SURFACE-ACTIVE CONDITIONAL INHIBITORS FOR THE ELECTROPLATING OF COPPER ON A SURFACE

Electrolytes for the electroplating of copper comprising, as inhibitor, a poly(alkylene-biguanide) salt. The inhibition developed by poly(alkylene-biguanide) salts is conditioned by the concentration of an accelerator on the surface of the copper. The surfactancy of poly(alkylene-biguanide) salts enables them to be instantly transferred from the electrolyte/air interface to the electrolyte/copper interface during contact between the cathode and the electrolyte. The electrolytes according to the invention are suitable for obtaining smooth and bright electroplated coatings and for depositing copper on surfaces having submicron-scale concavities useful in microelectronics.

Skip to: Description  ·  Claims  · Patent History  ·  Patent History
Description

The present invention relates to a new family of additives for copper electroplating baths. More specifically, the present invention relates to a new family of surface-active inhibitors for copper electroplating baths used in the manufacture of smooth and bright copper electrodeposits and in the manufacture of semiconductor copper interconnects.

Each electrolytic copper deposit must present specific properties according to its utilization. Several mineral or organic additives are usually added to copper electroplating baths to intend to obtain deposits having these required properties. The minor components of copper electroplating baths are often referred to as additives.

To obtain a smooth and bright electrolytic copper deposit on a surface initially comprising grooves, it was empirically found that it was necessary to add a combination of additives to the electrolyte such as, an accelerator associated with an inhibitor and a brightener. Such a combination of additives promotes an accelerated copper deposition starting from the bottom of concavities of micronic and submicronic size on the surface, such as the concave parts of roughness, grooves or microtrenches, which makes it possible to fill them preferentially. At the same time, these additives slow down the electrodeposition of copper on the convex parts of surface. The combination of these two effects makes it possible to obtain smooth electrolytic copper deposits.

This result was explained by the fact that when an accelerator of the deposit is chemically adsorbed on the surface of copper, its surface concentration increases as this surface rapidly shrinks as occurs during the filling of a concavity, and that on the contrary, the surface concentration of this accelerator decreases as this surface increases as occurs during the deposition on a convex microstructure. (J. Osterwald and J. Schulz-Harder, Galvanotechnik, 1975, vol. 66, 360; J. Osterwald, Oberflache—Surface, 1976, vol. 17, 89).

The manufacture of semiconductor interconnections also includes the filling of concavities of submicronic size in the form, for example, of trenches by electrolytic deposition of copper. This filling should not leave any void in the concavities. Such a result was already obtained by a copper deposition which accelerates starting from the bottom of concavities. Therefore, obtaining smooth electrolytic copper deposits on the one hand, and filling concavities of submicronic size by copper electrolytic deposition on the other hand, set the same technical problem as already established (T. P. Moffat et al. IBM J. RES & DEV. 2005, 49, (1) 19-36; G. B. McFadden et al. Journal of The Electrochemical Society 2003, 150, C591-C599). Consequently, combinations of additives previously known to generate smooth electrolytic copper deposits were used to manufacture semiconductor copper interconnections by filling concavities of submicronic size. To obtain a copper deposition rate which increases starting from the bottom of submicronic trenches, one thus proposed to use a well-known association of polyalkyleneglycols, which act as inhibitors in the presence of chloride ions i.e. they increase the copper deposition overvoltage, with thiols bearing a sulfonic acid group, or with disulfides bearing two sulfonic acid groups, which act as accelerators, i.e. they decrease the copper deposition overvoltage.

This combination of additives presents the disadvantage of including necessarily chloride ions, generally in the concentration range of 10−3 to 10−2 mole/l. However, in the presence of chloride ions, the copper surface reacts with Cu2+ ions to form cuprous chloride by a disproportionation reaction. (W-P Dow et al. Journal of The Electrochemical Society, 2005, 152 C67-C76). This disproportionation reaction, which consumes the already deposited copper, can partly remove the initial copper layer when this layer is a thin and divided copper seed layer obtained by vapor phase deposition of copper, or may change its surface properties.

This combination of additives also presents the disadvantage of generating electrolytes which do not perfectly and instantaneously wet the initial surface of copper.

It would thus be desirable to have new compositions of copper electroplating baths devoid of chloride ions, making it possible to wet the surface of copper instantaneously, then preferentially fill the grooves and concavities of submicronic size present on the copper surface.

More generally, the additives known to present a useful activity in acidic copper electroplating baths form, in fact, a small number of families, known for a long time, such as, for example, polyethers, thiols or disulfides comprising sulfonic acid groups, as well as phenazinic nitrogen heterocycles, such as Janus Green B. It would be thus very desirable to have new families of active additives for copper electroplating baths in order, for example, to increase the number of possible combinations when one seeks to obtain a copper electrodeposit presenting optimal properties for a particular application.

The present invention relates to the discovery of the inhibiting properties of the electrolytic deposition of copper and of the surface-active properties at the copper-electrolyte interface of poly (alkylene-biguanide) salts.

The present invention also relates to copper electroplating baths devoid of chloride ions, comprising poly (alkylene-biguanide) salts.

The present invention also relates to copper electroplating processes implementing baths comprising poly (alkylene-biguanide) salts.

The present invention further relates to smooth copper electrodeposits and semiconductor interconnects obtained from baths comprising poly (alkylene-biguanide) salts.

More specifically, the present invention has as a first object, poly (alkylene-biguanide) salts used as additives in neutral or acidic copper electroplating baths, characterized in that said poly (alkylene-biguanide) salts are represented by the following general formula:


R—NH—C(:NH)—NH—[(CH2)p—NH—C(:NH)—NH—C(:NH)—NH—]n—(CH2)p—NH2, xAH  (B)

wherein:

    • p is an integer of at least 2 up to 12
    • n is an integer of at least 2 up to 100
    • R is a group represented by the formula: NC— or H2N—C(:O)
    • AH is an acid
    • x is comprised between (n+1) and 2(n+1) when AH is a monoacid, and between (n+1)/2 and (n+1) when AH is a diacid.

Preferably, in the present invention, one implements poly (alkylene-biguanide) salts having the general formula (B) wherein, R— represents NC—, p=4, 6, 8, 10 or 12, and n is comprised between 5 and 50.

Most preferably, in the present invention one implements poly (alkylene-biguanide) salts having the general formula (B) wherein, R— represents NC—, p=6, and n is comprised between 6 and 25.

It is possible to increase the number n of monomers in poly (alkylene-biguanide) salts in a known way. For example, a solution of the hydrochloride of an oligomer of general formula (B) in an alcohol can be heated at a temperature from about 70° C. to 150° C. to produce an oligomer with a higher degree of condensation by reaction between the amine and cyanamide endings.

The preparation of poly (alkylene-biguanide) and of their salts is well-known and consist, for example, in polycondensing an α, ω alkylene diamine hydrochloride with the sodium salt of dicyandiamide.

The structure of poly (alkylene-biguanide) salts depends on the pH of the solution in which they are dissolved. Their structure comprises one proton per biguanide group G, i.e. GH' when the pH of the solution is comprised between approximately 3 and 11, and two protons, i.e. GH22+ when the pH of the solution is lower than approximately 3.

Preferably, poly (alkylene-biguanide) salts are added to the copper electroplating baths, in the form of aqueous solutions. Among the preferred salts, one can cite the salts of strong oxygenated acids, such as sulfates, bisulfates, acidic phosphates and sulfonates, such as methanesulfonate.

A poly (alkylene-biguanide) salt in aqueous solution can be transformed into another salt by known anion exchange processes, such as electrodialysis, or alternatively, by isolating the poly (alkylene-biguanide) in the form of the free base and then forming a solution of a new salt of said free base with a new acid.

Some poly (alkylene-biguanide) compounds are commercialized as antiseptics in the form of neutral aqueous solutions of their hydrochloride, such as poly (hexamethylene-biguanide) hydrochloride or PHBG.

Compounds of general formula (B) can be used in copper electroplating baths within the range of concentration of 10−7 mole/l to 10−3 mole/l, preferably of 5×10−7 mole/l to 10−5 mole/l. They can be used preferably in association with copper deposition accelerators taken among mercapto-alkyl sulfonic acids and dialkyl disulfide disulfonic acids, or among their salts, such as sodium mercaptopropyl sulfonate (MPS) or bis-sulfopropyl disulfide (SPS).

Compounds of general formula (B) can be used in neutral or acidic copper electroplating baths, the pH of which being preferably comprised between 3 and 0. The operating temperature of these baths is comprised between 0° C. and 100° C., preferably between 20° C. and 80° C.

The applicant discovered that compounds of general formula (B) inhibited copper electrodeposition at very low concentrations, within the range from, for example, 5×10−7 mole/l to 10−5 mole/l as shown by current vs potential curves corresponding to copper electrodeposition in the presence of these compounds in examples and Figures. This inhibiting property which is subject of the present invention can be compared to those of known inhibitors comprising polyethylene glycol and chloride ions as shown in FIG. 3.

On the other hand, a molecule including only one biguanide group, such as N,N dimethyl biguanide hydrochloride, did not show inhibiting properties of the electrolytic deposition of copper in baths containing copper sulfate and sulfuric acid, of pH comprised between 0.5 to 4.

A second object of the present invention concerns copper electroplating baths characterized in that they contain at least one additive of general formula (B).

It was shown previously, that accelerators such as 3-mercatopropyl sulfonate (MPS) or di n-propyl disulfide 3,3′ disulfonate (bis-sulfopropyl disulfide or SPS) in a copper electroplating bath were chemisorbed on a copper surface. For example, in a bath containing copper sulfate, sulfuric acid, a polyethylene glycol, chloride ions, 10−3 mole/l and SPS, 5×10−5 mole/l, it was shown that the copper surface coverage by SPS was 5.4% at equilibrium at room temperature, and that for SPS concentrations lower than 5×10−5 mole/l the copper surface coverage by SPS at equilibrium was proportional to the SPS concentration in solution. The time necessary to fill concavities of submicronic size by electrolytic deposition of copper is in general 10 to 30 seconds; this time is less than that required by the accelerator chemisorbed on the copper surface to reach an equilibrium with the solution, so that its surface concentration on the copper surface increases transitionally during the filling of a concavity of submicronic size because of the fast reduction of the surface of that concavity. The accelerator, whose concentration increases, then progressively displaces an inhibitor of the polyethylene glycol+chloride ion type and the resulting suppression of inhibition induces an increasingly fast filling starting from the bottom of the concavity, which makes it possible to carry out a complete and void-free filling. (T. P. Moffat et al. The Electrochemical Society Interface, 2004, 46-52).

An inhibitor likely to be appropriate to fill grooves during the formation of a smooth electrolytic deposit of copper, or appropriate for filling concavities of submicronic size during the manufacture of semiconductor copper interconnections, must thus show the following properties and characteristics:

    • Not be notably displaced from the copper surface by an accelerator of the MPS or SPS type at low concentration, for instance 10−5 mole/l.
    • Be actually displaced from the copper surface by this same accelerator at higher concentrations in solution, i.e. multiplied by 4 to 10, i.e. for example at a concentration comprised between 4×10−5 mole/l and 10−4 mole/l. This increased concentration of the accelerator in solution is then in equilibrium with a concentration of the accelerator on the surface of copper similar to that which is reached transitionally during the fast filling of grooves or during the fast filling of concavities of submicronic size in the manufacture of semiconductor copper interconnections.
    • Properties to which is added the requirement that there must be no chloride ions present.

The applicant showed that mercaptopropyl sulfonate (MPS) at a concentration higher than 5×10−5 mole/l suppressed the inhibition of the electrolytic deposition of copper induced by compounds of general formula (B) at room temperature as well as at 70-75° C.

More precisely, the applicant showed that mercaptopropyl sulfonate (MPS) at a concentration of 10−5 mole/l did not oppose to the inhibition of the electrolytic deposition of copper induced by compounds of general formula (B), whereas at a concentration of 5×10−5 mole/l to 10−4 mole/l, MPS suppressed this inhibition.

It was shown for example that at room temperature, MPS at a concentration of 10−5 mole/l in fact increased the inhibition of the electrolytic deposition of copper induced by one of the compounds of general formula (B), whereas at a concentration of 5×10−5 mole/l, MPS suppressed this inhibition.

Table 1, established from the data of example 4 represented on FIG. 4, shows the overvoltage for the electrolytic deposition of copper induced by one of the compounds of general formula (B) at a concentration of 1.25×10−6 mole/l in the presence of 10−5 mole/l of MPS at room temperature and the decrease of this overvoltage resulting from an increase in the concentration of the accelerator MPS by a factor of 5, i.e from 10−5 mole/l to 5×10−5 mole/l.

TABLE 1 i (mA/cm2) 0 1 2 3 4 5 6 7 8 9 10 MPS = 10−5 M η(mV) 0 −200 −223 −231 −233 −237 −239 −240 −241 — −242 MPS = 5 10−5 M η(mV) 0 −28 −62 −92 −110 −120 −128 −133 −138 −142 −147 Δ(mV) 0 172 161 139 123 117 111 107 103 — 95

On a copper surface, flat areas and concavities are positioned electrically in parallel.

An electrolyte comprising for example the compound of formula (B) of the example 4 and 10−5 mole/l of MPS should give rise:

    • To overvoltages for the electrolytic deposition of copper on flat areas of the surface listed in the first line of Table 1 according to the current density imposed.
    • To overvoltages for the electrolytic deposition of copper in concavities of the surface listed in the first line of Table 1 at the beginning of their filling, evolving to the overvoltages listed in the second line of Table 1 at the end of their filling.

If the electrolyte is flowing fast enough on the surface so that the diffusion of copper ions does not limit the deposition process, the current density is initially identical over the whole surface, including in the concavities. Table 1 shows that with an electrolyte containing the compound of formula (B) of example 4 and 10−5 mole/l of MPS, the current distribution must change during filling in favor of the bottom of concavities where the overvoltage becomes lower because of the progressive increase of the surface concentration of the accelerator.

In submicronic structures in the form of trenches, the trench surface is in general 4 to 10 times larger than that of the trench opening. With respect to the surface occupied by the trench on the surface, the current densities listed in Table 1 would be multiplied by 4 to 10 at the beginning of the filling.

Generally, the inhibition of copper electrodeposition by the compounds of the invention is a partial inhibition, the magnitude of which is conditioned by respective concentrations of the compounds of general formula (B) and of accelerator of MPS or SPS type, as well as by the potential of the copper electrode and by temperature.

More precisely, a concentration of 10−5 mole/l of mercaptopropyl sulfonate (MPS) did not suppress, or only slightly suppressed, to the partial inhibition of the electrolytic deposition of copper induced by one of the compounds of general formula (B), whereas a concentration of 5×10−5 to 10−4 mole/l of MPS suppressed most of this inhibition. This was observed by the applicant over a wide range of electrode potential ranging from 0 to −140 mV/Ag/AgCl at 70-75° C. For a given concentration of an accelerator of the MPS or SPS type, for example 10−5 mole/l, and a given concentration of one of the compounds of general formula (B), ranging for example between 5×10−7 mole/l and 2.5×10−6 mole/l, one can choose the deposition rate on the flat areas of the surface by varying the current which makes the electrode potential change over the range 0 to −140 mV/Ag/AgCl at 70-75° C.; the cathode potential vs electrolyte being all the more negative as the intensity is higher.

In submicronic concavities of the surface, the increase of the accelerator concentration must result in a progressive suppression of the inhibition and in a deposition rate increase with filling whatever the cathode potential in a range of 0 to −140 mV/Ag/AgCl.

Compounds of general formula (B) associated with an accelerator of the mercaptopropyl sulfonate (MPS) or sulfopropyl disulfide (SPS) type thus present the properties of conditional inhibitors required to induce a process of accelerated filling starting from the bottom of submicronic concavities by progressive suppression of inhibition during filling.

Copper electroplating baths implementing compounds of general formula (B) and an accelerator of the mercaptopropyl sulfonate (MPS) or sulfopropyl disulfide (SPS) type thus present the required properties to obtain smooth and bright copper deposits or to fill submicronic concavities in the manufacture of interconnections in micro-electronics.

Another object of this invention is to provide electroplating baths able to wet the surface of copper instantaneously.

The applicant showed that compounds of general formula (B) are partially localized at the electrolyte-air interface and cover instantaneously a significant part of the copper surface during the immersion of a copper electrode in the electrolyte.

The injection of a compound of general formula (B) in the electrolyte during the electrolytic deposition of copper at a constant electrode potential resulted in a progressive inhibition of this electrodeposition. The decrease of current vs time which resulted from the arrival of the inhibitor of general formula (B) on the electrode made it possible to measure precisely the rate at which this inhibitor reached the surface of the electrode by diffusion from the bulk of the electrolyte. For example, when a compound of general formula (B), p=6; n=8; R═NC—; AH═H2SO4 was injected, the necessary time to observe a 50% decrease of the current was about 18 seconds at room temperature for an amount of this inhibitor injected corresponding to a concentration of 6.25×10−7 mole/l, whereas it was about 8 seconds for an amount corresponding to a concentration of 1.25×10−6 mole/l, and about 2 seconds for an amount corresponding to a concentration of 5×10−6 mole/l.

By using a small auxiliary copper wire and a potentiostat, one can make the potential of a copper electrode equal to an imposed potential vs a reference electrode as of the very first instant of its contact with the electrolyte. By recording the current every 0.005 second during the immersion of a copper electrode in electrolytes, the applicant found that, for example, when the electrolyte contained one of the inhibitors of general formula (B) at a concentration of 5×10−6 mole/l, the current was already not more than 61% of the value it had in the absence of this inhibitor after only 0.02 second. This ratio was 55% after 0.05 second, 52% after 0.1 second and then decreased up to 18% after 20 seconds.

An amount of the inhibitor of general formula (B), sufficient to decrease the copper ions reduction rate by half was thus present instantaneously on the copper surface during its immersion in the electrolyte, followed by the slow diffusion of an additional amount of this inhibitor. The amount of this inhibitor which was instantaneously present on the copper surface was therefore already present at the electrolyte-air interface, since its diffusion from the bulk of the electrolyte would have required approximately two seconds.

This result can be explained by the fact that compounds of general formula (B) are amphiphilic oligomers and for this reason present surface-active properties, i.e. these compounds present a concentration excess at the interface of their solutions with air compared to their concentration in the electrolyte.

During the immersion of a copper electrode in an aqueous solution containing one of the compounds of general formula (B), the initial electrolyte-air interface, which presented a concentration excess of this compound compared to its concentration in the electrolyte, was instantaneously transformed into an electrolyte-copper interface, which consequently presented approximately the same concentration of the compound of general formula (B) which existed at the electrolyte-air interface initially.

The discovery of the process of instantaneous transfer of compounds of general formula (B) from the electrolyte-air interface to the electrolyte-copper interface by the applicant is likely to favor an instantaneous wetting of the copper surface by electrolytes containing compounds of general formula (B).

The amphiphilic character of compounds of general formula (B) can be modified according to known general rules which describe the surface-active properties of amphiphilic compounds, in particular by modifying the length of the hydrophobic chain (CH2)p or by modifying the concentration of mineral salts and of the acid present in the electrolyte, as well as the temperature.

The use of compounds of general formula (B) as inhibitors in copper electroplating baths thus makes it possible to deliver instantaneously a given and adjustable amount of these inhibitors to the copper surface as soon as this surface is immersed in the electrolyte. This instantaneous initial input of these inhibitors to the copper surface is followed by an additional amount controlled by their diffusion which can take from a few seconds to a few tens of seconds according to the concentration of these inhibitors in the electrolyte, the geometry of the concavity to be filled and the temperature.

The present invention has for third object, copper electroplating processes characterized in that they use baths containing at least one additive of general formula (B).

The electrolytic deposition of copper from baths implementing molecules of general formula (B) can be done by application of a direct current with a current density ranging from 0.1 to 100 mA/cm2 at the cathode, preferably ranging from 1 to 60 mA/cm2.

In an embodiment preferred by the applicant, a voltage is first applied between the anode, immersed beforehand in the electrolyte, and the cathode, not yet in contact with the electrolyte. Then, the cathode and the electrolyte are put in contact.

The present invention has for fourth object, electrolytic copper deposits characterized in that they were obtained by using baths containing at least one additive of general formula (B).

The electrolytic copper deposits obtained, for example, from baths containing poly (hexamethylene biguanide) sulfate have a smooth and bright aspect and an old rose color.

The following examples illustrate the invention.

EXAMPLE 1 Preparation of a poly (hexamethylene-biguanide) sulfate Solution

100 grams of a 20% w/w commercial solution of poly (hexamethylene biguanide) hydrochloride comprising mainly eight biguanide groups per polymer molecule were placed in a flask and concentrated under a 20 mm Hg vacuum at 50° C. The resulting residue was dissolved in absolute ethanol, then the ethanol was evaporated under vacuum. This operation was repeated twice. The residue was then dissolved in 100 ml of anhydrous methanol and the resulting solution degassed with nitrogen. 20 grams of a 25% w/w solution of sodium methylate in anhydrous methanol previously degassed with nitrogen were then added dropwise to this methanolic solution upon stirring under nitrogen. The resulting solution with a fine suspension of sodium chloride was then left at 4° C. overnight, and then centrifuged to eliminate the sodium chloride. A titration showed that the resulting methanolic solution contained 0.6 mole/l of a strong base corresponding to 0.066 mole/l of the compound of formula (B) (p=6; n=8) as the free base.

0.17 ml of the preceding methanolic solution were added to 10 ml of 0.005 mole/l aqueous sulfuric acid. Approximately one gram of this solution was evaporated under vacuum to remove the methanol, then the volume of the solution was adjusted to 10 ml with distilled water.

This solution contained 0.001 mole/l of poly (hexamethylene-biguanide) sulfate of formula (B) (p=6; n=8; x=4.5; AH═H2SO4) that will be designated by the abbreviation In1 the following examples.

EXAMPLE 2 Copper Electrodeposition Overvoltage Induced by In1

A 20 ml electrochemical cell with a magnetic stirrer was used. It was equipped with an Ag/AgCl reference electrode, a platinum grid as the counter-electrode, and the section of a 3.1 mm diameter copper wire surrounded by Teflon as the working electrode. The copper electroplating baths were degassed with nitrogen before carrying out measurements.

The cell was connected to a potentiostat. The copper cathode potential was scanned linearly over time towards negative potentials at a scanning rate of 1.66 mV/second, and the resulting current measured in milliamperes was recorded.

FIG. 1 shows the current vs potential curves obtained with the three following baths:

    • a05d: 0.25 mole/l SO4Cu solution; pH adjusted to 0.5 with H2SO4
    • m05: idem a05d+In1: 1.25×10−6 mole/l
    • d05: idem a05d+In1: 5×10−6 mole/l
    • Temperature: Room temperature

The copper electrodeposition overvoltage induced by compound In1 from example 1 was about 80 mV for an In1 concentration of 1.25×10−6 mole/l, and about 140 mV for an In1 concentration of 5×10−6 mole/l.

EXAMPLE 3 Copper Electrodeposition Overvoltage Induced by poly (hexamethylene biguanide) hydrochloride at Various pHs

By dilution of the 20% commercial solution of poly (hexamethylene biguanide) hydrochloride comprising mainly eight biguanide groups per polymer molecule, one prepared a 0.001 mole/l solution of poly (hexamethylene biguanide) hydrochloride of formula (B) (p=6; n=8; x=9; AH═HCl) designated hereafter by the abbreviation In2.

The test protocol described in example 1 was repeated with the four following baths:

    • E3: 0.25 mole/l SO4Cu solution; In2: 5×10−6 mole/l; pH adjusted to 3 with H2SO4
    • E2: idem E3 but pH adjusted to 2 with H2SO4
    • E1: idem E3 but pH adjusted to 1 with H2SO4
    • E05: idem E3 but pH adjusted to 0.5 with H2SO4
    • Temperature: Room temperature

FIG. 2 shows the current vs potential curves obtained.

The copper electrodeposition overvoltage induced by a 5×10−6 mole/l concentration of compound In2 was about 200 mV. It was almost independent of pH from pH=3 to pH=0.

The copper electrodeposition overvoltage induced by compounds In1 and In2 can be compared with the overvoltage induced in a known way by polyethylene glycol in the presence of chloride ions. For comparison, the test protocol of example 2 was repeated with the two following baths:

    • og05: 0.25 mole/l SO4Cu solution+PEG 3400: 88×10−6 mole/l+Cl−: 10−3 mole/l; pH=0.5
    • og05 mps: idem og05+sodium mercaptopropyl sulfonate: 10−5 mole/l

FIG. 3 shows the corresponding current vs potential curves.

EXAMPLE 4 Progressive Suppression of the Inhibiting Effect of In1 by Mercaptopropyl Sulfonate (MPS)

The inhibition of copper electrodeposition by the compound In1 can be suppressed by increasing concentrations of mercaptopropyl sulfonate (MPS) as shown when the protocol of example 2 was repeated with the three following baths:

Temperature: Room temperature
m 05: 0.25 mole/l SO4Cu solution; In1: 1.25×10−6 mole/l; pH adjusted to 0.5 with H2SO4
o05: idem m05+MPS: 10−5 mole/l
n05: idem m05+MPS: 5×10−5 mole/l

FIG. 4 shows the corresponding current vs potential curves.

A 10−5 mole/l concentration of MPS increased the copper electrodeposition overvoltage induced by 1.25×10−6 mole/l of In1 whereas a concentration of 5×10−5 mole/l of MPS suppressed this inhibition.

EXAMPLE 5 Progressive Suppression of the Inhibiting Effect of In1 by Mercaptopropyl Sulfonate (MPS) at 70-75° C.

Example 4 was repeated with the following baths:

Temperature: 71° C.

tx05: 0.25 mole/l SO4Cu solution; pH adjusted to 0.5 with H2SO4
tx05b: idem tx05+1 nl: 5×10−6 mole/l
tx05d: idem tx05b+MPS: 5×10−5 mole/l
tx05e: idem tx05b+MPS: 10−4 mole/l

FIG. 5 shows the corresponding current vs potential curves.

The addition of 5×10−5 mole/l of MPS almost suppressed the inhibition induced by In1. The experiment was repeated with a double concentration of In1 with the three following baths:

Temperature: 74° C.

tb05: 0.25 mole/l SO4Cu solution; In1: 10−5 mol/l; pH adjusted to 0.5 with H2SO4
tb05e: idem tb05+MPS: 5×10−5 mole/l
tb05g: idem tb05+MPS: 10−4 mole/l

FIG. 6 shows the corresponding current vs potential curves. A concentration of 5×10−5 mole/l of MPS suppressed the inhibition induced by In1 at 74° C.

EXAMPLE 6 Inhibiting Effect of In1 in the Presence of Mercaptopropyl Sulfonate (MPS) at 70-75° C. Measurement by Chronoamperometry

Using the cell of Example 1, the current corresponding to the electrodeposition of copper was recorded as a function of time for a given and constant copper electrode potential with respect to the reference electrode. The solution was stirred by a small magnetic bar. After 70 to 80 seconds the inhibitor In1 was introduced in the bath and the resulting inhibition was recorded.

Temperature: 74° C.

Imposed potential: E=−40 mV/Ag/AgCl
Baths composition:

  • cata 14: 0.25 mole/l SO4Cu solution; MPS: 10−5 mole/l; pH adjusted to 0.5 with H2SO4
    • At 70 seconds: Injection of 6.25×10−7 mole/l of In1
  • cata 10: 0.25 mole/l SO4Cu solution; MPS: 5×10−5 mole/l; pH adjusted to 0.5 with H2SO4
    • At 80 seconds: Injection of 6.25×10−7 mole/l of In1

FIG. 7 shows the corresponding current vs time curves. The abscissa represents the time starting with the setting of the potential which was established about two seconds after the copper electrode immersion in the bath. The ordinate represents the current in milliamperes.

At 74° C. and for a copper electrode potential E=−40 mV/Ag/AgCl, a concentration of 10−5 mole/l of MPS did not prevent the inhibition induced by a concentration of 6.25×10−7 mole/l of In1, whereas a concentration of 5×10−5 mole/l of MPS effectively prevented the inhibition induced by a concentration of 6.25×10−7 mole/l of In1.

EXAMPLE 7 Inhibiting Effect of In1 as a Function of In1 and MPS Concentrations

Example 6 conditions were applied to the following two baths:

Temperature: 74° C.

Imposed potential: E=−40 mV/Ag/AgCl

  • cata 11: 0.25 mole/l SO4Cu solution; MPS: 5×10−5 mole/l; pH adjusted to 0.5 with H2SO4
    • At 70 seconds: Injection of 1.25×10−6 mole/l of In1
  • cata 23: 0.25 mole/l SO4Cu solution; MPS: 10−4 mole/l; pH adjusted to 0.5 with H2SO4
    • At 70 seconds: Injection of 1.25×10−6 mole/l of In1

Curves corresponding to examples 6 and 7 have been combined in FIG. 8.

At 74° C. and for a copper electrode potential E=−40 mV/Ag/AgCl, a concentration of 5×10−5 mole/l of MPS did not suppress the inhibition induced by a concentration of 1.25×10−6 mole/l of In1, but suppressed the inhibition induced by 6.25×10−7 mole/l of In1. A concentration of 10−4 mole/l of MPS only partially suppressed the inhibition induced by 1.25×10−6 mole/l of In1.

EXAMPLE 8 Influence of the Copper Cathode Potential in the Competition Between Inhibitor In1 and Accelerator MPS

Experiments of examples 6 and 7 were repeated with a cathode potential E=−140 mV/AgCl instead of E=−40 mV/Ag/AgCl

Temperature: 74° C.

  • cata 13: 0.25 mole/l SO4Cu solution; MPS: 10−5 mole/l; pH adjusted to 0.5 with H2SO4
    • At 55 seconds: Injection of 6.25×10−7 mole/l of In1
  • cata 12: 0.25 mole/l SO4Cu solution; MPS: 5×10−5 mole/l; pH adjusted to 0.5 with H2SO4
    • At 60 seconds: Injection of 6.25×10−7 mole/l of In1
  • cata 19: 0.25 SO4Cu solution; MPS: 10−5 mole/l; pH adjusted to 0.5 with H2SO4
    • At 70 seconds: Injection of 1.25×10−6 mole/l of In1
  • cata 20: 0.25 mole/l SO4Cu solution; MPS: 10−4 mole/l; pH adjusted to 0.5 with H2SO4
    • At 80 seconds: Injection of 1.25×10−6 mole/l of In1
  • cata 21: 0.25 mole/l SO4Cu solution; MPS: 10−5 mole/l; pH adjusted to 0.5 with H2SO4
    • At 75 seconds: Injection of 2.5×10−6 mole/l of In1
  • cata 22: 0.25 mole/l SO4Cu solution; MPS: 10−4 mole/l; pH adjusted to 0.5 with H2SO4
    • At 75 seconds: Injection of 2.5×10−6 mole/l of In1

FIGS. 9 and 10 show the curves corresponding to these six experiments.

At 74° C. and for a copper electrode potential E=−140 mV/Ag/AgCl,

    • A concentration of 10−5 mole/l of MPS did not suppress the inhibition induced by 6.25×10−7 mole/l of In1, whereas a concentration of 5×10−5 mole/l of MPS completely suppressed this inhibition (cata13/cata12)
    • A concentration of 10−5 mole/l of MPS did not suppress the inhibition induced by 1.25×10−6 mole/l of In1, whereas a concentration of 10−4 mole/l of MPS totally suppressed this inhibition (cata19/cata20) contrary to what was observed for E=−40 mV/Ag/AgCl (cata23)
    • However, a concentration of 10−4 mole/l of MPS only partially suppressed the inhibition induced by 2.5×10−6 mole/l of In1 (cata21/cata22)

This indicated that the displacement of the inhibitor In1 by MPS was therefore easier at an electrode potential of E=−140 mV than at a potential of E=−40 mV/Ag/AgCl

EXAMPLE 9 Influence of the Copper Cathode Potential on the Competition Between the Inhibitor In1 and the Accelerator MPS

The influence of the copper cathode potential is illustrated by the recording of the inhibition obtained with two identical baths at different potentials.

Temperature: 70° C.

  • cata 11: 0.25 mole/l SO4Cu solution; MPS: 5×10−5 mole/l; pH adjusted to 0.5 with H2SO4 Cathode potential: E=−40 mV/Ag/AgCl At 70 seconds: Injection of 1,25×10−6 mole/l of In1
  • cata 8 0.25 mole/l SO4Cu solution; MPS: 5×10−5 mole/l; pH adjusted to 0.5 with H2SO4 Cathode potential: E=−80 mV/Ag/AgCl At 70 seconds: Injection of 1.25×10−6 mole/l of In1

FIG. 11 shows the corresponding curves.

The inhibiting effect of In1 was observed at both potentials, but the inhibition is less for a copper cathode potential E=−80 mV than for a potential E=−40 mV/Ag/AgCl.

EXAMPLE 10 Inhibition of the Copper Deposition by In1 Followed by the Suppression of this Inhibition by an Increased Concentration of Accelerator MPS at 70-75° C.

Starting from a bath containing 10−5 mole/l of MPS, an inhibition was induced by addition of In1, then suppressed by addition of an additional amount of MPS.

Temperature: 74° C.

Imposed potential: E=−40 mV/Ag/AgCl

  • cata 31: 0.25 mole/l SO4Cu solution; MPS: 10−5 mole/l; pH adjusted to 0.5 with H2SO4
    • At 40 seconds: Injection of 6.25×10−7 mole/l of In1
    • At 95 seconds: Injection of 9×10−5 mole/l of MPS
  • cata 32: 0.25 mole/l SO4Cu solution; MPS: 10−5 mole/l; pH adjusted to 0.5 with H2SO4
    • At 40 seconds: Injection of 1.25×10−6 mole/l of In1
    • At 95 seconds: Injection of 9×10−5 mole/l of MPS

FIG. 12 shows the corresponding current vs time curves. A concentration of 10−5 mole/l of MPS did not prevent the inhibition induced by a concentration of 6.25×10−7 mole/l of In1 or of 1.25×10−6 mole/l of In1 at 74° C. whereas a total concentration of MPS of 10−4 mole/l of MPS effectively suppressed the inhibition induced by a concentration of 6.25×10−7 mole/l of In1 or of 1.25×10−5 mole/l of In1 at 74° C.

EXAMPLE 11 Inhibition of the Copper Deposition by In1 Followed by the Suppression of this Inhibition by an Increased Concentration of Accelerator MPS at 70-75° C. Diluted Copper Sulfate Solution

The experiment of example 10 was repeated with solution comprising 0.1 mole/l of copper sulfate at two different imposed potentials:

Temperature: 74° C.

  • cata 36: 0.1 mole/l SO4Cu solution; MPS: 10−5 mole/l; pH adjusted to 0.5 with H2SO4 Electrode potential: E=0 V/Ag/AgCl
    • At 60 seconds: Injection of 6.25×10−7 mole/l of In1
    • At 110 seconds: Injection of 9×10−5 mole/l of MPS
  • cata 40: 0.1 mole/l SO4Cu solution; MPS: 10−5 mole/l; pH adjusted to 0.5 with H2SO4 Electrode potential: E=−0.40 V/Ag/AgCl
    • At 20 seconds: Injection of 6.25×10−7 mole/l of In1
    • At 70 seconds: Injection of 9×10−5 mole/l of MPS

FIG. 13 shows the corresponding current vs time curves. Conclusions of example 10 also apply to solutions comprising containing 0.1 mole/l of copper sulfate.

EXAMPLE 12 Chronoamperometric Recording of the Electrode Immersion in a Bath Containing In1 and Comparison with the Case where In1 is Added During the Deposition

The effect of In1 is compared according to whether it is present during the electrode immersion or added during the copper deposition.

Immersion of the electrode at a controlled potential: (cata41) In order that the copper electrode be at the imposed potential as of its first contact with the electrolyte, a small fine copper wire was attached to the part of the electrode which will not be immersed. The small copper wire was then immersed in the electrolyte, and then the potentiostat was started in order to impose a potential E=−40 mV/Ag/AgCl to the small copper wire against the electrolyte. Then, the copper electrode was immersed and was placed at a potential E=−40 mV/Ag/AgCl as of its first contact, and the resulting current was recorded as a function of time at the rate of one point every 0.005 second.

Temperature 71° C.

Imposed potential: E=−40 mV/Ag/AgCl

  • cata 30: 0.25 mole/l SO4Cu solution; MPS: 10−5 mole/l; pH adjusted to 0.5 with H2SO4
    • At 20 seconds: Injection of 6.25×10−7 mole/l of In1
    • At 95 seconds: Injection of 9×10−5 mole/l of MPS
  • cata 41: 0.25 mole/l SO4Cu solution; MPS: 10−5 mole/l; In1: 6.25×10−7 mole/l pH adjusted to 0.5 with H2SO4
    • At 130 seconds: Injection of 9×10−5 mole/l of MPS

FIG. 14 shows the corresponding curves.

After 20 seconds, In1 induced the same inhibition whatever its mode of introduction in a bath containing 10−5 mole/l of MPS, and this inhibition was suppressed in the same way by an additional amount of MPS.

One observed that the initial current at the very first instant of the immersion in the bath comprising In1 was only about one half of the current observed during the immersion in a bath which did not comprise In1. The increase of the current which was multiplied by a factor of 1.8 during the first 4 to 5 seconds was interpreted as resulting from a faster diffusion of MPS than of In1 to the electrode surface. This was followed by a decrease until a constant current was reached, after about 15 seconds, resulting from the equilibrium between MPS and In1 at the electrode surface.

EXAMPLE 13 Kinetics of In1 Inhibition During the Immersion of a Copper Electrode at Controlled Potential

The chronoamperometric recording procedure of the electrode immersion in the bath described in example 12 was used.

Temperature: Room temperature
Imposed potential (small wire, then electrode): E=−100 mV/Ag/AgCl
Recording of the resulting current vs time at the rate of one point every 0.005 second.

Electrolytes:

a05b: 0.25 mole/l SO4Cu solution; pH adjusted to 0.5 with H2SO4
d05d: Idem a05b+5×10−5 mole/l of In1
d05f2: Idem a05b+10−5 mole/l of In1

FIG. 15 shows the I(t) curves corresponding to a 20-second recording.

FIG. 16 shows the I(t) curves corresponding to the first second following the immersion.

Table 2 below gives the ratio between the current observed during the immersion in the presence of In1: d05d or d05f2 and the current observed in the absence of In1: a05b.

TABLE 2 T(seconds): 0.02 0.05 0.1 0.2 0.3 0.4 0.5 0.6 0.8 1.0 20 I d05d/I a05b: 0.618 0.55 0.52 0.503 0.5 0.494 0.487 0.483 0.468 0.467 0.181 I d05f2/I a05b: 0.476 0.407 0.388 0.36 0.344 0.344 0.326 0.319 0.302 0.302 0.087

The initial decrease of the current during about 0.2 second in FIG. 16 corresponds to the establishment of the Cu2+ concentration gradient. Table 2 and FIGS. 15 and 16 show that 0.05 second after the immersion, the initial inhibition was 45% and 60% for concentrations of 5×10−6 mole/l and 10−5 mole/l of In1, respectively. Then, the inhibition increased slowly during the next 20 seconds.

EXAMPLE 14

Conditions of example 2 were repeated except that, in the bath composition In2 was replaced by N,N dimethyl biguanide hydrochloride at a concentration of 10−4 mole/l. The overvoltage for the copper electrodeposition was not notably modified compared to the same bath devoid of inhibitor for all pHs examined: pH=3, 2, 1 and 0.5 The same conclusion was reached with a bath containing 2×10−4 mole/l of N,N dimethyl biguanide hydrochloride.

EXAMPLE 15 Copper Electrolytic Deposition from Baths Containing In1

Two identical copper sheets were used as cathode and anode, each face of the sheets had an area of 1.45 cm2. In1 was added at a concentration of 4.8 10−6 mole/l in an electrolyte comprising 0.5 mole/l of copper sulfate and 1 mole/l of sulfuric acid and a 30 mA electrolytic current was applied during 15 minutes at room temperature. The copper electrolytic deposit observed with a microscope was smooth, bright and of old rose color.

EXAMPLE 16

The experiment described in example 15 was repeated after adding 2.1×10−5 mole/l of sodium 3-mercaptopropyl sulfonate to the electrolyte. The copper deposit obtained, observed with a microscope, was smooth, mid-bright and of old rose color.

Claims

1. An electrolyte for the electroplating of copper on a cathode, characterized in that it contains an effective amount of a poly (alkylene-biguanide) salt having the general formula (B) wherein:

R—NH—C(:NH)—NH—[—(CH2)p—NH—C(:NH)—NH—C(:NH)—NH—]n—(CH2)p—NH2, xAH
p is an integer comprised between 2 and 12
n is an integer comprised between 2 and 100
R— represents NC— or H2N—C(:O)—
AH represents an acid
x is comprised between (n+1) and 2(n+1) if AH is a monoacid, and between (n+1)/2 and (n+1) if AH is a diacid.

2. The electrolyte for the electroplating of copper on a cathode according to claim 1, characterized in that in formula (B) of the poly (alkylene-biguanide) salt,

p is equal to 6
n is comprised between 6 and 20
R— represents NC—.

3. The electrolyte for the electroplating of copper on a cathode according to one of claim 1 or 2, characterized in that in formula (B) AH represents sulfuric acid.

4. The electrolyte for the electroplating of copper on a cathode according to any of claims 1 to 3, characterized in that the poly (alkylene-biguanide) salt concentration is comprised between 5 10−7 mole per liter and 10−5 mole per liter.

5. The electrolyte for the electroplating of copper on a cathode according to any of claims 1 to 4, characterized in that it includes an accelerator of the electrolytic deposition of copper taken among mercapto-alkyl sulfonic acids and dialkyl disulfide disulfonic acids, or among their salts.

6. A process for the electroplating of copper on a cathode, characterized in that the deposit is obtained starting from an electrolyte according to any of claims 1 to 5.

7. The process for the electroplating of copper on a cathode according to claim 6, characterized in that the cathode comprises submicronic concavities.

8. The process for the electroplating of copper on a cathode according to one of claim 6 or 7, characterized in that a voltage is applied between the anode and the cathode before said cathode is contacted with the electrolyte.

9. A copper electrodeposit on a cathode, characterized in that it has been obtained from an electrolyte according to any of claims 1 to 5.

10. A copper electrodeposit on a cathode, characterized in that it has been obtained by a process according to any of claims 6 to 8.

Patent History
Publication number: 20100243467
Type: Application
Filed: Apr 2, 2007
Publication Date: Sep 30, 2010
Inventor: Daniel Michelet (Nice)
Application Number: 12/294,947
Classifications
Current U.S. Class: Utilizing Organic Compound-containing Bath (205/296)
International Classification: C25D 3/38 (20060101);