Patent Application
20240133028
The present invention concerns electrodeposition of cobalt on a conductive surface. More precisely, it concerns an electrolyte and a method for cobalt electrodeposition that can be used to fabricate electrical interconnects in integrated circuits.
Semiconductor devices comprise different levels of integration and two categories of conductive metallic interconnects: trenches a few tens of nanometers wide, which run on the surface of the device and which connect the electronic components, and through vias which connect the different levels and whose diameter is around several hundred nanometers.
Fabrication of interconnects comprises etching cavities on the substrate, followed by depositing a metallic seed layer on the surface of the cavities to allow a subsequent step of filling the cavities electrochemically with a conductive metal.
Conventional methods for filling interconnects with cobalt use electrolytes containing a cobalt salt and numerous organic additives. The combination of these additives is generally necessary to obtain a good-quality cobalt mass, more particularly without material voids and of good conductivity.
The filling of cavities can follow two mechanisms depending on the composition of the electrolyte used: a bottom-up filling or a conformal filling. The filling method by a bottom-up mechanism is opposed to a filling method in which the cobalt deposit grows at the same rate at the bottom and on the walls of the hollow patterns.
In order to obtain a bottom-up filling, electrolytes of the prior art comprise several additives including a suppressor and an accelerator. Such a system makes it possible to avoid the formation of voids in the cobalt deposit and premature closure of the cavity openings during filling. The suppressor limits the deposition of cobalt at the upper level of the cavities, on their walls as well as on the flat surface of the substrate onto which the cavities open, while the accelerator diffuses at the bottom of the cavities to promote the deposition of cobalt. The presence of an accelerator is even more necessary for cavities of narrow width and large depth since it makes it possible to increase the rate of cobalt deposition at the bottom of the cavities.
Electrodeposition baths designed for bottom-up filling have several disadvantages that ultimately limit the smooth operation of the electronic devices manufactured and which make them too expensive to manufacture. They actually generate cobalt interconnects contaminated by the organic additives made necessary to limit the formation of holes in the cobalt during filling. Moreover, the filling speeds obtained with these chemicals are too slow and not compatible with industrial-scale production.
In application US 2016/0273117, for example, the electrolyte contains numerous additives including a suppressor and an accelerator with complementary functions to ensure bottom-up filling. The inventors found that the resistivity of the cobalt deposited with this electrolyte was very high and that holes were formed in the cobalt during filling. This is why it is necessary to anneal the deposit to remove them.
There is therefore a need to provide electrolysis baths that lead to cobalt interconnects with improved performances, in particular relative to their conductivity. To attain this goal, it is desirable to produce cobalt deposits having extremely low quantities of impurities and that are free of material voids, even in the absence of annealing step. It is also desirable to propose electrolytes which, while avoiding the formation of holes in the cobalt, make it possible to reach a sufficiently high deposition speed to make device manufacture profitable.
The inventors found that the combination of an alpha-hydroxy carboxylic acid and a nitrogen compound such as polyethyleneimine or benzotriazole meets these needs.
Alpha-hydroxy carboxylic acids have certainly already been used in electrochemical methods for cobalt deposition such as, for example, in application WO 2019/179897, but these methods follow a conformal filling mechanism at the end of which holes persist in the metal in the absence of annealing of the deposit.
Thus, the invention concerns a method for creating cobalt interconnects by bottom-up filling of cavities that uses an electrolyte of pH comprised between 1.8 and 4.0, comprising cobalt II, chloride ions, an alpha-hydroxy carboxylic acid and a additive chosen from among polyethyleneimines and benzotriazole.
More precisely, the invention concerns an electrolyte for the electrodeposition of cobalt in the form of an aqueous solution comprising from 1 to 5 g/L of cobalt II ions, from 1 to 10 g/L of chloride ions, a strong acid in a sufficient quantity to obtain a pH comprised between 1.8 and 4.0, and organic additives including at least one first additive chosen from alpha-hydroxy carboxylic acids and mixtures thereof and at least one second additive chosen from polyethyleneimines and benzotriazole.
The electrolyte of the invention allows obtaining continuous cobalt deposits of high purity whose duration of production can be less than that of the prior art.
Indeed, the filling kinetics of conventional methods must be slower to prevent the formation of holes, and the method must comprise an annealing step when holes are formed. Moreover, the method can comprise two separate steps of cobalt electrodeposition: a step of performing the filling of cavities at a fairly slow speed, and a second step of electrodeposition using a second electrolyte comprising cobalt ions for depositing the so-called “overburden layer” on the entire substrate surface.
The method of the invention advantageously makes it possible to perform the filling of cavities and deposition of the overburden layer in a single electrodeposition step. It also makes it possible to avoid annealing the cobalt deposit before performing the polishing step combining chemical and mechanical attack of the overburden layer.
In addition, cobalt deposits produced in the context of the invention have the advantage of forming interconnects having a very low amount of impurities, preferably less than 1000 atomic ppm.
“Electrolyte” means the liquid containing precursors of a metal coating used in an electrodeposition method.
“Continuous filling” means a mass of cobalt free of voids. In the prior art, material holes or voids can be observed in a cobalt deposit between the walls of the cavities and the cobalt deposit (sidewall voids) and holes located at a distance from the walls of cavities in the form of seams. These voids can be observed and quantified by transmission or scanning electron microscopy by making cross sections of the structures. The continuous deposit of the invention preferably has a mean void percentage less than 10% by volume, preferably less than or equal to 5% by volume. The void percentage within the structures to be filled can be measured by scanning electron microscopy with a magnification between 50,000 and 350,000.
“Mean diameter” or “mean width” of the cavities means the dimensions measured at the opening of the cavities to be filled. The cavities are, for example, in the form of cylinders or flared channels.
According to a first embodiment, the invention concerns an electrolyte for the electrodeposition of cobalt characterized in that the electrolyte is an aqueous solution comprising from 1 to 5 g/L of cobalt II ions, from 1 to 10 g/L of chloride ions, a strong acid in a sufficient quantity to obtain a pH comprised between 1.8 and 4.0, and organic additives including at least one first additive chosen from alpha-hydroxy carboxylic acids and mixtures thereof and at least one second additive chosen from polyethyleneimines and benzotriazole.
The mass concentration of cobalt II ions can range from 1 g/L to 5 g/L, for example from 2 g/L to 3 g/L. That of the chloride ions can range from 1 g/L to 10 g/L.
Chloride ions can be introduced by dissolution in water of cobalt chloride or one of its hydrate salts, such as cobalt chloride hexahydrate.
The electrolyte preferably comprises at most two organic additives, these additives being the first and second additive.
All of the organic additives contained in the electrolyte are preferably sulphur free. For example, the alpha-hydroxy carboxylic acid is preferably sulphur free.
The electrolyte preferably does not contain any sulphur compound. Also, the composition is preferably not obtained by dissolution of a cobalt salt such as cobalt sulfate or one of its hydrates because it generates a sulphur contamination of the cobalt deposit, which we wish to avoid.
The total concentration of organic additives in the electrolyte is preferably comprised between 5 ppm and 50 ppm.
The concentration of the first additive is preferably comprised between 5 and 200 ppm and the concentration of the second additive is preferably comprised between 1 and 10 ppm.
The first additive is chosen, for example, from citric acid, tartaric acid, malic acid, mandelic acid and glyceric acid.
In a particular embodiment of the invention, the alpha-hydroxy carboxylic acid is tartaric acid.
According to one implementation of the invention, the second amine additive is a linear or branched poly(ethyleneimine) homopolymer or copolymer. The poly(ethyleneimine) is in the form of an acid, part or all of its amine functions being protonated.
A linear poly(ethyleneimine) will be chosen, for example, having number-average molecular weight Mn comprised between 500 g/mol and 25,000 g/mol.
A branched poly(ethyleneimine) can also be chosen having a number-average molecular weight Mn comprised between 500 g/mol and 70,000 g/L, which comprises primary amine, secondary amine and tertiary amine functions.
Thus, the poly(ethyleneimine) can be a poly(ethyleneimine) of CAS number 25987-06-8, having, for example, a number-average molecular weight Mn comprised between 500 g/mol and 700 g/mol, and preferably a weight-average molecular weight Mw comprised between 700 g/mol and 900 g/mol. Such a poly(ethyleneimine) exists under reference 408719 sold by the Sigma-Aldrich company.
The poly(ethyleneimine) can also be a poly(ethyleneimine) of CAS number 9002-98-6, having, for example, a number-average molecular weight Mn comprised between 500 and 700 g/mole. Such a poly(ethyleneimine) exists under reference 02371 sold by Polysciences, Inc.
The number-average molecular weight and the weight-average molecular weight can be measured independently from one another by a conventional method known to the skilled person, such as gel permeation chromatography (GPC) or light scattering (LS).
According to one implementation of the invention, the amine is benzotriazole.
The pH of the electrolyte is preferably comprised between 1.8 and 4.0. In a particular embodiment, the pH is comprised between 1.8 and 2.6.
The pH of the composition can optionally be adjusted with a base or an acid known to the skilled person. The acid used can be hydrochloric acid. The electrolyte may not contain a buffer compound, such as, for example, boric acid. Preferably, the electrolyte does not contain boric acid.
Although there is no restriction in principle regarding the nature of the solvent (provided that it sufficiently solubilizes the active species of the solution and does not interfere with electrodeposition), it will preferably be water. According to one embodiment, the solvent predominantly comprises water by volume.
The conductivity of the electrolyte is preferably comprised between 2 mS/cm and 10 mS/cm.
The invention also concerns an electrochemical method for deposition on a substrate provided with a conductive surface comprising a flat part and cavities, by filling said cavities bottom-up, said method comprising:
In one advantageous embodiment, the duration is sufficient to perform the filling of the cavities and the coating of the flat part of the conductive surface by a cobalt deposit having a thickness ranging from 50 nm to 400 nm.
In one advantageous variant, it is not necessary to perform a step of annealing the cobalt deposit obtained at the end of the polarization step, so that the polarization step can be immediately followed by a polishing step combining chemical and mechanical attack (also called mechanochemical) of the cobalt deposit obtained at the end of the polarization step. According to one embodiment, the deposition method of the invention therefore comprises:
The polarization step in the presence of the electrolyte of the invention can last as long as necessary to fill the cavities without covering the flat surface. In this case, the deposition method can comprise a second polarization step during which a second cobalt deposit is formed using an electrolyte other than that of the invention.
Alternatively, the polarization step in the presence of the electrolyte of the invention can last as long as necessary to fill the cavities and coat the flat surface, the thickness of the cobalt deposit above the flat surface being at least 20 nm.
The part of the cobalt deposit that coats the flat surface, also called overburden layer, can have a thickness comprised between 50 nm and 400 nm. It is advantageously of constant thickness over the entire substrate surface. The layer is also homogenous, shiny and compact.
Under certain conditions, the method of the invention is a so-called “bottom-up” method as opposed to the “conformal” methods of the prior art. In this case, the speed of cobalt deposition is higher at the bottom of the cavities than on the walls.
The cobalt deposit obtained at the end of the polarization step advantageously has an impurity content less than 1000 atomic ppm. The predominant impurity is oxygen, followed by carbon and nitrogen. The total carbon and nitrogen content is preferably less than 300 ppm.
The cobalt deposit obtained at the end of the electrodeposition step is advantageously continuous, in the sense that it comprises a void percentage of less than 10% by volume or by area, preferably less than or equal to 5% by volume or by area, without undergoing a heat treatment at a temperature ranging from 50° C. to 500° C., preferably comprised between 150° C. and 500° C.
The void percentage in the cobalt deposit can be measured by electron microscope observation known to the skilled person, who will choose the method that seems most appropriate. One of these methods can be scanning electron microscopy (SEM) or transmission electron microscopy (TEM) by using a magnification comprised between 50,000 and 350,000. The void volume can be assessed by measuring the void area observed over one or more cross sections of the substrate comprising the filled cavities. When several areas are measured over several cross sections, the mean of these areas will be calculated to assess the void volume.
A low content of impurities combined with a very low void percentage makes it possible to obtain a cobalt deposit whose resistivity is reduced. Also, the resistivity of the cobalt deposit obtained at the end of the polarization step can be less than 30 μ∩·cm without undergoing a heat treatment at a temperature ranging from 50° C. to 500° C.
The cobalt deposition rate can be between 0.1 nm/s and 3.0 nm/s, preferably between 1.0 nm/s and 3.0 nm/s, and more preferably between 1 nm/s and 2.5 nm/s.
The cavities to be filled can be fashioned according to a Damascene or Dual Damascene method known to the skilled person comprising a succession of steps comprising: —etching trenches on the upper part of a silicon wafer; —depositing an insulating dielectric layer generally consisting of silicon oxide on the etched surface; —depositing a thin layer of a barrier material used to prevent the migration of cobalt into the silicon; —optionally depositing a thin metallic layer, called the seed layer.
The barrier layer and the seed layer generally have, independently of one another, a thickness comprised between 1 nm and 10 nm.
The conductive surface which is in contact with the electrolyte is a surface of a metal layer comprising, for example, at least one compound chosen in the group made up of cobalt, copper, tungsten, titanium, tantalum, ruthenium, nickel, titanium nitride and tantalum nitride.
The conductive surface of the substrate can be the surface of an assembly comprising a tantalum nitride layer of a thickness comprised between 1 nm and 6 nm, itself covered and in contact with a layer of metallic cobalt comprised between 1 nm and 10 nm, preferably between 2 nm and 5 nm, on which the cobalt will be deposited during the electrical step.
The substrate can therefore be obtained by successive deposits of SiO2, tantalum nitride and cobalt. Cobalt can be deposited on tantalum nitride by chemical vapor deposition (CVD) or by atomic layer deposition (ALD).
The resistivity of the assembly comprising the metal layer and the cobalt deposit can range from 7 to 10 ohm/cm. It is preferably comprised between 7.5 and 8.5 ohm/cm.
The cavities designed to be filled with the cobalt according to the method of the invention preferably have a width at their opening (i.e., at the surface of the substrate) of less than 100 nm, preferably comprised between 10 and 50 nm. The depth can range from 50 to 250 nm. According to one embodiment, they have a width comprised between 30 nm and 50 nm, preferably between 35 nm and 45 nm, and a depth comprised between 125 nm and 175 nm.
The intensity of the polarization used in the electrical step preferably ranges from 2 mA/cm2 to 20 mA/cm2. The cobalt deposition rate is comprised between 0.1 nm/s and 3.0 nm/s when the intensity of the polarization current ranges from 8.5 mA/cm2 to 18.5 mA/cm2, which is very advantageous in comparison with methods of the prior art for which a much lower rate is observed in this current range.
The electrical polarization step of the method of the invention may comprise a single or several different polarization mode steps.
The conductive surface can be contacted with the electrolyte either before polarization or after polarization. It is preferred that the contact with the cavities be done before energizing, so as to limit corrosion of the surface by the electrolyte.
The electrical step can be performed by using at least one polarization mode chosen from the group made up of ramp mode, galvanostatic mode and galvano-pulsed mode.
For example, the electrical step comprises one or more steps of cathode polarization in ramp mode in a current range from 0 mA/cm2 to 10 mA/cm2, for a duration preferably comprised between 10 s and 100 s.
The electrical step can also comprise one or more steps of polarization in galvanostatic mode with a current ranging from 5 mA/cm2 to 20 mA/cm2.
According to one example, the electrical step comprises at least one step of polarizing the cathode in ramp mode with a current preferably ranging from 0 mA/cm2 to 10 mA/cm2, followed by a step in galvanostatic mode by imposing a current of 5 mA/cm2 to 20 mA/cm2.
The method of the invention can comprise a step of annealing the cobalt deposit obtained at the end of filling described previously, but it advantageously does not have this step. An annealing heat treatment is generally conducted at a temperature comprised between 350° C. and 550° C., for example around 450° C., preferably under reducing gas such as 4% H2 in N2.
The method may comprise a preliminary step of treatment with a reducing plasma so as to reduce the native metal oxide present on the conductive surface of the substrate. The plasma also acts on the surface of the trenches which allows improving the quality of the interface between the seed layer and the electrodeposited cobalt. It is preferred that the electrodeposition step be performed immediately after the plasma treatment to minimize the reformation of native oxide.
The method of the invention is especially applied in the manufacture of semiconductor devices when creating conductive metallic interconnects such as trenches running on the surface and vias connecting different levels of integration.
The invention is further illustrated by the following examples
Trenches are filled by electrodeposition of cobalt on a cobalt seed layer. The deposition is conducted using a composition containing cobalt dichloride, an alpha-hydroxy carboxylic acid and polyethyleneimine (PEI) at pH 2.2.
A. Materiel and Equipment
Substrate
The substrate used in this example was made up of a 3.3×3.3 cm trench-etched silicon coupon that has been successively coated with a layer of silicon oxide, a layer of TaN 2 nm thick and a layer of metallic cobalt 3 nm thick. The resistivity of the substrate is approximately 600 ohms per square. The width of the cavities to be filled is equal to 40 nm at their opening and their depth is equal to 150 nm.
Electrodeposition Solution:
In this solution, the Co2+ concentration is equal to 2.3 g/L obtained from CoCl2 (H2O)6. The tartaric acid has a concentration equal to 15 ppm. The PEI has a concentration equal to 5 ppm. The pH of the solution is adjusted to 2.2 by addition of hydrochloric acid.
Equipment:
In this example, electrolytic deposition equipment is used composed of two parts: the cell designed to contain the electrodeposition solution equipped with a fluid recirculation system in order to control the hydrodynamics of the system and a rotating electrode equipped with a sample holder adapted to the size of the coupons used (3.3 cm×3.3 cm). The electrolytic deposition cell had two electrodes:
Connectors allowed the electrical contact of the electrodes that were connected by electrical wires to a potentiostat providing up to 20 V or 2 A.
B. Experimental Protocol:
Electrical Method:
Three tests are conducted, called Test 1, Test 2 and Test 3, by applying different electrical methods. The three methods comprised two, three or five steps out of the following steps:
The first electrical protocol (Test 1) comprised three steps, steps a), b) and c).
The second electrical protocol (Test 2) comprised five steps, steps a) to e). During step e), the cathode was polarized in galvanostatic mode at 90 mA (11.4 mA/cm2) with a rotation of 100 rpm for 40 seconds.
The third electrical protocol (Test 3) comprised two steps, steps a) and e). During step e), the cathode was polarized in galvanostatic mode at 90 mA (11.4 mA/cm2) with a rotation of 65 rpm for 133 seconds.
C. Results Obtained:
As can be seen in
In Test 2, an analysis by scanning electron microscopy (SEM) reveals filling with no hole defects on the walls of the trenches (sidewall voids) reflecting good nucleation of the cobalt and no holes in the structures (seam voids) reflecting optimal bottom-up filling with no annealing.
Trenches identical to those of Example 1 are filled using a composition containing cobalt dichloride, an alpha-hydroxy carboxylic acid and benzotriazole at pH 2.2.
A. Materiel and Equipment
Substrate
The substrate used is strictly identical to that of Example 1.
Electrodeposition Solution:
In this solution, the Co2+ concentration is equal to 2.3 g/L obtained from CoCl2(H2O)6. The tartaric acid has a concentration equal to 15 ppm. The benzotriazole has a concentration equal to 10 ppm. The pH of the solution is adjusted to 2.2 by addition of hydrochloric acid.
Equipment:
The equipment is identical to that of Example 1.
B. Experimental Protocol:
Electrical Method:
The electrical method was identical to that of Test 2 of Example 1 and comprised the five steps a) to e).
C. Results Obtained:
An analysis by scanning electron microscopy (SEM) reveals filling with no hole defects on the walls of the trenches (sidewall voids) reflecting good nucleation of the cobalt and no holes in the structures (seam voids) reflecting optimal bottom-up filling with no annealing.
Trenches identical to those of Example 1 are filled using a composition containing cobalt dichloride and alpha-hydroxy carboxylic acid at pH 2.2.
A. Materiel and Equipment
Substrate
The substrate used is strictly identical to that of Example 1.
Electrodeposition Solution:
In this solution, the Co2+ concentration is equal to 2.3 g/L obtained from CoCl2(H2O)6. The tartaric acid has a concentration equal to 15 ppm. The pH of the solution is adjusted to 2.2 by addition of hydrochloric acid.
Equipment:
The equipment is identical to that of Example 1.
B. Experimental Protocol:
The electrical method was identical to that of Test 2 of Example 1 and comprised the five steps a) to e).
C. Results Obtained:
An analysis by scanning electron microscopy (SEM) reveals filling comprising holes in the structures (seam voids) which require an additional annealing step to be able to remove them, reflecting a growth by closing structures from bottom to top, similar to a zipper.
An electrodeposition of cobalt in trenches identical to those of Example 1 is performed using a composition of the prior art according to the teaching of application US 2016/0273117 A1 containing cobalt sulfate, boric acid, thiourea and polyethyleneimine (PEI) at pH 4.
A. Materiel and Equipment
Substrate
The substrate used is strictly identical to that of Example 1.
Electrodeposition Solution:
In this solution, the Co2+ concentration is equal to 2 g/L obtained from CoSO4. The boric acid has a concentration equal to 20 g/L. The thiourea has a concentration equal to 150 ppm. The PEI has a concentration equal to 10 ppm. The pH of the solution is adjusted to 4 by addition of sulfuric acid.
Equipment:
The equipment is identical to that of Example 1.
B. Experimental Protocol:
The method is identical to that of Test 3 of Example 1 and comprised the two steps a) and e).
C. Results Obtained:
As can be seen in
At the same time, an analysis of the film obtained in Test 3 of Example 1 and the film obtained in this example allowed comparing their resistivity. The results are reported in Table 1 below.
The resistivity of the film deposited in Test 3 of Example 1 is better than that of Comparative Example 4, which is more desirable at the industrial level. A lower resistivity is synonymous with a better film quality with fewer impurities.
| Number | Date | Country | Kind |
|---|---|---|---|
| FR2101582 | Feb 2021 | FR | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP22/53023 | 2/8/2022 | WO |
