycol, POE mercaptan, long-chain carboxylic acid esters, alkanolamine alkanolamides, tertiary acetylenic glycols, POE silicones, N-alkylpyrrolidone, and alkyl polyglycosides. More specific examples include, but are not limited to, monoglycerides of long-chain fatty acids, polyoxoethylenated alkylphenol, polyoxyethylenated alcohol, and polyoxyethylene cetyl stearyl ether. For a more complete description of anionic and nonionic surfactants, see, for example, Chapter 1 of "Surfactants and Interfacial Phenomena" by Milton J. Rosen, 3rd Edition, Wiley-Interscience, 2004.

[0036] Examples The following examples illustrate the invention with a focus on polyurethane formulation, coagulation control, and polishing performance.

[0037] Materials

[0038] In the examples, Component A represents CRISVON (trademark) 8166NC of DIC and methylene diphenyl diisocyanate (MDI) for generating a "hard segment" in thermoplastic polyurethane. In particular, the polyurethane is a polyester-type low-elasticity polyurethane, which is processed in a coagulation process to form an upper porous layer as a polishing layer. The analysis specifications of Component A are as follows. Non-volatile solid weight %: 29.0 - 31.0%. Viscosity at 25°C: 60,000 - 80,000 MPa·s; 300% modulus of elasticity - 17 MPa; tensile strength - 55 MPa; elongation up to at least 500% fracture and melting point of 195°C.

[0039] The chemical composition of Component A was determined by proton and carbon-13 NMR and was as follows: [Table 2]

[0040] The first surfactant was sodium dioctyl sulfosuccinate (''DSS'') RESAMINE CUT-30 purchased from Dainichiseika. The second surfactant was polyoxyethylene alkyl ether (''EOPO'') PL-220 purchased from Kao Chemical.

[0041] Component A: Polyurethane Component B: Dioctyl sodium sulfosuccinate surfactant Component C: Polyoxyalkylene alkyl ether surfactant Component D: Dimethylformamide (DMF)

[0042] The formulations used various combinations of Components A - D formed in various coagulation processes: [Table 3]

[0043] The control of pore growth and final pore morphology was achieved by using surfactant components B and C at various concentrations. The coating solution was a mixture of components A, B, C, and D used for coating, with water added for the subsequent DMF substitution coagulation process.

[0044] The coagulation film of the polyurethane formulation was prepared by a laboratory drawdown test to investigate the surfactant ratio for producing a porous material. An impregnated nonwoven polyester felt was used as the substrate. The polyurethane was diluted to the designed solids content % with DMF, mixed with surfactants, degassed, equilibrated to the designed temperature, and then drawn down. Coagulation was carried out in a DMF / water bath, followed by washing and drying.

Table 4

[0045] Two surfactants controlled the coagulation mechanism and enabled primary pore growth. The DSS surfactant promoted the growth of primary pores deep at the bottom of the coated layer. The height of the "beak" of the primary pores became deeper as the concentration of the DSS surfactant increased.

[0046] The combination of DSS and EOPO surfactants regulated the coagulation of polyurethane, and instead of a pure teardrop shape without a cylindrical section, increased primary pores with a cylindrical upper section were formed. An EOPO surfactant concentration exceeding 2.0 phr hindered the growth of primary pores, leaving only a uniform layer of micropores. This was probably due to the affinity of EOPO for the soft segments of the polyurethane chains, promoting the solvation of the polyurethane and reducing the degree of phase separation.

[0047] The best ratio of DSS / EOPO surfactants was 4:1 phr / phr. Example 2 : Candidate formulation: Component A Polymer concentration: 20 wt%, 22 wt% in DMF Surfactant: DSS and EOPO Concentration of surfactant mixture: DSS concentration = 4.0 phr EOPO concentration = 1.0 phr Coating thickness: 65 mils (1.65 mm), 90 mils (2.23 mm) DMF concentration: 7 wt% Coagulation bath temperature: 25℃、30℃、35℃ Sample size: 12 drawdown Method: Laboratory drawdown test, standard conditions Results: The coagulation temperature affected the pore morphology and the growth of the "beak" (i.e., the pores). The solids concentration affected the pore morphology, especially the shape of the teardrops. The growth of the pores could reach the bottom of the drawdown with a thicker coating, but better control of the pore morphology was needed. Example 3 : Candidate formulation: Component A Polymer concentration: 20 wt% in DMF Surfactant: DSS and EOPO Concentration of surfactant mixture: DSS concentration = 4.0 phr EOPO concentration = 1.0 phr Coating thickness: 65 mils (1.65 mm) DMF concentration: 0 wt%, 7 wt%, 14 wt% Coagulation bath temperature: 20℃、30℃、40℃ Sample size: 9 drawdowns Method: Laboratory drawdown test, standard conditions Result: The coagulation temperature had a great influence on the pore morphology and the growth of "hairs". Increasing the DMF concentration hindered the formation of primary pores. The main process conditions in coagulation control and pore morphology were determined as follows. Coagulation bath temperature Polymer solid content % DMF / water concentration Coating thickness Example 4 : Candidate formulation: Component A Polymer concentration: 20 wt% in DMF Surfactant: DSS and EOPO Concentration of surfactant mixture: DSS concentration = 3.2, 4.0, 4.8 phr EOPO concentration = 0.8, 1.0, 1.2 phr Coating thickness: 65 mils (1.65 mm) DMF concentration: 7 wt% Coagulation bath temperature: 25℃ Method: Laboratory drawdown test, standard conditions Sample size: 11 drawdowns

Table 5

[0048] Table 5 summarizes the ratio of surfactants for Example 4. Table 6 provides the results for the polishing pads manufactured under the conditions of Table 5. The data were summarized as follows based on SEM analysis.

Table 6

[0049] Pore formation was observed to have pore morphology within ±1.5 sigma variation. The ratio of surfactants had a significant effect on the height of the "hairs" compared to other parameters. For the substrates in Table 6, the pore structures of Samples 1, 5, and 8 with a DSS to EOPO weight percent concentration ratio of 4 to 1 provided the best pore morphology.

[0050] Example 5 Film Tensile Properties

Table 7

[0051] The above data show excellent toughness and breaking energy of the porous substrate. Note: The above properties are representative of film substrates tested in accordance with (ASTM D886).

[0052] Polishing Protocol

[0053] Pad polishing performance was determined on a 300 mm blanket wafer attached to an Applied Material Reflexion® LK 300 mm CMP polishing tool. The polishing removal rate experiments were conducted on a copper wafer of a Novellus 300 mm sheet 20K Cu electron plate, on a TEOS wafer of a Novellus 300 mm blanket 20k tetraethyl orthosilicate (TEOS) sheet wafer, on a tantalum (Ta) wafer of a Sematech 300 mm sheet 1K tantalum, on a Black Diamond™ and Coral™ low-k dielectric wafer of a CNSE 300 mm sheet 5K BD (k = 3.0), and on a BD2S wafer of a SVTC 300 mm sheet 5K BD2 (k = 2.7).

[0054] All polishing experiments were carried out using the ACuPLANE® LK393c4 Cu barrier slurry from Rohm and Haas Electronic Materials CMP Inc. All wafers were polished under standard conditions of a low pressure of 12.4 kPa (1.8 psi), a flow rate of 300 mL / min of the chemical mechanical polishing composition, a table rotation speed of 93 rpm, and a carrier rotation speed of 87 rpm, usually for 60 seconds. A 3M-A82 diamond pad conditioner commercially available from 3M was used to condition the polishing pad. The specifications of the 3M-A82 disk are listed in Table 8. The polishing pad was conditioned by a conditioner that uses a downward force of 2.0 pounds (0.9 kg) for 10 minutes at a platen speed of 73 rpm / conditioner speed of 111 rpm with the high pressure rinse (HPR) of the conditioner turned on. During polishing, the pad was fully conditioned by a conditioner that uses a downward force of 2.0 pounds (0.9 kg) for only 3.2 seconds at a platen speed of 73 rpm and a conditioner speed of 111 rpm with the high pressure rinse (HPR) turned on.

[0055] The TEOS removal rate was determined by measuring the film thickness before and after polishing using a KLA-Tencor SPECTRAFX200 measurement tool. The removal rates of copper (Cu) and tantalum (Ta) were determined using a KLA-Tencor RS100C measurement tool. The defect map scan was performed using a KLA-Tencor SP2 measurement tool, and the defect review was performed using a KLA-Tencor eDR-5210 measurement tool.

Table 8

[0056] Polishing Examples Examples of four pads and their respective characterizations are summarized below. All pads were manufactured with the same polyurethane / surfactant formulation with different process parameters for Sample 1.

Table 9

[0057] All of the above pads had the pore structures of FIGS. 3A, 4, and 5. In particular, the primary pores had a spring arm shape that promoted the compressibility of the pads. The above data shows that increasing the line speed increases the compression rate of the polishing pad. When the compression rate increases, the contact area during polishing increases. This increased contact area enabled the pad to operate in a softer structure that was less likely to generate defects.

[0058] Example 6 Polishing Performance

[0059] The results of the removal rate and defects are summarized in the following table.

[0060] Example 7: Polishing Performance of Pads 1, 4 (different batches):

Table 10

Table 11

[0061] The above table shows the excellent polishing stability of 500 wafers when polishing copper, TEOS, and tantalum substrates. Example 7 Removal Rate [Table 12] [Table 13]

[0062] Regarding the copper removal rate, the pads of Examples 1-4A and 1-4B showed copper rates of 694 and 680 A / min, respectively, which were approximately 12% and 14% lower than that of the commercially available pad A. The TEOS removal rates of 1-4A and 1-4B were 1416 and 1414 A / min, which were the same as those of the commercially available pads A and B. The similar removal rates among the commercially available pad A, pad B, and the pads of Examples 1-4A and 1-4B suggest that good contact area or wear, and the affinity among the pad, abrasive, and wafer facilitate effective oxide and copper removal. Example 8 Defect Performance [Table 14] [Table 15] [Table 16] [Table 17]

[0063] The polishing pads 1-4A and 1-4B showed far fewer total defects than commercially available pads A and B. The total number of defects was on average 73, 146, 19, and 37, and the scratch and chattering marks were on average 35, 10, 4, and 1 for pads A, B, 1-4A, and 1-4B, respectively. This showed a measurable and dramatic reduction in polishing defects. The high compressibility pads manufactured at the highest line speed tended to have the fewest total defects.

[0064] An enhanced scanning recipe was created to increase resolution and differentiate performance. The results are summarized in the graph on the right, showing total defect counts of 241, 736, 34, and 85, and scratch and chattering marks of 126, 360, 5, and 9 for pads A, B, 1-4A, and 1-4B, respectively. The polishing pads 1-4A and 1-4B showed a reduction of over 99% in scratches and chattering on average compared to commercially available pad B, and over 95% compared to commercially available pad A. The high compressibility pads manufactured at the highest line speed had the fewest chattering mark defects. Example 9 Pad Analysis after Polishing

[0065] To evaluate pad wear, SEM analysis was performed on the polished pad surface. The sampling areas included the center, middle, and edge of the pad. For the polishing pads 1-4A and 1-4B, all primary pores remained open and there were no fragments. No significant hanging materials were seen from pad break-in, conditioning, or wafer polishing. Furthermore, there was no significant difference in the surface pore morphology from the center, middle, or edge of the pad. This suggests that consistent wear occurred across the entire pad. Additionally, high-resolution SEM images (magnifications of 500x and 1000x) showed a distinct secondary pore structure. The smaller micropores remained open after polishing and no accumulation of fragments was observed. This indicated an effective slurry flow through the porous structure. There was no difference between the center, middle, or edge of the pad.

[0066] A uniform distribution of primary pores and interconnected micropores represents one reason for the excellent performance of pads that, in combination with excellent defect performance, provide a satisfactory removal rate. The present invention demonstrated a new highly compressible structure that provides excellent polishing performance. In particular, it showed an ultra-low defect rate, good removal rates for copper, TEOS, and barrier metals, and a long pad life. In particular, the pad is polished at excellent copper and TEOS rates while maintaining stability for multiple wafers. Further, the pad has significantly fewer scratch and chattering defects than conventional polishing pads. The use of the manufacturing process was a determining factor in the final primary and secondary pore structures. Further, the manufacturing process provided a robust and reproducible pad pore morphology and polishing performance.

Claims

1. A porous polyurethane polishing pad, comprising: a porous substrate having large pores extending upward from a base surface and opening to an upper surface, wherein the large pores are interconnected with micropores, a part of the large pores opening to a top polishing surface, at least a part of the large pores extending to the top polishing surface and having a lower section and an upper section with a vertical orientation, where vertical is the orthogonal direction perpendicular to the base surface and towards the polishing surface, and an intermediate section having a spring arm shape, connecting the lower and upper sections, the lower section having a teardrop shape and provided with a vertical bisector, the upper section having a cylindrical shape and provided with a vertical bisector, the intermediate section extending from the vertical bisector of the lower section to the vertical bisector of the upper section, the intermediate section having an angle of 15 to 90 degrees as measured from an upward vertical direction, the intermediate section having an average diameter smaller than an average diameter of the lower section of the large pores and larger than an average diameter of the upper section offset horizontally from the lower and upper sections, the offset resulting in the intermediate section extending beyond the lower section, medium pores having a columnar shape and a vertical orientation occurring adjacent to the intermediate section, and small pores having a columnar shape and a vertical orientation occurring between the medium pores, the medium pores forming a "keba" layer having a "keba" height of 10 to 30 mils (0.25 to 0.76 mm), the large pores, the medium pores, and the small pores having horizontally offset upper and lower sections being combined such that during polishing, compressibility in the polishing pad and contact area to the top polishing surface are increased, the polishing pad being measured such that with a 5 mm diameter probe, 60.5 grams are added to a flat sample, after waiting for 60 seconds, thickness (T1) is measured, then after waiting for an additional 60 seconds, an additional 98 grams are added to a total of 158.5 grams, and after waiting for 60 seconds more, thickness (T2) is measured, and the compression rate (%) = (T1 - T2) / T1, and having a compression rate of at least 5%, the above polishing pad.

2. The polishing pad according to claim 1, wherein most of the middle section creates a horizontal separation gap between the lower section and the upper section of the large pores.

3. The polishing pad according to claim 1, having an embossed surface that forms a groove extending to the periphery of the polishing pad.

Citation Information

Patent Citations

  • Polishing cloth

    JP2009101504A

  • Method for manufacturing foamed body

    JP2011011278A

  • Polishing pad

    JP2013208670A

  • Polishing cloth

    JP2014030868A