KR101616535B1 - Customized polishing pads for cmp and methods of fabrication and use thereof - Google Patents

Abstract

The present invention relates to a polishing pad for chemical mechanical planarization of a substrate, and a method for manufacturing and using the polishing pad. The pads of the present invention are adapted according to polishing issues, including, but not limited to, the materials to be polished, the chip design and architecture, the chip density and pattern density, the equipment platform, and the type of slurry used. Such pads can be designed with specific polymer nanostructures with long range regularity or short range regularity that can be adjusted at the molecular level to achieve good thermal-mechanical properties. In particular, such pads can be designed and manufactured to have both uniform and non-uniform spatial distribution of chemical and physical properties within the pad. Additionally, these pads can be designed to adjust the coefficient of friction by creating a low shear integral pad with multiple layers of polymeric material forming a surface treatment with the addition of a solid lubricant, a parallel interface with the polishing surface . These pads may also have a controlled void ratio, an inserted abrasive, an ingrown groove on the polishing surface made in situ for slurry delivery, and a transparent area for endpoint detection.

Description

Technical Field [0001] The present invention relates to a polishing pad suitable for chemical mechanical planarization and a method of manufacturing and using the polishing pad. BACKGROUND OF THE INVENTION < RTI ID = 0.0 > [0001]

The present invention relates to a polishing pad for chemical mechanical planarization (CMP) and a method of making and using the polishing pad.

CMP uses a slurry called a reactive liquid medium with a polishing pad to provide a chemical-mechanical action to remove material from the substrate surface during the planarization process. For example, one area using CMP is to planarize individual layers (dielectric or metal layers) during fabrication of integrated circuits (ICs) on semiconductor substrates. CMP is used to remove undesirable topographic features of the integrated circuit layer, such as metal deposits after the etching process, to remove extra oxide from the shallow trench isolation step, or to remove inter-level dielectric layers and inter- Planarize the inter-metal dielectric layers. The main purpose of CMP used in integrated circuit fabrication is to maintain the flatness at each step of forming sequential dielectric and metal layers and forming photolithographic images.

During the CMP process, the chemical interaction of the substrate and the slurry forms a chemically modified layer on the polishing surface. At the same time, the abrasive contained in the slurry mechanically interacts with the chemically modified layer to remove material. Generally, the polishing pad is made of a rigid, microporous polymeric material, such as polyurethane, and performs a number of functions including uniform slurry delivery and dispensing, removal of reaction products, and uniform distribution of pressure on the wafer . The interaction of the pad with the slurry with respect to the creation and removal of a thin surface layer at the nanoscale or micron scale determines the material removal rate (removal rate), flatness, surface non-uniformity, surface defects and selectivity of material removal. In this regard, local pad material properties / tribological properties / mechanical properties are important both for local planarization and overall planarization during the CMP process.

As mentioned earlier, one of the applications of CMP is the semiconductor industry where CMP is used at different process stages. Current state-of-the-art CMP pads, such as open-pored polymer pads and closed pore polymer pads, are incompatible with achieving optimized tribological, chemical and friction properties. Although these pads may be suitable for conventional integrated circuit manufacturing processes and newly developing 90 nm CMOS technology, high yields can not be achieved with such pads. This problem is not solved by design (i.e., system on chip (SOC)) and fabrication process (i.e., silicon on) as well as material differences and variations (i.e., STI: copper and low k dielectric) Insulator (SOI)). The impact of this problem on manufacturing processes in technologies less than 90 nanometers is that chip yield, device performance and device reliability are significantly degraded.

The CMP process can remove the excess dielectric during the semiconductor manufacturing process. Due to the complexity in design, the primary impact is an increase in the number of materials being polished at the same time. For example, STI and copper CMP (Cu CMP) show the problem of CMP of different materials. During STI CMP, the dishing of oxides and the erosion of nitrides are generally observed, in which case a CMP process with selectivity for the rate of removal of such materials is required due to material differences. Likewise, for Cu CMP of less than 90 nanometer technology, copper is uniformly removed through mechanical action by pads and abrasives, while wear occurs while local abrasion of the dielectric causes local anomalies do. A high level of planarization is compromised by excessive distribution and wear, which causes difficulties in meeting resistive properties for different pattern densities. Currently, the problem of feature loss due to loss of planarization due to distribution and wear accounts for more than 50% of the yield loss for technologies below 90 nanometers. Distribution and wear are affected by pad properties such as hardness, toughness and porosity.

As another example, changing the pattern density also presents a problem for CMP of an integrated circuit. For example, since pattern density is correlated with chip size, low pattern density is associated with small chip size, and conversely, there is a high pattern density for larger chip size. It is desirable to change pad characteristics such as hardness, surface structure, and surface texture as a function of pattern density change. In general, the complexity is also indicated because the pattern density in a single chip is also varied.

Since a large number of variables are given in the manufacture of integrated circuits such as integrated circuit design, material differentiation and pattern density, it is possible to systematically solve problems in order to achieve high polishing quality in consideration of various possible results in the polishing process There is a need for an abrasive pad.

Suitable polishing methods require a variety of pad engineering techniques. Given the size, pad engineering can be considered a process that is suited to the size of the macro-level as well as the size of the nano-micron level (the macro level is about 1 cm). For example, it may be desirable to have a pad that fits the nanostructure at a nanoscale size (i.e., the distribution, size, and type of the solid region throughout the pad). There are a number of engineering opportunities at the macroscopic level. CMP pads can be designed and manufactured to have a spatial distribution of chemical and physical properties tailored for performance for a given type of substrate. In this regard, it may be desirable to have a polishing pad optionally having considered properties such as material type as well as physical properties such as hardness, porosity, toughness and compressibility prior to manufacture. It may also be desirable to include features on the pad. One of these characteristics relates to pad surface engineering through the addition of a solid lubricant to the body of the pad. Another characteristic relates to controlling pad porosity through the use of different amounts and sizes of porogen as well as manufacturing process temperatures. Another characteristic relates to functional grading of the pad by adjusting the polymer composition of the pad in different zones along the polishing surface. Another characteristic is that an interface is added to the pad body to produce a pad having a low shear force. Another characteristic is to add abrasive to the pad by distributing the selected abrasive material within the pad. Another characteristic is to form in situ grooves on the polishing surface to optimize slurry delivery. Another characteristic is the formation of selectively transparent regions in the pad for endpoint detection. The various optimized polishing pads described here address the need for pads with optimized designs as well as control of manufacturing processes in order to realize optimized designs. Optimized design and control of the manufacturing process yields a single integrated pad that is particularly suitable for providing superior CMP performance of the substrate.

For CMP, it is generally known that uniformity of pad characteristics such as pad coefficient, pore size distribution, and chemical composition of materials is important for stable operation in boundary lubrication systems. Not only are design schemes based on these basic pad characteristics achieved, but optimized polishing requirements such as low COF are also described.

A pad having one or any combination of the following characteristics is described.

1) pad microstructure

The choice of pad microstructure may have an effect on the polishing characteristics. Conventionally, various polymers including polyolefins, polyurethanes and polycarbonates have been used as polishing pad materials. Urethanes were most commonly used to make CMP pads among the polymers. In the present invention, the pad microstructure is controlled through the selection of suitable polymer components. First, an isocyanate prepolymer is synthesized or commercially available. The isocyanate prepolymer is then reacted with a mixture of polyamine and polyol chain extender and polyamine and polyol curing agent to complete the polymer formation. As a result, a uniform spatial distribution of alternating hard and soft regions with long-range regularity is obtained at the nano-micron size level. This pad microstructure allows for a flattened and extended strike-cut curve. This pad structure also allows for excellent control of the frictional, thermal and optical properties. Thus, these properties can also be spatially distributed to achieve optimized polishing functionality.

As a result of this polymerization scheme, many properties of the polymer pad, such as the storage modulus (E ') of the polymer and the loss factor (E') of the polymer can be increased while the glass transition temperature (Tg) modulus (E ') the ratio of the loss modulus (E ") to (tanδ), KEL (tanδ * 10 12 (E' (1 + tan 2 δ))), surface tension, compressibility, as a function of thermal changes, temperature DELTA E 'and compressibility can be reduced and the surface tension can be controlled.

2) controlled porosity

Control of the pad porosity, that is, control of pore size, density and shape, can affect factors such as slurry transport, microstructure and abrasive distribution, which is an important criterion for uniform performance of CMP, such as removal rate And the number of wafer non-uniformities (WIWNU) within the wafer. Additionally, pads manufactured without controlling porosity can cause non-uniform shear forces on the substrate from different areas of the pad, and thus can cause non-uniform COF over the entire process range. The shear force non-uniformity affects two additional criteria of CMP performance, flatness and defects.

A number of suitable polishing pads described herein can be fabricated such that the porosity formed in the pads is controlled with respect to the distribution of pore size and shape, pore density, and porosity.

3) Functional grading of mechanical properties

The functional grading of the material represents different regions of the polymer material along the polishing surface that may be radially symmetric or non-symmetrical. Functional grading of the mechanical properties of the pad can be used to adjust pad tribological and polishing characteristics in a given systematic manner, leading to an increase in planarization length and efficiency. In addition, functional grading can be useful to overcome the loss of outer edge yield during CMP. One cause of the outer edge yield loss is the uneven distribution of pressure as observed by the wafer during the polishing process. The non-uniform pressure distribution from the center to the edge is inherent in the way the wafer is mounted on the polishing head. If radially symmetrical functional grading is used to compensate for the uneven pressure distribution, a reduction in the outer edge yield and a reduction in the number of defects can be achieved. Functional grading of mechanical properties (hardness, compressibility, pore size and distribution) can be used to compensate for non-uniformity of pressure distribution.

4) Surface-Engineering

The surface treatment of the pad is achieved through the addition of a solid lubricant and / or a polymer lubricant in the pad. This method of surface treatment through the addition of a lubricant can be used to effectively reduce the coefficient of friction while maintaining the desired removal rate. These pads can be used for most polishing applications, because in most cases a lower coefficient of friction is desirable. In particular, such surface-treated pads can be used for all process steps in copper CMP, including bulk, soft landing, and barrier removal steps, eliminating the need for three different pads for each process step .

5) Integral pad with low shear force

The low-shear integral pad has at least one interface that can be parallel to the polishing surface. The interface may be selectively formed at a location between materials having the same or different properties and may exhibit a reduction in shear force at the pad / substrate boundary. The reduction of the shear force allows the reduction of the coefficient of friction during polishing while maintaining the desired removal rate. A pad of low shear force representing the interface parallel to the polishing surface is schematically shown in Fig.

6) Pad with abrasive insert

Pads with abrasive inserts can be made by incorporating abrasives in the pad during pad manufacturing by techniques such as liquid casting / molding, injection molding, sintering, and the like. The pad into which the abrasive is inserted may have the advantage of eliminating the need for adding abrasives to the slurry by providing the abrasive through the pad composition. The pad into which the abrasive is inserted may comprise individual abrasive particles and may also comprise a block copolymer and in the case of a block copolymer the composition of the abrasive polymer has a different composition in the block copolymer depending on the distance.

7) In situ groove pad

Generally, a method for forming an in situ groove includes patterning a silicon lining, placing the silicon lining on a mold or on a mold, adding a CMP pad material to the silicon lining, and allowing the CMP pad to solidify . ≪ / RTI > In some variations, the silicon lining may be made of a silicone elastomer, and in some variations, patterning the silicon lining comprises patterning the silicon lining using lithography or embossing. The method of forming the in situ grout may further include attaching the silicon lining to the mold using, for example, adhesive, tape, clamp, pressure fitting technique and a combination of the above.

In some variations, the mold is a metal. For example, the mold may be made of a material selected from the group consisting of aluminum, steel, ultra-mold material, and combinations thereof. In some variations, the mold is patterned in addition to the patterning of the silicon lining (i. E., A combination of patterning is used). In some variations, the CMP pad material comprises a thermoplastic material. In another variation, the CMP pad material comprises a thermoset material. In some variations, the CMP pad material is polyurethane.

Also described is a CMP pad that incorporates an ingenious groove design. For example, what is described herein is a CMP pad that includes a groove in an inverted log, a concentric circular groove, and an axially curved groove. In some variations, the axially curved grooves are discontinuous. Further, the concentric circular groove and the axially curved groove can be intersected.

The grooves created here can be made by the group of methods consisting of silicon lining molding, laser lighting, water jet cutting, 3-D printing, thermoforming, vacuum forming, micro contact printing, hot stamping and combinations thereof.

8) Pad with transparent window for end point detection

The polishing pad is provided to include a transparent region and a method of manufacturing such a pad is described. The pad is useful in a method for detecting the end point of a substrate polishing process, such as a CMP process, wherein optical measurements are used to evaluate the surface of the substrate. Optical measurements can measure light transmitted through a polishing pad from a light source to a slurry beneath the substrate surface or substrate surface, from the slurry below the substrate surface or substrate surface to the detector, or in two cases mentioned. The transparent area of the polishing pad is sufficiently transparent to the spectrum or wavelength of light. It is preferable that the material is sufficiently transparent at a wavelength of at least one light from ultraviolet rays such as 100 nm to 1,000 nm, visible light and infrared spectrum. The transparent region need not be transparent over all spectra, but it is transparent to one or more wavelengths within a broad spectrum.

Optical transparency is achieved by reducing the scattering center through reduction of the porous element. In one example, the polishing pad has a transparent region lacking pores to be sufficiently transparent to a predetermined wavelength or wavelengths of light, and a microporous region having a lower transparency than a transparent region with respect to a predetermined wavelength or wavelength of light Containing polymer. The region having poor transparency has a sufficient porosity to have a predetermined compressibility or hardness in the region.

In one example, the transparent region is sufficiently transparent to light comprising a wavelength in the range of about 100 to 1,000 nm, or about 200 to 800 nm, or about 250 to 700 nm. In one example, the region with poor transparency comprises the same material as the transparent region, wherein the region with poor transparency has a higher porosity than the transparent region. In one example, the transparent region comprises the first polymer, the pores are sufficiently deficient, and the region of poor transparency comprises the second polymer and is substantially microporous. In one example, the first and second polymers are the same polymer. In one example, as the distance from the transparent region increases to the maximum pore density for the pad, the porosity density of the region that is less transparent increases gradually. In this example, most of the pads are close to the maximum pore density or near the maximum pore density, and significant changes in pore density are found within about 2 cm at the boundary of the transparent region, and also around the transparent region, such as within about 1 cm . In one example, the pore structure is selected from the group consisting of inorganic salts, foam formers, supercritical fluids, chemical foams, micelles, block copolymers, porogen materials, microballons, Of the pore-forming agent.

Numerous variables in the manufacture of integrated circuits, such as integrated circuit design, material differentiation and pattern density, are conferred so that high polishing quality can be achieved taking into account the various possible consequences of the polishing process.

Figures 1A and 1B show an exemplary deposition layer formed in the underlying layer,
Figures 2A and 2B show the erosion of the metal deposited in the trenches of the dielectric layer,
Figure 3 schematically shows the components of a CMP apparatus,
4 is a diagram showing an example of a Stribeck curve,
5 is a view showing an example of a prestonian float,
Figure 6 schematically illustrates how pore forming materials or pore formers can be used to produce a constant pore size, pore density, and distribution in a matrix,
Figure 7 schematically illustrates a radially discrete symmetrical functionally graded pad,
Figure 8 schematically illustrates pores combined with a functionally graded pad,
Figure 9 schematically illustrates a radially asymmetrical functionally graded pad,
Figure 10 schematically illustrates a radially continuous symmetrical functionally graded pad,
11A and 11B schematically illustrate a low shear integral pad having one interface 11A or a plurality of interfaces 11B with grooves on the surface,
Figure 12 shows the effect of the interface on stress reduction for CMP,
13 is a view showing stress-strain characteristics of polycrystalline copper,
Figs. 14A and 14B show the wafer pressure profile for the grooved pad 14A and the grooved pad 14B,
15 is a cross-sectional view of an exemplary silicon lining mold,
16A-16C illustrate a unique groove design for a 20 inch pad 16A, a 24 inch pad 16B, a 30 inch pad 16C,
Figure 17 schematically illustrates an example of a possible shape for a transparent region,
18 is a view showing a transparent base together with a groove with a lower transparency,
Figures 19 and 20 illustrate an example of a case where the window is equal to the thickness of the remaining portion of the pad (19), or thinner than the remaining portion of the pad (20)
21 schematically illustrates a fabrication process that may be used to create a transparent region in a CMP pad,
22 is a diagram illustrating an example of compensation for eliminating adverse effects due to changed hardness in a transparent region,
Figure 23 shows a transparent pad,
Figure 24 schematically illustrates a plurality of steps in a copper CMP,
25 shows a comparison of thermal changes for two commercially available pads (prior art) and three new pads (the present invention)
Figures 26a-26d illustrate a Prestonian float for two commercially available pads 26c, 26d and two adapted pads 26a, 26b,
Figures 27a-27d illustrate a Stribeck curve for two commercially available pads 27c, 27d and two adapted pads 27a, 27b,
Figures 28A and 28B illustrate a die measurement scheme in which nine die are selected per wafer measurement and Figures 28B and 28B illustrate respective individual die elements;
29 is a comparison of oxide thicknesses as a function of the layout pattern density in one die as a function of pressure and velocity for three polishing times (30 seconds, 60 seconds, 120 seconds) of a commercially available pad,
30 is a graph comparing the oxide thicknesses as a function of the layout pattern density in all nine dies of FIG. 28A for three polishing times (30 seconds, 60 seconds, 120 seconds) of a commercially available pad;
31 is a comparison of oxide thicknesses as a function of the layout pattern density in one die as a function of pressure and speed for three polishing times (30 seconds, 60 seconds, 120 seconds) of the fitted pad,
32 is a graph comparing the oxide thicknesses as a function of the layout pattern density in all nine dies of FIG. 28A for three polishing times (30 seconds, 60 seconds, 120 seconds) of the fitted pad,
33 is a view showing X-ray diffraction (XRD) data,
Fig. 34 is a diagram showing the lattice constant calculated from the X-ray diffraction data as compared with an un-polished wafer (bulk)
35 is a diagram showing the total width (FWHM) at the middle of the maximum height of 222 peaks in comparison with an unground wafer (bulk)
Figure 36 shows Strybeck curve data and a prestonian float for two pads used in copper CMP, rather than an integral pad with a low shear force, having solid lubricant,
Figure 37 is a plot of neat pads and pad alignment analysis for commercially available pads,
38 shows polishing stability analysis over time for a commercially available pad and a new pad,
Figure 39 is a diagram showing a Stribeck curve for two commercially available pads (A, B) and a new pad (C)
Figures 40A and 40B show the results of copper dishing 40a and copper erosion 40b,
41A-41C illustrate one example of a commercially available one-layer pad, a sub-surface treated pad, and a sub-surface treated to selectively remove (41c) the oxide polishing 41a, the nitride polishing 41b, A low integrated pad, and a low shear integral pad,
42A-42C illustrate one example of a commercially available one-layer pad, a sub-surface treated pad, and a sub-surface treated to selectively remove (42c) the oxide polishing 42a, the nitride polishing 42b, A low integrated pad, and a low shear integral pad.

The various polishing pads described herein are not limited to those listed below for the substrate to be polished, but specific aspects, including the structure, materials, and characteristics of the substrate, are those considered in the appropriate design of the polishing pad. The pads are fabricated using manufacturing means to control the characteristics of the pads according to the adapted design, thereby creating a single integrated and adapted pad.

The substrate means any material or device on which a polishing process such as CMP is carried out. In this regard, a variety of adapted polishing pads as described herein include, but are not limited to: 1) wafers such as silicon, quartz, silicon carbide, gallium arsenide, germanium, 2) Cleaning the oxide in an aluminum etch process, cleaning metal deposits (copper and tantalum barriers) in a double etch process, fabricating a uniform FinFET structure, fabricating an SoC device, or removing excess oxide at the STI stage A layer deposited or grown on a wafer formed in a semiconductor manufacturing process, 3) a rigid disk used for storage media such as nickel plated aluminum, glass and other magnetic materials commonly used in storage media, and 4) a fiber optic cable and optical interconnect Optical devices used for Internet and digital optical networks, such as connections, 5 Materials such as metal materials, ceramics, inorganic materials, polymers, epoxy-based carbon fiber composites and nanocomposite substrates, and 6) micro-processing techniques such as lithographic techniques, laser ablation, hot embossing, It may be useful to process various types of substrates including microstructures and devices, nanostructures and devices that are created. Briefly, polishing pads adapted for various views are useful for a variety of materials, devices, and systems that require finishing, uniformity, flatness, and low-defect surfaces for finishing.

A polishing pad adapted for various aspects described herein is intended to be adapted for use in the semiconductor industry for CMP of integrated circuits (ICs) on a wafer substrate. For this purpose, a polishing pad for CMP of an IC structure is adapted by obtaining one or more characteristics of the IC structure on the substrate, such as IC size, pattern density, IC architecture, film material, film topography, Pad nature, pad type, hardness, porosity, toughness, compressibility, surface architecture, surface composition, lubricant addition, interface formation in the pad, and the like, having long range regularity and short range regularity based on one or more characteristics of the IC structure. The characteristics of the pad, such as the abrasive additive, are selected. Such adaptive design and in situ fabrication for a single integrated pad can exhibit uniform performance for CMP processes in integrated circuits.

The uniform performance of the CMP process for an integrated circuit includes, but is not limited to, maintaining the constant polishing rate across the different regions of the substrate and maintaining a constant rate of friction of the frictional coefficients in the boundary lubrication zone Which should be performed on a number of criteria that can be used to evaluate the polishing quality. One criterion for polishing performance is the removal rate (RR). As will be explained in more detail below, removal rates are affected by a number of devices and consumable factors. Examples of pad properties such as compressibility, porosity, and surface composition affect slurry delivery, which can, for example, affect removal rates in practice. Another criterion of polishing performance is substrate flatness that minimizes or eliminates the occurrence of dishing and erosion of dielectric material during dielectric material or copper polishing processes in the STI stack. Pad hardness, toughness and porosity are examples of pad characteristics that affect substrate flatness. Pads with controlled porosity, i.e., pads with controlled pore size and density and pore distribution, can planarize the substrate better. The number of substrate non-uniformities, such as scratches and chips, is another measure of polishing performance. Examples of pad characteristics that affect the number of substrate non-uniformities include hardness and surface composition, which affect slurry delivery. Finally, defects are another criterion for evaluating the polishing process. The CMP process is a chemically and mechanically harsh process, and defects due to stress in integrated circuits reduce device yield. An example of a pad characteristic that affects defects is pad strength. Strong pads increase defects but have excellent flatness. If the control over the pad characteristics described above is not appropriate, it affects pad performance. For example, if the control of the pad porosity is not appropriate, a non-uniform shear force will occur across the polishing surface, and non-uniform COF will increase defects. In addition, poor control of other pad parameters can result in poor pad performance in a manner similar to pad performance problems caused by inadequate porosity control. Polishing performance, removal rate, substrate flatness, occurrence of non-uniformity, and defect criteria are examples of criteria that affect the cost of a CMP process.

Several variables in integrated circuit design have an impact on pad design and polishing performance. One of these variables can be the pattern density of the integrated circuit. The pattern density affects the degree of film removal and therefore affects the uniformity within the integrated circuit and across the wafer. The integrated circuit 10 manufactured in Figure 1 has features 12 that lie underneath, such as metal lines, that create an upper region 16 and a lower region 18 in the topography of the deposited film 14. In particular, topography has a different width across the chip and is highly dependent on the pattern density in the copper-based double etch structure due to the nature of the electroplating in the trenches with different chemical actions in relation to the additives used in the electroplating process. Generally, the top region 16 of the topography polishes faster than the bottom region 18. As shown in FIG. 1A, the initial step height 20 is associated with the deposited film 14 before polishing. 1B, the final stage height 22 is associated with the deposited film 14 after polishing. Different removal rates for the upper and lower regions 16 and 18, represented by the initial stage height 20 and the final stage height 22, are a form of advantage for planarization. The larger the difference, the better the flatness after the CMP process.

Another example of a variable in the manufacture of integrated circuits inside the integrated circuit and affecting uniform polishing performance across the wafer is the film material. In particular, dishing and erosion may occur in a CMP process involving a plurality of film materials due to the fact that different materials often have different polishing rates. 2A schematically illustrates an integrated circuit 30 that is fabricated having metal lines 32 deposited on the trenches of the dielectric layer 34. As shown in FIG. 2B, the dishing of the metal line 32 is shown as a change in the height 36 of the metal line 32 from the flatness with the dielectric layer 34. The erosion of the dielectric layer 34 is also shown as a change in the height 38 of the dielectric layer 34 from the intended height. Dishing and erosion may be present in a double etch process for shallow trench isolation (STI), tungsten plugs, and copper-based interconnects. If copper is used, additional film material is used as the barrier layer between copper and the dielectric material.

One characteristic of the pad that can be selected is porosity (i.e., pore size and density). Typical pore density is between about 5 and 20% of the polishing pad. The pore density is zero, i.e. the non-porous pad does not allow constant slurry flow and thus presents a problem with removal rate uniformity. Pore size is generally a good measure of pad performance. Approximately 40 microns can be the preferred size for good pad performance. If slurry reduction is not a big concern, larger pore sizes such as 80 microns can be used. Larger pore sizes provide a more constant removal rate, while smaller pore sizes may be used where slurry flow rate reduction is desired.

Another property of the pad that can be selected based on the integrated circuit size is the pad surface architecture, such as a groove, surface configuration, or irregularities. In particular, a large degree of unevenness can be used for small integrated circuit size and large density of integrated circuit larger than low density. Numerous crystals can be made based on the description of integrated circuit size, pattern density and material to be polished, but for crystals below 90 nm, such crystals become extremely complex. For small integrated circuit sizes, pattern densities of less than about 30% are common, while for large integrated circuits, pattern densities of about 50% are common. Therefore, a higher pattern density is associated with a larger integrated circuit size.

In the field of CMP technology, the fields that can be described as "pad engineering" have been studied at a very limited level. Pad engineering from a general point of view, as well as nano- and micro-scale scales as well as nano- and micro-scale scales for selectively controlling and individually controlling various aspects of the polishing process, such as control of lubricity, uniformity of removal rate, thermal behavior and stress control cm can be explained as the use of basic materials in addition to the scientific concept of a macro size scale. The conventional open pore and closed pore polymer pads currently used in the art have a number of limitations, which are more pronounced in the lower node technology. Many of these limitations can be overcome by the novel "pad engineering" method. Molecular structure engineering of pad microstructures, functional grading of pads, surface engineering in pad design through addition of solid lubricant, interface in pad parallel to polishing surface to have an effect of reducing shear force on substrate to be polished A variety of new pad engineering designs are described, such as the fabrication of low shear integral pads having a plurality of polymer layers, abrasive insert pads, and pads containing transparent regions for endpoint detection.

 3 schematically shows the main components of the CMP apparatus 50. As shown in FIG. Slurry 52 is dispensed onto the polishing pad through slurry dispenser 54 as shown in FIG. Alternatively, the slurry 52 may be led from the bottom of the pad to the surface of the pad. The polishing pad 56 is mounted on a rotatable table 58, and a rotatable table shaft 60 extends from a rotatable table. The substrate 66 is held by the substrate chuck 62 and the substrate chuck shaft 64 extends from the substrate chuck. The arrows indicate the direction of the vector force acting to rotate the polishing pad 56 and the substrate chuck 62 and the substrate (not shown). A downward force is exerted controllably on the substrate chuck 62 through the substrate chuck shaft 64 to provide controllable contact between the polishing pad 56 and the substrate 66.

To understand some of the factors affecting the CMP process, it is useful to understand the Stribeck curve shown in FIG. The Stribeck curve shows the relationship of the coefficient of friction (COF) to the Sommerfeld number (So), and the coefficient of friction (COF) and the number of sommerfeld (So) are given by:

COF = F _shear / F _normal (1)

Where F _shear is the shear force and F _normal is the normal force.

So [= μV / (pδeff)] (2)

Where μ is the slurry viscosity, V is the relative velocity between the pad and the wafer, p is the pressure, and δeff = αR _a + [1-α] δ _groove

Where R _a is the pad average roughness, δ _groove is the pad groove depth, and α is the scale factor, α = A _{up - feature} / A _{flat pad} , where A is the corresponding area.

There are three regions in the general Stryvek curve shown in FIG. In the region indicated by "boundary lubrication ", the polishing pad and substrate are in close contact with the slurry abrasive particles, and the coefficient of friction is kept constant as the number of sommerfeld increases. In this region, a large value of the friction coefficient and the removal rate is obtained. Such consistency is desirable for process stability. Drift in the boundary lubrication zone is the result of changes in the wafer / slurry / pad interface during the CMP process. In the partial lubrication zone, the substrate and pads are separated by an oil film layer having a thickness roughly equal to the roughness of the pad. Since the removal rate in the partial lubrication zone is lower than the removal rate in the boundary lubrication zone, the pad life is increased in some lubrication zones. However, the rate of change of the downward slope to the partial lubrication zone indicates that there is less stability, control and possibility than the boundary lubrication zone. In the hydrodynamic lubrication zone, the removal rate is very low due to the larger film thickness.

An ideal Prestonian plot is shown in Figure 5, and the removal rate (RR) is given by:

RR = K _Pr xpx V (3)

K _Pr is the preston constant, p is the actual pressure between the pad and the substrate, and V is the pad and substrate relative velocity.

Ideally, the Prestonian float is linear as a function of pressure and velocity. The deflection from the ideal linear behavior is due to the slurry flow and the polishing pad tribology. For example, it has been found that when comparing different slurries under certain conditions, some exhibit non-idealistic Prestonian behavior at high pressures. This slurry is referred to as pressure sensitive slurry. Additionally, polishing pad tribology, which is influenced by such parameters as pad hardness, thickness, compressibility, porosity and surface composition, can also be responsible for non-idealistic Prestonian behavior.

CMP pads with low coefficient of friction can be fabricated through a number of novel neopad designs including low surface treatment of the pad and low shear design. The uniformity of the coefficient of friction can be controlled by the pad microstructure through the use of small and numerous hard segments distributed in the urethane matrix. The expansion of the boundary lubrication zone is also directly related to the pad microstructure.

The pad fit can be performed systematically based on the polishing process. Because CMP technology involves a number of parameters, adaptation must be performed according to various aspects that affect the process. Methods according to the material to be polished as well as integrated circuit characteristics for pad conditioning are described below, which may be used systematically to design the pad according to particular requirements. The adaptation scheme of the present invention based on polyurethane / polyurea is based on the ratio of the loss factor to the storage modulus of the pad (tan delta), the loss factor (E "), the storage modulus (E '), The present inventors have found that the control of the factors described above through the selection of suitable materials and through a particular manufacturing process, can be achieved by controlling the critical parameters such as glass transition temperature, hard segment and soft segment distribution, micropore size and distribution. Respectively.

Ⅰ. Polymer formation and control of pad microstructure for polymer pads used in chemical mechanical planarization

A variety of materials have been attempted for use in the manufacture of the customized polishing pads of the present invention. Although the pads are substantially polymeric and have a purpose-sized size and a density of hard and soft regions, another embodiment of the present invention is a pore-forming material, a solid lubricant, an insert abrasive, a one perpendicular to the polishing surface Layers such as layers, in situ grooves, and transparent zones for end point detection can be added in the continuous polymer.

The adapted polishing pad of the present invention is made of a polymer. Examples of polymers that have been attempted for the preparation of various adapted polishing pads described herein include polyurethane, polyurea, epoxy polymer, phenolic polymer, polycarbonate, polyamide, polyimide, polyester, polysulfone, , Polyacrylate, polystyrene, polyaryl ether ketone, polyethylene terephthalate, polyvinyl, polypropylene, polyethylene, polysilane, and polysiloxane. In addition, suitable polymers for the various adapted polishing pads described herein may be copolymers, blends, composites, networks, composites, fusions, laminates of those selected from the exemplified types of polymers. Other polymers suitable for use in the pad may be used, as will be apparent to those skilled in the art.

The manner in which these materials are used requires an understanding of the relationship between the structures of the polymers and the resulting physical properties of the polymeric materials used in the pads. Examples of such properties include, but are not limited to, hardness, toughness, porosity, compressibility, and the like.

For example, polymers having significant scientific, engineering, and commercial performance for CMP polishing pads include polyurethanes, polyureas, and copolymers thereof. Such polymers may be prepared using starting materials such as isocyanates, polyols, polyamines, as well as chain extenders and crosslinkers. The reaction of an isocyanate functional group with an alcohol forms a urethane bond that is the basis for the polyurethane polymer. The reaction of an isocyanate functional group with an amine forms the urea bond, which is the basis for the polyurea polymer. For the polyurethanes, at least diol and diisocyanate monomers are required for the polymerization reaction and, alternatively, at least three hydroxyl or isocyanate groups in the polyol or polyisocyanate provide the reaction site for cross-linking. For polyureas, at least diamine and diisocyanate monomers are required for the polymerization reaction, and alternatively, three or more amine or isocyanate groups in the polyamine or polyisocyanate provide a reaction site for cross-linking. Examples of cross-linking agents that react with hard-locks or amine groups include diisocyanate cross-linkers such as toluene-diisocyanate (TDI), diphenylmethane-diisocyanate (MDI), and polymethylene polyphenylisocyanate (PAPI). The type of cross-linking agent and the degree of cross-linking of the polymer chains can affect material properties such as, for example, hardness, toughness and porosity. The size and molecular weight of hydrophilic molecules, such as polyamines and polyols, affect material properties such as flexibility, melting point, and surface energy.

Allows control of hardness and mechanical properties and has a high storage modulus (E ') and a loss modulus (E ") and has a low thermal change, a glass transition temperature (T _g ), a KEL value, Polyurethane and polyurea having a ratio of change (DELTA E '), compressibility, loss factor to storage modulus (tan delta) can be used for pad fabrication.

Casting / Molding, Pad Materials and Microstructure Control

Various methods for casting and molding are suitable for making various polishing pads adapted as a single structure. Some exemplary manufacturing methods for casting and molding a polishing pad as a single structure that allows for spatial control of the physical properties designed within the pad are described below.

Liquid casting of polymers

Liquid casting of the polymer can be used to make the pad for CMP. Liquid casting is a manufacturing technique that can be suitable for manufacturing polymeric components from the simplest designs to entangle polymeric parts. Polymer disc-like shapes can be made using this technique, and thus polymer pads for chemical mechanical planarization can be made using liquid casting. Liquid casting allows spatial control of the pad material properties during manufacture and can therefore be a suitable choice for making pads for CMP. In using this method to make a polymer pad for CMP, a mold with the appropriate dimensions is first made. In addition, liquid casting can be performed to produce a CMP pad in which grooves can be formed in the pad using two possible options of X-SITO or In-situ. Generally, X-situ groove formation is used in the industry. However, this method is very expensive. The in situ groove mold may be adapted to provide a groove in the pad once the polymer is hardened and hardened. Depending on whether the polymer is cured in place, a suitable material is injected into the mold. If the polymer is not already cured, appropriate monomers, crosslinking agents, pore-forming agents, initiators, and accelerators are added to the mold and the reaction is terminated after the desired temperature is reached. When using liquid casting, a first section or layer is injected and cured once, and a second section or layer can be injected if necessary. In addition, in the liquid casting method, an insert abrasive as well as a solid lubricant may be added to the polymer blend to achieve the desired abrasive performance, as described below.

Multiple injection molding

Another method for making fitted pads is known as multiple injection molding. Multiple injection molding is a sequential process in which two or more polymeric materials are used, and each material is injected into the mold at different times. This method can be used to form pads having different areas across the pad as well as adapted pads having two or more layers. This method can also be used to achieve a pattern of any spatially designed polymeric material from a circular pattern that is most simply defined in a single layer or a plurality of layers to a most complex and non-uniform pattern.

Multiple live feed (or in situ) injection molding

Molds containing a plurality of in situ injection ports may be used to make the fitted pad. In this method, a mold having at least two ports, which are generally independent, is selected to inject the polymer. Often at the same time to fill the mold, at least two different polymers are injected through the port during the same injection step. Depending on the desired spatial variation for the adapted pad, fluid flow and heat transfer calculations are performed to select the appropriate injection time and injection flow rate for the different polymers and materials being fed into the mold. In this manner, it is possible to produce a conformed pad having two or more layers as well as different areas of the polymer material across the diameter of the pad.

Reactive Injection Molding (RIM)

Special polymer systems (e.g., polyurethane) may be applied to the molding step using reactive injection molding techniques. In this molding process, instead of injecting the pre-synthesized polymer, the monomer mixture of the component and the appropriate crosslinking agent as well as the promoter and chain extender are added and the final mixture is polymerized during molding. A plurality of pods may be used to create a fitted pad whose chemical structure changes over different regions of the pad. A plurality of ports may be used to inject two or more monomer units (and corresponding chain extenders) as well as other selected materials such as bubblers, solid lubricants, and insert abrasives. This can give a functional grade of the chemical composition and mechanical and physical properties of the polymer. By adding different materials to the mold differently, the method can be used to produce a conformed pad whose properties change substantially or gradually from one layer or region to an adjacent layer or region. In this way, the reactive injection molding can also be used to produce a conformed pad with uniform properties in a plane along the radial and / or thickness direction of the pad.

Lamella Injection Molding

By using a mixture of pre-extruded polymers, for example in layers, in an injection molding manner as described above, a polishing pad adapted to be spatially varied in properties can be produced. This method of producing a simple physical mixture of polymers is directly and easily applied to varying requirements depending on the manufacturer. The spatial property changes depend on the mechanical and physical properties of the individual polymers as well as other selected materials such as solid lubricants or intercalated abrasives that can be added onto the continuous polymer. This method can be used to generate fine-scale grading in horizontal or vertical regions or layers.

Injection molding of gas to produce pads with micropores.

One method for producing a conformed pad having micropores in one or more sections of the pad includes injecting a gas during the injection molding step to change porosity in the adapted polishing pad. May be dispersed and injected into the mold from different ports at different flow rates to achieve a spatial distribution of the gaseous components within the pads. The produced pads may be different in amount of gas contained at different points and thus a planned change in hardness and / or density can be achieved.

Microcellular (Mucell molding)

In this technique, a polymer fluid is mixed with a gas to form a solution mixture. Using two or more solutions with different chemical properties (i.e., starting materials of different chemical properties, such as two different polymers) will result in a spatial change in physical properties.

One shot and two shot polymer synthesis technology

The manner in which the polymer is prepared prior to molding or casting has an effect on the polishing pad properties and consistency. For example, two known methods for forming polyureas and polyurethanes are known as one shot and two shot techniques. In the one shot technique, all reactive components (e.g., monomers, chain extender, crosslinker) can be reacted together. This approach is difficult to control due to factors such as local variations in the concentration of the reactants and localized non-uniform temperature changes that will widely vary the properties of the polymer product. In the two-shot technique, an isocyanate is pre-reacted with a polyamine or polyol chain extender in the first step to form a high molecular weight prepolymer. This functional prepolymer is then also reacted with a polyamine or polyol curing agent and / or chain extender to complete polyurea or polyurethane formation. This method is more easily controlled, but often requires a higher processing temperature of about 100 ° C. When a material with high consistency is required, a process capable of giving consistency is preferable.

CMP pad synthesis

In this study, the uniformity of the size, density and type of hard areas in the CMP pad can be controlled through the selection of appropriate relative concentrations of polyurethane and polyurea in the final product. To-shot technique can be used. In the first step, poly or die functional isocyanate prepolymers are synthesized or obtained from commercial suppliers. The distribution of the molecular weight of the isocyanate prepolymer enables a uniform size and distribution of the hard regions in the pad. In the second step, about 60-80 wt% of the synthesized or commercially available isocyanate prepolymer is reacted with about 1 to 15 wt% of one or more of the polyamine and polyol chain extender to complete the polyurea / polyurethane formation And about 5 -25 wt% of one or a mixture of polyamine and polyol curing agents. In the second step, about 0.1 - 3 wt% stabilizer may be added to prevent ultraviolet deterioration, about 0.1 - 5 wt% of pore-forming agent may be added to form micropores, About 0.1 - 20 wt% of a solid lubricant and about 0.1 - 10 wt% of an insert abrasive may be added. In some cases, the chemical composition of the polyol used as a chain extender is the same or similar to the polyol used in the synthesis of the isocyanate prepolymer.

As a result, a uniform distribution of the size, type and density of the hard regions in the nano-micron size range can be achieved. The individual hard segment segments contain zones for urethane or urea linkages in each polyurethane and polyurea formation. The hard regions may be comprised of one or more individual hard segments. The type of hard region depends on the relative concentration of urea and urethane constituting the hard region. The density of the hard areas can be well controlled using planned process control. For example, the length and functionality of the prepolymer can affect the density of the hard region. The size of the rigid region can be governed by the relative amount of urethane relative to urea because urethane has a single hydrogen bond while urea has two hydrogen bonds and hydrogen bonds with other regions around the urethane or urea bond Thereby making it possible to generate a block of a larger hard segment and to increase the size of the hard segment. The size of the individual rigid segments or segments containing the rigid regions can be controlled by adjusting the size of the isocyanate monomer used to synthesize the isocyanate prepolymer. For example, a large monomer can form a large hard segment, and therefore a combination of large hard segments can produce a larger hard segment. In addition, the temperature at which the polymerization takes place can affect the size and density of the hardened region. At higher reaction temperatures, smaller areas can be formed, increasing the density of the areas, and vice versa. As described above, uniform distribution of the size and density of the hard regions can be achieved through strict control of the molecular weight distribution of the polymer components. Strict control of the temperature distribution in the reaction vessel and mold is also important to achieve a uniform distribution of the size and density of the hard regions, since the temperature affects the size and density of the hard regions. Generally, the ratio of polyol to polyamine is about 20% to 40% polyamine and about 60% to 80% polyol. Generally, the number of hard chain segments per region is from about 1 to about 20. This distribution of the size and density of the hard regions allows a flattened and extended Stavey curve in the boundary lubrication region. Thus, the density and size of the rigid region can be varied over different areas of the pad to achieve an adapted abrasive functionality.

The consistency in the type, size and density of the hard regions allows uniformity in bulk properties. The more consistent and spatially uniform the type, size and density of the hard regions, the greater the consistency in the tribological properties. For example, the thermal properties can be better controlled by using blocks of uniformly alternating spaced polyamine / polyols, while the random distribution exhibits local differences upon heating.

Polishing pads for CMP can be manufactured individually. In pad manufacturing, all pad materials are divided into two treatment batches. The first group of raw materials includes isocyanate prepolymers, abrasives, lubricants, and pore formers such as microballoons or gases. The second treatment group includes a curing agent, a UV stabilizer and a mixture of polyol and polyamine chain extender. The first treatment group is first mixed at a temperature between about 80 [deg.] F and 100 [deg.] F and a vacuum to remove air that can be trapped in the mixture as the porogen is added and achieve homogeneity. The first treatment group is then heated to the required temperature between about 120 < 0 > F and 200 < 0 > F. The second treatment group is maintained at approximately room temperature and mixed for approximately 15 minutes. The first and second treatment groups are then added together in the correct amounts. Liquid casting is used to form the pad. Thus, after mixing, the material is injected into the top of the rotating mold at a temperature between about 150 [deg.] F and 220 [deg.] F. The temperature uniformity of the mold allows a uniform distribution of the type, size and density of the hard regions in the pad and allows uniformity of tribological properties. The pads are also formed by a compression centrifugal casting, vacuum casting or pressure forming method described below.

For compression centrifugal casting, a mixture of the first and second treatment groups is injected into the mold. The mixture is left to react for about 2 to 3 minutes. The mold is then covered by a flat stainless steel plate and placed in a compression molding machine. Compression is performed at a temperature of about 200 ℉ - 300 ℉ and at about 100,000 psig. After compressing for about 10 minutes, the pad is separated from the mold. The pad is then cured at about 100 [deg.] F to 200 [deg.] F for about 6 to 12 hours. The temperature uniformity of the compression molding machine makes it possible to maintain a uniform distribution of the type, size and density of the hard areas in the pad and allows uniformity in tribological properties. For example, a uniform temperature can be maintained by bringing the outside of the mold into contact with a fluid maintained at a constant temperature.

The compression centrifugal casting has pores having an elliptical shape. Elliptical pores can act as microgrooves during polishing and thus eliminate the need to introduce higher groove densities. The elliptical orientation is coplanar with the polishing surface and has an aspect ratio of at least about 1: 2 or higher. When distributed evenly, elliptical pores act as periodic discontinuities on the polishing surface to create natural microstructures, which improves the efficiency of inspection (i.e., reduces inspection time). The elliptical pores also act as a slurry reservoir, which prevents slurry loss at the wafer during polishing at very low slurry flow rates. In some instances, it has the advantage of a slurry usage reduction of at least 40% as compared to a spherical pore structure. As another advantage, elliptically shaped micropores can provide stability of friction coefficient and removal rate (maintains a higher removal rate and / or an adjustable removal rate).

After the first and second treatment groups are mixed for the pressure forming method, the mixture is immediately injected into the mold and the mold is placed in the sealed chamber. Next, the enclosed chamber is placed under a pressure of approximately 3 - 10 times the atmospheric pressure and the temperature of the mold is maintained uniformly at about 150 ° F - 220 ° F. This pressure causes the air trapped in the pad to be released. After about 2 to 5 minutes, the pressure chamber is depressurized and the mold is removed. After about 15 minutes the pad is separated from the mold. The pad is then cured at about 100 [deg.] F to 200 [deg.] F for about 6 to 12 hours.

During curing of the individual CMP pads, a surface layer in the range of less than about 2 micrometers is formed on the surface of the polishing pad. This surface layer may be important to protect the pad surface from damage caused during handling of the CMP pad. The pad needs to be inspected before use, and the inspection can be performed using a diamond tester. In some cases, the surface layer for efficient inspection of the pads prior to polishing of the substrate is less than about 2 micrometers.

Some materials used for pad fabrication are described below. These materials fall into several categories. Categories include isocyanate prepolymers and monomers, polyols and polyamine monomers and chain extenders, curing agents (crosslinking agents), stabilizers, pore formers, solid lubricants, and abrasives.

The isocyanate monomers and prepolymers that can be used for the preparation of the pads are listed in Table 1.

The polyol monomers and chain extenders that can be used for the preparation of the pads are described in Table 2.

The polyamine monomer and chain extender, which can be used for the preparation of the pads, are shown in Table 3.

The curing agents that can be used for the preparation of the pads are listed in Table 4.

Stabilizers that can be used for the preparation of the pads are listed in Table 5.

Pore formers that can be used to produce micropores in the preparation of the pad are described in Table 6.

Solid lubricants that can be used for pad manufacture are listed in Table 7.

The insert abrasives that can be used for pad fabrication are listed in Table 8.

A non-limiting portion of the possible combinations of materials described above that can be used for pad fabrication are shown in Table 9, where the first number is the number of representative compounds in each of the above tables, the number in parentheses corresponds to wt% ).

Ⅱ. Controlled porous pad

As described above, one of the exemplary characteristics that affect the performance of CMP in the process of semiconductor wafers is porosity. Control of porosity can be achieved by carefully controlling the distribution of pore-forming agents in the polymeric material and by carefully controlling the uniformity of the temperature during the manufacturing process. Failure to control pad porosity, i.e., failure to control the size, density, and distribution of pores in the pad can affect factors such as slurry delivery and abrasive dispensing, thereby increasing the removal rate (RR) The performance of the pad and the number of variations in the wafer. Additionally, the pads produced without controlling porosity also showed uneven shear forces across the polishing surface and thus were found to exhibit non-uniform coefficient of friction over the entire treatment range. Unevenness in the shear force can introduce defects on the substrate that affect the planarization efficiency and lead to a reduction in product yield.

A variety of adapted polishing pads described herein are fabricated to have a uniform porosity in the pad in relation to porosity, i.e., the distribution of pore size, pore density, and porosity. Figure 6 schematically shows how a pore-forming material or pore-former can be used to produce a uniform pore size, pore density, and distribution in a matrix. In this example, the pore-forming material or pore-former has different properties under different external conditions such as temperature or pressure. Initially, a pore-forming material or pore-former may be added to the matrix, and then the pore-forming material or pore-former may be expanded to a predetermined pore size by applying heat uniformly. The distribution and density of the pores can be controlled by the amount of pore-forming material or porogen that is added to the matrix, wherein the matrix is a polymer.

In some variations of the adapted polishing pad, the pore size range is from about 20 nm to about 80 microns, and in other variations of the adapted polishing pad, the pore size range is from about 50 nm to about 15 microns, In another variation of the pad, the size may be between about 100 nm and about 10 [mu] m. The porosity density modification of the adapted polishing pad is determined by the concentration of pore-forming material or porogen that is added to the polymer prior to casting or molding. It is contemplated that the pore density can be varied such that the polishing pad has a pore density between about 1% and about 20% of the total pad.

Various materials can be useful for forming pores in a controlled manner within the polymer matrix during manufacture. Some exemplary materials include foam formers, chemical foams, supercritical fluids, block copolymers, micelles, and porogen materials described below.

A. Polymer A hollow microballoon (microballon)

The polymeric hollow-component microcomponent material is a spherical ball generally ranging in size from 10 to 100 micrometers made of polymer. For example, inorganic materials such as Expancel, PVDF, phenolic resin, and silicate and zirconate having a gas such as isobutane gas contained in a spherical capsule may be used. When this material is added to the polymer body before molding, the gas inside it expands to a predetermined size through the control of the heat. These hollow microcomponents are available in both expanded and unexpanded forms and one of these forms may be used for pad formation. In the expanded form, the hollow micro-component is pre-expanded and does not cause a change in size during the actual polymer processing. In the unexpanded form, the hollow micro-component expands during the pad elevation process. A large amount of pore size is controlled by using such a microballoon material. The pore density is controlled by the amount of microballoons added.

B. Chemical foaming agent

For example, hydrocarbons, sodium hydrogencarbonate, which generates carbon dioxide during heating, and chemical blowing agents such as azo diboronide and oxybisbenzenesulfonyl hydrazide, which produce nitrogen gas, can be added to the polymer treated group. When heating the polymer, these chemical blowing agents decompose to provide a gas that will form pores in the molded part. Another example of a blowing agent includes a solid that can be dissolved using a solvent after molding of the polymer.

C. Supercritical Fluids

In the Mucell method, supercritical fluid is dissolved in the polymer feed to produce a single-phase solution. Once the polymer feed is allowed to cool in the mold, the gas forms microbubbles in the size range of 0.1 to 10 mu m.

D. Micelles

The micelle structure can be introduced into the polymer feed stream. The micelles (liquid or solid) can then be dissolved using selective solubility using a selectively soluble solvent for the micelle, such as hexane, leaving a porous zone within the polymer matrix. Materials such as dodecylphosphocholine, C ₁₆ SO ₃ Na can be used to introduce micelles into the polymer.

E. Porogen materials

Porogen materials can be used to produce pores in the polymer matrix. The porogen material is made of another polymer, for example polystyrene with a low decomposition temperature. After a predetermined amount of porogen material is added to the polyurethane matrix and the pad is formed, the porogen material can be removed by heat treatment of the entire pad.

Ⅲ. Functionally graded pads

The adapted polishing pad described herein is a functionally graded polishing pad. These pads consist of a conformal polishing pad with a polishing surface for polishing the substrate, which is one, substantially planar, and includes at least two regions having different physical properties. At least two regions may have boundaries or discontinuous boundaries formed by a mixture of polymeric components. The at least two regions each comprise a polymeric material having a different composition and the region between the regions may comprise a mixture of polymeric materials of different compositions.

Figure 7 schematically shows two radially symmetrical graded pads in which two different polymer compositions, one for each region, are used, the first outer annular ring of the pad is of the centrifugal liquid casting type . The center of the pad ring is then filled with a second polymeric material. Since two different materials are used, there are two different regions or regions of the pad that have different physical properties. Proper bonding at the interface between the two materials requires a selection of materials that can be compatibilized with respect to each other.

In addition to functional grading, changes in functionally graded pads can have the same or different pore sizes and densities in different polymer zones. 8A is a schematic view of a functionally graded pad having a rigid inner zone consisting of a short chain prepolymer and a soft outer zone comprised of a long chain prepolymer. Figure 8b is a schematic diagram showing pores of different sizes that can be formed in different zones with the same pore density. Figure 8c is a schematic diagram showing pores of the same size that can be formed in different zones with different pore densities. The at least two zones may each comprise a polymeric material of different composition. If the polishing head (retaining ring) exerts a greater pressure on the outer zone than the pressure exerted on the inner zone of the polishing pad in order to cause a high removal rate in the zone abraded by the outer edge, Lt; / RTI > This method of compensating for uneven pressure distribution can achieve a reduction in edge yield loss and a minimization of pattern density effects. The functional grading of polymeric materials can be guided by the grading of mechanical properties (hardness, compressibility, pore size and pore distribution) and can be used to equalize the unevenness of pressure distribution.

9 can be made using a manner in which a set of irregular patterns with selected areas of the type such as ellipses 202,204, 206 and flag shapes 208 are functionally graded on the polishing pad 200 Which schematically illustrates a more complicated pattern having a more complex pattern. In each marked zone, the individual polymers may be different polymers of the abovementioned kind or at least two different polymers. Likewise, such a pattern can be achieved using a suitable mold shape.

Figure 10 is a schematic diagram of an example of a functionally graded and adapted polishing pad that is a continuously graded pad wherein a first polymer is injected from the outer periphery 212 of the mold while a second polymeric material Lt; RTI ID = 0.0 > 214 < / RTI >

The graded pad may have a selected polymer and / or composition to provide different values for the restitution factor in different regions of the pad. The outer ring or outer annular portion of the polishing surface of the circular pad may have a higher restitution factor than the interior of the pad to provide more uniform wafer polishing. The outer annular top may be formed by increasing the amount of hardener and / or by changing the chemical composition of the polymer formulation (e.g., changing the type of hardener when forming the outer annular top) compared to the formulation for the inner portion of the pad Which will be described later. In this manner, the hardness of the polishing surface may remain substantially or essentially unchanged, but the pad may provide improved flatness and / or improved device yield from the wafer when the pad is operating. The restoration factor can be estimated by measurement using Bashore Rebound% and the Bashore rebound of the polishing pad can be about 0.05 - 0.6 for the pad of the present invention. Another evaluation of the restitution factor can be performed using a compression set test at approximately 158 DEG F (70 DEG C) and 22 hours. Also, the value obtained from the compression set test for the pad of the present invention is between about 0.05 and 0.6.

Likewise, graded pads may have selected polymers and / or formulations to provide different values for compressibility in different regions of the pad. To provide a more uniform wafer polishing, the outer annular top or outer ring of the polishing surface of the circular pad may have a higher compressibility than the inner portion of the pad. The outer annular top may be formed by increasing the amount of hardener and / or by changing the chemical composition of the polymer formulation (e.g., changing the type of hardener when forming the outer annular top) compared to the formulation for the inner portion of the pad . Also in this manner, the hardness of the polishing surface may remain substantially or essentially unchanged, but the pad may provide improved flatness and / or improved device yield from the wafer when the pad is operating. Compressibility is defined as the reciprocal of the bulk modulus. Volume elastic modulus is defined as the magnitude of the pressure required to cause a unit volume change.

In graded pads as described above, for example, the properties vary across the radius of the paddle from the axis of rotation of the pad and are often at least about 75% of the volume of the pad formed or the internal surface area of the polishing surface, The volume of the pad having the second value or the remaining amount of the surface area of the polishing surface. For example, the outer edge of the belt polishing pad or the outer perimeter of the circular pad is more sensitive to movement from, for example, equipment vibrations, edge effects, and greater torques, Lt; RTI ID = 0.0 > a < / RTI >

IV. Low-shear integral pad

Another suitable polishing pad is a low shear integral pad. The adapted low shear abrasive pad is a multi-layer or integral pad made of at least two materials so that the interface between the two layers acts as a stress sink to reduce the coefficient of friction at the pad / substrate interface. The materials on both sides of the interface may be the same or different. Except that layers of material are injected one at a time, the interface is formed using the manufacturing method described above. After implanting the first layer of material, the material is allowed to cure for 0.5 to 2 minutes before the second layer is implanted. After the last layer is injected, the entire pad is compressed, vacuum formed or pressure molded as described above.

The pads having a plurality of layers may be of a single configuration and the plurality of layers may be covalently bonded to each other through an integrated interface, or the pads may be formed by squeezing or attaching pre-cured layers to each other. Many of the pads described herein are monolithic and thus have an integrated interface through which layers are covalently bonded to one another. The single pad may have additional layers added to the pad, such as a double-sided tape to attach the hardened pad to the platen of the chemical mechanical polishing machine, and such additional layer does not add significant impact to the performance characteristics of the pad in use.

11A is a schematic diagram of a two-layer, adapted low shear stress pad 100 having layers 102 and 104 and an interface 103. Fig. An integrated pad with one interface has two layers, a pad with two interfaces has three layers, and an N-layer pad has N-1 interfaces.

This can be clearly seen in the pad shown in FIG. 11B. This exemplary pad 300 has five layers of materials 302, 304, 306, 308, 310 and four interfaces 305, 305, 307, 309. The material layers 302, 304, 306, 308, 310 may be made of the same or different materials, and physical properties and properties such as porosity and grade may be the same or different. Four interfaces 305, 305, 307 and 309 are formed to act as stress sinks, thereby effectively reducing the shear force and the coefficient of friction.

This effect of the interface acting as a stress sink is schematically shown in Fig. In schematic view 100, the shear force S at the pad / substrate boundary 104 is orthogonal to the normal force N applied to the substrate 102, because it acts as a stress sink in the single layer pad 106 This is because the selected interface does not exist. In schematic 300, the shear force S2 is determined to be between the pad layers 308 and 308 when the polishing substrate 302, the pad with the layers 306 and 308, is designed to have the interface 307 Lt; RTI ID = 0.0 > 307 < / RTI > S1 is considerably smaller than S because the shear force S1 at the pad / substrate interface 304 is reduced due to the shear force S2 at the interface 307 of the low-shear integral pads 306, 308.

Ⅴ. Subsurface Engineered Pads

Another suitable polishing pad is a sub-surface treated pad. The various sub-surface treated pads described herein have properties imparted through a combination of structural characteristics designed within the polymer pad in combination with a dispersion of the solid lubricant within the pad depth of at least about 1% from the polishing surface. The use of a pad with a solid lubricant dispersed in a polymer matrix effectively minimizes the coefficient of friction without sacrificing removal rate. Solid lubricants are materials such as powders or foils used to protect against damage during relative movement and to reduce friction and wear. Some desirable properties of solid lubricants are that they are thermally stable, chemically inert, inert, mechanically stable, and hardness not exceeding about 5 Moh's. Solid lubricants meeting these criteria generally have better performance at high loads and speeds, higher resistance to degradation, and higher stability over extreme temperatures, pressures, radiation, and other reactive environments, It has advantages. There are many classes of solid lubricants including inorganic solids, polymers, soft metals, composites. In addition, the sub-surface treated pads can be used in combination with the functionally graded pads, porosity controlled pads, and low shear force pads described above.

In addition to the general properties of the solid lubricants described above, the solid lubricants contemplated for use in the pads have a coefficient of friction between about 0.001 and about 0.5 and a particle size between about 10 nm and about 50 micrometers. In addition, many differently adapted pads having at least one solid lubricant within at least about 1% of the depth of the pad from the polishing surface can be made. A combination of lubricants may be used instead of a single lubricant.

Examples of inorganic solid lubricants having the above-mentioned desirable characteristics include layered solid such as graphite, graphite fluoride, niobium sulfide, tantalum sulfide, molybdenum sulfide, tungsten sulfide, magnesium silicate hardoxides, hexagonal boron nitride, cerium fluoride . Such a layered solid is a crystalline solid layered in sheet form, and the slip surface appears between the sheets. Other inorganic solids suitable as solid lubricants include calcium fluoride, barium fluoride, lead oxide and lead sulfide. Although not structurally layered, these solid lubricants have a surface that easily slides along each other at the molecular level, thereby exhibiting lubricity at the macroscopic level.

Examples of polymeric solid lubricants are: 1) polyhalogenated hydrocarbons such as PTFE and related members, 2) polyamides such as nylon 6,6 and related members, 3) PEK (polyether ketone), PEEK Ether ketone), PEKK (polyether ketone ketone) and PEKEKK (polyether ketone ketone ketone), 4) poly (p-borazylene) or poly (p-vinyleneborazylene) ≪ / RTI > and the like. These polymeric solid lubricants have generally low surface energies, are stable like unfused dispersions, have low coefficient of friction, and are thermally and chemically stable. For example, PTFE has a substantially small static and dynamic coefficient of friction of about 0.04, is known to be chemically inert, and is stable up to about 260 ° C. Polymeric solid lubricants like calcium fluoride, an inorganic solid lubricant, have a surface that easily slides with one another.

Other solid lubricants contemplated for use include various materials having suitable properties formed into nanospheres, nanotubes, or other nanoparticle structures useful for lubrication. For example, these nanoparticles of carbon are known as buckminster florin or "buckyball". Various solid lubricant materials, such as, for example, molybdenum sulfide, inorganic materials such as tungsten sulfide, or polymer materials such as PTFE or boron nitride polymers, can be made into nanostructures useful as solid lubricants. Since such structures generally have nanopores, they may include other solid or liquid lubricants that produce solid lubricants of various properties. Additionally, composites and fusions made of inorganic and polymeric solid lubricants as well as solid lubricants made of polymers, mixtures, networks, composites, fusions of polymers and copolymers are also possible.

The adapted sub-surface treatment pads may be used for all processing steps in copper CMP, including bulk, soft landing, and barrier removal steps. In particular, the effect of a single pad approach to copper CMP reduces consumable costs to make it desirable in terms of cost for technical processes below 90 nm.

As shown in Fig. 13, copper has a very high deformation before fracture occurs. In addition, copper causes considerable plastic deformation before it is destroyed. In the case of strain induced defects in the dielectric, the natural bonding properties of the material lead to brittle fracture. Such brittle fracture occurs at significantly lower strain values, for example, less than 2%. Due to the high plasticity of copper, some problems need to be solved for copper CMP. The first problem is the selective elongation of the material in the stressed zone causing plastic deformation. The plastic deformation thus induced is a permanent deformation due to long-term stress. Under conditions in which selective elongation occurs due to contact between the polishing pad and the copper region, this region undergoes plastic deformation and has different properties from the inner copper region. The second problem is the local strain hardening of copper that appears before fracture. This problem of copper elongation and strain hardening is predominant due to the constraints of copper in vias and trenches. Finally, depending on how the pad interacts with the copper layer, the copper residue may be left after CMP is terminated and the bond introduced into the substrate being polished. Minimizing stress build-up can be achieved by lowering the coefficient of friction by lowering the effective shear force acting on the wafer / slurry / pad interface. For CMP, it is generally known that uniformity of pad properties such as pad count, pore size distribution, and chemical composition of materials is important in providing CMP processing that operates in a highly stable boundary lubrication region. Additionally, a significant reduction in shear force is required to reduce or eliminate stress induced defects in order to achieve uniformity for copper CMP processing. In order to reduce the shear force, a high level of lubrication uniformity is also required. For various conformed sub-surface treated polishing pads as described herein, the use of a pad with a solid lubricant dispersed in a polymer matrix in the pad depth of at least about 1% from the polishing surface can effectively reduce shear forces without sacrificing removal rate Thereby minimizing or reducing the strain hardening of copper thereby.

VI. A suitable pad with abrasive inserts

Unlike the commercially available "fixed abrasive pad ", the abrasives in the pads into which the abrasive materials described herein are distributed are distributed over the polymer matrix rather than distributed on the surface. If multiple layers of pads are desired, the abrasives to be inserted may be distributed in all layers or distributed in some layers. The advantage of a pad with abrasive inserts over a fixed abrasive pad is the stability of the treatment over time. The pad wears during polishing. In the case of a pad in which an abrasive is inserted, the same polishing conditions can be expected since the abrasive distribution is designed to be well controlled within the pad depth. In contrast, commercially available fixed abrasive pads exhibit gradual wear in shape, size and distribution density as the polishing process proceeds. This represents a non-uniform polishing rate, non-uniform process control due to the need for frequent pad replacement and high cost.

Ceramic or glass particles (alumina, silica, ceria) may be inserted into the pad for CMP. These particles may have a size of about 100 nm to 30 [mu] m depending on the desired performance. In some cases, the adhesion between the particles and the pad matrix can be minimal. This allows the particles incorporated in the pad to be exposed and released into the slurry. These polymer pads enable the polishing action without the use of a slurry containing an abrasive. In practice, the entire polishing process can be carried out using pads with distilled water and abrasive inserts.

As known in the art, a new class of abrasive materials, called nano-abrasive particles, has recently been developed. These particles range in size from a few tens of nanometers to a hundred nanometers. The polymer pads can be functionalized by incorporating these nanosilver slurry particles directly into the pads using the manufacturing method described above. A variety of nano abrasive particles including ceramics and glass such as zirconia, silica, ceria, and carbon nanotubes (plurin rings) as well as clay particles can be used.

The distribution of the abrasive material inserted into the pad can be adapted to the pattern density of the particles through functional grading of the self abrasive in different areas on the polishing surface. Grading can also be achieved through the grading properties of abrasives, such as the size distribution, density and shape of the abrasive. This grading can be done individually or in combination with other means for grading (i.e., different porosity and pad material).

The above-mentioned micron scale particles and nano abrasive can be added to a polymer in a dry state or to a suitable liquid such as a solvent. Such particles may optionally have a group such as an oligomer or polymer group attached to one of the polymers or to the surface that selectively or preferentially allows the incorporation of the particles into the polymer if a discontinuous phase is formed on the selected polymer. For two or more phases of polymer, the group bonded or attached to the surface of the particles may be of sufficient similarity to one phase and the abrasive grains are completely different from the other phases that gather on the desired phase when the polymer melt coagulates . The group bonded or attached to the particle surface can also be selected to be different from the polymer in which the particles are placed. This helps release abrasive particles from the polymer as the pad wears to expose new particles.

Block Copolymers for Making Self Abrasive Pads

A double block copolymer can be used that can serve as a matrix and another block can serve as an abrasive material to make a self-abrasive pad. The block is selected to provide a predetermined continuous phase, abrasive phase and immiscibility so that the abrasive phase is formed within the continuous phase. In one embodiment, the pad may be configured so that the ratio of the other block (continuous matrix) is higher in a state where one block is discontinuous. The discontinuous block can be selected to act as an abrasive for the material to be polished. Mineral or metal particles may be optionally added in the block to make the second phase abrasive. For example, an abrasive material may be chemically bonded to some or all of the monomer molecules in a discontinuous phase of the block copolymer when formed in the pad, or the abrasive particles may be incorporated into the polymer melt. The abrasive particles have one or more characteristics (e.g., surface action and thermodynamic conditions) that facilitate selective incorporation into one of the phases. For example, the abrasive particles may be selected to exhibit a higher abrasive particle concentration than the continuous phase and discontinuous phase. Mixtures of abrasive particles may also be used. In some instances, the type of each abrasive particle in the mixture appears to be higher in concentration than in the other phases (although the particles can be selected so that there is a higher concentration in the continuous phase, preferably a discontinuous phase) . However, it is preferred that one or more types of particles in a continuous phase exhibit a higher concentration, and / or that one or more types of particles are substantially homogeneous across both phases, such that one or more types of particles in the discontinuous phase exhibit a higher concentration A mixture of particles can be selected to be distributed.

It may not be necessary to incorporate the abrasive particles into one of the blocks since one of the blocks may be the abrasive itself. As a low proportion of material, a particular silicon block can be incorporated into a portion of the copolymer and the carbon backbone can act as a high proportion of material. When a pad is used for polishing, the silicon portion of the polymer may be exposed and act as an abrasive. Such abrasive materials made of silicone polymers can be tailored to have similar consistency with silica particles that are currently used as abrasive particles in many techniques.

The polymer incorporating the previously described nanosized abrasive particles may be a polymer forming one continuous phase, or alternatively the polymer may be a special block copolymer that forms a discontinuous phase as described above. Nanoparticles can be selected to provide uniform dispersion or selective accumulation in a discontinuous phase or a continuous phase, and mixtures as described above can be used.

Flame Spraying

Another technique that can be used to provide a polishing surface in addition to the manufacturing techniques described above is the flame spray technique used to make the polymer coating on the pad. The flame sprayed polymer may have ceramic or glass particles that are incorporated into the material upon formation of the coating such that a self-abrasive surface can be formed. The pads on which the polymer and accompanying abrasive particles are sparkled are generally polymers, such as polyurethane or polycarbonate, which have no discontinuous phase. The pad may be as desired having at least one discontinuous phase as described above. In this case, the flame sprayed polymer / abrasive layer is first worn, and once this layer is worn, the pad placed under the polymer / abrasive layer is worn. For example, when the initial wear rate must be different from the later wear rate when worn or very weary layers on the wafer, or wearable layers or layers with different properties, or using pads, this type of Configuration can be used.

VII. In Situ Groove Pad

The grooves in the CMP pad prevent hydroplaning of the wafer being polished over the surface of the pad and contribute to providing a distribution of slurry to the pad surface and contribute to ensuring that sufficient slurry reaches the interior of the wafer, It is believed to provide a channel for removing abrasive debris from the pad surface in order to control defects and to contribute to controlling the compliance and local stiffness of the pad in order to control the gender and minimize edge effects. Figures 14a and 14b schematically illustrate the effect of grooves on the hydrodynamic pressure generated around the pad / wafer area. For example, FIG. 14A shows the wafer pressure profile (indicated by the area of the triangle marked with an oblique line) when a grooveless polishing pad is used. 14B shows how the pressure around the wafer is released along the groove. That is, the grooves conform to the pressure generated at all groove pitches and contribute to provide a uniform slurry distribution along the wafer / pad area.

Generally, any suitable method of creating an in situ groove in a CMP pad can be used. Unlike the method of generating primarily current mechanical x-groove grooves by nature, the method for creating in-situ grooves described herein has a number of advantages. For example, the method of creating the in-situ grooves described herein is generally inexpensive, takes less time, and requires fewer manufacturing steps. In addition, the method described here is more useful for obtaining complex groove designs. Finally, the method for creating in situ grooves described herein can produce a CMP pad with better tolerance (e.g., better groove depth, etc.).

In one variation, a method for creating an in situ groove includes using a silicon lining placed in a mold. The mold may be made of any suitable material for molding. For example, the mold may be made of aluminum, steel, ultra-mold materials (e.g., metal / metal alloys with extreme smooth edges and extreme tolerances for molding finer features), combinations thereof, It can be a metal. The mold can be any suitable dimension, and generally the dimensions of the mold will depend on the dimensions of the CMP pad to be produced, for example a mold having a diameter of 22 inches and a thickness of 2 inches for a 20 inch pad. In general, the pad dimension depends on the size of the wafer to be polished. For example, the dimensions of a CMP pad for polishing 4, 6, 8, and 12 inch wafers may be 12, 20, 24, or 30.5 inches, respectively.

Generally, the silicone lining is made of a silicone elastomer or a silicone polymer, but any suitable silicone lining can be used. The silicon lining is then embossed or etched in a pattern corresponding to the desired groove pattern or design. The lining may then be adhered to or otherwise held on the mold by gluing or otherwise. The lining can also be placed in the mold before it is patterned. Lithographic techniques for etching patterns in silicon lining contribute to providing more accurate groove sizes. For example, C. Dekker, Steriolithograhpy tooling for silicone molding, Advanced Materials & Process, vol. 161 (1), pp 59-61, Jan. 2003; D. Smock, Modern Plastics, vol. 75 (4), pp 64 - 65, April 1998. For example, grooves in the micron to sub-micron range can be obtained. Also, grooves having large dimensions in the millimeter range can be obtained relatively easily. In this way, the silicon lining serves as a "molding pattern ". However, in some variations, the mold may be patterned with a complementary groove pattern. In this way, the mold and lining, or the mold itself, can be used to create a CMP pad groove design.

FIG. 15 shows a cross-sectional view of an exemplary silicon lined processed mold 200 as described herein. Shown in the figure are an upper mold plate 202, a lower mold plate 204 and a silicon lining 206. The silicon lining 206 has an embossed or etched pattern 208. Although the silicon lining 206 is shown in FIG. 15 along the top mold plate 202, it need not be. In fact, the silicon lining 206 may also be attached to or held on the lower mold plate 204. The silicon lining can be attached or held on the mold plate using any suitable method. For example, the silicone lining may be adhered or held on the mold plate with an adhesive, tape, clamp, press fit or otherwise.

Using this method, the CMP pad can be formed of a thermoplastic or thermoset material, or similar material. In the case of thermoplastic materials, a melt is typically formed and injected into a mold that is silicon lined. In the case of thermosetting materials, the reaction mixture is typically fed to a mold that is silicon lined. The reactive mixture may be added to the mold in one step or two or more steps. Regardless of the material used, however, the final shape of the pad can generally be obtained by hardening, cooling or solidifying the pad material prior to removal from the mold. In one variation, the material is polyurethane and a polyurethane pad is made. For example, the polyurethane pellets can be placed in a mold that is melted and silicon lined. The mold to be silicon lined can be etched with the desired pad pattern as described above. The polyurethane is cooled and then taken out of the mold. The pad has a pattern corresponding to the pattern of the mold to be silicon lined.

The advantages of creating in situ grooves using these silicon lining methods are numerous. For example, if there is a break in the silicone lining or wear or tear in the silicone lining, the silicone lining can easily be replaced and the silicone lining itself generally has a very long life, have. Similarly, it is easier to remove the pad from the mold to be silicon-lined compared to the pattern engraved in the mold. Therefore, the grooves made using the molds to be silicon lined are more accurate and the damage to the pad during removal can be minimized. In a similar manner, the groove size made using the mold to be silicon lined can be better controlled and better formed. For example, very small dimensions (e.g., lateral and horizontal grooves in the micron to sub-micron range) can be achieved. Better control and better formation of the groove dimensions may be particularly advantageous for special purpose pads such as low K dielectric, Cu removal, shallow trench isolation, system on chip, and the like.

An ingenious groove design is also described. The original groove design was developed based on flow visualization studies. This study contributed to the identification of the flow pattern of the slurry at the top of the pad. In this manner, the preferred trajectory of the groove was calculated. Figures 16a-16c illustrate a non-limiting, exemplary illustration of a suitable groove design for 20, 24, and 30 inch pads. As shown, at a smaller radius value (i.e., near the inner portion of the pad), the grooves are designed with concentric circular grooves and concentric circular grooves extending radially to follow the identified flow pattern. At radially larger values (i.e., near the outer portion of the pad), radially extending grooves that were initially linear near the inner portion of the pad prevent the slurry from flowing out of the pad and also increase the density of the grooves around the periphery And may be a curve. The distribution of the grooves at the polishing surface can also be used to control the distribution of the slurry across the polishing surface, such as by maintaining a constant slurry density across the polishing surface. An increase in groove density near the periphery can be made to maintain a substantially constant groove density across the polishing surface that may be important to maintain uniformity of polishing performance across the surface of the pad. In some cases, as shown in Figure 16A, additional radial grooves that do not extend into the pad to maintain a constant groove density across the polishing surface may be added. Typical groove widths range from about 50 to about 500 microns, and typical groove depths range from about 100 to about 1000 microns.

Such an ingenious groove design can be made by a suitable method. For example, the grooves may be made using the in situ method described above, or alternatively the grooves may be formed by laser lighting or cutting, water jet cutting, 3-D printing, thermoforming and vacuum molding, micro- It can be made using the same X-ray or mechanical method.

A. Laser Writing (Laser Cutting)

Laser lighting or laser cutting may be used to create the ingenious groove design described herein. Generally, the laser cutter consists of a downwardly directed laser mounted on a mechanically controlled positioning mechanism. For example, a sheet of plastic is placed below the working area of the laser apparatus. As the laser passes back and forth across the surface of the pad, the laser evaporates the material to form a small channel or cavity at the point where the laser hits the surface. The groove / cutting portion thus formed is generally accurate and precise, and does not require a surface finishing treatment. Generally, any pattern of grooves can be programmed in the laser cutting machine. For more information on laser writing, see J. Kim et al., J. Laser Application, vol. 15 (4), pp 255-260, Nov. 2003.

B. Waterjet cutting

Water jet cutting can also be used to create the ingenious groove design described herein. This method uses a pressurized water jet (e.g., a high pressure of 6,000 pounds per square inch) to create a groove in the pad. Often, abrasives, such as garnet, that promote better edge tolerance and good edge finishing treatments are mixed into the water. To obtain the desired pattern of grooves, the water jets are typically programmed (e.g., using a computer) to follow the desired geometric path. A further description of water jet cutting is found in Abrasive water jet, Rivista De Metalurgica, vol. 34 (2), pp 217 - 219, Mar - April 1998.

C. 3-D Printing

Another method, 3D printing (or 3-D printing), can be used to create the ingenious groove design described herein. In 3-D printing, the parts are formed into layers. First, the necessary part of the computer (CAD) model is created, and then the slicing algorithm maps the information for each layer. Each layer begins with a thin distribution of the powder spread across the surface of the powder bed. The selected binder material then selectively binds the particles where the article is formed. A portion of the piston and process to support the powder bed is then removed to form the next layer. After the same process is repeated for each layer, the final heat treatment is performed. Because 3-D printing can perform precise, local control of material composition, microstructure and surface texture, a variety of new and unattainable geometric shapes of grooves can be achieved in this way. More detailed information on 3-D printing can be found in Anon et al., 3-D printing speeds prototype dev., Molding Systems, vol. 56 (5), pp 40-41, 1998.

D. Thermoforming and Vacuum Forming

Other methods that can be used to create the ingenious groove designs described herein include thermoforming and vacuum forming. Generally, this method is used only for thermoplastic materials. In thermoforming, the flat plastic sheet is heated and then brought into contact with the mold using vacuum pressure or mechanical pressure. Generally, thermoforming techniques create pads with good tolerances, rigorous design criteria and precise details in the groove design. In addition, thermoforming pads are typically comparable to injection molds, sometimes with much better quality, but at a much lower cost. More detailed information on thermoforming can be found in Rev. M. Heckele et al., Which is incorporated herein by reference. on Micro Molding of Thermoplastic polymers, J. Micromechanics and Microengineering, vol. 14 (3), pp R1 - R14, Mar. 2004.

In vacuum molding, heated plastic is vacuum-sucked into a mold to form sheet plastic into a desired shape. Vacuum molding can be used, for example, to mold a plastic of a predetermined thickness of 5 mm. Significantly complex molding and complicated groove patterns can be achieved relatively easily by vacuum molding.

E. Micro contact printing

Using micro-contact printing, a high-resolution printing technology, grooves can be embossed / printed on top of the CMP pad. This method is sometimes characterized as "soft lithography ". This method uses an elastomeric stamp to transfer the pattern to the CMP pad. This method is a non-photolithographic method for forming and manufacturing microstructures that are convenient, inexpensive and can be used as grooves. This method can be used to create patterns and structures with feature sizes in the nanometer and micrometer range (e.g., 0.1 to 1 micron).

F. Hot stamping and printing

Hot stamping can also be used to create the ingenious groove design described herein. In this method, the thermoplastic polymer may be a hard master (e. G., A metal having an embossing pattern therein, having sufficient strength to withstand high temperatures and allowing the polymer pad to be embossed when pressed into a hard master, Material can be used for hot embossing. When the polymer is heated to a viscous state, the polymer can be molded under pressure. After being made to correspond to the shape of the stamp, it is cured by cooling below the glass transition temperature. Other types of groove patterns can be obtained by changing the initial pattern on the master stamp. Additionally, this method allows the creation of nanostructures that can be replicated on large surfaces using molding of thermoplastic materials (e.g., by making stamps with nanorilyed structures). Such nanostructures can be used to provide local grades / grooves.

VIII. Integrated optical transparent window for end point detection during CMP

There is provided a polishing pad having at least one region that is sufficiently transparent to the wavelength of at least one light used for endpoint detection and a method of making such a polishing pad. The polishing pad may be used in an optical detection or monitoring method in any suitable chemical mechanical planarization system. For example, if the polishing pad is mounted on a rotatable plate as described in U.S. Patent No. 6,280,289, which is incorporated herein by reference, or as described in U.S. Patent No. 6,179,709, which is incorporated herein by reference, The polishing pad can be modified by the present invention to include a transparent area that allows optical detection on or near the surface of the substrate being polished. Optical detection and monitoring methods are useful for endpoint detection, such as the measurement of light reflected from a substrate surface as described in the patent document. It is also possible to monitor at the interface between the polishing pad and the substrate surface. Optical measurements may be made, for example, by measuring the distribution of the slurry layer between the substrate surface and the polishing pad, as described in U.S. Patent No. 6,657,726, which is incorporated herein by reference. The invention also relates to a method for detecting light emitted by a luminescent material as a function of position below the surface of the substrate, And may include a concentration-sensitive luminescent material. Such a system is described in U.S. Patent Application No. 60 / 654,173, which is incorporated herein by reference. All of these systems and methods can be used to pass light from a light source through a pad to the substrate surface or to a slurry interface or to pass light from a substrate surface or slurry interface through a pad to a detector, At least one zone of the polishing pad that is transparent enough to allow light to pass therethrough.

In one example, the novel method of the present invention includes a process of making a local transparent zone pad. This method basically maintains the same chemical (polymer) composition for the entire pad, while reducing the porosity in the areas that need to be transparent in the pad manufacturing process, or not adding it at all, . This new method of creating a local transparent area pad allows a significant increase in pad life and a substantial improvement in the polishing performance of the pad with the window. Additionally, the method of the present invention may include compensating for differences in properties such as the hardness of the pad and window. For example, in order to increase the transparency sufficiently for optical termination point detection, the removal of micropores from one region of the pad creates a harder region with less pores, and thus, in view of the increased hardness, A soft polymer material may be used. Such compensation provides better control or more uniform polishing of the wafer.

Another property that can affect transparency is the size and density of the hard areas within the CMP pad. Larger hard areas create a pad that scatters light and therefore is less transmissive to the light used for end point detection. Therefore, reduction of the size and number of hard regions in the pad is required to obtain sufficient transparency for end point detection.

The concept of such a "localized transparent zone" is readily realized to form a plurality of windows during pad fabrication, such as liquid casting or reactive injection molding, to provide a suitable pad for the optical path of a plurality of detector assemblies on the polishing apparatus . This multiple window design can be used to provide accurate endpoint detection and an immediate polishing wafer surface profile.

The polishing pad provided is described in terms of transparent areas and areas of reduced transparency. Although the entire pad can be transparent, it is not desirable that the entire pad is transparent, as the transparent area, as described above, is substantially lacking in the characteristics of the porous structure. Generally, the pad has a transparent region in the region where the transparency is low, but is not limited to any shape. For example, various geometric shapes may be used within the circular pad. FIG. 17 is a schematic illustration of a non-limiting example of a possible geometric shape for a transparent region in a region of reduced transparency; the transparent region may be a cylindrical shape 102, a rectangular shape 104, or a ring shape 106. In addition, in the case of having a square or rectangular window, the orientation of the window can be changed. Other configurations are also possible, such as transparent base pads with grooves with reduced transparency as shown in Fig. The formation of the grooves is described in the section on in situ groove pads. 19 and 20 illustrate examples where the window is the same thickness as the rest of the pad or thinner than the rest of the pad. The transparent zone can be of any size and shape, and the area of the transparent zone can be up to 100% of the total pad area and is generally smaller (i.e., less than about 50% of the total area) with less transparency. In some instances, the total clear area is less than about 40%, less than about 30%, less than about 20%, less than about 10%, or less than about 5% of the total area. The polishing pad may have a plurality of transparent areas, in which case the total area of all the transparent areas is generally less than the total area of the areas of reduced transparency. Although one or more transparent regions may divide the pad into two or more opaque regions, generally one or more transparent regions are provided within one continuous opaque region.

A transparent zone is a portion or zone of a pad that is sufficiently transparent to light of a given wavelength. If the light is transmitted through the pad area in a sufficient amount to allow for the necessary optical detection or monitoring described herein, the area is sufficiently transparent. The transparent areas do not have to be completely transparent and can tolerate some scattering and absorption of the incident light. Although the transmission of light can be varied as a function of wavelength over a certain range, a preferably transparent region transmits light over a wide range of wavelengths. If desired, the transparent zone can transmit only a single wavelength. The light containing the spectrum of wavelength does not necessarily have to be transmitted at all wavelengths, but it must be necessary to use an appropriate optical detection method. The transparent zone is sufficiently transparent to some or all of the wavelengths from infrared to ultraviolet. For example, the transparent region is sufficiently transparent to some or all of the wavelengths in the range of 100 to 1000 nm, and also in the range of about 200 to 800 nm or about 250 to 700 nm, and sufficiently transparent in one aspect means that at least , About 20%, and also at least about 50%, or at least about 75%, of the composition is transmitted through the area.

The transparent zone contains a polymer of suitable transparency, the transparent zone lacking sufficient porosity. Pores scatter light, so if the pore density is too high, a lot of light is scattered and the area is not sufficiently transparent. The remainder of the pad is poorly transparent and can sufficiently transmit light useful for optical end-point detection. The rest of the pad has low transparency because the pores have a pore density that causes the incident light to scatter and reduce the transparency of the area. In one example, a region of low transparency has a significant porosity or is substantially porous, thus less than about 20%, less than about 10%, less than about 5%, or less than about 1% of the light transmitted by the transparent region . As described herein, different regions of the lower transparency portion of the pad do not block the same amount of light, but due to the porosity, the pore density can be varied across the lower transparency portion to block sufficient light in all regions.

The following examples illustrate examples of novel polishing pads and methods of making such pads.

Example 1: Process for forming pads with windows

A window forming fabrication process may be used. The manufacturing process is designed so that the respective product streams of the hardener, diol, prepolymer and microballoons are added separately before mixing or during mixing in a continuous process. This is schematically shown in Fig. When using such a manufacturing process, each required feedstock stream can be easily controlled to supply a predetermined amount of a curing agent, a diol, a prepolymer, and a microballoon.

Another goal that can be achieved using this manufacturing process is to form windows in place, while these processes provide notable adjustment and flexibility. Each part of the mold that needs to be filled to make the pad during the manufacturing process can be traversed by the insertion nozzle at a predetermined rate. In order to obtain transparency for a particular localized zone, the microballoons stream may be blocked or the flow rate may be reduced while the feeder crosses the specified zone. The transparency is obtained because the polymer matrix composed of the polyurethane formed by the reaction of the curing agent, the diol (or amine) and the prepolymer is transparent. The opacity of the pad is due to the introduction of microballoons in the area without windows.

Although it is possible to omit the microballoons to make the entire pad transparent, such transparent pads may not have the desired flexibility for polishing purposes. Not adding microballoons can increase the hardness by about 5 to 10 on Shore D scale. It is therefore desirable to create a localized transparent zone for endpoint detection purposes and to create a compensation rating scheme for the pad to counteract the adverse effects due to hardness changes in the transparent zone. An example of such a pad is shown in Fig. Compensation grading can be achieved very effectively by means of a predetermined grading scheme. This compensation leveling scheme can be achieved by the adjustment capability that is allowed in the manufacturing process described in this example. For example, the hardness in a clear zone can be controlled through the addition of a soft material. This manufacturing process can also be used to create one or more windows if desired.

Example 2: Characteristics of a polishing pad with a transparent window for visible light

For example, this approach is described as creating a graded CMP polyurethane pad with a 0.75 x 2.25 inch window used for polishing the wafer. The polyurethane polishing pad is made to have a desired hardness, pore size and density. The pads have a porosity density of about 15% to 25% of the pad material having a hardness of about 65 to 75 and a pore size of 35 to 55 占 퐉 in Shore D scale. In general, the hardness for the pad is from 45 to 75 on a Shore hardness D scale, and in one example the hardness for the window is preferably about 70 with a Shore hardness D scale. In order to achieve the desired goal of optical endpoint detection using commercially available CMP equipment, it is preferred that the pad window be transparent to visible light because visible light is used in this detection scheme. Figure 23 shows a completely transparent pad.

The features described in paragraphs II through VIII of the adapted pad for CMP, which may be made by the materials and processes described in paragraph I of the specification, may be combined to achieve certain characteristics for the pad. Table 10 lists the characteristics described in paragraphs II to VIII.

In addition to the polymer generation schemes described in Section I, the properties listed in Table 10 may be combined to form a fitted pad with the additional properties listed in Table 10, along with the controlled microstructure. The combinations of characteristics that can be combined with the controlled microstructure are as follows (numbers represent the properties listed in Table 10): 1, 2, 3, 4, 5, 6, 7, 1 & 2, 1 & 3, 1 & 3 & 4, 5 & 6, 5 & 7, 6 & 7, 1 & 2 & 3, 1 & 2 & 3, 2 & 4, 2 & 5, 2 & 3, 2 & 7, 3 & 4, 3 & 1 & 3 & 6,1 & 3 & 7, 1 & 4 and 5, 1 and 4 & 6, 1 & 4 & 7, 1 & 5 & 6, 1 & 5 & 7, 1 & 6 & 7, 2 & 3 & 4, 2 & 3 & 5, 2 & 3 & 6, 2 & 3 & 7, 2 & 4 & 5, 2 & 4 & 6, 2 & 4 & 7, 2 & 5 & 6, 2 & 5 & 7, 2 & 6 & 7, 3 & 4 & 5, 3 & 4, & 6, 3 & 4 & 7, 3 & 5 & 6, 3 & 5 & 7 3 & 6 & 7, 4 & 5 & 6, 4 & 5 & 7, 4 & 6 & 7, 5 & 6 & 7, 1 & 2 & 3 & 4, 1 & 2 & 3 & 5, 1 & 2 & 3 & 6, 1 & 2 & 3 & 7, 1 & 2 & 4 & 5, 1 & 2 & 4 & 6, 1 & 2 & 4 & 7, 1 & 2 & 5 & 6, 1 & 2 & 5 & 7, 1 & 2 & 6 & 7, 1 & 3 & 4 & 5, 1 & 3 & 4 & 6, 1 & 3 & 4 & 7, 1 & 3 & 5 & 6, 1 & 3 & 5 & 7, 1 & 3 and 6 & 7, 1 & 4 and 5 & 6, 1 & 4 & 5 & 7, 1 & 4 and 6 & 7, 1 & 5 & 6 & 7, 2 & 3 & 4 & 5, 2 & 3 & 4 & 6, 2 & 3 & 4 & 7, 2 & 3 and 5 & 6, 2 & 3 & 5 & 7, 2 & 3 and 6 & 7, 2 & 4 and 5 & 6, 2 & 4 & 5 & 7, 2 & 4 and 6 & 7, 2 & 5 & 6 & 7, 3 & 4 & 5 & 6, 3 & 4 & 5 & 7, 3 & 5 & 6 & 7, 4 & 5 & 6 & 7, 1 & 2 & 3 & 4 & 5, 1 & 2 & 3 & 4 & 6, 1 & 2 & 3 & 4 & 7, 1 & 2 & 3 & 5 & 6, 1 & 2 & 3 & 5 & 7, 1 & 2 & 3 & 6 & 7, 1 & 2 & 4 and 5 & 6, 1 & 2 & 4 & 5 & 7, 1 & 2 & 4 and 6 & 7, 1 & 2 & 5 & 6 & 7, 1 & 3 & 4 & 5 & 6, 1 & 3 & 4 & 5 & 7, 1 & 3 & 4 & 6 & 7, 1 & 4 & 5 & 6 & 7, 2 & 3 & 4 & 5 & 6, 2 & 3 & 4 & 5 & 7, 2 3 & 4 & 5 & 6 & 7, 1 & 2 & 3 & 5 & 6 & 7, 1 & 2 & 3 & 5 & 6 & 7, 1 & 2 & 3 & 5 & 6 & 7,

Ⅸ. How to fit

The fit of the pad may be based on the desired pad properties. For example, the formation of urethane hard pads can be achieved through the use of high levels of cross-linking, the use of toluene-diisocyanate (TDI) based systems instead of diphenylmethane-diisocyanate (MDI), the use of short polyols and polyamine chains Can be accomplished. Pads with low glass transition temperatures can be made by using polyether polyols, reducing the size of the hard segment and increasing the number. Pads with improved breaking strength can be made using polyester polyols. Transparent pads increase the number of hard segments, do not allow phase separation of short soft chains, reduce the size of pores, use lowly directional polyols, and have a linear relationship (i.e., a stoichiometric relationship) . ≪ / RTI > Hydrophilic pads can be made through the addition of hydrophilic low molecular weight polyols.

A. A method for adapting pads according to the material to be polished

Polishing of the oxide such as SiO ₂ may be achieved through the addition by inserting the abrasive, such as SiO ₂ particles in the polymeric material of the polishing pad.

Copper polishing involves a three-step process. 24 is a schematic view of a copper polishing process. The first step is bulk copper removal. The second step is the elimination of the low K barrier which may require a low coefficient of friction. The final third step is the removal of the tantalum / tantalum nitride barrier layer. Generally, three different pads are used for each individual step. For the inventive pad (Neopad's pad), a single functional pad can be achieved as described herein (i.e., a single pad can be used for all three steps). This can be achieved using the addition of a solid lubricant such as boron nitride and / or Teflon 占 and the use of a low shear integral pad. Pads comprising solid lubricants such as boron nitride or Teflon 占 which can be solid lubricants for copper polishing and low shear integral pads permit a lower coefficient of friction that can be desirable for copper polishing processes.

The shallow trench insulator (nitride and oxide stack) can be polished by using an integrated pad of low shear force as well as the addition of an insert abrasive such as cerium oxide. Since cerium oxide can selectively polish the nitride, cerium oxide enables selective polishing in the polishing process. The interface in a low shear integral pad allows a lower coefficient of friction. A functional grading of the abrasive may also be used to polish the shallow trench insulator.

A soft pad having a higher compressibility than the pad used to polish the oxide may be used to polish tungsten, a brittle material. The reduction in pad hardness can be achieved by using a soft polymer such as one prepared using a long chain polyol material and by increasing the porosity in the pad.

Optical materials that are very fragile and may require low removal rates may require methods such as whisper or cluster polishing with a very low coefficient of friction. This can be accomplished by the addition of a solid lubricant and / or the use of a low shear force integral pad having a plurality of interfaces.

What is present on the substrate as having an optical material, strained silicon, vertical gate, FinFet structure or silicon on insulator (SOI) is very fragile, may require low removal rates, Clustering of the cluster may be required. This can be accomplished by the addition of a solid lubricant and / or the use of a low shear force integral pad having a plurality of interfaces.

If system-on-chip (SoC) is present, polishing can be achieved by the addition of a solid lubricant such as boron nitride or Teflon® and the use of a low shear integral pad. Functional grading may be required if large pattern densities are present.

B. A method for adapting the pad according to the IC characteristics to be polished.

Rigorous control of pad characteristics, such as long-range regularity, pore size and distribution, is required for high IC pattern densities of greater than 70% on a substrate. Strict control of these properties can be achieved through control of the pad manufacturing process, such as controlling the uniformity of the temperature during the manufacturing process and creating a uniform mixture of polymer starting components.

For a high IC pattern density of the substrate, a functional grading of the pad may be required to adjust the density range. For a high pattern density range, such as from about 50% to 100% of the substrate, a radially continuous or discrete symmetric functional grading as schematically shown in Figures 7 and 10, respectively, may be used. For higher pattern density ranges, such as about 80% to 100% of the substrate, grading that is not radially symmetric, as shown in Figure 9, can be used for a more tailored distribution of pad characteristics to the polishing surface .

 For ICs with a smaller line width, it is preferable that the friction coefficient of the pad is further reduced. This can be achieved by the addition of a solid lubricant and / or the use of a low shear integral pad.

The chip size can determine a desired characteristic of the CMP pad. For large chips such as system-on-chips, functional grading can be important for high chip yield.

The size of the technology node, such as transistors and metal lines, can determine the desired characteristics of the polishing pad. For low technology nodes below 90 nm, the use of solid lubricants in the pad material and low shear integral pads can be important because they provide a low coefficient of friction. The lower the structure, the greater the possibility of breakage of the structure during polishing, so a low coefficient of friction can be important and a low coefficient of friction may therefore be required. The inventive pad can be designed to achieve very consistent polishing performance at 65 nm, 45 nm, 32 nm and smaller technology nodes as well as 90 nm and higher technology nodes.

Characteristics of Fitted Pads for CMP

Pad thermal characteristics (temperature change)

Changes in temperature during the polishing operation can affect the polishing performance. The temperature change depends on a number of variables including the slurry flow rate as well as the loss factor (E ") and the storage coefficient (E ') of the pad. Variations in the polishing temperature lead to changes in the removal rate, For example, a change in the polishing temperature of 2 DEG C may exhibit a polishing rate change of about 20%. In the present invention, the new methods include polishing (E ') and loss factor (E "), while maintaining a uniform distribution of the type, size and density of the hard regions within the polymer matrix of the pad, (20 占 폚 - 40 占 폚) of the storage coefficient that should be less than 20%, to have a ratio of the loss coefficient to the smaller storage coefficient (tan?), To reduce the hard zone size,And by increasing the density of the hard areas (which may increase the values of the storage modulus E 'and the loss modulus E ").

The novel inventive fit pad has a thermal change within 3 ° C maximum, while a commercially available pad has a thermal change of 10 ° C or higher. The temperature as a function of the pressure and velocity (px V) for the three adapted inventive pads and the two commercially available pads is shown in Fig.

DMA / TMA characteristics

The thermal-mechanical properties of the pad may be important for the polishing performance of the CMP pad. The main characteristic is above the glass transition temperature (T _g), loss modulus (E "), storage modulus (E '), tanδ (E " / E'), KEL (tanδ * 10 12 (E '(1 + tan 2 δ))), surface tension, compressibility and thermal change characteristics. A lower glass transition temperature is preferred for uniform and improved polishing. (E ") and storage factor (E ') are preferred because this leads to an increase in polishing performance. The higher loss factor E " and the storage factor E' Can be accomplished by reducing the size of the hard regions within the pad and increasing the density. Also, a lower tan delta value is preferred over the polishing temperature range because the smaller the value, the lower the temperature variation is possible.

The DMA / TMA characteristics are shown in Table 11. The storage modulus (E ') is greater than about 400 MPa for the novel adapted pad of the present invention compared to less than 300 MPa for commercially available pads. The loss factor (E ") is greater than about 250 MPa for the novel adapted pad of the present invention as compared to less than 250 MPa for commercially available pads. The pad of the present invention has a glass transition temperature (T _g ) of less than about -30 ° C. When the glass transition temperature is different from the working temperature, the temperature of the pad characteristic A low glass transition temperature is preferred because the effect is reduced. At or about the glass transition temperature, the polymer properties are subject to abrupt changes and have a high temperature dependency.

It may also be important to lower the change in storage coefficient (DELTA E ') as a function of temperature. Lower values can minimize changes in the properties of the polymer (i.e., the material properties remain the same and such properties are desirable). As shown in Table 11 for the pad characteristics, a small change in the storage modulus is obtained between 40 [deg.] C and 20 [deg.] C for the pad of the present invention compared to commercially available pads (greater than 30%). A reduction in storage coefficient can be achieved by maintaining a uniform distribution of the type, size, and density of the hard regions within the polymer matrix of the pad.

Other properties listed in the table are tan? Greater than 1 for commercially available pads and less than about 0.7 for the inventive pads, 100-1000 (1 / Pa) for commercially available pads, for typically from about 100 (1 / Pa) is less than ^{KEL (tanδ * 10 12 (E} '(1 + tan 2 δ))), approximately about the large pad of the present invention than 34 mN / m with respect to commercially available pads Surface tension less than 25 mN / m, between 1% and 5% for commercially available pads, and compressibility less than about 1% for the pad of the present invention. The preferred pad properties of the polishing pad are listed in Table 12.

Fitting can also be used in view of the type of slurry used during the polishing process. Depending on the slurry used, the surface tension of the pad can be adjusted to accommodate the viscosity and wetting characteristics of the slurry. Acceptance of the wetting property can be achieved through the use of a polymer material which can be miscible with the slurry used. A viscous slurry can hold a greater amount of slurry and may require a softer pad.

The equipment platform on which the polishing takes place can also affect the fit. Different equipment platforms (i.e., AMAT, Ebara) have different areas of the pad exposed to different areas of the wafer for different times. Functional grading can be used to adapt to different areas of high and low pressure occurring in different equipment platforms. The size of the pads can also be adjusted to accommodate different equipment platforms.

Ⅹ. Exemplary Pad Performance

A comparison of CMP performance in some important aspects of the pad of the present invention compared to commercially available pads and the contents of an exemplary non-limiting manufacturing method are described below.

Example 1

A compliant pad (A) designed to polish the oxide contains urethane of hardness 70 D (Shore D scale). The pads are formed using liquid casting and formed using the above-described method. To prepare the pad, isocyanate having a hardness of 70 D, a polyol chain extender, a curing agent, a stabilizer used for ultraviolet ray protection, and a pore-forming agent are used. The injection is carried out at about 150 - 160.. After injection, the material is left to cure for about 15 minutes. The pad is then detached from the mold and placed in an oven for post-cure treatment and treated at a uniform temperature between about 100 [deg.] F and 200 [deg.] F for about 12 hours. The thickness of the pad is 80 mils and the diameter of the pad is 20 inches. A double-sided tape is attached to the backside to make the pad ready for polishing. The adapted pad (B) is similar in configuration to the matched pad (A), but the hardness is lower than 65D.

26 shows a two-layer pad of the present invention (Figs. 26A and 26B) and a prestonian float for a commercially available pad (Figs. 26C and 26D). A removal rate is obtained as a function of pressure and rate changes during polishing of the interlayer dielectric layer. As described above, a linear relationship is expected with respect to ideal Prestonian behavior. Comparing the pads of the present invention (Figs. 26A and 26B) with the commercially available pads (Figs. 26C and 26D), it can be seen that the commercially available pads do not show a linear relationship with respect to the pads of the present invention. The main difference between the inventive pad and the commercially available pad is that the pad of the present invention is made in a manner that controls the shape, size and density of the pores through the above described manufacturing method.

The Stivac curve for two adapted pads of the present invention compared to two commercially available pads is shown in Figures 27a-27d. As described above, a certain relationship is obtained in the predetermined boundary lubrication region. It can be seen from this graph that a very uniform boundary lubrication behavior is obtained for the fitted pad. By comparison, it can be seen that commercially available pads (Figures 27c and 27d) are deflected from ideal boundary lubrication behavior. As discussed above with respect to the data for the Prestonian float, the main difference between the inventive pad and the commercially available pad is that the pad of the present invention is made in a manner that controls the porosity of the pad.

Example 2

A pad for polishing the oxide was prepared in a manner similar to that described in Example 1. [ In addition, the pads are functionally graded to improve polishing performance. In Figures 28 to 32, a planarization efficiency and planarization length comparison of the pad of the present invention adapted was made using patterned wafers. 28A shows a die measurement method in which nine dies are selected for each wafer measurement. Figure 28B shows the structural elements of each individual die. 29 and 31 compare the oxide thicknesses for three polishing times (30s, 60s, 120s) as a function of the density of the layout pattern in one die for commercially available pads and the inventive fit pad The results are shown. The overall axis of Figures 29 and 31 is for polishing performed as a function of pressure and velocity. Die 2 is selected because it is in the middle of the wafer and shows the effect from the inner edge of the pad as well as the outer edge of the pad. The slope is approximately 0.5-0.6 for a commercially available pad and approximately 0.2-0.3 for a fitted pad of the present invention, indicating that the pad of the present invention has a greater planarization length. Comparing FIG. 29 and FIG. 31, the oxide thicknesses are compared as a function of the layout pattern density for all of the dies, and the planarization length for the compliant, graded pad of the present invention of FIG. Which is much smaller than the commercially available pad of FIG. 30, which is smaller than the slope of the line to the pad.

Example 3

Three pads were fabricated for copper CMP. The three new pads (inventive pads) have unique microstructures, are radially graded, can be sub-surface treated with boron nitride as a solid lubricant, and can be low-shear integral pads . The three new pads are: 1) a surface treated pad (new pad A), 2) a low shear pad (new pad B), and 3) a low shear surface treated pad (new pad C).

Additionally, the copper lines of the performance evaluated wafers were analyzed using X-ray diffraction (XRD) and compared to the un-polished wafers to check whether a substantial change in the copper was caused by stress.

Fig. 33 shows X-ray diffraction data. The lattice constant measurements were performed on the polished wafers using five test pads (commercially available pads A and B and the new pads A, B and C) and compared to the measurements obtained from the un-polished wafers Respectively. The lattice constant of the unpolished copper film is 3.06086 A. The measured lattice constants for the polished wafers using Fujimi slurry and Cabot slurry are also indicated. The measurement error range for all X-ray diffraction experiments is approximately ± 0.0001 Å. In order to compare the test errors, the errors for the un-polished coating are indicated by squares across the data shown in the figure.

For both slurries, the measured lattice constant values of the polished copper coatings using commercially available pads are very large compared to the measured lattice constant of the unpolished coating. The shift direction indicates tensile stress. The measurement lattice constant values of the abraded coatings using the inventive pad are smaller than those obtained with the abraded coatings using commercially available pads. The value of the lattice constant measured for the film polished with the new pad A (surface treated pad) is less than 3.6091 A (both slurries). Similar results were obtained for the films that were polished using the new pad B. For a film that is polished using the new pad C, the measurement lattice constant (3.6086 A) is equivalent to the lattice constant value for the un-polished coating, which exhibits stress-free polishing, when a Foujima slurry is used. When carbotherm slurry is used on the new pad C, the measured value of the lattice constant for the polished coating is 3.6090 A. From the results of comparing the lattice constants of the coated films using the pads A, B, and C of the present invention with respect to the carboB slurry, the effect of using the surface treatment and the low shear integral pad is directly in an additive relationship . Nonetheless, the techniques of this approach, which make the surface treated and low shear pads, can independently reduce stress induced processes in copper CMP. This technique can also eliminate the stress inducing process when employed individually or in an integrated manner for the pad design.

Fig. 34 shows a pad (a new pad B) formed by a combination of a pad having a solid lubricant and a pad having a low shear force, as well as a commercially available pad A and a pad B, The lattice constants calculated from the X-ray diffraction data are compared with respect to the polished wafer and the unpolished wafer (BULK) using a pad having a solid lubricant (new pad C) although not a monolithic pad having a low shear force. This data shows both for commercially available slurries A (fumi) and sludge B (carboats). The lattice constant is a fundamental property that gives an average distance between atoms in a crystal arrangement. If the material is fundamentally changed at the atomic or molecular level, a change in lattice constant can be detected. It is clear from the lattice constant data that the copper of the wafers polished with the inventive inventive pad is comparable to the copper of the un-polished wafers, and any substantial Indicating that no change has occurred. In contrast, the wafers polished with commercially available pads A and B are not on an equal level and indicate that a change has occurred in the copper of the polished wafer using a commercially available pad.

34 shows the comparison of the total width (FWHM) at the middle of the maximum height of 222 peaks for the polished pads and the unpolished wafers using pads A, B and C of the present invention as well as the commercially available pads A and B . The data in Fig. 34 show all of the commercially available slurries A and B. If the polishing process causes a non-uniform deformation with respect to copper, the peak becomes narrower or wider and thus the overall width at the middle of the maximum height is an indicator of whether the copper has caused non-uniform deformation during the polishing process. In Figure 34, two cases of a pad with a low shear force (new pad B) combined with a solid lubricant and a pad with a solid lubricant (new Pad C), though not a low shear integral pad, It can be seen that it is very good with regard to alleviating the non-uniform deformation of the copper.

FIG. 36 compares the Stivac curve data and the Prestonian float for two inventive pads for use in copper CMP, which is not a low shear integral pad but has a solid lubricant. The difference between the two inventive pads is the amount of boron nitride. For the first pad, the pad contained 5 wt% boron nitride. For the second pad, the pad contained 8 wt% boron nitride. In the Stavey curve, both pads work in the boundary lubrication region and are equally visible. However, it can be seen that the removal rate for a pad having a solid lubricant of 8% in a prestonian float is significantly greater than that of a pad having a solid lubricant of 5%. This clearly explains how adding a solid lubricant to the sub-surface of the pad can increase the removal rate while keeping the friction coefficient low. If X-ray diffraction data is obtained that indicates that no serious damage has occurred to the copper structure of the wafer, this proves the preferred characteristics of the inventive pad described herein. This feature includes high removal rate and low shear pad operation which allows efficient polishing of copper CMP without damage caused by undesirable stress on the copper structure of the wafer.

Figure 37 shows a quantitative analysis of pad tampering by comparing pad A, which is commercially available, and pad C of the present invention. The normal removal rate was monitored as a function of time. Commercially available pad A takes about 30 minutes to reach steady state. In contrast, the pad of the present invention reaches a steady state in a significantly less than about 15 minutes. These results are directly due to the pad microstructure. It is believed that multiple uniform hard segments enable the formation of microreservoirs of consistent size. These microreservoirs are produced in a relatively short time and can provide a continuous supply of slurry once formed.

A time-dependent polishing stability analysis was performed for commercially available pad A and pad C of the present invention for three slurry (carbo-slurry) flow rates of 40 cc / min, 60 cc / min and 80 cc / min. The parameters studied are the removal rate (Fig. 38 (a)) and the friction coefficient (Fig. 38 (b)) on a single wafer polished for 150 seconds. The removal rate of commercially available pad A shows a significant change over time. In particular, the change at the lowest slurry flow rate (40 cc / min) exceeds 2.5. The pad C of the present invention exhibits a remarkably small change in removal rate. The change at a flow rate of 40 cc / min is about 2, but the change to the higher slurry flow rate is minimal. 38 (b) shows a very large change in the coefficient of friction (05 - 0.8) of the commercially available pad A as opposed to the friction coefficient (0.5 - 0.65) obtained in the pad C of the present invention Respectively. The consistent friction characteristics and uniform removal rate for Pad C of the present invention are characteristic of pads having a sub-surface treated solid lubricant.

39 shows the Stavey curves for two commercially available pads A and B and pad C of the present invention. Uniform lubrication behavior has been observed for pad C of the present invention indicating that it operates in the preferred boundary lubrication region. In contrast, the Stivac curve for the two commercially available pads A and B does not exhibit the expected linear trend for performance in the preferred boundary lubrication region. The main difference between the inventive fit pad and the commercially available pad used to produce the data shown in Figure 38 is the difference in pore size uniformity and the addition of a solid lubricant to the sub- . The combination of the pad characteristics and the solid lubricant provides a smaller and more uniform coefficient of friction and provides desirable results as shown in the Stavey curve.

Figure 40 shows bulk copper polishing results for 854 mask patterned copper wafers for commercially available pad A and pad C of the present invention using a commercially available slurry (JSR slurry). The uniformity in the die was studied through quantitative characteristics of dishing and erosion. Measurements were performed on the center die, annular die and edge die to understand the overall effect. Figure 39 (a) shows copper dishing results for a 100 탆 line structure of a mask set. Two sets of measurements were made: over 20% wafers wafers and another wafer over 60% wafers. The number of dishing for over 20% wafers obtained using commercially available pad C exceeds 400 A for all three dies. In contrast, the results for wafers over 20% obtained using Pad A of the present invention show that they are superior in die uniformity when showing a significantly lower number of dishing (less than 100 A) for all dies. The excellent dishing performance for the pad of the present invention is directly attributable to the pad microstructure. In addition, when comparing the number of dishing for three dies of a wafer polished by more than 20% using Pad C of the present invention, the change in die and die is very small (less than 10 A). Improved center-edge performance is a result of symmetrical and functional grading in the radial direction of the pad where the outer ring of the pad is more flexible than the inner. Similar results were obtained for the number of dishing for wafers with> 60% polishing. 40 (b) shows the results of erosion for the 9/1 [mu] m feature in the mask set. The erosion number (300 - 500 Å) obtained for wafers polished by more than 20% with respect to the commercially available pad A is significantly larger than the erosion number (150 Å or less) for the pad of the present invention. The erosion number for over 60% abrasive also shows a similar comparative tendency.

(1) summary of data of pad C of the present invention and commercially available pad A for (1) bulk copper polishing (platform P1) (2) pad C of the present invention for barrier layer polishing ) Shows a comparative trend for a number of important planarization parameters including dishing, erosion and planarization efficiency indicating that Pad C of the present invention performs better than commercially available Pad A . In addition to studies on bulk polishing, barrier layer polishing parameters were obtained by comparing commercially available pad A with pad C of the present invention. For all of the important planarization parameters, pad C of the present invention exhibits a much better performance than commercially available pad A. These results suggest that the pad of the present invention can be used for both bulk polishing as well as barrier layer polishing and the functionality of a single pad can be achieved.

Further, quantitative measurement of the stress (? _Acc ) accumulated in the film can be obtained by using the following equation.

σ _acc = E / (1-υ) ε (4)

Where E = modulus of elasticity

υ = Poissonby

ε = lattice strain

In Equation 4, the lattice strain is calculated as a unit change of the lattice constant based on the reference value. In the calculations, the ungrinding coating lattice constant serves as a reference. Using Equation 4 to calculate the accumulated stress? _Acc with a modulus of elasticity (E = 120 MPa) and a Poisson's ratio against copper (? = 0.34), values in the range of about 25 MPa to about 50 MPa Is obtained. For the polished coatings using the inventive pads, the accumulated stresses are significantly lower and the lowest accumulated stresses (? _Acc < 2 MPa) are obtained for the polished films using a low shear surface treated pad . In addition, the accumulated stress (σ _acc > 25 MPa) when measured against a commercially available pad is large and can affect the mechanical integrity as well as the electrical properties of the copper coating.

Table 14 shows the DMA characteristics of the pads used for copper CMP. The adapted pad of the present invention has a very high loss and storage coefficient at 20 占 폚 and 40 占 폚, a very low change in storage modulus between 20 占 폚 and 40 占 폚, a lower glass transition temperature, And has high wettability.

Example 4

Sub-surface treated pads with solid lubricant and sub-surface treated pads that are not low-shear monolithic pads, two-layer monolithic pads, solid lubricants combined with two-layer monolithic pads used for shallow trench insulator polishing Were compared with commercially available single layer pads. The two-layer integrated pad has one interface that acts as a stress sink. For comparison, two commercially available slurries of slurry A (Figures 41a-41c) and Slurry B (Figures 42a-42c) were used. This result is compared to the shallow trench heater polishing shown in Figs. 40A and 40B and Figs. 41A and 41B, which shows a comparison of Preston's constant as an index of wear rate versus friction coefficient. The comparison was made for both oxides 41a and 42a and for the nitrides 41b and 42b and the selectivity was compared for the two pads (Figures 41c and 42c).

In Fig. 40A using slurry A, the coefficient of friction for three adapted pads for oxide polishing is about half the coefficient of friction for conventional pads, while the removal rate is maintained at about the same level. Similarly, in Figure 42B, which shows the results for nitride polishing, the coefficient of friction of the fitted pad is about 33% less than the coefficient of friction of the single layer pad while the removal rate is about the same for each pad. Figure 41c shows that the selectivity of the fitted pad is comparable to that of a conventional pad.

Similarly, Figures 42a and 42b using slurry B show that the coefficient of friction for polishing oxides and nitrides using a matched pad is about 20% less than the coefficient of friction for conventional pads, while the friction rate is comparable. Figure 42c shows that the selectivity of the fitted pad is comparable to that of a conventional pad.

This result shows that in the example of the integrated pad of the present invention, which has been manufactured and evaluated to have at least one interface acting as a stress sink, the friction coefficient is reduced while maintaining the desired wear rate.

The above-described contents are various features that can be combined with various apparatuses and methods of the examples described below, and these examples naturally do not limit the scope of the present invention but supplement the above-mentioned contents.

(Examples of the present invention)

Claims 1. An article comprising a single polishing pad for polishing a substrate, the pad comprising a polymer having different properties in a first zone and a second zone in the pad, To increase the flatness or yield to the substrate as compared to a single pad that is constant in a region corresponding to the different regions or under comparable conditions under the same operating conditions as the single polishing pad.

2. An article according to claim 1, characterized in that the characteristic is porosity.

3. The article of claim 2, wherein the polymer has a different second property in the third and fourth zones and the second property is a hardness.

4. The article of claim 3, wherein said first and third zones are the same zone and said second and fourth zones are the same zone.

5. An article according to claim 1, wherein said characteristic is a hardness.

6. The article according to claim 5, wherein the pad has a circular profile and a rotation axis, the first zone has a circular profile about the axis of rotation, the second zone has a ring- And the first zone has a hardness higher than the hardness of the second zone.

7. The article of claim 6, wherein the difference in hardness between the first zone and the second zone is at least about 5 on a Shore hardness D scale.

8. An article according to claim 7, wherein said difference is at least about 10 on a Shore hardness D scale.

9. The article of manufacture according to claim 6, wherein the circular profile of the pad has an area measurement and the first area comprises at least about 75% of the area measurement of the circular profile of the pad. .

10. An article according to claim 9, wherein the interface between the first zone and the second zone and the second zone occupy the remaining area measurements of the circular profile of the pad.

11. The article of claim 5, wherein the polymer has a different second characteristic in the third and fourth zones and the second characteristic is a continuity of the polymer.

12. An article according to claim 11, wherein said third zone has an interface in said single polishing pad and said fourth zone is located apart from said interface.

13. An article according to claim 12 comprising a solid lubricant on the abrasive surface of said pad.

14. The article of claim 13, wherein the solid lubricant has a coefficient of friction between about 0.0001 and about 0.5.

15. The article of claim 13, wherein the pad comprises at least 5 wt% solid lubricant.

16. An article according to claim 1, wherein said first and second zones are disposed in said single polishing pad.

17. An article according to claim 16, wherein said first and second zones are additionally disposed on a polishing surface of said single polishing pad.

18. An article according to claim 17, wherein said characteristic is porosity.

19. The article of claim 18, wherein the polymer has a second characteristic different in the third and fourth regions, and the second characteristic is a hardness.

20. An article according to claim 17, wherein the characteristic is a hardness.

21. An article according to claim 1, wherein said first and second zones are disposed on a polishing surface of said single polishing pad.

22. The article of claim 21, wherein the characteristic is a hardness.

23. The article of claim 22, wherein the first zone is near the rotation axis of the single polishing pad and the second zone is near the outer edge of the pad, and the hardness of the second zone is greater than the first Lt; / RTI > is less than the hardness of the zone.

24. An article according to claim 1, wherein said characteristic is compressible.

25. An article according to claim 1, wherein said characteristic is a restoration coefficient.

26. An article comprising a polishing pad having a first characteristic that is non-uniform along a radius perpendicular to the axis of rotation of the pad, wherein the polishing pad has an improved first characteristic value on the semiconductor wafer due to the non- Flatness. ≪ / RTI >

27. An article according to claim 26, wherein the difference in value is determined by the device density on the substrate.

28. The article of claim 27, wherein the difference in value is additionally determined by the size of the technology node on the substrate.

29. The article of claim 26, wherein the difference in value is determined by the size of the technology node on the substrate.

30. The article of claim 26, wherein the characteristic is a hardness.

31. An article according to claim 30, wherein the porosity, the second characteristic, is different from the first radius or is different along the same second radius.

32. The article of claim 26, wherein the characteristic is porosity.

33. An article according to any one of the preceding claims, characterized in that it comprises a solid lubricant on the polishing surface of the pad.

34. An article according to any one of the preceding claims, wherein said characteristic is not transparency.

35. The article of claim 34, wherein the pad has a region that is more transparent than an adjacent region.

36. An article according to any one of the preceding claims, wherein the characteristic is a pore density.

37. An article according to any one of the preceding claims, wherein said characteristic is a pore size.

38. An article according to any one of the preceding claims, wherein said characteristic is selected based on the material to be polished.

39. An article according to claim 38, wherein said material comprises copper.

40. An article according to any one of the preceding claims, wherein the property is selected based on the slurry to be used with the article.

41. An article according to any one of the preceding claims, characterized in that the characteristic is selected based on the polishing equipment used with the article.

42. An article according to any one of the preceding claims, wherein said substrate is a semiconductor wafer and said pad comprises a chemical mechanical planarization pad.

43. A method of planarizing a layer of a semiconductor wafer having a patterned feature that causes the layer to have a high area and a low area, the method comprising the steps of: varying porosity, hardness, / RTI > contacting the layer with a polishing pad having a thickness of less than about < RTI ID = 0.0 > 0 < / RTI > and / or a restitution factor, and planarizing the layer of semiconductor wafer by removing the layer at a higher rate than the rate at which the polishing pad removes the layer ≪ / RTI >

44. A method of planarizing a layer of a semiconductor wafer having a patterned feature that causes a high area and a low area in the layer, said method comprising the steps of: contacting said layer with an article according to any one of inventions 1 to 37; And planarizing said layer. ≪ Desc / Clms Page number 20 >

45. A polymer polishing pad formed from a synthetic polymer and having an integrated interface between a first polymer layer and a second polymer layer of the pad.

46. The pad according to claim 45, wherein the first and second polymer layers are the same polymer.

47. The pad according to claim 46, wherein the first polymer layer has a first porosity, the second polymer layer has a second porosity, and the first porosity and the second porosity are not the same. .

48. The pad of claim 46, wherein the first polymer layer has a first porosity, the second polymer layer has a second porosity, and the first porosity and the second porosity are the same.

49. The pad according to any of the inventions 45 to 48, characterized in that the first polymer layer and the second polymer layer are formed from the same reactant but are reacted under different conditions to provide different polymers to the first and second polymers .

50. The pad according to any one of claims 45 to 49, wherein the pad further comprises a solid lubricant.

51. The pad according to any one of claims 45 to 50, wherein the pad is a single pad.

52. The pad according to any one of claims 45 to 51, wherein the interface is effective to reduce the friction coefficient of the pad as compared to a comparable pad or a pad that is the same as the polymer polishing pad but does not have an interface Pads.

53. A polymer pad for chemical mechanical planarization, wherein the pad comprises a polyurethane thermosetting resin and has a tan? Of less than about 1.0.

54. The pad according to claim 53, wherein tan? Is less than about 0.5.

55. The pad of claim 53 or 54, wherein the pad has an E 'value greater than about 400 MPa.

56. The pad of any of claims 53-55, wherein the pad has an E "value greater than about 250 MPa.

57. A pad according to any of the 53-56 invention, wherein the polyurethane has a Tg value of less than about -30 占 폚.

58. A pad according to any of the 53-57 invention, wherein the polyurethane further has a urea bond.

59. A pad according to any of the 53-58 invention, wherein the pad has a DELTA E '(20 DEG C - 40 DEG C) of less than about 20%.

60. A pad according to any of the 53-59 invention, wherein the pad has compressibility of less than about 1%.

61. The pad according to any one of claims 53 to 60, wherein the pad has a surface tension of less than about 25 mN / m.

62. The pad of any of claims 53 to 61, wherein the pad has a KEL value of less than about 100.

63. A polymer pad for chemical mechanical planarization comprising a polyurethane thermosetting resin and having an E 'value of greater than about 400 MPa.

64. A polymer pad for chemical mechanical planarization comprising a polyurethane thermosetting resin and having an E "value of greater than about 250 MPa.

65. A polymer pad for chemical mechanical planarization comprising a polyurethane thermosetting resin and having a Tg value of less than about -30 占 폚.

66. A polymer pad for chemical mechanical planarization comprising a polyurethane thermosetting resin and having compressibility of less than about 1%.

67. A polymer pad for chemical mechanical planarization comprising a polyurethane thermosetting resin and having a surface tension of less than about 25 mN / m.

68. A polymer pad for chemical mechanical planarization comprising a polyurethane thermosetting resin and having a KEL value of less than about 100.

69. A pad according to any of the 53-58 invention, wherein said pad comprises an interface.

70. The pad according to claim 69, wherein the interface is an integrated interface.

71. A pad according to any of the 53-57 invention, wherein the pad comprises a solid lubricant on the polishing surface of the pad.

72. The pad according to any one of 53 to 71, wherein the pad has, on the polishing surface of the pad, an area having characteristic values different from those of the same characteristic value in different areas of the polishing surface.

73. A pad according to any of the 53-52 invention, wherein the pad comprises a region having a higher transmittance to light than an adjacent region.

74. A polishing pad comprising a single pad for chemical mechanical polishing formed from a thermosetting polymer, the pad comprising a hard polymer region and a soft polymer region on an abrasive surface of the pad, the polymer comprising a hard segment and a soft segment Wherein the rigid segment forms a rigid polymer region and the soft segment forms a soft polymer region when cured, wherein the polymer comprises a polyurethaneurea.

75. The article of claim 74, wherein the hard region has a size of less than about 20 nm.

76. The article of manufacture of claim 74 or 75, wherein said soft zone has a size of less than about 100 nm.

77. An article according to any of the aspects 74 to 76, wherein said soft region has a size greater than 10 nm.

78. The article of any of the embodiments 74-77, wherein the soft region is larger than the hard region.

79. An article according to any one of claims 74 to 78, wherein said rigid region has a total of from 1 to about 20 urethane and urea groups.

80. An article according to claim 79, wherein said rigid region has a total of from 2 to about 6 urethane and urea groups.

81. The article of any of the embodiments 74-80, wherein the pad is a polymer melt or a mixture of polymer melts or reactants forming a polymer in a mold having an appropriate dimension to form a single chemical mechanical polishing pad, Wherein the chemical mechanical polishing pad is a single chemical mechanical polishing pad formed by placing a mixture of reactants in a mold.

82. The article of any one of < RTI ID = 0.0 > 74 < / RTI > to 81, wherein the pad has first and second polymer zones on the polishing surface of the pad, And wherein the first zone has a value that is different from a value for the property in the second zone.

83. An article according to claim 82, wherein said characteristic is one selected from the group consisting of hardness, porosity, pore size, compressibility, restitution factor and continuity.

84. An article according to any of the embodiments 74-83, wherein the pad comprises an integral interface.

85. An article according to any of the aspects 74 to 84, wherein said pad comprises a solid lubricant.

86. An article according to any of the aspects 74 to 85, wherein the pad comprises an abrasive.

87. A method of making a chemical mechanical polishing pad comprising the steps of forming a mixture or polymer melt of a reactant forming a polymer, positioning the melt or mixture in a mold, and forming a mixture of a hard polymer region and a soft polymer region And curing the melt or mixture to form the chemical mechanical polishing pad.

88. A polishing pad formed of a closed cell porous polymer and having a polishing surface of the pad in which most of the pores are elongated in a direction parallel to the polishing surface of the pad.

89. The polishing pad according to claim 88, wherein the cell of the closed cell porous polymer is elongated in a direction parallel to the polishing surface of the pad.

90. The polishing pad according to claim 88 or 89, wherein the cells of the closed cell porous polymer are formed of microballoons.

91. The polishing pad according to any one of claims 88 to 90, wherein the elongated pores have a ratio of length to width of greater than about 2.

92. A method of making a polishing pad having a closed cell porous polymer, the method comprising the steps of: incorporating a microballoons into a polymer melt or a mixture of reactants forming a polymer; and applying a pressure sufficient to compress the microballoons ≪ / RTI > compressing the melt or mixture.

93. An article comprising a single chemical mechanical polishing pad formed of a thermosetting polymer, the pad comprising a hard polymer region and a soft polymer region on the polishing surface of the pad, the polymer comprising a hard segment and a soft segment Wherein the rigid segment forms a rigid polymer region and the soft segment forms a soft polymer region when cured, the polymer comprising a poly (urethane urea) containing a repeating alkoxy unit.

94. An article according to claim 93, wherein said rigid region has a width of less than about 100 nm in either direction.

95. The article of claim 94, wherein the hard region has a width of less than about 20 nm.

96. An article according to any of the aspects 93 to 95, wherein said soft region has a width greater than about 100 nm.

97. An article according to any of the aspects 93 to 96, wherein said hard zone has from 1 to about 20 total urethane and urea groups.

98. An article according to claim 97, wherein said hard zone has a total of from 2 to about 6 urethane and urea groups.

99. The article according to any one of claims 93 to 98, wherein the pad is a polymer melt or mixture of reactants or a polymer melt forming a polymer in a mold having an appropriate dimension to form a single chemical mechanical polishing pad Wherein the chemical mechanical polishing pad is a single chemical mechanical polishing pad formed by placing a mixture of reactants in a mold.

100. The article of any one of claims 93 to 99, wherein the pad has first and second polymer zones on the polishing surface of the pad, the first and second zones being both the hard and soft zones And wherein the first zone has a value that is different from a value for the property in the second zone.

101. The article of claim 100, wherein the property is one selected from the group consisting of hardness, porosity, pore size, compressibility, restitution factor and continuity.

102. An article according to any of the aspects 93-101, wherein the pad comprises an integral interface.

103. An article according to any of the aspects 93-102, wherein the pad comprises a solid lubricant.

104. An article according to any of the aspects 93 to 103, wherein the pad comprises an abrasive.

105. The article of claim 1, wherein said first region of a single polishing pad is present near the axis of rotation of said single polishing pad and said second region is near an outer edge of said pad.

106. An article according to Clause 105, characterized in that the polishing surface contains a solid lubricant.

107. An article according to claim 106, wherein the solid lubricant is boron nitride.

108. An article according to Claim 105 comprising a transparent region local to the polishing pad for endpoint detection.

109. The article of item 105, wherein the article has an in-situ groove on a polishing surface.

110. An article according to claim 109, wherein the groove has a depth in the range of 10 占 퐉 to 100 占 퐉.

111. The article of claim 109, wherein the groove has a depth in the range of 25 占 퐉 to 40 占 퐉.

112. The article of claim 109, wherein the groove has a width in the range of 10 占 퐉 to 100 占 퐉.

113. The article of claim 109, wherein the groove has a width in the range of 25 占 퐉 to 40 占 퐉.

114. An article according to claim 109, wherein the groove in the vicinity of the axis of rotation is a concentric circular groove and overlaps a linear groove extending in the radial direction and is straight in the vicinity of the axis of rotation and away from the axis of rotation, Is a curve and the groove of the curve intersects the concentric circular groove.

115. An article according to claim 114 comprising additional curved grooves near the periphery of the abrasive surface away from the axis of rotation.

116. The article of claim 114 or 115, wherein the grooves are formed to have a constant groove density across the polishing surface.

117. The article of claim 114 or 115, wherein the groove is formed to maintain a constant slurry density across the polishing surface.

118. The article of claim 105, wherein the pad comprises micropores through the polishing pad.

119. An article according to claim 118, wherein the density of the micropores in the first zone is different from the density in the second zone.

120. The article of claim 105, wherein the hardness of the second zone is less than the hardness of the first zone.

121. The article of claim 105, wherein the pad comprises an integral interface perpendicular to the polishing surface and the interface is between the two polymer layers.

122. The article of matter of 121, wherein the first polymer layer and the second polymer layer are the same polymer.

123. An article according to claim 121, wherein the first polymer layer and the second polymer layer are different polymers.

124. An article according to claim 105, wherein the polymer is formed from a thermosetting material.

The combinations of the above combinations may have any of the physical, chemical, and / or DMA characteristics described above.

Although exemplary modifications of the adapted polishing pad have been described, various modifications to the pads of the present invention can be made without departing from the spirit and scope of the invention as described in the specification. The contents of the plurality of adapted polishing pads described herein should not be construed as being limited by the specific examples and drawings described above. Moreover, it will be appreciated by those skilled in the art that numerous equivalent polishing pads may be made from these examples and figures.

Claims (48)

An article comprising a single polishing pad for polishing a substrate,
Wherein the pad has a first zone of a polishing surface comprised of a first polymer and a second zone of the polishing surface comprised of a second polymer, wherein the first and second polymers have hardnesses different from each other in Shore hardness D scale Wherein the interface between the first zone and the second zone comprises a mixture of the first polymer and the second polymer,
The difference in hardness between the first zone and the second zone is at least 5 with a Shore hardness D scale,
Wherein the first polymer layer formed of the first polymer of the pad has a first porosity, the second polymer layer formed of the second polymer of the pad has a second porosity, the first porosity and the second porosity are not the same ≪ / RTI > 2. The method of claim 1, wherein the pad has a circular profile and a rotation axis, the first zone having a circular profile about the axis of rotation, the second zone having a ring- And the first zone has a hardness greater than the hardness of the second zone. delete The article of claim 1, wherein the difference is at least 10 with a Shore hardness D scale. 3. The article of claim 2, wherein the circular profile of the pad has an area measurement, wherein the first area comprises at least 75% of the area measurement of the circular profile of the pad. 6. The article of claim 5, wherein the interface between the second zone and the first zone and the second zone occupies a remaining area measurement of the circular profile of the pad. 2. The method of claim 1 wherein the hardness is different along a radius perpendicular to the axis of rotation of the pad and the polishing pad provides improved planarization of the semiconductor wafer due to the difference in value of the hardness along the radius Goods. 8. The article of claim 7, wherein the difference in hardness values is determined by the device density on the substrate. 9. The article of claim 8, wherein the difference in hardness value is additionally determined by the size of a technology node on the substrate. 8. The article of claim 7, wherein the difference in the value of hardness is determined by the size of the technology node on the substrate. The article of claim 1 wherein the polishing surface of the pad contains a solid lubricant. 8. The article of claim 7, wherein the hardness is selected based on the material to be polished. 13. The article of claim 12, wherein the material comprises copper. 8. An article according to claim 7, wherein said hardness is selected based on a slurry used with said article. 8. The article of claim 7, wherein the hardness is selected based on the polishing equipment used with the article. 8. The article of claim 7, wherein the substrate is a semiconductor wafer and the pad comprises a chemical mechanical planarization pad. A method of planarizing a layer of a semiconductor wafer,
The semiconductor wafer having a patterned feature that causes a high area and a low area in the layer,
The method includes contacting the layer with a polishing pad according to claim 7 and removing the layer from the high area at a rate that is faster than the rate at which the polishing pad removes the layer from the low area, ≪ / RTI > wherein the step of planarizing the layer of the substrate is performed. A method of planarizing a layer of a semiconductor wafer,
The semiconductor wafer having a patterned feature that causes a high area and a low area in the layer,
Said method comprising the steps of contacting said layer with an article according to claim 1 and planarizing said layer. A polymer polishing pad provided on the article according to claim 1,
Wherein the polymeric pad is formed of a synthetic polymer and has an interface integrated between the first and second polymer layers of the pad. delete delete 20. The method of claim 19, wherein the first polymer layer and the second polymer layer are formed of the same reactant, but are reacted under different conditions to provide a different polymer to the first polymer layer and the second polymer layer pad. 20. The polishing pad of claim 19, wherein the pad further comprises a solid lubricant. 20. The polishing pad of claim 19, wherein the interface is effective to reduce the coefficient of friction of the pad as compared to a comparison pad that does not have the interface but is otherwise identical to the polymer polishing pad. A polymer pad provided on an article according to claim 1,
The polymer pad having a chemical mechanical planarization pad,
Polyurethane thermosetting resin and having a tan delta of less than 1.0. 26. The pad of claim 25, wherein the tan delta is less than 0.5. 26. The pad of claim 25, wherein the pad has a value of E 'greater than 400 MPa. 26. The pad of claim 25, wherein the pad has a value of E "greater than 250 MPa. 26. The pad of claim 25, wherein the polyurethane has a Tg of less than -30 占 폚. 26. The pad of claim 25, wherein the polyurethane further has a urea linkage. 26. The pad of claim 25, wherein the pad has a DELTA E '(20 DEG C-40 DEG C) of less than 20%. 26. The pad of claim 25, wherein the pad has compressibility of less than 1%. 26. The pad of claim 25, wherein the pad has a surface tension of less than 25 mN / m. 26. The pad of claim 25, wherein the pad has a value of KEL of less than 100. 26. The pad of claim 25, wherein the pad comprises an interface between the first polymer layer and the second polymer layer of the pad. 36. The pad of claim 35, wherein the interface is an integral interface. 26. The pad of claim 25, wherein the pad comprises a solid lubricant on an abrasive surface of the pad. 26. The pad of claim 25, wherein the pad comprises a region of higher light transmittance than an adjacent region. 2. The article of claim 1, wherein the first region of the single polishing pad is present near the axis of rotation of the single polishing pad and the second region is near the outer edge of the pad. 40. An article according to claim 39, wherein the abrasive surface contains a solid lubricant. 41. The article of claim 40, wherein the solid lubricant is boron nitride. 40. An article according to claim 39 comprising a local transparent zone configured for endpoint detection on the polishing pad. 40. The article of claim 39, wherein the hardness of the second zone is less than the hardness of the first zone. 40. The article of claim 39, wherein the pad comprises an integral interface perpendicular to the polishing surface, the interface being between the two polymer layers. 45. The article of claim 44, wherein the first polymer layer and the second polymer layer are different polymers. 40. The article of claim 39, wherein the polymer is formed from a thermoset material. 10. An article according to claim 1,
Wherein the pad is uniform in areas corresponding to the different areas of the single polishing pad, but is improved in comparison to a single comparison pad, which is the same as the single polishing pad in other respects, Flatness and yield of the article. 10. An article according to claim 1,
Wherein the pad has in situ grooves on the polishing surface.

Images