TW200825160A - Planarization composition - Google Patents

Abstract

The present invention provides CMP abrasive slurry that is substantially free of aluminum oxide and comprises liquid and solids wherein the solids comprises: (a) in an amount of at least about 90 weight percent based on the solids, at least one non-spherical component having formula Al2O3 * xH2O where x ranges from 1 to 3; and (b) up to about one weight percent based on the solids portion of submicron alpha-alumina. The CMP abrasive slurry may be used to polish metallic or dielectric surfaces in computer wafers.

Description

200825160 (1) Description of the Invention [Technical Field of the Invention] The present invention relates to a novel slurry for chemical mechanical planarization (CMP). The present invention can be applied to a high speed integrated circuit of an interconnect structure having sub-micron design features and high conductivity with high throughput. [Prior Art] φ In the fabrication of integrated circuits and other electronic devices, a plurality of layers of conductive, semiconductive, and dielectric materials are deposited on or removed from the surface of the substrate. The conductivity, semiconductivity, and dielectric properties of the thin layer can be deposited via a variety of deposition techniques. Deposition techniques commonly used in modern processing include physical vapor deposition (PVD), also known as sputtering, chemical vapor deposition (CVD), plasma enhanced chemical vapor deposition (PECVD), and current electrochemical plating (ECP). ). As the layers of material are deposited and removed sequentially, the uppermost surface of the substrate may become non-planar across its surface and needs to be planarized. Flattening a surface, or "flattening" a surface, is a procedure in which matter is removed from the surface of the substrate to form a generally uniform flat surface. Flattening is used to remove undesired topography and surface defects such as rough surfaces, cohesive materials, lattice damage, scratches, and contaminated layers or materials. Planarization is also used to form features on the substrate by removing excess deposition material used to replenish features and to provide a flat surface for use in subsequent metallization and processing stages. Chemical mechanical planarization, or chemical mechanical polishing (CMP), is a technique commonly used to planarize substrates in 200825160 (2). CMP utilizes a chemical composition, typically a slurry or other fluid medium, to selectively remove material from the substrate. Considerations in the design of CMP slurries can be found in Rajiv K. Singh et al., "Fundamentals of Slurry Design for CMP of Metal and Dielectrics Materials", MRS Bulletin, pages 752-760 (October 2002). In conventional CMP techniques, a substrate carrier or planarization head is placed on a carrier assembly and configured to contact a planarization pad in a CMP φ device. The carrier assembly provides a controllable pressure to the substrate to force the substrate against the planarization pad. The pad moves relative to the substrate under external driving force. Thus, the CMP apparatus performs a planarization or rubbing movement between the substrate surface and the planarization pad while dispersing the planarization composition, or slurry, to simultaneously apply chemical and mechanical activity. The slurry conventionally used in CMP procedures contains abrasive particles in the reactive solution. Alternatively, the abrasive article can be a fixed abrasive article, such as a fixed abrasive planarizing pad, which can be used with a CMP composition φ or slurry that does not contain abrasive particles. The fixed abrasive article typically comprises a backing sheet to which a plurality of geometrically abrasive composite elements are attached. The most widely used abrasives in semiconductor CMP programs are cerium oxide (SiO 2 ), aluminum oxide (A1203), cerium oxide (Ce02), chromium oxide (Zr02), and titanium oxide (Ti02), all of which can be used. Tobacco or sol-gel processes are described in U.S. Patent Nos. 4,959,1, 3, 5, 3, 490, 490, and 5, 5, 6, 3, 46 and WO 97/40,030. It has recently been reported to contain a composition or slurry of manganese oxide (?n203) (European Patent No. 8,166,457) or tantalum nitride (S i N) (European Patent No. 7,026,050). 200825160 (3) US 6,508,952 discloses a CMP slurry comprising any commercially available abrasive in the form of particles, such as SiO 2 , A 1203, ZrO 2 , Ce 02, SiC, Fe 203, TiO 2 , Si 3 N 4 , or a mixture thereof. These abrasive particles typically have high purity, high surface area, and narrow particle size distribution and are therefore suitable for use as abrasives in abrasive compositions. U.S. Patent 4,549,374 discloses the use of a slurry of a polishing agent made by dispersing smelting clay in deionized water to planarize a semiconductor wafer. The pH of the slurry is adjusted by the addition of a base such as NaOH and KOH. The demand for electrical processing speed has continued to increase, requiring higher circuit density and performance. It is now necessary to fabricate wafers with 8 or more layers of circuit patterns. In principle, the requirements for more layers do not change the nature of flattening, but do require more rigorous specifications from the flattening approach. Defects such as scratches and dishing must be reduced or eliminated. One issue that further increases the demand for technology is the trend toward 300 mm wafers. Compared to 8" or 200 mm wafers, the larger the wafer, the more difficult it is to be uniform over a larger length gauge. In addition to the added number of layers, the increased circuit density can be reduced between individual paths. The space is achieved. The path cannot be too close, because the electrical overflow of the SiO2 dielectric (wafer oxide) may cause the connection to be short-circuited. Recently, a small, high-density circuit pattern can be fabricated on the integrated circuit. Progress has placed a higher demand on the isolation structure. US Patent Application Publication No. 2003/0 1 29838 (filed on December 28, 999) discloses the following non-plate-like abrasive materials: iron oxide, barium titanate, phosphoric acid Stone, dioptase, iron, brass, fluorite, water-6-200825160 (4) iron oxide, and azurite. US Patent 5,693,23 9 teaches a CMP planarization composition, Containing water; 1 - 50% by weight of ^-oxidized or ^-Ming oxide; the remainder of the solid is a composition having substantially lower frictional properties selected from the group consisting of: aluminum hydroxide, bismuth - alumina, 5-alumina, amorphous alumina and Amorphous cerium oxide. See also U.S. Patents 4,956,15; 6,03 7,260; and 6,475,607. However, we believe that in the solid portion 0 of the CMP slurry, even if <5 wt% alumina is present It is possible to scratch the metal surface of the wafer. A publication of a flattening pad containing 5 to 50% by weight of bauxite abrasive particles is taught by the publication of the Japanese Patent Publication No. 2000-246649. The reference teaches that if the weight percentage of the bauxite exceeds 50, Then, the cushioning property of the mat is lowered. The slurry used together with the flattening mat contains 1-15% by weight of fine particles such as bauxite. See also Japanese Laid-Open Patent Publication No. 2000-246620. U.S. Patent No. 5,906,949 teaches a a CMP slurry containing abrasive particles of predominantly φ from bauxite for planarizing a dielectric film such as Si〇2 under alkaline pH conditions. We believe that Example 3 of this patent results in a Alumina surface coated alumina. U.S. Patent No. 6,562,091 teaches that spherical alumina does not scratch the wafer during CMP processing; the spherical particles preferably have a diameter of less than about 50 nanometers. Angular cerium oxide The prior art may scratch the surface of the wafer during CMP processing. [Abstract] 200825160 (5) The present invention provides a CMP abrasion slurry substantially free of anhydrous alumina (formula ai2o3) and comprising a liquid portion and a solid portion Wherein the solid portion comprises: (a) at least about 90% by weight of the solid portion, at least one non-spherical component having the formula ai2o3 · χΗ20 (where X is from 1 to 3); and (b) The submicron ^ α -alumina which can be up to about 1 weight ° / 〇 is calculated from the solid portion. [Embodiment] DETAILED DESCRIPTION OF THE INVENTION The present invention utilizes a component having the formula Α 2 〇 3 · χΗ 2 〇 (where χ is from 1 to 3). When X in the present formula is 1, the resulting product is referred to as see-through stone and has a Mohs' hardness of about 6.5-7. When X is greater than 1 to 2, also 艮Ρ, 1·1, 1·2, 1·3, 1.4, 1. 5, 1.6, 1.7, 1.8, 1.9 or 2, the resulting product is called bauxite Or pseudo soft bauxite and have a Mohs hardness of about 2.5-3. When the X in the present formula is 3, the obtained product is called gibbsite, doyleite, nordstrandite (all have a Mohs of about 2.5-3) Hardness), or bayerite. Preferably, the component is bauxite or pseudo soft bauxite. The phrase "at least one non-spherical component having the formula Α1203 ·χΗ20", as used herein, includes but is not limited to a mixture of the following phases: ΑΙ2Ο3*1 2Η2〇 and Αΐ2〇3·1·6Η2〇, A12〇3 * 1.2Η2〇 and 200825160 (6) Α1203 ·2Η20, Α1203 ·1·6Η20 and Α1203 ·2Η20, and Αΐ2〇3·1·5Η2〇 and Αΐ2〇3·3Η2〇. A useful commercially available mixture is about 80% by weight bauxite and 20% by weight gibbsite. The above formula 〇12〇3 · The X in the χΗ20 can be easily determined by commercially available thermal analysis instruments (TGA, TGA/DTA, TGA/DSC). In Figure 3-6, the sample in powder form without any special pretreatment (dry or humidified) was heated from room temperature to about 100 °C/min in a flow of anhydrous air φ at 100 ml/min. 1 2 〇0 °C. In Fig. 7, a sol sample was dried in a fume hood for about two days and then heated as described above. Obviously, in some cases, the X determined by such conventional thermal analysis techniques may be greater than 2 or less than 1 for bauxite or pseudo soft bauxite, and for gibbsite, tri-striped aluminum Stone, smectite or bayerite may be greater than or less than 3. Another useful technique for identifying such alumina hydrate phases is powder X-ray diffraction (XRD). The bauxite is usually produced by hydrothermal treatment of gibbsite or the like at a pressure of φ and a temperature of about 250 ° C, or via a general formula of Al(OR) 3 (where R It is prepared by a method of hydrolyzing an organoaluminum compound of an alkyl group. The term "non-spherical" as used herein means that at least one dimension (height, length, and/or width) of the morphology of the particles is substantially greater than the other dimensions. Thus, the non-spherical particle morphology can be in the form of a plate, a sheet, a needle, a capsule, a layer, or any other shape having at least one dimension substantially greater than another dimension. These morphologies are distinguished from spheroidal particles which are substantially circular in appearance and do not have an extensible surface as can be seen, as disclosed in U.S. Patent No. 6,5,62,091, the disclosure of which is incorporated herein by reference. The advantages of the non-spherical particles of the present invention over the spherical particles of U.S. Patent No. 6,562,091 are as follows. First, for a given uranium solid load in the slurry, the non-spherical particles provide a much larger effective contact, i.e., a flat surface. This can result in a higher rate of material removal. The reason for this is that the substantially spherical particles are only in point contact with the surface to be polished at the end. In contrast, the 0 non-spherical particle system of the present invention which is expected to be in a flat configuration during planarization advantageously contacts the surface to be polished through the largest face. In addition, the applied pressure from the polisher is transferred to the wafer through the surface rather than the point of transmission, so improved polishing uniformity and overall flatness can be expected. These improvements include reduced erosion, shallow dishing, and field oxide loss. Second, the larger planarization area of the non-spherical particles allows for a lower abrasive content to be used in the slurry. This has a positive impact on particle-related defects such as scratches and particle residue. Third, the non-spherical particles can positively contribute to the non-φ Prestonian behavior of the fully formulated slurry, i.e., the slurry does not exhibit a flattening rate that increases with the applied line shape. . This is important for low voltage planarization such as lower than 2-3 psi and flattening of next generation copper and low or ultra low k dielectric devices for planarization pressures as low as less than 1 psi. Figures 3-7 show examples of thermal analysis-thermogravimetric analysis (TGA) and differential thermal analysis (DTA) or differential scanning calorimetry (DSC) charts of alumina hydrates useful in the present invention. They were heated from room temperature to 1 200 °C in a 100 ml/min dry air stream using a TA instrument SDT Q600 analyzer at a heating rate of 20 °C/min. The results in Figure 3-7 -10 (8) 200825160 show a clear three-stage weight loss (TGA curve - left Y-axis), corresponding to the endothermic peak associated with water loss as indicated by the DTA or DSC curve (right axis) . The first weight loss varies from 1 to 25% by weight and is typically accompanied by a DTA/DSC peak between about 60 °C and 120 °C. The second weight loss is more consistent, from about 12 to 16% by weight, with a very sharp DTA/DSC peak in the range of 460 °C to 5 15 t. The third weight loss, in all cases less than 2%, is very gradual, occurs at temperatures above φ 600 °C, and has a very broad endothermic peak from 740 °C to about 905 °C. Although the overall weight loss at 1 200 °C observed in Figures 3-6 is consistent with X in the range of 1_2 in the above formula A1203 · χΗ20, Figure 7 shows a total of 3 8.5 % corresponding to x > Weight loss. This demonstrates that X値, as determined by routine thermal analysis, may be significantly mutated for similar samples and sensitive to sample processing prior to measurement. Unlike the bauxite surface coating of Example 3 of U.S. Patent No. 5,906,949, the non-spherical particles of the present invention comprise bauxite from the core of the particle to the entire surface. Useful bauxite is commercially available from Sas 〇1. An example of a DISPERAL® acid dispersible alumina alumina system that is available in Table 1 below. -11 - 200825160 (9) Table 1 Typical chemical and physical properties DISPERAL DISPERAL S DISPERAL HP14 DISPERAL 40 ai2o3 (%) 77 75 77 80 Na2〇 (%) 0.002 0.002 0.002 0.002 Particle size (d50) (micron) 25 15 35 50 Microcrystal size [120] (nano) 10 10 14 40 Dispersed particle size (nano) 80 100 100 140 Table 2 below shows examples of the DISPERAL® and DISPAL8 liquid bauxite alumina systems available. Table 2

Typical Chemical and Physical Properties DISPERAL Dispersion 20/30 DISPERAL AL 25 DISPAL 11N7-12 DISPAL 14N4-25 DISPAL 18N4-20 DISPAL 23N4-20 ai2o3(%) 30 25 12 25 20 20 N〇3(%) 0.006 ___ 0.015 0.240 0.300 0.380 nh3(%) — 2·· ___ - ___ Dispersion pH 4 10 7 4 4 4 Dispersed particle size (nano) 200 200 180 140 120 100 DISPERAL® and DISPAL®7jc dispersible water available in Table 3 below An example of an alumina alumina system. -12- 200825160 (10) Table 3 Typical chemical and physical properties DISPERA LP2 DISPERAL HP 14/2 DISPAL 11N7-80 DISPAL 14N4-80 DISPAL 18N4-80 DISPAL 23N4-80 ai2o3 (%) 72 75 80 80 80 80 Na20 (%) 0.002 0.002 0.002 0.002 0.002 0.002 N03(%) 4.0 1.3 0.1 0.7 1.1 1.6 Particle size (d50) (micron) 45 35 40 50 50 50 Microcrystal size [120] (nano) ... 13 35 25 15 10 Dispersed particles Size (nano) 25 100 160 120 110 90 Useful bauxite is also commercially available from SAPOL's CATPPALtm. CATAPAL A, B, C1 or D is spray dried aluminum oxide having an increasing microcrystalline size from 40 angstroms to 70 angstroms. Figure 1 shows the TEM of Sasol alumina. Figure 2 shows another embodiment of the invention. In Figure 2, the non-spherical abrasive particles 10 comprise at least a portion of the core 12 coated with an aluminum hydroxide layer 14. The available core material '12' includes the U.S. Patent Application Serial No. 10/79,293, filed on March 5, 2004, which is incorporated herein by reference. Layered clays such as kaolin, vermiculite and montmorillonite (which can be ablated) and preserved clay-like modified bodies such as acid leaching kaolin, mica, talc, graphite flakes, glass flakes, and synthetic polymer flakes Can be used. These non-spherical particles are primary particles in the slurry. Thus, the phrase "aspherical particles" is used in non-spherical cohesives that do not encompass spherical particles herein. -13- 200825160 (11) In addition to having an aspherical morphology, the abrasive particles of the present invention are preferably softer than cerium oxide, aluminum oxide or hydrazine hydride typically used in CMP. Therefore, the non-spherical abrasive particles have a Mohs hardness of about 1-5 to 6. Table 4 below lists several metals and abrasive particles as a reference: Table 4 Material Mohs Hardness Microhardness [kg mm-2] Copper 2.5-3.0 80 Giant 6.5 30 Crane 7.5-8.0 350 Hydrate Si02 4-5 400-500 Si〇2 6-7 1200 Copper oxide 3.5-4.0 One kaolin (aqueous) 2-3 • Kaolin (calcined) 4.0-6.0 α _ Oxidation 9.0 2000 Zr02 6.5 Diamond 10.0 10000

The non-spherical uranium uranium with a Mohs hardness of between about 1 and 6 is hard enough to provide the mechanical action required for the CMP slurry, while avoiding such phenomena as scratches, shallow discs, and excessive Defects such as flattening. Typically, the non-spherical particle honing agent constitutes up to 20% by weight of the slurry, although up to 60% by weight of the abrasive solids content can also be prepared. More typically, an amount of less than 15% by weight and more preferably from 0.5 to 8% by weight of the abrasive content can be utilized. For the core material 12, kaolin clay particles are preferred. Although hydrous kaolin can be used, it has been found that if the kaolin is calcined, a flattening rate can be achieved by -14-200825160 (12). However, the overall performance of hydrous kaolin is better than that of the calcining bureau. The calcination of kaolin is carried out by a strong endothermic reaction accompanying dehydroxylation leading to metakaolin. Calcined kaolin clay under conditions that are more severe than those used to convert kaolin into metakaolin, that is, the calcination to produce an exothermic reaction to the shale soil, which causes the calcination in the form of spinel. Kaolin and the use of more extreme conditions lead to mullite φ (mullite). Generally, calcining hydrous kaolin at temperatures of 1 200 °F and higher results in dehydroxylation of hydrous kaolin into metakaolin. The calcination temperature of 1400 - 2200 °F can be used to produce kaolin clay that has been calcined through its characteristic exotherm to the spinel form of kaolin. At higher temperatures, such as above 1 900 °F, mullite formation occurs. Any and all of these forms of kaolin clay can be used as the uranium blasting agent of the present invention. All of these materials are commercially available from the assignee of the present application. Engelhard Corporation, Iselin, New Jersey ° φ Hydrous kaolin is typically made by a combination of unit operations that alter particle size distribution and removes coloring impurities from kaolin. . The use of an aqueous suspension of kaolin in water facilitates the operation of these units. Examples of unit operations that alter particle size distribution are centrifugation, delamination or honing devices and selective flocculation. Examples of unit operations that can result in the removal of coloring impurities are flotation and magnetic separation. Further, the reducing impurities and the oxidative bleaching can be used to make the coloring impurities colorless. In addition, the water can be substantially removed from the kaolin by filtration, and then the high solids filtered product slurry can be sprayed dry. The spray dried portion can be added back to the high solids filtration product slurry to further increase the solids content of the slurry in steps -15-(13)(13)200825160. The filtered product can be obtained without dispersing and thus the filter cake can be dried and pulverized to obtain an acid-dried kaolin product known in the art. In addition, the kaolin can be modified by thermal or chemical treatment. Typically, the kaolin is comminuted before and after the calcining operation. The treated kaolin can be pulverized to further improve the particle size distribution through the above unit operation. Other non-spherical abrasive particles which can be used as the core material 12 are brucite (magnesium hydroxide), hydrotalcite, and Nano talc. The foregoing materials are all commercially available. Other useful non-spherical abrasive particles are disclosed in commonly assigned U.S. Patent No. 6,187,710, the disclosure of which is incorporated herein in its entirety. This patent teaches in a specific example a clay mineral made of three layers of elemental plates comprising a central layer of octahedral oxygen around a metal ion (octahedral layer) surrounded by two layers of tetrahedrons - containing sand Surrounded by an atomic layer (tetrahedral layer), the size of the clay particles varies from 0.1 μm to 1 μm. In the octahedral layer, up to 30 atom% of metal ions are replaced by lower valence ions, and in the tetrahedral layer, up to 15 atom% of sand ions are replaced by lower valence ions. This patent teaches in another specific example that the ruthenium (锗) in the tetrahedral layer can be replaced by trivalent ions. Preferably, the octahedral layer contains aluminum, chromium, iron (ΠΙ), gu (I π ), manganese (111 ), gallium, sharp, molybdenum, tungsten, indium, yttrium, and/or yttrium as three Valence ion. As the divalent ion, magnesium, zinc, nickel, cobalt (II), iron (II), ruthenium (II) and/or ruthenium are preferably contained in the octahedral layer. In the tetrahedral layer, lanthanum and/or cerium is contained as a tetravalent component and preferably contains aluminum, boron, gallium, chromium, iron (III), cobalt (m), and / -16- (14) (14). ) 200825160 or manganese (in ) as a trivalent component. The tin hydroxide layer material 14 may be the one described above. The portion of the aluminum hydroxide coating typically has a thickness of up to about 0.5 microns. Any known coating method can be used to apply aluminum hydroxide 14 to the core material 12. In general, the 'CMP slurry composition includes the uranium granules required for mechanical action and at least one of the following: oxidizing agents, acids, bases, complexing agents, surfactants, dispersing agents, and other chemicals to provide A chemical reaction on the polished surface such as an oxidation reaction. Non-limiting examples of useful bases include KOH, NH4OH, and R4NOH ° which may also add π acid, examples of which may be H3P04, CH3COOH, HCl, HF, and the like. Useful supplemental oxidants are h202, ΚΙ03, ΗΝ〇3, Η3Ρ〇4, K2Fe(CN)6, Na2Cr207, KOC1, Fe(N03)2, NH2OH and DMSO. Divalent acids such as oxalic acid, malonic acid and succinic acid can be used as additives in the planarization composition of the present invention. Additional suitable acid compounds which may be added to the slurry composition include, for example, formic acid, acetic acid, propionic acid, butyric acid, valeric acid, caproic acid, heptanoic acid, caprylic acid, capric acid, lactic acid, nitric acid, sulfuric acid, malic acid. , tartaric acid, gluconic acid, citric acid, citric acid, pyrocatechinic acid, pyrogallic acid, gallic acid, citric acid, and mixtures thereof. Suitable corrosion inhibitors that can be added to the slurry composition include, for example, benzotriazole, 6-tolyltriazole, 1-(2,3-dicarboxypropyl)benzotriazole, and the like mixture. If added, the carboxylic acid can also impart sulphide inhibition properties to the slurry composition. -17- 200825160 (15) In order to increase the selectivity of the macro and molybdenum compound relative to cerium oxide, an additive compound may be added to the composition of the tear paste. Suitable fluorine-containing compounds include, for example, hydrogen fluoride, perfluoric acid, alkali metal fluoride salts, alkaline earth metal fluoride salts, ammonium fluoride, tetramethylammonium fluoride, ammonium difluoride, diammonium difluoride, and tris Fluorinated diamethylene diammonium, and mixtures thereof. Suitable chelating agents which may be added to the slurry composition include, for example, ethylenediaminetetraacetic acid (EDTA), N. hydroxyethylenediaminetriacetic acid φ (NHEDTA), nitrogen triacetic acid (NTA), Ethylene triamine valeric acid (DPTA), acetic acid diglycolate, and mixtures thereof. The tongs help to soften the metal surface or even help to protect low profile features or specially composed surfaces. The mourning of the protection mechanism can lead to significant improvements. Suitable amines which may be added to the slurry composition include, for example, transamines, monoethanolamine, diethanolamine, triethanolamine, diethylene glycol amine, N-hydroxyethylpiperidin, and mixtures thereof. Suitable surfactant compounds which may be added to the slurry composition include, for example, any of a wide variety of nonionic, anionic, cationic or amphoteric surfactants known to those skilled in the art. The pH of the slurry is important to the performance of all slurry components. The acidity level of the solution controls the reaction rate of the surface, the formation constant of the metal complex, the surface oxidation rate, the solution ionic strength, the aggregate size of the slurry particles, and more. The discussion of various acid, base, and pH buffers is a forward-looking area for CMP development. The alumina bauxite slurry can be conveniently prepared by dispersing the bauxite abrasive agent in water and adjusting the pH by adding an acid or a base as needed. This mixture of -18-200825160 (16) is then agitated for a period of time to ensure that the desired solids are dispersed and form a slurry of particles. An active CMP slurry component such as an oxidizing agent or other complexing agent, a chelating agent, a passivating agent, and a surfactant is added to the particle slurry. Additional active ingredients may be added if desired to ensure optimal performance of the fully formulated CMP slurry. The pH of the final slurry can then be adjusted via the addition of an acid or a base. The removal of excess metal or other impurities from the increasingly smaller space between individual paths presents an increasing challenge to CMP processing. Copper metal has a smaller intrinsic resistance and capacitance than the Cu/Al alloy used today as a conductive medium. Therefore, only a lower voltage is required to transmit the signal through the copper wire, which reduces the tendency of the electrical spill. In fact, the circuit paths can be placed closer together by using only Cu. However, the use of Cu also has disadvantages. Copper does not adhere well to oxide surfaces. Copper is also prone to host oxidation. Unlike wo3 or ai2o3, the CuO or Cu02 surface layer will still allow 02 and H20 to penetrate into the host metal. Furthermore, the Cu atoms are movable and can migrate into the φ of the SiO2 wafer material, eventually causing the transistor in the circuit to fail. Therefore, a thin layer of low dielectric material is disposed between the wafer oxide layer and the conductive Cu layer, which is typically composed of giant, nitrided or titanium nitride. This buffer layer enhances Cu adhesion, prevents oxidation of the bulk Cu metal, prevents bulk oxide from being contaminated by Cu ions, and further reduces dielectric between circuits (i.e., allows circuits to be spaced closer together). One of the uses of CMP technology is to fabricate shallow trench isolation (STI) structures in integrated circuits formed on semiconductor wafers or wafers such as germanium. The purpose of the STI structure is to isolate discrete device components (e.g., transistors) in a given pattern layer of -19-(17)(17)200825160 to prevent current leakage between them. The STI structure is typically formed by depositing a tantalum nitride layer on the thermally grown oxide layer after thermally growing an oxide layer on the tantalum substrate. After deposition of the nitrided chopped layer, shallow trenches are formed through the tantalum nitride and the thermally grown oxide layer and partially through the germanium substrate using, for example, any well known photolithographic masking and uranium etching process. A layer of dielectric material, such as hafnium oxide, is then typically deposited using chemical vapor deposition to completely fill the trench and cover the tantalum nitride layer. Next, the portion of the ceria layer covering the tantalum nitride layer is removed by a CMP process and the entire surface of the object is planarized. The tantalum nitride layer serves as a planarization stop layer to protect the underlying thermally grown oxide layer and the tantalum substrate from exposure during CMP processing. In some applications, the tantalum nitride layer is later removed by, for example, dipping the article in an HF acid solution, leaving only the ruthenium dioxide-filled trench for use as an STI structure. Additional processing is then typically performed to form a polysilicon gate structure.

The use of Cu and the accompanying low dielectric buffer layer requires enhanced performance of the planarization technique. The new technology is called Cu-CMP, but it is not significantly different in principle from the previous planarization method. The CMP program must be able to remove the soft Cu metal overload, but limit the Cu dishing, scratches, and removal of the low dielectric buffer layer. At the same time, there must be tighter tolerances because of the more closely spaced circuit patterns. The ability to make thin, flat, and defect-free layers is extremely important. As is also known in the art, a method of forming interconnections in a semiconductor structure is the so-called dual damascene method. The dual damascene method begins with depositing a dielectric layer (typically an oxide layer) deposited on a single crystal such as 矽-20-(18) (18)200825160. The oxide layer is etched into a trench by uranium having a pattern corresponding to the vias and traces used to interconnect the components contained in the circuit. The via is an opening in the oxide through which the different layers of the structure are electrically interconnected and the line pattern is defined by the trenches in the oxide. A metal is then deposited to fill the openings in the oxide layer. Subsequently, excess metal is removed via flattening. The program can be repeated many times as needed to form the desired interconnection. Thus, a dual damascene structure has a trench in the upper portion of the dielectric layer and a via that terminates at the bottom of the trench and passes through the lower portion of the dielectric layer. The structure has a step between the bottom of the trench and the sidewall of the via at the bottom of the trench. The abrasive particles of the present invention can be used in copper CMP in applications where copper is used in interconnect metal layers, other than logic devices (such as microprocessors) or memory devices (such as flash memory). For example, improving the thermal and electrical characteristics of the device package can include the use of a copper layer that requires planarization. The interconnected copper layers in the integrated circuit device and the copper layers in the package may be different resulting in different requirements for layer thickness, planarity, shallow dishing, and defect rate to be removed. In addition, microelectromechanical systems (MEMS) may have copper layers that require CMP to be flattened. The abrasive particles of the present invention can also be used in the CMP slurry used in this application. A review of CMP processing is available in ''Advances in Chemical- Mechanical Planarization,' Rajiv K. Singh and Rajiv Bajaj, MRS Bulletin, October 2002, pages 743-747. In general, although the CMP procedure is rather simple, To achieve detailed understanding is still limited by the many input variables in the flattening process. These variables include silt-21 - (19) 200825160 slurry variables such as particles and chemicals, pad variables, tool variables such as downward pressure and linearity Speed, and substrate variations such as pattern density. This article provides a good review of the program variables and the outstanding application of CMP technology and is incorporated herein by reference. [FIG. 1 is a TEM of a specific example of the invention. 2 shows a specific example of the present invention. Fig. 3-7 is a thermal analysis diagram (TGA/DTA or TGA/DSC) of a bauxite which can be used in the present invention. [Explanation of main components] 1 0 : non-spherical abrasive Particle 12: Core 1 4: Aluminum Hydroxide Layer - 22-

Claims (1)

200825160 (1) 'X. Patent Application Area 1. A chemical mechanical planarization (CMP) abrading slurry substantially free of anhydrous alumina and comprising a liquid and a solid, wherein the solid comprises: (a) calculating the solid An amount of at least about 90% by weight of at least one non-spherical component having the formula Al2〇3·χΗ20 (wherein X is from 1 to 3); and (b) up to about i% by weight of the submicron calculated by the solid portion 0 α - Alumina. 2. A CMP milled uranium slurry according to claim 1, which consists essentially of the at least one non-spherical component having the formula Α1203·χΗ20 (wherein X is from 1 to 3). 3. The CMP uranium slurry as claimed in claim 1 wherein the non-spherical component is boehmite. 4. The CMP abrasion slurry of claim 1, wherein the non-spherical component comprises bauxite coated kaolin. φ 5 . A method of planarizing a metal comprising the steps of: polishing the metal using a CMP abrasion slurry of the scope of claim 1 of the patent. 6. The method of claim 5, wherein the planarization occurs under acidic pH conditions. 7. The method of claim 5, wherein the slurry is used to polish copper. twenty three-