COBALT CAPPING SURFACE PREPARATION IN MICROELECTRONICS MANUFACTURE

Abstract

Cleaning compositions and methods in connection with cobalt-based capping of interconnects in integrated circuit semiconductor devices.

Description

CROSS-REFERENCE TO RELATED APPLICATION

This application claims priority from U.S. App. Ser. No. 60/909,654, filed Apr. 2, 2007, entitled Cobalt Capping Surface Preparation in Microelectronics Manufacture.

FIELD OF THE INVENTION

The present invention generally relates to the preparation of wafers and conductive surfaces of interconnect features thereof prior to electroless cobalt deposition and to the cleaning of wafers and cobalt capping layers after electroless deposition.

BACKGROUND OF THE INVENTION

The demand for semiconductor integrated circuit (IC) devices such as computer chips with high circuit speed and high circuit density requires the downward scaling of feature sizes in ultra-large scale integration (ULSI) and very-large scale integration (VLSI) structures. The trend to smaller device sizes and increased circuit density requires decreasing the dimensions of interconnect features and increasing their density. An interconnect feature is a feature such as a via or trench formed in a dielectric substrate which is then filled with metal, typically copper, to yield an electrically conductive interconnect. Copper, having better conductivity than any metal except silver, is the metal of choice since copper metallization allows for smaller features and uses less energy to pass electricity. In damascene processing, interconnect features of semiconductor IC devices are metallized using electrolytic copper deposition.

In the context of semiconductor integrated circuit device manufacture, substrates include patterned silicon wafers and dielectric films such as, for example, SiO₂ or low-K dielectrics. Low-K dielectric refers to a material having a smaller dielectric constant than silicon dioxide (dielectric constant=3.9). Low-K dielectric materials are desirable since such materials exhibit reduced parasitic capacitance compared to the same thickness of SiO₂ dielectric, enabling increased feature density, faster switching speeds, and lower heat dissipation. Low-K dielectric materials can be categorized by type (silicates, fluorosilicates and organo-silicates, organic polymeric etc.) and by deposition technique (CVD; spin-on). Dielectric constant reduction may be achieved by reducing polarizability, by reducing density, or by introducing porosity.

Copper can diffuse rapidly into the silicon wafer substrate, SiO₂ or low-K dielectric films, and in subsequently deposited materials such as SiN. Moreover, copper can also diffuse into a device layer built on top of a substrate in multilayer device applications. Such diffusion can be detrimental to the device because it can cause electrical leakage in substrates or form an unintended electrical connection between two interconnects resulting in an electrical short. Moreover, copper diffusion out of an interconnect feature can disrupt electrical flow therethrough. Metal deposited on an integrated circuit substrate also has a tendency to migrate when electrical current passes through interconnect features in service. Electron flux moves the metal atoms from one place of the interconnect feature, creating the void, to a different location, forming hillock. This migration can damage an adjacent interconnect line, and disrupt electrical flow in the feature from which the metal migrates.

In response, barrier layers that surround the copper interconnect have been developed that inhibit copper diffusion out of the feature. For example, it is known to deposit “caps” on interconnect metallization. A “cap” or capping layer refers to a metal layer deposited over an interconnect feature that has been metallized with copper. This is opposed to conventional diffusion barrier layers that are deposited prior to copper metallization, which therefore line the surface of the silicon substrate or dielectric film. These diffusion barrier layers typically comprise tungsten, tantalum, titanium, and alloys thereof.

With respect to the cap barrier layer, cobalt, nickel, and alloys thereof are typical capping layers over copper-metallized interconnect features. See, for example, U.S. Pat. No. 5,695,810, which discloses Co—W—P as a barrier material on a semiconductor wafer. Caps have been developed that incorporate multiple layers of cobalt-based and nickel-based alloys. See U.S. 2005/0275100.

In a typical damascene process, a diffusion barrier lined interconnect feature is metallized by electrolytic copper deposition, which superfills (using superfilling additives) and then overfills the trench or via with a thick copper deposit. See, for example, U.S. 2006/0141784. After interconnect metallization, chemical mechanical polishing is employed to planarize and smooth the metallization deposited within the interconnect features. See, for example, Dubin et al. (U.S. Pat. No. 5,891,513) (describing CMP). Finally, a capping layer, which may be a cobalt-based alloy, a nickel-based alloy, or both is deposited by electroless deposition.

During any of the device manufacturing stages, the possibility of non-uniform, as opposed to planar, surfaces (i.e., copper interconnect surface roughness and variations in the wafer surface) arises. Additionally, processing steps, such as CMP, may leave residues on the surfaces of the both the dielectric material and the copper interconnect. Further, CMP may result in roughness of the copper surface and may leave a layer of copper oxides on the metallization layer or even on the wafer surface. These defects related to both surface non-uniformity and residues may cause, during electroless deposition of the cap, cobalt particle nucleation and growth on both the cobalt cap and on the wafer surface. Moreover, surface roughness can entrap contaminants during wet processing, cause defects and voids thereby promoting electromigration failure, affect the signal propagation across the circuitry, and promote nodular, dendritic growth of the electroless deposit at the interface between the cobalt cap and copper interconnect. These defects can significantly reduce selectivity of the capping layer, increase current leakage, and in extreme cases even result in electrical shorts.

Further, defects present on the wafer surface can affect initiation and deposition of capping materials on structures that are densely packed and those that are isolated. Accordingly, there is a need for a capping method characterized by even initiation and growth resulting in good thickness distribution amongst these structures.

SUMMARY OF THE INVENTION

Briefly, therefore, the present invention is directed to a composition (and associated method) for cleaning an interconnect feature comprising metallization and a dielectric surface of a semiconductor integrated circuit device prior to electroless deposition of a cobalt-based capping layer, the composition comprising a reducing agent, a proton source, a particle suspension agent, and a surfactant, wherein the composition has a pH less than about 6.0.

In another aspect, the invention is directed to a composition (and related method) for cleaning an electrolessly deposited cobalt-based capping layer on an interconnect feature comprising metallization of a semiconductor integrated circuit device, the composition comprising a reducing agent, a proton source, a surfactant, and a particle suspension agent.

Other objects and features will be in part apparent and in part pointed out hereinafter.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

The present invention is directed to methods and compositions for preparing conductive surfaces of interconnect features for electroless cobalt deposition and to the cleaning of cobalt-based capping layers after electroless deposition.

One aspect of the present invention is directed to a cleaning composition that is applied to the surface of a copper interconnect feature and the surrounding dielectric film after chemical mechanical polishing and prior to electroless deposition. The term “pre-cleaning” is used to describe this composition and refers to cleaning of the substrate prior to electroless deposition. The cleaning composition is used to improve the performance of the subsequently deposited electroless cobalt-based capping layers. Features improved by using the cleaning composition of the present invention include film thickness uniformity, decreased surface roughness, and improved selectivity. In this regard, the cleaning composition is formulated to remove surface oxides from the copper interconnect, to remove residues left over from CMP that may be present on both the interconnect and the wafer surface, and to inhibit copper ion re-deposition from solution onto other areas of the surface, such as dielectric surface. All of these defects can result in nucleation sites for particle growth during electroless deposition, such that their removal prior to electroless deposition of the cap is desirable.

In another aspect, the present invention is directed to a cleaning composition that may be used to clean and improve the surface quality of capping layers after electroless cobalt or electroless nickel deposition.

The term “post-cleaning” is used to describe this composition and refers to cleaning of the deposit and/or substrate after electroless deposition.

The present invention is further directed to a method for improving the film thickness uniformity, decreasing surface roughness, and improving the selectivity of electroless cobalt capping layers over copper interconnect features by using the cleaning compositions described generally above.

The pre-cleaning and post-cleaning compositions of the present invention may include a reducing agent, a proton source, a particle suspension agent, and a surfactant.

Persulfate and other components which are known to etch Cu and/or dielectric are specifically avoided because it is a specific goal of this invention to avoid effects which are inclined to increase the amount of particles, microparticles, and fragments in the aqueous cleaning phase. Such particles, microparticles, and fragments increase the risk stray Co deposition, electrical shorts, and other problems as discussed herein. So the composition is preferably substantially persulfate-free and etchant-free, preferably absolutely persulfate-free and etchant-free.

In one embodiment, the pre-cleaning composition further comprises a pH adjuster such as an acid to lower and maintain the pH below about 6.0, such as between about 2.0 and about 4.0. Typically, the post-electroless deposition cleaning composition has a pH greater than about 2.0, such as between about 2.0 and about 11.0; for example, 2.5 to 7.0 in one preferred embodiment.

Applicable reducing agents for use in both the pre-cleaning composition and the post-cleaning composition include hypophosphorous acid, glyoxylic acid, and formic acid. Salts of these reducing agents may be used, such as TMAH, sodium, potassium, and ammonium salts of hypophosphorous acid, glyoxylic acid, and formic acid. A particular reducing agent may be used alone, or in combination with other reducing agents. In one embodiment, for example, the reducing agent is hypophosphorous acid. In one embodiment, the reducing agents is glyoxylic acid.

The reducing agent may be added to the cleaning composition in a concentration of at least about 0.05 g/L. In the context of the pre-cleaning composition, the reducing agent reduces oxides and leaves the surface electron-rich and in a reduced state, thereby more conducive to subsequent metal deposition. Also in the context of the pre-cleaning composition, the reducing agent helps to inhibit oxidation, with a goal of reducing dissolution of copper from the substrate. Copper released from interconnects serves as stray Co initiation sites and are therefore advantageously avoided. In the context of the post-cleaning composition, the reducing agent inhibits oxidation and the formation of Co oxides, which negatively affect electrical properties.

The concentration of the reducing agent is at least about 0.05 g/L, more typically at least about 0.5 g/L. The reducing agent may be added to the cleaning composition in a concentration of less than about 500 g/L. At very high concentrations, the cleaning solution may be too corrosive to copper metallization. Typically, the concentration of the reducing agent is less than about 100 g/L, more typically less than about 10 g/L. Accordingly, the reducing agent concentration may be between about 0.05 g/L and about 500 g/L, typically between about 0.5 g/L and about 10 g/L, more typically between about 1 g/L and about 5 g/L.

Applicable proton sources include polyphosphoric acid, solutions of sulfur dioxide (sulfurous acid), sulfuric acid, phosphoric acid, organic sulfonic acids such as methanesulfonic acid, ethanesulfonic acid, methanedisulfonic acid, and other water soluble organic acids such as acetic acid, carbonic acid, and oxalic acid. The cleaning solutions are preferably acidic to achieve better cleaning, especially for copper oxide removal. It has been observed that alkaline cleaning solutions tend to leave too much copper oxide on the surface, which may cause particle formation and reduced selectivity. Preferably, these compounds are added as their acids to the pre-cleaning composition. A particular proton source may be used alone, or in combination with other proton sources. In one embodiment, for example, the proton source is methanesulfonic acid. In one embodiment, the proton sources are polyphosphoric acid and sulfuric acid.

Sources of polyphosphoric acid and polyphosphate salts include ammonium polyphosphate and sodium polyphosphate, and others of the formula M_(n+2)P_nO_(3n+1), where M represents a counter ion including hydrogen, and n is greater than or equal to two, for example, n=2 to 600. These include tripolyphosphates, tetrapolyphosphates, orthophosphates (P), and pyrophosphates (P—P). A suitable polyphosphate structure is linear and unbranched as follows, where M is a H+ or a monovalent metal cation, such as sodium and potassium:

Other suitable polyphosphate structures are branched, such as in the following fragment:

Some commercially available sources of polyphosphates and polyphosphoric acids include certain quantities of cyclophosphate (metaphosphate) compounds. These include compounds of the formula (MPO₃)_n, where M is a H+ or a metal cation, and n=3 to 600. Examples include cyclotriphosphate and cyclotetraphosphate of the following structures (a) and (b), respectively:

The proton source may be added to the cleaning composition in a concentration sufficient to yield the desired pH. Typically, the concentration of the proton source is at least about 0.05 g/L, more typically at least about 0.5 g/L, even more typically at least about 1 g/L. Typically, the proton source is less than about 500 g/L, more typically, less than about 50 g/L, even more typically less than about 10 g/L. Accordingly, the proton source concentration may be between about 0.05 g/L and about 500 g/L, typically between about 0.5 g/L and about 50 g/L, more typically between about 1 g/L and about 10 g/L.

The cleaning compositions of the invention include one or more surfactants which typically have a hydrophilic head group and a hydrophobic tail.

In one embodiment, the surfactant is an anionic surfactant. Hydrophilic head groups associated with anionic surfactants include carboxylate, sulfonate, sulfate, phosphate, and phosphonate.

In another embodiment, the surfactant is a cationic surfactant. Hydrophilic head groups associated with cationic surfactants include quaternary amine, sulfonium, and phosphonium. Quaternary amines include quaternary ammonium, pyridinium, bipyridinium, and imidazolium.

Alternatively, the surfactant may be non-ionic, including surfactants that comprise polyether groups as the hydrophilic tail, based on, for example, ethylene oxide or propylene oxide also may be applicable.

The surfactant may also be zwitterionic. Hydrophilic head groups associated with zwitterionic surfactants include betaine. The surfactant component may also comprise a combination of two or more of the foregoing types of surfactants, and/or two or more surfactants of the same type.

With respect to all of the various surfactants applicable to this invention, the hydrophobic tail typically comprises a hydrocarbon chain. The chain typically comprises between about six and about 24 carbon atoms, more typically between about eight to about 16 carbon atoms, provided the surfactant is otherwise sufficiently soluble. For those surfactants having a polyether chain, the chain typically comprises between about two and about 24 EO or PO groups, more typically between about 6 and about 15 EO or PO groups, provided the surfactant is otherwise sufficiently soluble.

Applicable species of surfactants within the above description include naphthalene sulfonic acid, naphthalene phosphate, alcohol sulfates such as Niaproof 08 (sodium ethylhexyl sulfate), diphenyl oxide disulfonic acids such as Calfax 10LA-75, triethanolamine salts of lauryl sulfate such as Calfoam TLS-40, ammonium laureth sulfates such as Calfoam EA 603, alkylbenzene sulfonates such as Calsoft L-40C and Calsoft AOS-40, dodecylbenzene sulfonic acids such as Calsoft LAS-99, alkyldiphenyloxide disulfonate salts such as Dowfax 3b2, and soluble, low molecular weight polypropylene glycol containing compounds such as PPG 425.

The surfactant may be added to the cleaning composition in a concentration of at least about 0.01 g/L. The surfactant concentration is at least about 0.01 g/L to achieve sufficient wetting of the surface and to release particles from the wafer surface. Typically, the concentration of the surfactant is at least about 20 ppm, more typically at least about 200 ppm. The surfactant may be added to the cleaning composition in a concentration of less than about 50 g/L. Typically, the concentration of the surfactant is less than about 1000 ppm, more typically less than about 500 ppm. Accordingly, the surfactant concentration may be between about 0.01 g/L and about 50 g/L, typically between about 10 ppm and about 500 ppm, more typically between about 50 ppm and about 100 ppm.

Particle suspension agents are present in the post-cleaning composition and optionally present in the pre-cleaning composition. These particle suspension agents are employed to capture particles such as dirt, CMP slurry, dielectric fragments, and Cu fragments; and, in the context of post-cleaning, these particles with Co coated thereon, and spontaneously generated Co particles. These agents act on these particles and microparticles by changing their zeta potential so they prefer to be in the aqueous phase rather than on the substrate surface, and thus are removed from the substrate surface as the substrate is removed from the cleaning composition aqueous phase. These agents include sulfonated polymers, sulfonated compounds, and quaternary amines. Exemplary particle suspension agents for use in the post-electroless deposition cleaning composition include sodium polyvinylsulfonate, poly-4-styrenesulfonic acid, poly(2-acrylamido-2-methyl-1-propanesulfonic acid), and poly(acrylamide-co-diallyl-dimethylammonium chloride).

The particle suspension agent may be added to the cleaning composition in a concentration of at least about 0.01 g/L. Typically, the concentration of the particle suspension agent is at least about 0.1 g/L. The particle suspension agent may be added to the cleaning composition in a concentration of less than about 100 g/L. At concentrations higher than the maximum, the particle suspension agents may make the composition too viscous.

Typically, the concentration of the particle suspension agent is less than about 1 g/L, more typically less than about 0.5 g/L. Accordingly, the particle suspension agent concentration may be between about 0.1 g/L and about 100 g/L, typically between about 0.1 g/L and about 5 g/L, more typically between about 0.5 g/L and about 2 g/L. The molecular weight of the particle suspension agents range from 5,000 to 3000,000, preferably 200,000 to 2000,000.

The cleaning compositions are preferably manufactured and stored in reduced oxygen or oxygen-free environment. Reducing dissolved oxygen in the cleaning composition both extends the shelf life of the composition and improves its cleaning performance. Dissolved oxygen in the cleaning composition may cause copper oxidation and erosion and re-deposition of copper oxides on interconnect metallization and wafer surface. These defects may increase the resistance of a feature and result in stray particle nucleation and growth during electroless deposition of the capping layer. Moreover, since dissolved oxygen in the cleaning compositions may cause non-uniformity on the surface, it may cause initiation delay of electroless deposition on interconnect features. This can adversely affecting the deposit thickness uniformity.

In one aspect the purpose of the invention is to impart a surface activity to the metal surfaces which is the same regardless of particular surface geometry. This is a challenge because the features range from independent small features to large bonding pads to interdigitized combs (fine parallel lines). This uniform surface activity is critical because it renders all the surfaces equally supportive of deposition. The surfactants in the pre-clean compositions of the invention help achieve these objectives.

In the process of using the cleaning compositions of the present invention, the following protocol may be employed:

(1) Electrolytic deposition of copper to metallize interconnect features.

(2) Chemical mechanical polishing to remove excess copper deposition and clean wafer surfaces.

(3) Pre-cleaning using the pre-electroless deposition cleaning composition of the present invention.

(4) Electroless deposition of cobalt alloy and/or nickel alloy caps.

(5) Post-cleaning using the post-electroless deposition cleaning composition of the present invention.

The substrates may be rinsed with deionized water between steps.

Electrolytic copper deposition may employ any conventional chemistry known in the art, such as ViaForm®, available from Enthone Inc. (New Haven, Conn.). Electroless deposition of cobalt alloy and/or nickel alloy caps processes may be those as disclosed in, for example, U.S. Pub. No. 2005/0275100, U.S. Pub. No. 2006/0083850, and U.S. Pub. No. 2006/0280860, the disclosures of which are hereby incorporated in their entirety. Other methods of damascene metallization include physical vapor deposition and chemical vapor deposition, as are known in the art.

The cleaning compositions used in steps (3) and (5) may be applied to the substrate in any manner sufficient to achieve adequate cleaning of the substrate surface. Exposure may be by flooding, dip, cascade, or spraying. Typical exposure times may be between about 10 seconds and about 5 minutes, such as between about 30 seconds and about 1 minute. The temperature of the cleaning solution may be between about 10° C. and about 40° C., such as between about 20° C. and about 25° C. To enhance cleaning, the wafer may be brushed during exposure of the cleaning compositions.

Having described the invention in detail, it will be apparent that modifications and variations are possible without departing from the scope of the invention defined in the appended claims.

EXAMPLES

The following non-limiting examples further illustrate the present invention.

Example 1

Cleaning Compositions

Four cleaning compositions were prepared comprising the following components and concentrations:

Composition 1 contained phosphinic acid (0.1 g/L to 30 g/L).

Composition 2 contained phosphinic acid (0.1 g/L to 30 g/L) and sodium ethylhexyl sulfate (20 ppm to 5 g/L).

Composition 3 contained methanesulfonic acid (0.1 g/L to 20 g/L) and sodium ethylhexyl sulfate (20 ppm to 5 g/L).

Composition 4 contained polyphosphoric acid (0.1 g/L to 30 g/L) and citric acid (1 g/L to 300 g/L).

Example 2

Thickness Uniformity Improvement Using Cleaning Compositions

The cleaning compositions of Example 1 were used to clean wafer substrates having copper metallized interconnect features prior to electroless deposition of a cobalt alloy capping layers and to clean the capping layers after electroless deposition. The features were metallized by electrolytic deposition from an electrolytic copper chemistry, such as ViaForm®, available from Enthone Inc. The wafers having copper metallized interconnect features were cleaned prior to electroless deposition of the cobalt cap using the cleaning compositions of Example 1 by immersing the wafers in the cleaning compositions for about 1 minute at 25° C. After cleaning, cobalt capping layers were deposited on the copper interconnect features. The cobalt alloy was a quaternary alloy comprising cobalt-tungsten-boron-phosphorus (CoWBP).

Thickness measurements of the deposits were taken from the various features, and the results are presented in Table 1. The conventional cleaner, which lacked surfactant, had deposition much thinner on the isolated feature and the pad than on the dense comb. The performance improved with Example 1, where the phosphinic acid was added instead of citric acid. Thickness was even more uniform with Example 2, which included both phosphinic acid and a surfactant. Example 3 showed good uniformity with MSA and a surfactant. And Example 4 suffered at the isolated feature, probably due to a lack of surfactant.

	TABLE 1
	Deposit Thickness (A)
				Isolated
	type	Pad	Dense comb	feature
	conventional	13 ± 9	85 ± 26	22 ± 18
	Example 1	96 ± 18	95 ± 13	46 ± 6
	Example 2	98 ± 14	80 ± 15	93 ± 7
	Example 3	95 ± 19	93 ± 15	85 ± 8
	Example 4	98 ± 7	85 ± 12	34 ± 21

When introducing elements of the present invention or the preferred embodiments(s) thereof, the articles “a”, “an”, “the” and “said” are intended to mean that there are one or more of the elements. The terms “comprising”, “including” and “having” are intended to be inclusive and mean that there may be additional elements other than the listed elements.

In view of the above, it will be seen that the several objects of the invention are achieved and other advantageous results attained.

As various changes could be made in the above compositions and processes without departing from the scope of the invention, it is intended that all matter contained in the above description shall be interpreted as illustrative and not in a limiting sense.

Claims

1. A method of cleaning an interconnect feature comprising metallization and a dielectric surface of a semiconductor integrated circuit device prior to electroless deposition of a cobalt-based capping layer, the method comprising:

exposing the semiconductor integrated circuit device to a cleaning composition prior to the electroless deposition of the cobalt-capping layer, wherein the cleaning composition comprises a reducing agent, a proton source, and a surfactant, wherein the composition has a pH less than about 6.0.

2. The method of claim 1 wherein the reducing agent is selected from the group consisting of hypophosphorous acid, glyoxylic acid, formic acid, a salt of hypophosphorous acid, a salt of glyoxylic acid, and a salt of formic acid.

3. The method of claim 1 wherein the reducing agent has a concentration between about 0.5 and about 10 g/L and is selected from the group consisting of hypophosphorous acid, glyoxylic acid, formic acid, a salt of hypophosphorous acid, a salt of glyoxylic acid, and a salt of formic acid.

4. The method of claim 1 wherein the proton source is selected from the group consisting of polyphosphoric acid, solutions of sulfur dioxide (sulfurous acid), sulfuric acid, phosphoric acid, methanesulfonic acid, ethanesulfonic acid, methanedisulfonic acid, acetic acid, carbonic acid, and oxalic acid.

5. The method of claim 1 wherein the proton source is present in a concentration between about 0.5 and about 50 g/L and is selected from the group consisting of polyphosphoric acid, polyphosphate salt, solutions of sulfur dioxide (sulfurous acid), sulfuric acid, phosphoric acid, methanesulfonic acid, ethanesulfonic acid, methanedisulfonic acid, acetic acid, carbonic acid, and oxalic acid.

6. The method of claim 1 wherein the surfactant is selected from the group consisting of naphthalene sulfonic acid, naphthalene phosphate, sodium ethylhexyl sulfate, diphenyl oxide disulfonic acid, triethanolamine salt of lauryl sulfate, ammonium laureth sulfate, alkylbenzene sulfonate, dodecylbenzene sulfonic acid, alkyldiphenyloxide disulfonate salt, and soluble, low molecular weight polypropylene glycol.

7. The method of claim 1 wherein the reducing agent is selected from the group consisting of hypophosphorous acid, glyoxylic acid, formic acid, a salt of hypophosphorous acid, a salt of glyoxylic acid, and a salt of formic acid; and the proton source is selected from the group consisting of polyphosphoric acid, solutions of sulfur dioxide (sulfurous acid), sulfuric acid, phosphoric acid, methanesulfonic acid, ethanesulfonic acid, methanedisulfonic acid, acetic acid, carbonic acid, and oxalic acid.

8. A method of cleaning an interconnect feature comprising metallization and a dielectric surface of a semiconductor integrated circuit device prior to electroless deposition of a cobalt-based capping layer, the method comprising:

exposing the semiconductor integrated circuit device to a cleaning composition prior to the electroless deposition of the cobalt-capping layer, wherein the cleaning composition comprises:

a reducing agent in a concentration between about 0.5 and about 10 g/L and selected from the group consisting of hypophosphorous acid, glyoxylic acid, formic acid, a salt of hypophosphorous acid, a salt of glyoxylic acid, and a salt of formic acid; and

a proton source is present in a concentration between about 0.5 and about 50 g/L and selected from the group consisting of polyphosphoric acid, polyphosphate salt, solutions of sulfur dioxide (sulfurous acid), sulfuric acid, phosphoric acid, methanesulfonic acid, ethanesulfonic acid, methanedisulfonic acid, acetic acid, carbonic acid, and oxalic acid; and

a surfactant;

wherein the composition has a pH less than about 6.0.

9. A method of cleaning an electrolessly deposited cobalt alloy or nickel alloy capping layer on an interconnect feature comprising metallization of a semiconductor integrated circuit device, the method comprising:

exposing the semiconductor integrated circuit device to a cleaning composition a reducing agent; a proton source; a surfactant; and a particle suspension agent.

10. The method of claim 9 wherein the reducing agent is selected from the group consisting of hypophosphorous acid, glyoxylic acid, formic acid, a salt of hypophosphorous acid, a salt of glyoxylic acid, and a salt of formic acid.

11. The method of claim 9 wherein the reducing agent has a concentration between about 0.5 and about 10 g/L and is selected from the group consisting of hypophosphorous acid, glyoxylic acid, formic acid, a salt of hypophosphorous acid, a salt of glyoxylic acid, and a salt of formic acid.

12. The method of claim 9 wherein the proton source is selected from the group consisting of polyphosphoric acid, solutions of sulfur dioxide (sulfurous acid), sulfuric acid, phosphoric acid, methanesulfonic acid, ethanesulfonic acid, methanedisulfonic acid, acetic acid, carbonic acid, and oxalic acid.

13. The method of claim 9 wherein the proton source is present in a concentration between about 0.5 and about 50 g/L and is selected from the group consisting of polyphosphoric acid, polyphosphate salt, solutions of sulfur dioxide (sulfurous acid), sulfuric acid, phosphoric acid, methanesulfonic acid, ethanesulfonic acid, methanedisulfonic acid, acetic acid, carbonic acid, and oxalic acid.

14. The method of claim 9 wherein the surfactant is selected from the group consisting of naphthalene sulfonic acid, naphthalene phosphate, sodium ethylhexyl sulfate, diphenyl oxide disulfonic acid, triethanolamine salt of lauryl sulfate, ammonium laureth sulfate, alkylbenzene sulfonate, dodecylbenzene sulfonic acid, alkyldiphenyloxide disulfonate salt, and soluble, low molecular weight polypropylene glycol.

15. The method of claim 9 wherein the reducing agent is selected from the group consisting of hypophosphorous acid, glyoxylic acid, formic acid, a salt of hypophosphorous acid, a salt of glyoxylic acid, and a salt of formic acid; and the proton source is selected from the group consisting of polyphosphoric acid, solutions of sulfur dioxide (sulfurous acid), sulfuric acid, phosphoric acid, methanesulfonic acid, ethanesulfonic acid, methanedisulfonic acid, acetic acid, carbonic acid, and oxalic acid.

17. The method of claim 9 wherein the reducing agent has a concentration between about 0.5 and about 10 g/L and is selected from the group consisting of hypophosphorous acid, glyoxylic acid, formic acid, a salt of hypophosphorous acid, a salt of glyoxylic acid, and a salt of formic acid; and wherein the proton source is present in a concentration between about 0.5 and about 50 g/L and is selected from the group consisting of polyphosphoric acid, polyphosphate salt, solutions of sulfur dioxide (sulfurous acid), sulfuric acid, phosphoric acid, methanesulfonic acid, ethanesulfonic acid, methanedisulfonic acid, acetic acid, carbonic acid, and oxalic acid.

18. The method of claim 9 wherein the particle suspension agent is present in a concentration of between about 0.1 and about 5 g/L and is selected from the group consisting of sulfonated polymers, sulfonated compounds, and quaternary amines.

19. The method of claim 9 wherein the particle suspension agent is selected from the group consisting of sodium polyvinylsulfonate, poly-4-styrenesulfonic acid, poly(2-acrylamido-2-methyl-1-propanesulfonic acid), and poly(acrylamide-co-diallyl-dimethylammonium chloride).