CN111771016A

CN111771016A - Method for electrodeposition of cobalt

Info

Publication number: CN111771016A
Application number: CN201980015252.2A
Authority: CN
Inventors: 文森特·梅费里克; 多米尼克·祖尔; 米卡卢·蒂亚姆; 路易斯·凯拉德
Original assignee: Avini
Current assignee: Alimont Solutions; MacDermid Enthone Inc
Priority date: 2018-03-20
Filing date: 2019-03-15
Publication date: 2020-10-13
Anticipated expiration: 2039-03-15
Also published as: EP3768880A1; TWI804593B; CN111771016B; WO2019179897A1; TW201940746A

Abstract

The present invention relates to a method for manufacturing a cobalt interconnect and an electrolyte capable of implementing the method. The electrolyte has a pH below 4.0 and contains cobalt ions, chloride ions and at most two low molecular weight organic additives. One of these additives may be an alpha-hydroxycarboxylic acid or a compound having a pKa value in the range of 1.8 to 3.5.

Description

Method for electrodeposition of cobalt

Technical Field

The invention relates to the electrodeposition of cobalt on conductive surfaces. More particularly, the present invention relates to a method for the electrodeposition of cobalt, which can be used for the fabrication of electrical interconnects in integrated circuits.

Background

The semiconductor device includes conductive metal interconnects such as trenches extending along the surface and integrated vias connecting various levels. The fabrication of the interconnects involves etching structures in a dielectric material and then depositing a metal seed layer over the entire surface of the structures to enhance their conductivity in a subsequent step of filling with a conductive metal, which is typically performed electrochemically.

The traditional method of filling interconnects with cobalt uses an electrolyte containing a cobalt salt and various organic additives including a suppressor (suppressor) and an accelerator (accelerator) with complementary functions to obtain the filling, known as bottom-up filling. Combinations of these additives are often required to obtain good quality cobalt agglomerates (mass), especially cobalt agglomerates without material voids. The inhibitor controls the deposition of cobalt at the cavity opening and on the planar surface of the substrate surrounding the cavity, either by adsorption to the cobalt surface or by complexing with cobalt ions. Thus, the compound may be a high molecular weight molecule, such as a polymer that is not capable of diffusing within the cavity or a cobalt ion complexing agent. The promoter itself diffuses to the bottom of the cavity, and in very deep cavities its presence is more necessary. It can increase the deposition rate of cobalt on the bottom of the cavity and on its walls. The method of filling using the bottom-up mechanism is in contrast to a filling method called "conformal", in which cobalt deposits grow at the same rate on the bottom and walls of the hollowed-out pattern.

These electrodeposition baths and their use have several drawbacks which ultimately limit the correct operation of the electronic devices manufactured and make their manufacture too costly. In fact, they produce cobalt interconnects that are contaminated with organic additives that are necessary to limit the formation of filled pores in cobalt. Furthermore, the filling rates obtained with these chemical methods are too slow to be compatible with industrial scale production.

Thus, there remains a need to provide an electrolytic bath that produces cobalt deposits with improved properties, i.e., with very low impurity levels, at a rate sufficient to make the manufacture of devices cost effective, and free of material voids to ensure good conductivity.

The present inventors have found that the family of alpha-hydroxycarboxylic acids is capable of achieving this goal.

These additives are known in "bottom-up" or "super conformal" filling processes. In these processes, the bath must contain several additives to accelerate deposition at the bottom of the cavity, slow down deposition on the flat area of the substrate and on the cavity walls. Such a system can prevent void formation in cobalt deposits within the cavity by closing the cavity opening early during the filling process.

However, the accelerator also acts on the planar surface of the substrate outside the cavity, and therefore a leveler (leveller) derived from an α -hydroxy carboxylic acid is typically used to offset/reduce its effect, reduce excessive deposition of cobalt on the surface, and avoid lengthy subsequent polishing steps. Thus, when the trenches and/or vias are almost completely filled, the α -hydroxycarboxylic acid plays its role at the end of the electrodeposition process.

The possibility of using alpha-hydroxy carboxylic acids to deposit cobalt at pH values below 4 has never been suggested in conformal cobalt electrodeposition processes, which makes the results of the present invention even more surprising.

Disclosure of Invention

The invention therefore relates to an electrolyte for the electrodeposition of cobalt, which is in the form of an aqueous solution having a pH between 2.0 and 4.0 and comprises cobalt II ions, chloride ions and at most two non-polymeric organic additives, such as only one α -hydroxycarboxylic acid or two α -hydroxycarboxylic acids.

The term "polymer" refers to a compound comprising at least two repeating units in its chemical formula.

The invention also relates to a method for filling cavities with cobalt by means of a conformal deposition mechanism, using the above-mentioned electrolyte.

The electrolyte and the process of the invention enable continuous cobalt deposits of high purity to be obtained within a production time compatible with industrial applications, which can be shortened compared with the prior art. This is because the method for fabricating the cobalt conductive line implements two distinct cobalt electrodeposition steps: a first electrodeposition step of filling the cavity with a first electrolyte solution containing cobalt ions, and a second electrodeposition step of depositing an "overburden" layer over the entire surface of the substrate using a second electrolyte solution containing cobalt ions. Furthermore, the substrate must be rinsed and dried at the end of the first electrodeposition step before the second step is carried out. The method of the invention advantageously enables the filling of the cavity and the deposition of the cover layer to be carried out in a single electrodeposition step.

Furthermore, the cobalt deposits produced in the context of the present invention have the advantage of a very high purity.

Prior art methods for creating cobalt interconnects use alkaline electrolytes (e.g. pH above 9) to keep the pH above 4 within the trench throughout the filling step by applying very low current densities and cobalt specific inhibiting compounds, which results in the formation of significant amounts of cobalt hydroxide in the resulting cobalt deposit, which reduces the conductivity of the cobalt interconnect and reduces the performance of the integrated circuit.

The electrodeposition process that can be carried out using the electrolyte of the present invention follows a conformal fill pattern and therefore does not require organic additives that are used in large quantities and that can generate contamination in prior art bottom-up fill processes.

Subsequently, working in a pH range below 4.0 has the advantage of limiting the formation of cobalt hydroxide and makes it possible to dispense with the use of buffer compounds such as boric acid, which are necessary to stabilize the pH during the polarization phase of the process using an alkaline electrolyte as described in the prior art. However, boric acid, which is commonly used to perform this function, decomposes into boron derivatives that contaminate the cobalt deposits. Contamination is even more pronounced because the concentration of the buffer compound in the electrolyte must be high to stabilize the pH of the electrolyte during electrodeposition.

Thus, the electrolyte and method of the present invention can greatly limit contamination of cobalt deposits by limiting the concentration of organic molecules (e.g., buffer species) and the formation of cobalt hydroxide during electrodeposition.

The electrolyte of the present invention also has the advantage of producing a cobalt line or via that does not contain voids.

Furthermore, the electrolyte and the method of the invention enable cobalt interconnects with very low impurity content, preferably below 1000ppm atomic, to be obtained while forming at faster deposition rates.

Definition of

The term "electrolyte" refers to a liquid containing a precursor of a metal coating used in an electrodeposition process.

The term "continuous fill" refers to a cobalt mass without voids. In the prior art, pores or voids of material ("sidewall voids") can be observed in the cobalt deposit between the pattern walls and the cobalt deposit. Voids equidistant from the pattern walls can also be observed in the form of holes or lines ("seams"). By making a cross-section of the deposit, these voids can be observed and quantified by transmission electron microscopy or scanning electron microscopy. The continuous deposit of the invention preferably has an average porosity of less than 10% by volume, preferably less than or equal to 5% by volume. The measurement of the porosity inside the structure to be filled can be carried out by electron microscopy at a magnification of between 50000 and 350000.

The term "average diameter" or "average width" of a cavity refers to the size measured at the opening of the cavity to be filled. For example, the cavity is in the form of a tapered channel or a cylinder.

The term "conformal filling" refers to a filling pattern in which cobalt deposits grow at the same rate on the bottom and walls of the openwork pattern. This filling is in contrast to bottom-up (referred to as "bottom-up") filling where cobalt is deposited at the bottom of the cavity at a faster rate.

The term "buffer substance" or "buffer compound" refers to a compound that is part of an electrolyte containing cobalt ions and chloride ions and has a pH in the range of 2.0 to 4.0. In the electrical step of the electrodeposition method, the compound is used in an amount sufficient to stabilize the pH of the electrolyte within ± 0.3 (preferably within ± 0.2) after contact with the conductive surface of the substrate to be covered with cobalt metal. Thus, a compound may be a buffer substance at a given concentration in a given electrolyte, and no longer be a buffer substance in the same electrolyte if its concentration is insufficient to avoid a pH change of the electrolyte during an electrical step. The expression "a substance in an amount sufficient to produce a buffering effect" may also be used.

Drawings

Fig. 1 and 2 are transmission electron microscope images of cavities filled according to the method of the invention (example 1 and example 2).

Fig. 3 is a scanning electron microscope image of a cavity filled according to a prior art electrodeposition method (comparative example 3).

Detailed Description

According to a first embodiment, the invention relates to an electrolyte for the electrodeposition of cobalt comprising, in an aqueous solution, cobalt II ions, chloride ions, an acid in an amount sufficient to obtain a pH between 1.8 and 4.0 (for example between 2.0 and 4.0), and at most two organic additives (preferably only one or at most two organic additives), at least one of which is selected from alpha-hydroxycarboxylic acids preferably free of sulphur.

In particular, the invention relates to an electrolyte for the electrodeposition of cobalt, characterized in that it is an aqueous solution comprising from 1g/L to 5g/L of cobalt II ions, from 1g/L to 10g/L of chloride ions, an amount sufficient to obtain an acid having a pH of between 1.8 and 4.0 (for example between 2.0 and 4.0), and up to two organic additives which are not polymers and preferably do not contain sulfur, at least one or even two of the organic additives being preferably an α -hydroxycarboxylic acid.

The electrolyte preferably comprises at most one organic additive, which may be a sulfur-free alpha-hydroxycarboxylic acid.

The molecular weight of the organic additive is preferably less than 250g/mol, preferably less than 200g/mol and more than 50g/mol, more preferably more than 100 g/mol.

The concentration of the additive or the sum of the concentrations of the two additives is preferably between 5mg/L and 200 mg/L. In this embodiment, each of the additives may be a sulfur-free α -hydroxycarboxylic acid.

The mass concentration of cobalt II ions may range from 1g/L to 5g/L (e.g., from 2g/L to 3 g/L). The mass concentration of the chloride ions can range from 1g/L to 10 g/L.

The relatively high cobalt ion concentration at strongly acidic pH has several advantages over prior art electrolyte baths having a basic or weakly acidic pH, which is lower in cobalt ion concentration.

In fact, contrary to what is taught in the prior art, the inventors have found that it is not necessary to work at a pH above 4 in order to limit the corrosion of cobalt deposits. By increasing the concentration of cobalt ions and lowering the pH, it is possible, without being bound by any theory, to stabilize the deposit of cobalt metal by substantially increasing the concentration of cobalt ions present in the aqueous solution. Thus, the inventors observed a faster deposition rate than the prior art, and observed larger size cobalt grains (grains) in the deposit, typically greater than 100 nm.

The chloride ions may be provided by dissolving cobalt chloride or a hydrate thereof (e.g., cobalt chloride hexahydrate) in water.

Since cobalt salts such as cobalt sulfate or hydrates thereof produce sulfur-containing contaminants of cobalt deposits, which are desirably avoided, the composition is preferably not obtained by dissolving cobalt salts such as cobalt sulfate or hydrates thereof.

The organic additive is preferably free of sulphur and is preferably selected from alpha-hydroxy carboxylic acids, such as citric acid, tartaric acid, glycolic acid, lactic acid, malic acid, mandelic acid, maleic acid, oxalic acid and 2-hydroxy butyric acid compounds.

The additional organic compound may be of any nature provided that it does not cause a bottom-up fill effect. The compounds may have various functions, such as the function of an accelerator, a suppressor, a growth promoter or a leveler, but the electrolyte of the invention is advantageously free of these compounds. For example, the electrolyte of the present invention is free of inhibitor polymers, especially polymers such as polyethylene glycol, polyvinylpyrrolidone or polyethyleneimine.

The hydroxylated carboxylate is, for example, tartrate, and the electrolyte preferably comprises at most one organic additive.

In the electrolyte according to the invention, preferably before and during poling, the cobalt II ions are advantageously in free form, i.e. not complexed with organic additives, which may be, for example, α -hydroxycarboxylic acids, glycine or ethylenediamine.

The absence of numerous complexes of cobalt with organic molecules has many advantages as follows: because the concentration of organic molecules in the bath can be very low, organic contamination of the cobalt metal deposit can be reduced; any uncontrolled change in pH that may destabilize the solution is also avoided throughout the deposition of cobalt in the structure. In addition, cobalt ions are not stabilized by the complex and are more easily reduced, so that the deposition rate of cobalt is faster. Finally, the very high concentration of cobalt ions protects the conductive surfaces of the cavity from corrosion. This effect is decisive when the substrate is covered with a cobalt layer (seed layer) of very small thickness.

When the organic additive or both organic additives are alpha-hydroxycarboxylic acids and the electrolyte pH is below 4.0, the additive does not complex with cobalt ions.

In the context of the method of the invention in the cavity to be filled, a distinction can be made between a flat part and several cutouts on the flat part. One of the goals pursued by the prior art is to slow down the deposition of cobalt on the flat portions using inhibitors that adsorb specifically on the flat surface of the substrate without penetrating into the hollows of the pattern. In the present specification, they are referred to as surface inhibitors.

The electrolyte according to the invention advantageously comprises one of the following features, alone or in combination:

-no promoter of cobalt growth is included at the bottom of the pattern;

the electrolyte is free of organic inhibitor molecules capable of slowing down the growth of cobalt on the flat portion of the substrate at the opening of the cavity by specifically adsorbing to the cobalt deposited in this position during electrodeposition;

-is free of polymer;

-is free of sulphur-containing compounds;

-free of buffering compounds useful in alkaline media, such as boric acid;

-a combination of additives not causing a bottom-up filling mechanism, in particular a combination of inhibitor and accelerator, or a combination of inhibitor, accelerator and levelling agent.

Among the surface inhibitors, the following compounds may be mentioned: carboxymethylcellulose, nonylphenol polyglycol ether, polyethylene glycol dimethyl ether, octanediol bis (polyalkylene glycol ether), octanol polyalkylene glycol ether, oleic acid polyglycolide, poly (ethylene glycol-propylene glycol), polyethylene glycol, polyethyleneimine, polyethylene glycol dimethyl ether, polyoxypropylene glycol, polypropylene glycol, polyvinyl alcohol, stearic acid polyglycolide, stearyl alcohol polyglycolide, butanol/ethylene oxide/propylene oxide copolymer, 2-mercapto-5-benzimidazolesulfonic acid, 2-mercaptobenzimidazole.

Accelerators are usually compounds containing a sulfur atom, for example the (3-sulfopropyl) ester of N, N-dimethyldithiocarbamate, the (3-sulfopropyl) ester of 3-mercaptopropanesulfonic acid, the ester of 3-mercapto-1-propanesulfonate (3-sulfo-1-propanesulfonate), the o-ethyl dithiocarbonate s-ester (dithiocarboxylic acid o-ethyl esters-ester) with the potassium salt of 3-mercapto-1-propanesulfonate, disulfopropyl disulfide, the sodium salt of 3- (benzothiazolyl-s-thio) propylsulfonate, pyridylpropylsultaine, sodium 3-mercapto-1-propanesulfonate, (3-sulfoethyl) ester of N, N-dimethyldithiocarbamate, the salts of N, N-dimethyldithiocarbamate, (3-sulfoethyl) ester of 3-mercaptoethylpropanesulfonic acid, sodium salt of 3-mercaptoethylsulfonic acid, pyridylethylsulfobetaine or thiourea.

In the first embodiment, the pH of the electrolyte is preferably between 2.0 and 4.0. In a particular embodiment, the pH is between 2.0 and 3.5, or between 2.0 and 2.4, or between 2.5 and 3.5, or between 2.8 and 3.2.

The pH of the composition may optionally be adjusted using bases or acids known to those skilled in the art. The acid used may be hydrochloric acid.

In a preferred embodiment of the invention, the electrolyte contains less than 500ppm of a buffer compound (having at least 1pKa) capable of preventing a change in the pH of the electrolyte during the polarization step of the electrodeposition process. The concentration of the buffer compound in the electrolyte is preferably below 400ppm, 300ppm or even 250 ppm. In the electrical step, for example for the separate addition of acid (e.g. hydrochloric acid), the pH of the electrolyte can be readjusted if necessary.

However, in a preferred embodiment of the invention, the electrolyte does not contain a large amount of buffer substance.

Although there is no fundamental limitation on the nature of the solvent (provided that it adequately dissolves the active substance in solution and does not interfere with electrodeposition), the solvent is preferably water. According to one embodiment, the solvent comprises essentially a volume dose of water.

According to a second embodiment, the invention relates to an electrolyte for the electrodeposition of cobalt, the pH of which is between 1.8 and 4.0 (for example between 2.0 and 4.0) and comprising, in an aqueous solution, cobalt II ions, chloride ions and between 5mg/L and 200mg/L of one or more compounds having at least 1pKa (range 1.8-3.5, preferably range 2.0-3.5, more preferably range 2.2-3.0).

This second embodiment may include the following features.

The molecular weight of the compound is preferably less than 250g/mol, preferably less than 200g/mol and more than 50g/mol, preferably more than 100 g/mol.

In some cases, the compound having a pKa value of at least 1 (in the range of 2.0 to 3.5) may be the same as at least one of the organic additives used in the first embodiment. It may be chosen in particular from citric acid, tartaric acid, malic acid, maleic acid and mandelic acid.

It may also be selected from the compounds fumaric acid (pKa ═ 3.03), glyceric acid (pKa ═ 3.52), orotic acid (pKa ═ 2.83), malonic acid (pKa ═ 2.85), L-alanine (pKa ═ 2.34), phosphoric acid (pKa ═ 2.15), acetylsalicylic acid (pKa ═ 3.5) and salicylic acid (pKa ═ 2.98).

The prior art method of filling with cobalt uses an alkaline electrolyte (e.g. pH above 9) while applying a very low current density and a cobalt specific inhibiting compound to keep the pH above 4 within the trench throughout the filling step, which results in a significant formation of cobalt hydroxide in the resulting cobalt deposit, which reduces the conductivity of the cobalt interconnects and reduces the performance of the integrated circuit.

The electrolyte of the invention and the method of the invention aim at solving this problem by significantly limiting the formation of cobalt hydroxide in such a way that it is present only in trace amounts in the deposited cobalt. A solution to this problem consists in using an electrolyte having a pH between 1.8 and 4.0 (for example between 2.0 and 4.0) and adding to the electrolyte an additive, preferably having at least one or even all of the following characteristics such as:

a buffer capacity enabling the pH of the electrolyte to be maintained at a value higher than 1.8 or 2.0 and lower than 3.5 (preferably lower than 2.5) throughout the poling of the substrate;

-a low molecular weight enabling the additive to diffuse in structures with small opening diameters; and

a very low concentration in the electrolyte, so that the amount of additive present in the electrolyte before the start of polarization almost completely diffuses into the cavities of the structure, and the additive has a local buffering capacity.

The electrolyte containing such an additive is able to limit in a selective manner (for example only in the cavities of the structures and not on the flat surface of the substrate) the increase in pH, the value of which is lower than 4.0, preferably lower than 3.0 and more preferably between 2.0 and 2.5. Thus, the additive may advantageously act as a buffer by exerting its effect locally, i.e. only in the cavity. Organic additives or compounds having a pKa of at least 1 (in the range of 1.8-3.5 or 2.0-3.5) can act as local buffers, the effect of which is only observed in the cavity.

This second embodiment may include other features which may correspond to some of the features of the first embodiment of the invention described above.

The invention also relates to an electrochemical method for filling a cavity, the method comprising:

-a step of bringing the conductive surface of the cavity into contact with one of the above-mentioned electrolytes, and

-a step of poling the conductive surface for a time sufficient to completely fill the cavity by cobalt deposition to be obtained, such filling being preferably conformal.

The method preferably comprises a step of annealing the cobalt deposit obtained at the end of the polarization step.

The invention also relates to an electrochemical method for depositing cobalt on a substrate comprising a conductive surface within a cavity hollowed out in the substrate and a conductive surface outside the cavity, the method comprising:

-a step of contacting the conductive surface with an electrolyte in the form of an aqueous solution having a pH between 2.0 and 4.0 and comprising cobalt II ions, chloride ions and only one or at most two organic additives, which are not polymers,

-a step of poling the conductive surface for a time sufficient for filling the cavity with cobalt and depositing a cobalt layer at least 50nm thick on the conductive surface outside the cavity,

-a step of annealing the cobalt obtained at the end of the polarization step.

The cobalt layer, also referred to as capping layer, may have a thickness between 20nm-300 nm. It advantageously has a constant thickness over the entire surface of the substrate outside the cavity. This layer is also uniform, glossy and dense (compact). The purity thereof is preferably less than 1000ppm atomic.

The method of the invention may be carried out using one of the above-mentioned electrolytes comprising one or two non-polymeric organic additives as described in the first embodiment, or comprising between 5mg/L and 200mg/L of one or more compounds having a pKa of at least 1 (in the range 1.8-3.5, preferably 2.0-3.5, and more preferably 2.2-3.0).

The electrolyte used in the context of the method of the invention may correspond to the electrolyte of the first embodiment described above or the second embodiment described above.

Throughout the implementation of the filling step of the method of the invention, the pH inside the cavity is advantageously kept below 3.5 or even below 3.0, depending on the nature of the electrolyte used.

The cavities may be designed in the context of the implementation of the Damascene (damascone) method or the Dual-Damascene (dumascone) method. In particular, the cavity may be obtained by carrying out the following steps:

-a step of etching the structure on the silicon substrate,

a step of forming a silicon oxide layer on the silicon surface of the structure to obtain a silicon oxide surface,

-a step of depositing a metal layer on said silicon oxide layer to obtain a conductive surface of the cavity.

The metal layer has a thickness of between 1nm and 10nm, for example. It is preferably deposited on a silicon oxide layer in contact with the silicon.

The method of the present invention is a conformal method as opposed to the "bottom-up" or "super-conformal" methods of the prior art. In the conformal filling method of the present invention, cobalt deposits grow at the same rate on the bottom and walls of the hollow pattern to be filled. This filling is in contrast to other prior art methods in which cobalt is deposited at a faster rate at the bottom of the cavity than at the walls of the cavity.

The electrodeposition process of the invention may use the bath described above as subject of the first aspect of the invention. All features described in relation to the first aspect of the invention are applicable to the electrodeposition method.

The total impurity content of the cobalt deposit obtained by the electrodeposition process of the invention is less than 1000ppm atomic. The impurities are primarily oxygen, and secondarily carbon and nitrogen. The total content of carbon and nitrogen is less than 300 ppm. The cobalt precipitate is advantageously continuous. It preferably has an average porosity of less than 10% by volume or surface area, preferably less than or equal to 5% by volume or surface area. The porosity in cobalt deposits can be measured by electron microscopy observations known to those skilled in the art, who will select the method they consider most appropriate. One of these methods may be Scanning Electron Microscopy (SEM) or Transmission Electron Microscopy (TEM) using magnifications between 50000 and 350000. The void volume can be assessed by measuring the surface area of the voids observed on one or more cross sections of the substrate containing the filled cavities. In the case where multiple surface areas are measured over multiple cross-sections, the average of these surface areas will be calculated to estimate the void volume.

The electrodeposition step is typically stopped when the cobalt deposit covers the flat surface of the substrate: in this case, the cobalt deposit comprises a cobalt deposit in a cavity and a cobalt layer covering the surface of the substrate, the cavity being hollowed out in the substrate. The thickness of the cobalt layer covering the surface may be between 50nm-250nm, and may for example be between 125nm-175 nm.

The low impurity content and extremely low porosity enable lower resistivity cobalt deposits to be obtained.

The cobalt deposition rate is between 0.1nm/s and 3.0nm/s, preferably between 1.0nm/s and 3.0nm/s, and more preferably between 1nm/s and 2.5 nm/s.

Through-vias (vias) and interconnects may be fabricated according to a damascene or dual damascene method known to those skilled in the art, the method comprising a series of steps including: -etching a pattern in or through a silicon wafer on a spindle perpendicular to the wafer surface to obtain a pattern with a hollowed-out vertical profile; -depositing an insulating dielectric layer, typically consisting of silicon oxide; -depositing a layer of material for preventing cobalt migration in silicon; -optionally depositing a thin metal layer (called seed layer); -filling the pattern by electrodeposition of cobalt; and-removing the excess cobalt by polishing.

Conductive cobalt lines may be formed on the front (FEOL lines) or back (BEOL lines) of the metallization structure of the semiconductor device.

A silicon substrate that has been etched according to a desired pattern and then covered with a silicon oxide layer, followed by a metal layer (which may be a seed layer of a metal), a barrier layer for cobalt diffusion, a liner, or a combination of at least two of the above may be used. The metal layer may have a thickness between 1nm-10nm (e.g., between 2nm-5 nm) and may comprise, for example, a single layer or multiple superimposed layers of various materials.

The conductive surface of the pattern is a surface of a metal layer containing at least one compound selected from the group consisting of cobalt, copper, tungsten, titanium, tantalum, ruthenium, nickel, titanium nitride, and tantalum nitride. In a particular embodiment, the conductive surface is a cobalt surface.

The metal layer may thus comprise a tantalum nitride TaN layer having a thickness between 1nm and 6nm, itself covered and in contact with a cobalt metal layer having a thickness between 1nm and 10nm (preferably between 2nm and 5 nm), on which cobalt is deposited during the electrical step.

The resistivity of the component comprising the metal layer and cobalt deposit can range from 7ohm/cm to 10 ohm/cm. The resistivity is preferably between 7.5ohm/cm and 8.5 ohm/cm.

According to the method of the invention, the width or diameter of the cavity intended to be filled with cobalt at its opening, which is at the level of the substrate surface where the cavity is created, is preferably between 10nm and 200nm or between 10nm and 100nm, preferably between 10nm and 40nm, and more preferably between 15nm and 30 nm. The depth of the cavity may range from 50nm to 250 nm. According to one embodiment, the cavity has a width between 10nm and 30nm and a depth between 125nm and 175 nm.

The electrolyte of the present invention has been shown to be able to fill very thin trenches or small vias with widths or diameters below 40nm at the openings, at speeds higher than the prior art methods, and consistent with conformal filling. The electrolyte is also capable of obtaining a continuous cobalt agglomerate without material voids and with a very low content of impurities, such as cobalt hydroxide.

The range of polarization intensity used in the electrical step is preferably 2mA/cm²-20mA/cm²However, in the prior art method using an alkaline electrolyte, the range is usually 0.2mA/cm²-1mA/cm²。

The electrical step of the method of the invention may comprise only one or more polarization steps, wherein the person skilled in the art will be able to select the variables according to his general knowledge.

The electrical step may be performed using at least one polarization mode selected from the group consisting of ramp mode (ramp mode), galvanostatic mode (galvano-pulsed mode), and current pulse mode (galvano-pulsed mode).

Thus, the electrical step may comprise at least one electrodeposition step in current pulse mode and at least one electrodeposition step in galvanostatic mode, the electrodeposition step in galvanostatic mode preferably being subsequent to the electrodeposition step in current pulse mode.

For example, the electrical step comprises a first step of polarizing the cathode in a current pulse mode in a current range of 3mA/cm²-20mA/cm²(e.g., 12 mA/cm)²-16mA/cm²Time (T)_on) Preferably between 5ms and 50 ms) and zero polarization (time (T)_off) Preferably between 50ms and 150 ms).

In a first step, the substrate may be contacted with the electrolyte solution before or after poling. Preferably in contact with the cavity prior to application of the voltage to limit corrosion of the metal layer in contact with the electrolyte.

In a second step, the cathode may be polarized in a constant current mode with a current range of 3mA/cm²-20mA/cm². The two steps are preferably of substantially equal duration.

The second step in galvanostatic mode may itself comprise two steps: a first step in which the intensity of the current is in the range of 3mA/cm²-8mA/cm²(ii) a Anda second step in which an intensity range of 9mA/cm is applied²-20mA/cm²The current of (2).

This electrical step can be used in particular when the pH of the electrolyte is between 2.5 and 3.5.

According to another example, the electrical step comprises a first step of poling the cathode in a ramp mode, the current of said ramp mode preferably being 0mA/cm²-15mA/cm²(preferably 0 mA/cm)²-10mA/cm²) Followed by a step in galvanostatic mode, applied in a range of 10mA/cm²-20mA/cm²(preferred range is 8mA/cm²-12mA/cm²) The current of (2). This electrical step can be used in particular when the pH of the electrolyte is between 2.0 and 2.5.

Preferably, the method of the invention comprises a step of annealing the cobalt deposit obtained at the end of the above-mentioned filling.

The annealing heat treatment may be carried out at a temperature between 350 ℃ and 550 ℃ (e.g., about 450 ℃), preferably in a reducing gas (e.g., N)₂4% of H₂) The process is carried out as follows.

The method may include a pre-treatment step to reduce native metal oxide present on the surface of the substrate by reducing the plasma. The plasma also acts on the surface of the trench, making it possible to improve the quality of the interface between the seed layer and the electrodeposited cobalt. It is preferred to perform a subsequent electrodeposition step immediately after the plasma treatment to minimize the reformation of native oxide.

The present application also describes an electrochemical method for conformal filling of a cavity, the method comprising:

-a step of contacting the electrically conductive surface of the cavity with an aqueous solution having a pH between 2.0 and 4.0 and comprising cobalt II ions, chloride ions and at most two organic additives comprising an alpha-hydroxycarboxylic acid, and

-a step of poling the conductive surface so as to fill the cavities with cobalt according to a conformal filling pattern.

Alpha-hydroxycarboxylic acids, due to their small volume, preferably diffuse within the cavity.

The method may satisfy one or more of the features previously described with respect to the electrochemical method of the present invention.

The method of the present invention may be used in the manufacture of semiconductor devices, in particular during the creation of conductive metal interconnects (e.g. trenches extending along the surface and integrated vias connecting various levels).

Examples

The invention is further illustrated by the following exemplary embodiments.

Example 1: electrodeposition of 26nm wide, 150nm deep structures with organic additives at pH3.0

The trenches, 26nm wide and 150nm deep, were filled with cobalt by electrodeposition on a cobalt seed layer. The deposition was carried out using a composition containing cobalt chloride and an alpha-hydroxycarboxylic acid at ph 3.0. Finally, the substrate is heat treated to improve the quality of the deposited metal.

A. Materials and equipment:

substrate:

the substrate used in this example was composed of a 4cm × 4cm silicon sample blank (silicon wafer). The silicon was covered with silicon oxide, which was in contact with a 4nm thick TaN layer, which itself was covered with and in contact with a 3nm thick cobalt metal layer. The trench to be filled thus has a width of 26nm, a depth of 150 nm. The measured resistivity of the substrate (with its cavities filled with cobalt) was about 300 ohm/square.

Electrodeposition solution:

in this solution, Co²⁺The concentration obtained from CoCl was equal to 2.47g/L₂(H₂O)₆. The concentration of tartaric acid is between 5ppm and 200ppm, for example equal to 15 ppm. The pH of the solution was adjusted to 3.0 by adding hydrochloric acid.

Equipment:

in this example, an electrodeposition apparatus composed of two parts was used: a tank intended to contain an electrodeposition solution, equipped with a fluid recirculation system to control the fluid dynamics of the system; and a rotating electrode equipped with a sample holder adapted to the size of the sample embryo used (4cm x 4 cm). The electrowinning cell comprises two electrodes:

-a cobalt anode;

-a structured silicon sample blank coated with the above layer, constituting a cathode;

-a reference connected to the anode.

The connectors enable electrical contact of the electrodes, which are connected by wires to a potentiostat supplying up to 20V or 2A.

B. The experimental scheme is as follows:

the preparation method comprises the following steps:

by H₂The silicon sample was treated with a plasma (0.5mbar, 70W, 5min) for a short period of time to reduce native cobalt oxide on the substrate.

The electrical method comprises the following steps:

the method comprises three steps:

a) in a first step, the cathode is polarized in a current pulse mode with a current range of 30mA (or 3.8 mA/cm)²) 150mA (or 19 mA/cm)²) E.g. 110mA (or 14 mA/cm)²) The pulse time of the cathodic polarization is between 5ms and 50ms, and the pulse time of the zero polarization between two cathodic pulses is between 50ms and 150 ms. This step was carried out for 50 seconds with a cycle of 50 rpm. The electrolyte is brought into contact with the substrate before the voltage is applied.

b) In a second step, the cathode is polarized in a galvanostatic mode, with a current range of 30mA (or 3.8 mA/cm)²) 60mA (or 7.6 mA/cm)²) For example, a current of 55mA (or 7 mA/cm)²). This step was carried out for 20 seconds with a cycle of 100 rpm.

c) In the final step, the cathode is polarized in galvanostatic mode, with a current range of 80mA (or 10 mA/cm)²) 150mA (or 19 mA/cm)²) For example, 110mA (or 13.75 mA/cm)²). This step was performed for 30 seconds with a cycle of 100 rpm.

After these three successive electro-grafting steps, a complete filling of the structures with high aspect ratio is obtained within 100s, on which a thick layer of cobalt is deposited.

And (3) annealing:

annealing in a reducing gas (N)₂4% of H₂Referred to as forming gas) at 500 c for 10 minutes.

C. The results obtained were:

after annealing, analysis by transmission electron microscopy (Mag 320k, EHT 100kV) showed that the pores on the trench walls ("sidewall voids") were filled with no defects, reflecting good nucleation of cobalt, and no pores in the structure ("seam voids"), reflecting that annealing eliminated this type of defect (see fig. 1). The resistivity of the 150nm thick cobalt layer on the structure was 8.0ohm cm.

The total impurity content of the cobalt deposit obtained was equal to 790ppm atomic. The total content of carbon and nitrogen is equal to 200 ppm.

The cobalt deposition rate was equal to 1.5 nm/s.

Example 2: electrodeposition of 16nm wide, 150nm deep structures with organic additives at pH3.0

By electrodeposition on a cobalt layer, trenches (16nm wide and 150nm deep) that were more challenging (aggressive) than example 1 were filled with cobalt. The deposition was still carried out using a composition containing cobalt chloride and tartaric acid at ph 3.0. The substrate is still heat treated in order to improve the quality of the deposited metal.

A. Materials and equipment:

substrate:

the substrate used in this example was obtained from a 4cm × 4cm silicon sample blank. The silicon was covered with silicon oxide, which was in contact with a 4nm thick TaN layer, which itself was covered with and in contact with a 3nm thick cobalt metal layer. The trench to be filled thus has a width of 16nm, a depth of 150 nm. The measured resistivity of the substrate was about 170 ohm/square.

Electrodeposition solution:

the solution was the same as the solution of example 1.

Equipment:

the apparatus is the same as that of example 1.

B. The experimental scheme is as follows:

the preparation method comprises the following steps:

the plasma treatment was the same as that of example 1.

The electrical method comprises the following steps:

the process is as described in example 1; in this case, the process is also carried out in three steps and is identical.

And (3) annealing:

this annealing step is exactly the same as the annealing step of example 1.

C. The results obtained were:

after annealing, analysis by transmission electron microscopy (Mag 320k, EHT 100kV) showed that the pores on the trench walls ("sidewall voids") were filled with no defects, reflecting good nucleation of cobalt, and no pores in the structure ("seam voids"), reflecting that annealing eliminated this type of defect (see fig. 2). The resistivity of the 150nm thick cobalt layer on the structure was 8.5ohm cm.

When the same process is carried out at a pH equal to 3.0, a cobalt deposit is obtained having a total impurity content equal to 790ppm atomic. The total content of carbon and nitrogen is equal to 200 ppm.

The cobalt deposition rate was equal to 1.5 nm/s.

Comparative example 3: electrodeposition of 26nm wide, 150nm deep structures with organic additives at pH5.3

The trenches, 26nm wide and 150nm deep, were filled with cobalt by electrodeposition on a cobalt seed layer. The deposition was carried out using a composition containing cobalt dichloride hydrate and tartaric acid at ph 5.3. Finally, the substrate is heat treated to improve the quality of the deposited metal.

A. Materials and equipment:

substrate:

the substrate used is exactly the same as the substrate of example 1.

Electrodeposition solution:

in this solution, Co²⁺The concentration obtained from CoCl was equal to 2.47g/L₂(H₂O)₆. Tartaric acid in a concentration of 5ppm to 200ppm, e.g. equal to15 ppm. The pH of the solution was adjusted to pH5 by adding hydrochloric acid.

Equipment:

the apparatus is the same as that of example 1.

B. The experimental scheme is as follows:

the electrical method comprises the following steps:

as described in example 1, the process is also carried out in three steps in this case and is identical.

And (3) annealing:

the annealing step is exactly the same as the annealing step of example 1.

C. Results and discussion:

analysis of the substrate by scanning microscopy (Mag 100k, WD 2.2nm, signal a ESB, EHT 2.0kV) showed poor fill quality and many "seam void" type hole defects, visible as black in fig. 3.

These results emphasize the need to use a pH lower than 4 in order to obtain good electrodeposition from the solutions used.

The resistivity of a 150nm thick cobalt layer on the substrate surface was 9.0ohm cm.

Example 4: electrodeposition of 16nm wide, 150nm deep structures with organic additives at pH 2.2

Trenches 16nm wide and 150nm deep were filled with cobalt by electrodeposition on a cobalt seed layer. This example is different from example 2 due to the conditions of the electrical step and its pH.

A. Materials and equipment:

substrate:

the substrate was the same as that of example 2.

Electrodeposition solution:

the solution was the same as the solution of example 1, but the pH was adjusted to 2.2.

Equipment:

the apparatus is the same as that of example 1.

B. The experimental scheme is as follows:

the preparation method comprises the following steps:

the plasma treatment was the same as that of example 1.

The electrical method comprises the following steps:

the method comprises the following two steps:

a) in a first step, the cathode is polarized in a current dynamic ramp mode (galvanodynamic ramp mode), in which the current is varied proportionally with time from a value greater than or equal to 0mA to a maximum value of 110mA (or 13.75 mA/cm)²). In this embodiment, the current is varied from 0mA to 80mA (or 10 mA/cm) at a rate of 1.33mA/sec²). This step was carried out for 60 seconds with a cycle of 50 rpm.

b) In a second step, the cathode is polarized in a galvanostatic mode, with a current range of 80mA (or 10 mA/cm)²) 150mA (or 19 mA/cm)²) E.g. 80mA (or 10 mA/cm)²). This step was carried out for 40 seconds with a cycle of 50 rpm.

After these two successive steps, a complete filling of the structures with high aspect ratio is obtained within 100s, and a cobalt layer is also formed on the planar surface of the substrate.

And (3) annealing:

annealing in a reducing gas (N) by a rapid thermal annealing process₂4% of H₂Referred to as forming gas) at 450 ℃ for 5 minutes.

C. The results obtained were:

TEM analysis performed after annealing showed that the holes on the trench walls ("sidewall voids") were filled with defects, reflecting good nucleation of cobalt, and no holes in the structure ("seam voids"), reflecting that annealing eliminated such defects. The resistivity of the 150nm thick cobalt layer on the structure was 7.5ohm cm.

The cobalt deposition rate was equal to 1.9 nm/s.

Comparative example 5: electrodeposition of 22nm wide, 75nm deep trenches with organic additives at pH 8.0

Trenches 22nm wide and 75nm deep with a free surface of copper obtained by depositing a layer of copper on a substrate were filled with cobalt. The trenches were filled with a composition based on triethylene tetramine, which was present in a ratio of 1:1 stoichiometric with cobalt.

A. Materials and equipment:

substrate:

the substrate used in this example consisted of a silicon sample blank having a length of 4cm and a width of 4cm, covered with a structured silicon oxide layer having trenches 22nm wide and 75nm deep. The silicon oxide is covered by and in contact with a 2nm thick cobalt layer, which is itself covered by and in contact with a copper layer with a thickness below 4 nm. The resistivity of the copper layer was 250 ohm/square.

Electrodeposition solution:

in this solution, the concentration of triethylene tetramine stoichiometrically present with cobalt is 1.32g/L (from a 60% commercial solution). CoSO₄(H₂O)₆Is equal to 1.5g/L (i.e. 0.31g/L of Co²⁺Concentration).

Thioglycolic acid was also added in a concentration equal to 10 ppm. The pH of the solution was equal to 8.0.

Equipment:

-an inert carbon graphite anode (anode),

a structured silicon test piece blank coated with a copper layer, which constitutes the cathode,

-a reference connected to the anode.

B. The experimental scheme is as follows:

the cathode is polarized in a current pulse mode with a current range of 3mA (or 0.38 mA/cm)²) 35mA (or 4.38 mA/cm)²) For example 9mA (or 1.14 mA/cm)²) The pulse frequency in the cathodic polarization is between 1kHz and 10kHz, and the pulses in the zero polarization between two cathodic pulsesThe frequency is 0.5kHz-5 kHz. The cathode cycle was set at 6 rpm.

The duration of the electrodeposition step was 14 minutes to obtain complete filling of the 25nm wide, 75nm deep trench.

C. The results obtained were:

TEM analysis shows that the filling obtained with metallic cobalt by the bottom-up mechanism is of good quality.

However, the compound deposited on the surface is entirely composed of cobalt oxide, whereas cobalt is desired.

In addition, when cobalt is complexed in an alkaline medium, the fill time is longer. Within 12min, a 20nm thick cobalt oxide layer formed on the substrate surface was obtained.

These results emphasize the need to work at an acidic pH in order to obtain metallic cobalt at a sufficient rate.

Claims

1. Electrolyte for the electrodeposition of cobalt, characterized in that it is an aqueous solution comprising cobalt II ions in a quantity ranging from 1g/L to 5g/L, chloride ions in a quantity ranging from 1g/L to 10g/L, an acid in a quantity sufficient to obtain a pH ranging from 1.8 to 4.0, and at most two organic additives, said organic additives being not polymers.

2. The electrolyte of claim 1, wherein the organic additive has a molecular weight of less than 250g/mol and greater than 50 g/mol.

3. The electrolyte of any one of the preceding claims, wherein the electrolyte comprises at most one organic additive.

4. The electrolyte of any one of the preceding claims, wherein the organic additive is selected from alpha-hydroxycarboxylic acids.

5. The electrolyte of claim 1, wherein the acid is hydrochloric acid and at least one of the organic additives is selected from organic compounds having a pKa of at least 1 in the range of 1.8 to 3.5.

6. The electrolyte of claim 5, wherein at least one of the organic additives is selected from the group consisting of citric acid, tartaric acid, malic acid, mandelic acid, maleic acid, fumaric acid, glyceric acid, orotic acid, malonic acid, L-alanine, acetylsalicylic acid, and salicylic acid.

7. The electrolyte of any one of the preceding claims, wherein the cobalt II ions are present in free form, i.e. not complexed with the organic additive.

8. The electrolyte of any one of the preceding claims, wherein the pH of the electrolyte is between 2.0 and 3.5.

9. The electrolyte as claimed in any one of the preceding claims, characterized in that the concentration of the additive or the sum of the concentrations of the two additives is between 5mg/L and 200 mg/L.

10. An electrochemical process for filling a cavity, the process comprising:

-a step of contacting an electrically conductive surface of the cavity with an electrolyte as claimed in any one of the preceding claims,

-a step of poling the conductive surface for a time sufficient to achieve conformal and complete filling of the cavity by deposition of cobalt, and

-a step of annealing the cobalt deposit obtained at the end of the polarization step.

11. The method of claim 10, wherein the cavity is obtained by performing the steps of:

-a step of etching the structure on the silicon substrate,

-a step of depositing a metal layer on the surface of the silicon oxide or on the above metal alloy layer, so as to obtain a conductive surface of the cavity.

12. The method of claim 11, wherein the metal layer comprises at least one compound selected from the group consisting of cobalt, copper, tungsten, titanium, tantalum, ruthenium, nickel, titanium nitride, and tantalum nitride.

13. The method according to any of claims 10-12, wherein the cavity has a width or diameter at the opening of less than 40nm, preferably between 15nm and 30nm, and a depth of between 50nm and 250 nm.

14. The method of any of claims 10-13, wherein the cobalt deposit has an impurity content of less than 1000ppm atomic and has an average porosity, measured by electron microscopy, of less than 10% by volume or surface area.

15. The method of any of claims 10-14, wherein the cobalt deposition rate is between 0.1nm/s and 3.0 nm/s.

16. An electrochemical process for depositing cobalt on a substrate, the substrate comprising a conductive surface within a cavity hollowed out in the substrate and a conductive surface outside the cavity, the process comprising:

-a step of poling the conductive surface for a period of time sufficient to effect filling of the cavity with cobalt and depositing a cobalt layer at least 50nm thick on the conductive surface outside the cavity,

-a step of annealing the cobalt obtained at the end of the polarization step.