KR20090087034A - Planarization composition for metal surfaces comprising an alumina hydrate abrasive - 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. ® KIPO & WIPO 2009

Description

Planarizing composition for metal surface containing alumina hydrate abrasive {PLANARIZATION COMPOSITION FOR METAL SURFACES COMPRISING AN ALUMINA HYDRATE ABRASIVE}

The present invention relates to a novel slurry for chemical mechanical planarization (CMP). The present invention can be applied to fabricating high speed integrated circuits with submicron design features and high conductivity interconnect structures with high manufacturing throughput.

In the manufacture of integrated circuits and other electronic devices, a composite layer of conductive, semiconducting, and dielectric materials is deposited or removed on the surface of the substrate. Thin layers of conductive, semiconducting and dielectric materials can be deposited by a number of deposition techniques. Common deposition techniques of recent methods 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 top surface of the substrate may be nonplanar across its surface and require planarization. Planarizing or "flattening" the surface is a method of removing material from the surface of the substrate to form a generally flatter surface. Planarization is useful to remove unfavorable surface topography and surface defects such as rough surfaces, aggregated materials, crystal lattice damage, scratches and contaminated layers or materials. Planarization is also useful for forming properties on a substrate by removing excess deposited material used to meet the properties and provide a flat surface for metallization and subsequent layers of the process.

Chemical mechanical planarization, or chemical mechanical planarization (CMP), is a common technique useful for planarizing a substrate. CMP utilizes chemical compositions, generally slurries or other flowable media, to selectively remove material from the substrate. Considerations for CMP slurry design are discussed in Rajiv K. Singh et al., "Fundamentals of Slurry Design for CMP of Metals and Dielectrics Materials", MRS Bulletin, pages 752-760 (October 2002). In conventional CMP techniques, a substrate carrier or planarization head is installed on the carrier assembly and placed in contact with the planarization pad of the CMP device. The carrier assembly provides an adjustable pressure on the substrate that presses the substrate against the planarization pad. It is removed from the substrate by the pad external thrust force. Thus, the CMP apparatus performs both planarizing or frictional activity between the surface of the substrate and the planarizing pad while simultaneously dispersing the planarizing composition or slurry to effect both chemical and mechanical activity.

Conventional slurries used in CMP processes contain abrasive particles in reactive solutions. Alternatively, the abrasive article may be a fixed abrasive planarization pad that can be used with a fixed abrasive particle, such as a CMP composition or slurry that does not contain abrasive particles. Fixed abrasive particles generally comprise a backing sheet to which a plurality of geometric abrasive composite members are attached.

The most widely used abrasives in semiconductor CMP processes are described in US Pat. No. 4,959,113; 5,354,490; 5,354,490; And 5,516,346 and WO 97 / 40,030, silica (SiO ₂ ), alumina (Al ₂ O ₃ ), cerium oxide (CeO ₂ ), zirconium oxide (which may be prepared by fuming or sol-gel methods). ZrO ₂ ) and titanium dioxide (TiO ₂ ). Recently, slurries or compositions have been reported comprising manganese oxide (Mn ₂ O ₃ ) (European Patent No. 816,457) or Silicon Nitride (SiN) (European Patent No. 786,504).

US Pat. No. 6,508,952 contains abrasives in any commercially available particle form, such as SiO ₂ , Al ₂ O ₃ , ZrO ₂ , CeO ₂ , SiC, Fe ₂ O ₃ , TiO ₂ , Si ₃ N _4, or mixtures thereof. A CMP slurry is disclosed. Such abrasive particles generally have high purity, high surface area, narrow particle size distribution and are therefore suitable for use as abrasives in polishing compositions.

U.S. Patent 4,549,374 discloses planarized semiconductor wafers with abrasive slurry prepared by dispersing montmorillonite clay in deionized water. The pH of the slurry is adjusted by adding alkalis such as NaOH and KOH.

The demand for electrical process speeds continues to increase the need for increasingly higher circuit densities and performance. Now, it is desirable to manufacture a chip having a circuit pattern of eight or more layers. In principle, the need for more layers does not change the nature of the planarization, but it requires a more rigorous explanation from the planarization method. Defects such as scratches and dishing should be reduced or eliminated. An issue that further increases technical demand is the movement for 300 mm wafers. Larger wafers are more difficult to maintain uniformity over larger length scales compared to 8 "or 200 mm wafers.

In addition to adding layers, increased circuit density can be achieved by reducing the space between the individual paths. Paths that cannot be too close as an electrical filter can occur through the SiO ₂ dielectric (wafer oxide), which effectively shortens the connection. Recent technological advances that allow the fabrication of very small high density circuits on integrated circuits have placed higher demands on single structures.

U.S. Patent Application Publication 2003/0129838, filed December 28, 1999, discloses the following non-planar abrasive materials: zinc oxide, strontium titanate, apatite, dioptase, iron, brass, fluorspar, hydrated iron oxide, and M. Copper light is started.

U.S. Patent 5,693,239 is water; Teaching a CMP planarization composition comprising 1-50% by weight of alpha-alumina or alpha-aluminum oxide; The remainder of the solid consists of a polishing composition selected from the group consisting of substantially less aluminum hydroxide, gamma-alumina, delta-alumina, amorphous alumina and amorphous silica. See also US Pat. No. 4,956,015; 6,037,260; 6,037,260; And 6,475,607. However, it is understood that the presence of less than 5% by weight of aluminum oxide in the solid portion of the CMP slurry can scratch the metal surface of the wafer.

Japanese Laid-Open Patent Publication: 2000-246649 teaches a flattening pad containing 5 to 50% by weight of boehmite abrasive particles. The document teaches that the pad has poor buffering properties when the boehmite weight percentage exceeds 50.

The slurry was used with a planarization pad containing 1-15% by weight of particulates such as boehmite. See Japanese Laid-Open Patent Publication 2000-246620.

U. S. Patent 5,906, 949 teaches a CMP slurry containing abrasive particles mainly boehmite to planarize a dielectric film, such as SiO ₂ , under pH basic conditions. Example 3 of this patent is believed to result in boehmite surface coated alumina.

US Patent 6,562,091 teaches that spherical boehmite does not scratch the wafer during the CMP process; The spherical particles are preferably less than approximately 50 nm in diameter. This is in contrast to the prior art, which teaches that angled silica particles can scratch the wafer surface during the CMP process.

[Overview of invention]

The present invention provides a CMP abrasive slurry substantially free of anhydrous aluminum oxide (formula Al ₂ O ₃ ) and comprising a liquid portion and a solid portion, wherein the solid portion is

(a) at least one non-spherical component represented by the formula Al ₂ O ₃ .xH ₂ O, wherein x is 1-3 in an amount of at least about 90% by weight based on said solid portion; And

(b) up to about 1 weight percent submicron alpha-alumina based on the solid portion

It includes.

[Brief Description of Drawings]

1 is a TEM of one embodiment of the present invention.

2 illustrates one embodiment of the present invention.

3-7 are thermograms (TGA / DTA or TGA / DSC) for boehmite useful in the present invention.

Detailed description of the invention

The present invention uses a component represented by the formula Al ₂ O ₃ xH ₂ O (where x is 1 to 3). When x is 1 in the above formula, the obtained product is known as aluminum hydroxide ore and has a Mohs hardness of about 6.5 to 7. When x is 1 to 2, i.e., 1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9 or 2, the product obtained is known as boehmite or pseudoboehmite and has a Mohs hardness of about 2.5 to 3. In the above formula, when x is 3, the obtained product is known as gibbsite, dorite, nordrandite (the moss hardness is all about 2.5 to 3), or bayerite. Preferably, the component is boehmite or pseudoboehmite.

As used herein, examples of the phrase "one or more non-spherical components represented by the formula Al ₂ O ₃ .xH ₂ O" include Al ₂ O ₃ 1.2H ₂ O and Al ₂ O ₃ 1.6H. ₂ O, Al ₂ O ₃ ㆍ 1.2H ₂ O and Al ₂ O ₃ ㆍ 2H ₂ O, and Al ₂ O ₃ ㆍ 1.6H ₂ O and Al ₂ O ₃ ㆍ 2H ₂ O, and Al ₂ O ₃ ㆍ 1.5H ₂ O and Al ₂ O ₃ ㆍ 3H ₂ O, but are not limited to these. One useful mixture available commercially is about 80 wt% boehmite and 20 wt% Gibbsite.

The value of x in the formula Al ₂ O ₃ xH ₂ O can be conveniently measured using a commercially available thermal analysis instrument (eg, TGA, TGA / DTA, TGA / DSC). 3-6, samples in powder form were heated from room temperature to about 1200 ° C. at a rate of 20 ° C./min in a flow of dry air at 100 mL / min without any specific pretreatment (dry or humidified). In Figure 7, the sol sample was dried in a fume hood for approximately two days and then heated as described above. In certain instances, x measured by this general thermal analysis technique may be 2 or less than 1 in the case of boehmite or pseudoboehmite, and in the case of Gibbsite, Doilite, Nordstrandite or Barelight It is obvious that it can be greater than or less than three. Another useful technique for identifying this state of alumina hydration is powder X-ray diffraction (XRD).

Boehmite is generally prepared by hydrothermal treatment of gibbsite or the like at a temperature of about 250 ° C. or by hydrolyzing an organoaluminum compound of the formula Al (OR) ₃ , wherein R is an alkyl group.

As used herein, the term "nonspherical" refers to particles having a shape that is substantially larger than one other in one or more dimensions (height, length and / or width). Thus, the non-spherical particle form may be plate, sheet, needle, capsule, thin plate, or any other myriad of forms in which at least one dimension is substantially larger than the other. Such morphology is distinguished from spherical particles that do not have a substantially round and markedly elongated surface in appearance as disclosed in US Pat. No. 6,562,091.

The advantages of these non-spherical particles over the spherical particles of US Pat. No. 6,562,091 are as follows. First, when a slurry is loaded with abrasive solids into the slurry, the non-spherical particles provide a larger effective contact surface, i.e., a planarization surface. This leads to higher material removal rates. The rationale for this is that the substantially spherical particles have a point of contact with the surface to be polished at the end. In contrast, such non-spherical particles, which are expected to be placed flat during the planarization process, advantageously contact the surface to be polished through the largest facet. In addition, because the pressure applied from the grinder is transferred through the surface onto the wafer rather than through one point, polishing uniformity and overall planarity are expected to be improved. Such improvements include reduced corrosion, dishing and field oxide loss. Second, the larger planarization area of the non-spherical particles leads to the use of lower abrasive content in the slurry. This has a positive effect on particle related defects such as scratches and particle residues. Third, the non-spherical particles will contribute positively to the non-Prestonian behavior of the polishing rate of the fully prepared slurry, ie the slurry will not exhibit a linear increase in the planarization rate depending on the pressure applied. This may be significant, for example, for low pressure planarization of less than 2-3 psi and also for planarization of next generation copper and low or extremely low k dielectric devices with planarization pressures of less than 1 psi.

3-7 show examples of thermal analysis-thermogravimetric analysis (TGA) and differential thermal analysis (DTA) or differential scanning calorimetry (DSC) diagrams of possible alumina hydrate abrasives useful in the present invention. They were obtained using a TA instrument SDT Q600 analyzer by heating the sample from room temperature to 1200 ° C. at a heating rate of 20 ° C./min in a 100 ml / min flow of dry gas. The results show a distinct three step weight loss (TGA curve-left Y axis) with the corresponding endothermic peak, as indicated by the DTA or DSC curve (right Y axis) with respect to moisture loss in FIGS. The first weight loss varies from about 1 to 25 weight percent and is generally associated with a DTA / DSC peak between about 60 ° C. and 120 ° C .; The second weight loss is more constant in the range of about 12-16% by weight and has an associated very sharp DTA / DSC peak in the range of 460 ° C to 515 ° C. The third loss, which is less than 2% in all cases, is very staged and runs at temperatures above 600 ° C. and has a very wide endotherm in the range from 740 ° C. to about 905 ° C. The total weight loss at 1200 ° C. observed in FIGS. 3 to 6 corresponds to x in the range of 1 to _{2 in the} formula Al ₂ O ₃ xH ₂ O, while FIG. 7 shows a total of 38.5% corresponding to x greater than Weight loss. This shows that the value of x measured by routine thermal analysis can vary significantly for similar samples and is sensitive to sample processing prior to measurement.

Unlike the boehmite surface coating of Example 3 of US Pat. No. 5,906,949, the non-spherical particles comprise boehmite substantially throughout the core and the surface of the particles.

Useful boehmite is commercially available from Sasol. Examples of useful acid DISPERAL ^® dispersible boehmite alumina systems are in Table 1 below.

General chemical and physical properties DISPERAL DISPERAL S DISPERAL HP 14 DISPERAL 40 Al ₂ O ₃ (%) 77 75 77 80 Na ₂ O (%) 0.002 0.002 0.002 0.002 Particle Size (d ₅₀ ) (microns) 25 15 35 50 Crystallite size [120] (nanometers) 10 10 14 40 Dispersed particle size (nanometer) 80 100 100 140

Examples of useful DISPERAL ^® and DISPAL ^® liquid boehmite alumina systems are shown in Table 2 below.

General chemical and physical properties DISPERAL dispersion 20/30 DISPERAL AL 25 DISPAL 11N7-12 DISPAL 14N4-25 DISPAL 18N4-20 DISPAL 23N4-20 Al ₂ O ₃ (%) 30 25 12 25 20 20 NO ₃ (%) 0.006 --- 0.015 0.240 0.300 0.380 NH ₃ (%) --- 2 --- --- --- --- PH of the dispersion 4 10 7 4 4 4 Dispersed particle size (nanometer) 200 200 180 140 120 100

Examples of useful DISPERAL ^® and DISPAL ^® water dispersible boehmite alumina systems are shown in Table 3 below.

General chemical and physical properties DISPERAL P2 DISPERAL HP 14/2 DISPAL 11N7-80 DISPAL 14N4-80 DISPAL 18N4-80 DISPAL 23N4-80 Al ₂ O ₃ (%) 72 75 80 80 80 80 Na ₂ O (%) 0.002 0.002 0.002 0.002 0.002 0.002 NO ₃ (%) 4.0 1.3 0.1 0.7 1.1 1.6 Particle Size (d ₅₀ ) (microns) 45 35 40 50 50 50 Crystallite size [120] (nanometers) --- 13 35 25 15 10 Dispersed particle size (nanometer) 25 100 160 120 110 90

Useful boehmite is also commercially available from Sasol as CATAPAL ^™ . CATAPAL A, B, C1 or D is spray dried alumina whose crystallite size increases from 40 angstroms to 70 angstroms. CATAPAL 200 has a crystallite size of 400 angstroms. 1 shows the TEM of Sasol boehmite.

Another embodiment of the invention is shown in FIG. 2. In FIG. 2, the non-spherical abrasive particles 10 comprise a core 12 at least partially coated with an aluminum hydroxide layer 14. Useful core materials 12 include those described in Applicant's pending patent application US Serial No. 10/792738 (filed March 5, 2004), which is incorporated herein by reference in its entirety. Thin clays such as kaolin, vermiculite and montmorillonite (which may be stripped) and variants of such clays that preserve the clay form are useful, such as kaolin, mica, talc, graphite shards, glass shards, and synthetic polymer shards soaked in acid.

These non-spherical particles are the major component in the slurry. Thus, as used herein, the expression “nonspherical particles” does not include nonspherical agglomerates of spherical particles.

In addition to having a non-spherical form, the abrasive particles are generally softer than the cerium oxide, silica or alumina used for CMP. Thus, the non-spherical abrasive particles have a Mohs hardness of about 1 to 5 to 6. For reference, Table 4 below describes various metals and abrasive particles:

matter Morse Microhardness [kg mm ^-2 ] Copper 2.5 to 3.0 80 tantalum 6.5 230 tungsten 7.5-8.0 350 Hydrated SiO ₂ 4 to 5 400-500 SiO ₂ 6 to 7 1200 Copper oxide 3.5 to 4.0 - Kaolin (hydration) 2-3 - Kaolin (calcin) 4.0-6.0 Alpha-alumina 9.0 2000 ZrO ₂ 6.5 - Diamond 10.0 10000

Non-spherical abrasives with Mohs hardness of about 1 to 6 are hard enough to provide the necessary mechanical action of the CMP slurry, but it is believed that defects such as scratching, dishing and excessive planarization action can be avoided simultaneously.

Generally, the non-spherical particle abrasives comprise up to 20% by weight of the slurry, although up to 60% by weight abrasive solids content can be prepared. Generally, an abrasive content of less than 15% by weight, more preferably from 0.5 to 8% by weight, is used.

Kaolin clay particles are preferred as the core material 12. While hydrated kaolin can be used, it has been found that better planarization rates are obtained when kaolin is calcined. However, the overall performance of hydrated kaolin is better than calcined kaolin, so hydrated kaolin is preferred. Calcination of kaolin to undergo strong endothermic reactions associated with dehydration results in metakaolin. Kaolin clays that are calcined under more stringent conditions than those used to convert kaolin to metakaolin, ie, kaolin clay that is calcined to undergo a characteristic kaolin endothermic reaction, results in spinel formation of calcined kaolin and also utilizes more extreme conditions. In case it leads to the formation of mullite. Generally, calcination of the hydrous kaolin at 1200 ° C. and higher temperature dehydrates the hydrous kaolin to metakaolin. A calcination temperature of 1400-2200 ° F. can be used to prepare kaolin clay that is calcined through its characteristic exotherm to spinel forming kaolin. At higher temperatures, such as 1900 ° F., the formation of mullite occurs. Some and all of these forms of kaolin clay can be used as abrasives of the present invention. All such materials are commercially available from Engelhart, Inc. (Iselin, NJ), the assignee of the present application.

Hydrated kaolin is generally prepared through a combination of unit operations that modify the particle size distribution and remove colored impurities from the kaolin. This unit operation is facilitated using an aqueous suspension of kaolin in water. Examples of unit operations that change the particle size distribution include centrifugation, bed separation or mill apparatus, and selective agglomeration. Examples of unit operations to remove colored impurities include flotation and magnetic screening. Moreover, reducing and / or oxidative bleaching can be used to color the colored impurities. In addition, filtration can be used to substantially remove water from the kaolin, and the high solid filtration product slurry can then be spray dried. The spray dried portion can be added back into the high solid filtration product slurry to further increase the solids content of the slurry. The filtration product cannot be dispersed, thus the filter cake can be dried and ground to obtain what is referred to as acid dried kaolin in the industry. In addition, the kaolin can be modified by thermal or chemical treatment. In general, kaolin is powdered before and after the calcination operation. The treated kaolin can be slurried through the unit operations described above to make modifications to the particle size distribution.

Other useful non-spherical abrasive particles for the core material 12 are brucite (magnesium hydroxide), hydrotalsides and nanotalcs. The foregoing materials are commercially available. Other useful non-spherical abrasive particles are generally disclosed in US Pat. No. 6,187,710, which is incorporated herein by reference in its entirety. This patent teaches, in one embodiment, a clay mineral consisting of a basic three-layered small plate consisting of metal ions (octahedral layers) surrounded by oxygen on one side, which layer contains two silicon-sided silicon atoms. It is surrounded by a tetrahedral layer, and the clay particles have a variety of dimensions from 0.1 micron to 1 micron. In the octahedral layer, at most 30% of the metal ions are replaced by lower valence ions, and in the tetrahedral layer at most 15% of the silicon ions are replaced by lower valence ions. The patent teaches that in another embodiment the silicon (germanium) of the tetrahedral layer can be replaced with trivalent ions. In the octahedral layer, aluminum, chromium, iron (III), cobalt (III), manganese (III), gallium, vanadium, molybdenum, tungsten, indium, rhodium and / or scandium are preferably present as trivalent ions. As divalent ions, magnesium, zinc, nickel, cobalt (II), iron (II), manganese (II) and / or beryllium are preferably present in the octahedral layer. In the tetrahedral layer, silicon and / or germanium are present as tetravalent components, preferably aluminum, boron, gallium, chromium, iron (III), cobalt (III) and / or manganese (III) are present as trivalent components. do.

The aluminum hydroxide layer material 14 may be as described above. Partial aluminum hydroxide coatings are generally less than about 0.5 microns thick. Any known coating method can be used to coat aluminum hydroxide 14 on the core material 12.

Generally, CMP slurry compositions include abrasives for mechanical action and one or more oxidants, acids, bases, complexing agents, surfactants, dispersants, and other chemicals to provide chemical reactions, such as oxidation on surfaces to be polished. .

Non-limiting examples of available bases are KOH, NH ₄ OH and R ₄ NOH. Acids may also be added, such as H ₃ PO ₄ , CH ₃ COOH, HCI, HF, and the like. Available as such auxiliary oxidants are H ₂ O ₂ , KIO ₃ , HNO ₃ , H ₃ PO ₄ , K ₂ Fe (CN) ₆ , Na ₂ Cr ₂ O ₇ , KOCl, Fe (NOs) ₂ , NH ₂ OH And DMSO. Divalent acids such as oxalic acid, malonic acid and succinic acid can be used as an adjunct to the planarizing compositions of the present invention.

Additional suitable acid compounds that can be added to the slurry composition include, for example, formic acid, acetic acid, propanoic acid, butanoic acid, pentanic acid, hexanoic acid, heptanoic acid, octanoic acid, nonanoic acid, lactic acid, nitric acid, sulfuric acid, malic acid. , Tartaric acid, gluconic acid, citric acid, phthalic acid, pyrocatechol acid, pyrogallol carboxylic acid, gallic acid, tannic acid and mixtures thereof.

Suitable corrosion inhibitors that can be added to the slurry composition are, for example, benzotriazole, 6-tolyltriazole, 1- (2,3-dicarboxypropyl) benzotriazole and mixtures thereof.

When carboxylic acid is added, erosion inhibiting properties can also be imparted to the slurry composition.

To increase the selectivity of tantalum and tantalum compounds relative to silicon dioxide, fluorine containing compounds can be added to the slurry composition. 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 nonfluoride, ethylenediammonium difluoride, Diethylenetriammonium trifluoride and mixtures thereof.

Suitable chelating agents that can be added to the slurry composition include, for example, ethylenediaminetetraacetic acid (EDTA), N-hydroxyethylenediaminetriacetic acid (NHEDTA), nitrilotriacetic acid (NTA), diethylenetriaminepentaacetic acid (DPTA ), Ethanol diglycinate and mixtures thereof. Chelating agents not only help to soften the metal surface but also protect the underlying positional properties or surfaces of certain compositions. Thinking about protection mechanisms can lead to significant improvements.

Suitable amines that can be added to the slurry composition are, for example, hydroxylamine, monoethanolamine, diethanolamine, triethanolamine, diethylene glycol amine, N-hydroxyl ethyl piperazine and mixtures thereof.

Suitable surfactant compounds that can be added to the slurry composition are, for example, any of a number of nonionic, anionic, cationic or amphoteric surfactants known to those skilled in the art.

The pH of the slurry is very important for the performance of all slurry components. The acidity of the solution can control the reaction rate on the surface, the formation constant of the metal complexing agent, the rate of surface oxidation, the solution ion strength, the mass size of the slurry particles, and the like. Investigation of various acids, bases and pH buffers is a promising field for CMP development.

Boehmite slurries can be conveniently prepared by dispersing boehmite abrasives in water, adjusting pH, and adding acids or bases. This mixture is then stirred for a time to obtain the desired solid dispersion to form a particle slurry. To this particle slurry, active CMP slurry components such as oxidizing agents or other complexing agents, chelating agents, passivating agents and surfactants are added. Other active ingredients may also be added according to the necessary criteria to ensure the best performance of the fully prepared CMP slurry. The pH of the final slurry can then be adjusted by adding acid or base.

Removing excess metal or other contaminants from the very small spaces between individual paths increases the challenge for the CMP process. Copper metal has a lower resistivity and capacitance than the Cu / Al alloys currently used as conductive media. Therefore, the smaller the potential, the more it is necessary to send a signal through the copper wire and to reduce the tendency for electrical abuse. In fact, by using only Cu, the circuit paths can be placed closer together.

However, the use of Cu also has disadvantages. Copper does not adhere well to oxide surfaces. Copper is also easy to bulk oxidize, unlike WO ₃ or Al ₂ O _3, such that the CuO or CuO ₂ surface layer still allows O ₂ and H ₂ O to penetrate into the bulk metal. Moreover, Cu atoms are mobile and can eventually migrate to the SiO ₂ wafer material which causes the transistor to fail in the circuit. Therefore, a thin layer of low dielectric material consisting of tantalum, tantalum nitride or titanium nitride is located between the wafer oxide and the conductive Cu layer. The buffer layer promotes Cu adhesion, prevents oxidation of the bulk Cu metal, prevents Cu ion contamination of the bulk oxide, and lowers the dielectric between circuits (ie, makes the circuits closer together).

One use of CMP technology is the fabrication of device isolation (STI) structures in integrated circuits formed on semiconductor chips or wafers, such as silicon. The purpose of the STI structure is to isolate discrete device elements (eg, transistors) in a given pattern layer to prevent current leakage between them.

STI structures are generally formed by thermally growing an oxide layer on a silicon substrate and then depositing a silicon nitride layer on the thermally grown oxide layer. After deposition of the silicon nitride layer, the device is formed through the silicon nitride layer and the thermally grown oxide layer, and partially the silicon substrate, for example using any known photolithography mask and etching process. A layer of dielectric material, such as silicon dioxide, is then deposited generally using a chemical vapor deposition process to completely fill the grooves and cover the silicon nitride layer. Next, the CMP process is used to remove portions of the silicon dioxide layer that cover the silicon nitride layer and planarize the entire surface of the article. The silicon nitride layer serves as a planarization stop to protect the underlying thermal growth oxide layer and the exposed silicon substrate during the CMP process. In some applications, the silicon nitride layer is later removed, for example soaking the article in an HF acid solution, leaving only grooved silicon dioxide to serve as an STI structure. An additional process is then generally performed to form the polysilicon gate structure.

The use of Cu and the accompanying low dielectric buffer layers require improved performance from planarization techniques. The new technique is called Cu-CMP, but in principle is not significantly different from previous planarization methods. The CMP process may remove the soft Cu metal overburden, but limits the removal of Cu dishing, scratching and low dielectric buffer layers. At the same time, the durability is more stringent due to the closerly located circuit pattern. The ability to produce thin, flat and defect free layers is of utmost importance.

As is known in the art, one method of forming interconnects in semiconductor structures is the so-called dual damascene process. The dual damascene process starts with the deposition of a dielectric layer, typically an oxide layer, deposited on a circuit element formed from a single crystal, such as silicon. The oxide layer is etched to form grooves having a pattern corresponding to the wires for interconnecting the bias pattern and the member of the circuit element. The bias is the inlet where the different layers of the structure are electrically interconnected, and the pattern of the wire is defined by grooves in the oxide. The metal is then deposited to fill the inlet of the oxide layer. The excess metal is then removed by planarization. The process is repeated as many times as necessary to form the required interconnect ions. Thus, the dual damascene structure has a groove in the upper portion of the dielectric layer and is removed from the bottom of the groove and passes through the lower portion of the dielectric layer. There is a step between the bottom of the groove and the side wall through the bottom of the furrow.

The abrasive particles of the present invention may be used for CMP of copper in applications other than logic (eg, microprocessor) or memory (eg, flash memory) devices in which copper is used in interconnect metal layers. For example, improving thermal and electrical properties in the packaging of the device may include the use of copper layers that need to be planarized. The interconnect copper layer in the integrated circuit device and the copper layer structure in the packaging may lead to different requirements, flatness, dishing and defects on the thickness of the layer to be removed. Microelectromechanical systems (MEMS) may have a copper layer that may require planarization with CMP. The abrasive particles of the present invention can also be used in the CMP slurries of the present application.

A review of the CMP process is provided in "Advances in Chemical-Mechanical Planarization," Rajiv K. Singh and Rajiv Bajaj, MRS Bulletin, October 2002, pages 743-747. In general, the CMP process looks very simple, but understanding in detail is mainly limited by a number of input variables in the planarization process. Among such variables are slurry variables (eg particles and chemicals, pad variables, tool variables (eg down pressure and line speed)) and substrate variables (eg pattern density). It provides a good rethinking of process variables and emerging applications and is incorporated herein by reference.

Claims (7)

A CMP abrasive slurry comprising liquid and solid substantially free of anhydrous aluminum oxide, wherein the solid is (a) at least one non-spherical component represented by the formula Al ₂ O ₃ .xH ₂ O, wherein x is 1-3 in an amount of at least about 90% by weight based on said solid; And (b) up to about 1 weight percent submicron alpha-alumina based on the solid portion CMP abrasive slurry comprising a. The CMP abrasive slurry according to claim 1, consisting essentially of said at least one non-spherical component represented by the formula Al ₂ O ₃ xH ₂ O, wherein x is 1-3. The CMP abrasive slurry of claim 1, wherein the non-spherical component is boehmite. The CMP abrasive slurry of claim 1, wherein the non-spherical component comprises kaolin coated with boehmite. A metal planarization method comprising polishing a metal using the CMP abrasive slurry of claim 1. The method of claim 5, wherein the planarization is performed at pH acidic conditions. 6. The method of claim 5, wherein the slurry is used to polish copper.

Images