JP5022529B2 - Copper filling method - Google Patents

Description

The present invention relates to a copper filling method and includes an additive for promoting or suppressing copper deposition by microcontact printing a copper deposition inhibitor on the surface of the wafer or substrate excluding fine recesses in advance before copper plating. Provided is a copper filling bath that can satisfactorily fill copper in fine recesses of a wafer or substrate using a copper plating bath that is not present.

In recent years, with the miniaturization and high density of copper wiring patterns formed on a wafer or printed circuit board, it has been required to completely fill copper (copper filling) into fine recesses such as via holes or trenches on the wafer or substrate. Yes.
Conventionally, a small amount of additive such as polyethylene glycol (PEG), chloride ion, bis (3-sulfopropyl) disulfide or its disodium salt (SPS), bis (2-sulfopropyl) disulfide or its disodium salt is electrically charged. By adding the copper plating bath to the copper plating bath, the copper filling to the fine recesses is performed. For example, this point is described in the background art of Patent Document 1 (see paragraph 4 of the same Document 1).

Further, Patent Document 2 discloses a method of performing copper filling by performing electrolytic copper plating after a substrate is preliminarily immersed and adsorbed in an aqueous solution containing a plating accelerator composed of disulfides such as SPS.
Incidentally, the copper plating method of Patent Document 1 itself is similar to this.
Japanese Patent No. 3124523 Japanese Patent Laid-Open No. 2001-291554

However, in the conventional method in which electroplating is performed using a copper plating bath containing an additive, the working mechanism is complicated, the bath management is not easy, and the additive is mixed into the deposited film, resulting in a decrease in film characteristics. There is a problem.
Further, in the method of Patent Document 2 in which the substrate is previously immersed in an aqueous solution of a plating accelerator such as SPS, the plating accelerator is attached not only to the inner wall surface and the bottom surface of the fine concave portion of the substrate, but also to the substrate surface. By removing the promoter adhering to the surface of the substrate, it is necessary to selectively promote copper deposition only by the inner wall surface of the fine recesses, and the process is complicated and the productivity is low.

  It is a technical object of the present invention to satisfactorily fill copper in fine concave portions of a wafer or a substrate using a copper plating bath that does not require the addition of an additive such as SPS.

In Japanese translations of PCT publication No. 2003-502507, a substrate is pretreated with a titanium seed layer by vapor deposition or the like, a Pd colloid catalyst ink is applied to a patterned stamp (material is polydimethylsiloxane), and the stamp is placed on the substrate. A copper plating method using microcontact printing is disclosed, in which after a catalyst ink is transferred by contacting the substrate, a substrate is immersed in an electroless copper plating bath to form a fine copper pattern.
In Japanese Patent Publication No. 2004-527133, an alkanethiol ink is applied to a micropatterned stamp, and this stamp is brought into contact with a substrate on which a copper plating layer is formed on one side to perform microcontact printing. A method of forming a fine pattern on a substrate is disclosed that includes etching copper that has not been masked with a thiol resist.
Further, a method for forming a fine pattern using the microcontact printing is disclosed in Japanese Patent Application Laid-Open No. 2002-60979, Japanese Patent Application Laid-Open No. 2004-200663, Japanese Patent Application Publication No. 2005-520213, and the like.

As a result of intensive studies on the influence of ethylene oxide adducts of benzotriazole such as polyethylene glycol-bis (1,2,3-benzotriazolyl ether) on copper plating, the present inventors use the adduct as a representative example. Ascertaining that a specific compound effectively suppresses copper deposition, the idea of utilizing this series of compounds for copper filling by applying the microcontact printing was conceived.
That is, the copper deposition inhibitor ink is applied to an elastomer stamp, and the stamp is brought into contact with a wafer having fine recesses, and the copper deposition inhibitor is selectively microcontacted on the surface of the wafer except for the micro recesses. It was found that when copper electroplating is performed after printing, copper deposition on the wafer surface is suppressed by the action of ink, while copper can be selectively filled in the fine recesses, thereby completing the present invention.

The present invention 1 includes an elastomer stamp containing polyethylene glycol-bis (1,2,3-benzotriazolyl ether), polyethylene glycol-1,2,3-benzotriazolyl ether, amino group, mercapto group or chloro group. Applying an ink comprising a copper precipitation inhibitor that is at least one compound selected from the group consisting of polyethylene glycol derivatives, thiols, and benzothiazoles ;
Forming a metal seed layer on the surface of the wafer or substrate in which fine recesses are formed;
A step of bringing the stamp into contact with the wafer or substrate to transfer the ink to the surface of the wafer or substrate, and selectively microcontact printing a copper deposition inhibitor film on the surface of the wafer or substrate excluding fine recesses;
The method comprises the steps of applying electrolytic copper plating to the wafer or substrate to suppress copper deposition on the wafer or substrate surface, and accelerating and filling copper into the fine recesses of the wafer or substrate. This is a copper filling method.

Invention 2 is a copper filling method according to Invention 1, wherein the elastomeric stamp is a stamp made of polydimethylsiloxane and the metal seed layer is a copper seed layer.

Conventionally, when forming a copper fine pattern on a wafer or substrate, copper filling is performed by adding a trace amount of additives such as PEG and SPS to an electrolytic copper plating bath. However, when this additive-containing copper plating bath is used, there is a problem that the mechanism of action is complicated, the bath management is not easy, and the additive is mixed into the deposited film and the film properties are deteriorated.
In contrast, the present invention applies copper deposition inhibitor ink to an elastomeric stamp, contacts the stamp with a wafer or substrate on which fine concave portions are formed, and deposits copper on the surface of the wafer or substrate excluding the fine concave portions. This is a copper filling method of selectively filling and depositing copper in fine concave portions of a wafer or a substrate by electrolytic copper plating after selective microcontact printing of an inhibitor.
In other words, in the present invention, copper filling using microcontact printing is possible, and the above-mentioned additive-free (additive-free) copper plating bath can be used, so that the bath management becomes extremely easy, and the inclusion of the additive makes it possible to coat the film. The characteristics are not deteriorated.

The present invention relates to an elastomer stamp coated with an ink containing a specific compound such as polyethylene glycol-bis (1,2,3-benzotriazolyl ether) as a copper deposition inhibitor, A copper filling method for promoting copper deposition in fine recesses while making copper contact inhibitor (μCP) contact with the substrate and then suppressing copper deposition on the surface of the wafer or the substrate by electrolytic copper plating It is.

Copper deposition inhibitor of the present invention is a compound that can suppress the deposition of copper when performing electrolytic copper plating on the wafer or substrate, a compound selected from the group consisting of the following compounds (1) to (5) It is used alone or in combination.
(1) Polyethylene glycol-bis (1,2,3-benzotriazolyl ether) (hereinafter referred to as PBTA)
(2) Polyethylene glycol-1,2,3-benzotriazolyl ether
(3) Amino glycol, mercapto group or chloro group-containing polyethylene glycol derivative
(4) Thiols
(5) benzothiazoles

（式(ａ)中、ベンゾトリアゾール環の１位の窒素原子にオキシエチレン鎖の末端が結合している；ｎは１〜２００の整数である）
但し、式(ａ)の化合物のオキシエチレン付加数ｎは５〜１００が好ましく、より好ましくは６８程度（従って、ポリオキシエチレンの平均分子量としては３０００程度）である。 The compounds (1) and (2) are oxyethylene adducts of benzotriazole, and the compound (1) is represented by the following formula (a), in which a benzotriazole ring is bonded to both wings of the oxyethylene chain.

(In formula (a), the end of the oxyethylene chain is bonded to the nitrogen atom at the 1-position of the benzotriazole ring; n is an integer of 1 to 200)
However, the oxyethylene addition number n of the compound of the formula (a) is preferably 5 to 100, more preferably about 68 (therefore, the average molecular weight of polyoxyethylene is about 3000).

（式(ｂ)中、ベンゾトリアゾール環の１位の窒素原子にオキシエチレン鎖の末端が結合している；ｎは１〜２００の整数である）
但し、式(ｂ)の化合物のオキシエチレン付加数ｎは５〜１００が好ましく、より好ましくは６８程度である。 Compound (2) is represented by the following formula (b), and has a benzotriazole ring bonded to only one wing of the oxyethylene chain.

(In the formula (b), the end of the oxyethylene chain is bonded to the nitrogen atom at the 1-position of the benzotriazole ring; n is an integer of 1 to 200)
However, the oxyethylene addition number n of the compound of the formula (b) is preferably 5 to 100, more preferably about 68.

The compound (3) is represented by the following general formula (c).
_{A- (CH 2 CH 2 O)} n-H ... (c)
(In the formula (c), A is SH, NH ₂ or Cl)
The thiols of the compound (4) include alkanethiols represented by the following general formula (d) and dithiols.
CH ₃ (CH ₂ ) _n SH (d)
(N in the formula (d) is an integer of 1 to 20)
Examples of the dithiols include ethanethiol and octanethiol.
Furthermore, examples of the benzothiazoles of the compound (5) include 2-mercaptobenzothiazole, 2-hydroxybenzothiazole and the like.

The present invention is a method of transferring copper to a fine recess by performing copper electroplating after transferring the copper deposition inhibitor to a wafer or substrate on which a fine recess is formed using microcontact printing.
The wafer or substrate of the present invention is premised on a pattern in which fine concave portions are formed, and the fine concave portions mean fine via holes or trenches.
The stamp applied with the ink made of copper precipitation inhibitor must have elasticity, and any elastomer can be used, but silicon-based or fluorine-based elastomers are suitable from the standpoint of having good elasticity and chemical resistance. In particular, polydimethylsiloxane is preferable for the silicon-based elastomer because of its excellent rubber-like elasticity, properties similar to glass, buffering property to ink molecules or gas permeability (see the present invention 2 ).
For example, in the case of PBTA, an ink composed of a copper precipitation inhibitor uses a solution containing ethanol as a solvent, and the concentration of the solution is suitably 0.1 to 100 mmol / L, and more preferably around 5 mmol / L. preferable. As will be described later, a film of a copper precipitation inhibitor is formed on the surface of the wafer or substrate by transfer, and this film is preferably a thin film such as self-assembled monolayers (SAMs). The ink concentration is suitably in the range of millimoles per liter.

On the other hand, in the copper filling method of the present invention , since the surface of the wafer or substrate on which the fine recesses are formed is subjected to electrolytic copper plating in a later step, from the viewpoint of imparting conductivity to the surface, vapor deposition, sputtering, or A thin metal seed layer is previously formed by electroless plating or the like.
The seed layer can be made of any metal such as copper, nickel, tin, solder, gold, silver, etc., but copper is preferred (see Invention 2 ). The seed layer is not limited to a single layer, and may be a multiple layer.
For example, when the metal seed layer is a copper seed layer, from the standpoint of adhesion to the wafer, a titanium nitride underlayer is first formed on the wafer, and a copper seed layer is formed on the underlayer. Is preferred. The film thickness of the copper seed layer is preferably about 0.001 to 1 μm.

Therefore, the processing when the copper filling method of the present invention is applied to a wafer will be described in the order of steps based on FIG.
First, as shown in step (i), for example, polydimethylsiloxane (PDMS) is applied on a silicon wafer and cured, and then peeled off from the wafer to produce an elastomer stamp made of PDMS.
Next, as shown in step (ii), an ink of a copper deposition inhibitor (PBTA is a representative example) is applied to a PDMS stamp.
Next, as shown in step (iii), the stamp coated with the copper deposition inhibitor ink is brought into contact with the wafer on which the copper seed layer is formed. At this time, the PDMS stamp adheres to the wafer due to its rubber elasticity, and the ink is transferred to the surface (that is, only the protruding portion) of the wafer except for the fine recesses.
The wafer is washed with ethanol and / or ion-exchanged water, etc. to remove excess ink, and then dried to selectively deposit copper on the surface of the wafer excluding fine recesses as shown in step (iv). A film of the inhibitor is formed by microcontact printing. In this case, basically, it is preferable that a self-assembled monolayer of a copper deposition inhibitor is formed on the surface of the wafer.
Incidentally, in order to form a self-assembled film of a copper deposition inhibitor, as described above, it is preferable that the concentration of ink is very thin.
Next, as shown in step (v), when copper electroplating is performed on the wafer to which the copper deposition inhibitor has been transferred, copper deposition on the wafer surface is suppressed by the action of the copper deposition inhibitor, and the wafer Since there is no transfer of the inhibitor in the fine recesses of copper, copper is promoted to deposit and fill the fine recesses, thereby achieving copper filling.

Hereinafter, synthesis examples of the copper deposition inhibitor (PBTA) of the present invention, examples in which the copper filling method of the present invention is applied to a wafer, IR measurement test examples on a wafer in which PBTA is microcontact printed, a self-assembled film of PBTA Example of measurement test of cathode polarization curve in copper plating bath, Example of spectroscopic analysis by glow discharge emission in the depth direction of copper film of copper film transferred with copper plating by transferring PBTA, wafer of transferred PBTA Test examples in which the time-dependent change in copper deposition was observed microscopically using an electron microscope will be sequentially described.
The present invention is not limited to the following examples, synthesis examples, and test examples, and it is needless to say that arbitrary modifications can be made within the scope of the technical idea of the present invention.

<< Synthesis example of PBTA >>
In Synthesis Examples 1 and 2, the number of added moles of oxyethylene (EO) bonded to the benzotriazole ring is changed.
(a) Synthesis example 1
In Synthesis Example 1, PBTA was synthesized by first synthesizing polyethylene glycol-ω, ω′-bis (p-toluenesulfonate) from an aromatic sulfonic acid chloride and polyethylene glycol (PEG), and then oxyethylene of this sulfonate ester. It is produced by reacting an adduct with 1,2,3-benzotriazole to bond an oxyethylene chain to the benzotriazole ring.
(1) Synthesis of polyethylene glycol-ω, ω'-bis (p-toluenesulfonate) [Reactants and solvents, etc.]
PEG3000 (average molecular weight 3000; EO68 mol): 60 g
p-Toluenesulfonyl chloride: 7.62 g
Triethylamine: 4.04g
Methylene chloride: 80ml
[Production method]
A 300 ml four-necked flask equipped with a stirrer and a dropping funnel was charged with 60 g of PEG3000 (EO 68 mol) and 50 ml of methylene chloride and stirred under water cooling. Then, 4.04 g of triethylamine was added. Furthermore, a solution prepared by dissolving 7.62 g of p-toluenesulfonyl chloride in 30 ml of methylene chloride was added dropwise under water cooling, and the mixture was stirred at room temperature for 7 days after the addition.
After completion of the reaction, the contents were washed with 100 ml of 3N aqueous hydrochloric acid saturated with sodium chloride, and then washed twice with 100 ml of saturated brine. The separated methylene chloride layer was dried over anhydrous sodium sulfate, and methylene chloride was distilled off under reduced pressure to obtain 60.8 g of a product. The product (polyethylene glycol-ω, ω′-bis (p-toluenesulfonate)) was identified by ¹ H-NMR.
(2) Synthesis of PBTA (polyethylene glycol (EO 68 mol) -ω, ω'-bis (1,2,3-benzotriazolyl ether) [reactants and solvents, etc.]
1,2,3-benzotriazole: 1.43 g
Polyethylene glycol (EO68 mol)
-Ω, ω'-bis (p-toluenesulfonate): 20 g
Potassium carbonate: 1.82g
1,4-dioxane: 75 ml
[Production method]
A 300 ml four-necked flask equipped with a stirrer and a condenser is charged with 1.43 g of 1,2,3-benzotriazole and 60 ml of 1,4-dioxane, heated to 40 ° C., and stirred with 1.82 g of potassium carbonate. added. Next, the temperature was raised to 90 ° C., and a solution of 20 g of polyethylene glycol (EO 68 mol) -ω, ω′-bis (p-toluenesulfonate) produced in the above step (1) dissolved in 15 ml of 1,4-dioxane was 30. Added dropwise in minutes. It stirred for 24 hours at the same temperature after completion | finish of dripping.
After completion of the reaction, the inorganic substance was filtered off, the filtrate was subjected to reduced pressure, and the solvent was distilled off. The obtained residue was dissolved in 100 ml of methylene chloride and washed twice with 100 ml of saturated brine. Subsequently, methylene chloride was distilled off under reduced pressure to obtain 16.2 g of the product. The product (PBTA = added mole number n of oxyethylene was 68) was identified by ¹ H-NMR.

(b) Synthesis example 2
In Synthesis Example 1, an oxyethylene chain is added to the benzotriazole ring as the sulfonic acid ester moiety is eliminated. In Synthesis Example 2, an oxyethylene chain is directly added to the benzotriazole ring. It is an example.
(1) Synthesis of ω, ω'-dichloropolyethylene glycol (EO 45 mol) [Reactants and solvents, etc.]
PEG2000 (average molecular weight 2000; EO45 mol): 60 g
Thionyl chloride: 7.85 g
Pyridine: 5.45g
[Production method]
A 300 ml four-necked flask equipped with a stirrer and a dropping funnel was charged with 60 g of PEG2000 (EO 45 mol) and 5.45 g of pyridine, and stirred under water cooling. Then, 7.85 g of thionyl chloride was added dropwise. After dropping, the temperature was raised to 60 ° C. and reacted for 20 hours, and then the contents were put into ether and the resulting precipitate was suction filtered.
This residue was dissolved in 150 ml of methylene chloride, washed with 50 ml of 3N aqueous hydrochloric acid saturated with sodium chloride, and then washed twice with 50 ml of saturated brine. The separated methylene chloride layer was dried over anhydrous sodium sulfate, and methylene chloride was distilled off under reduced pressure to obtain 52.3 g of a product. The product (ω, ω′-dichloropolyethylene glycol (EO 45 mol)) was identified by ¹ H-NMR.
(2) Synthesis of PBTA (polyethylene glycol (EO 45 mol) -ω, ω'-bis (1,2,3-benzotriazolyl ether)) [reactants and solvents, etc.]
1,2,3-benzotriazole: 2.32 g
ω, ω′-dichloropolyethylene glycol (EO 45 mol): 20 g
Potassium carbonate: 2.96g
DMF: 80ml
[Production method]
A 200 ml four-necked flask equipped with a stirrer and a dropping funnel was charged with 2.32 g of 1,2,3-benzotriazole, 2.96 g of potassium carbonate and 50 ml of DMF and stirred for 30 minutes. Next, after the flask was heated to 60 ° C., a solution prepared by dissolving 20 g of ω, ω′-dichloropolyethylene glycol (EO 45 mol) prepared in the above step (1) in 30 ml of DMF was added dropwise. After dropping, the mixture was heated to 150 ° C. and reacted for 20 hours, allowed to cool, and then insolubles were filtered off.
This filtrate was dissolved in 150 ml of methylene chloride and washed twice with 100 ml of saturated brine. The separated organic layer was dried over anhydrous sodium sulfate, and the solvent was distilled off under reduced pressure to obtain 17.3 g of a product. The product (PBTA = added mole number n of oxyethylene was 45) was identified by ¹ H-NMR.

<< Example of Copper Filling Method >>
(1) Example 1
First, PBTA (EO 68 mol) obtained in Synthesis Example 1 was dissolved in ethanol to prepare a PBTA / ethanol solution having a concentration of 5 mmol / L.
Next, SILPOT 184 and CATALYST SILPOT 184 (both manufactured by Toray Dow Corning) were mixed at a weight ratio of 10: 1 to prepare a solution of polydimethylsiloxane (PDMS), and the solution was placed on a smooth silicon wafer (mold). After coating, it was cured by heat treatment at 70 ° C. for 1 hour, and peeled off from the wafer to obtain a PDMS stamp.
On the other hand, a titanium nitride base thin film was formed on a silicon wafer on which fine via holes were formed by pattern formation, and a copper seed layer was formed thereon by sputtering.
The above PBTA / ethanol solution having a concentration of 5 mm (M = mol / L) is applied on the PDMS stamp, dried in a nitrogen atmosphere for 1 minute, and then brought into contact with the copper seed layer of the silicon wafer for 5 seconds. PBTA was transferred onto the surface of the copper seed, and PBTA was selectively microcontact printed on the surface of the protruding portion excluding the fine via hole of the wafer. This was ultrasonically washed with ethanol for 1 minute and with ion-exchanged water for 1 minute to form PBTA self-assembled films (PBTA-SAMs) on the copper seed.

Next, this wafer was subjected to electrolytic copper plating by the constant potential method, and copper filling was performed on the fine via hole of the wafer.
In this case, the composition of the electrolytic copper plating bath and the plating conditions are as follows.
[Electric copper plating bath]
(1) Bath composition Copper sulfate (as Cu ²⁺ ) 0.80 mol / L
Sulfuric acid 0.45 mol / L
(2) Plating conditions Potential 50mV (vs. Ag / AgCl)
Anode Platinum plate Bath temperature 25 ℃

(2) Comparative Example 1
Based on the above Example 1, except that the step of transferring the PBTA ink to the wafer on which the copper seed layer was formed was omitted, and the copper seed layer was subjected to electrolytic copper plating under the same conditions as in Example 1. did.

Therefore, first, an IR measurement test was performed on the wafer on which PBTA was microcontact-printed based on Example 1 described above.
<Example of IR measurement test by PBTA transfer>
A wafer on which PBTA was microcontact-printed was subjected to IR measurement, and a comparison test with the reflection spectrum of pure PBTA powder was performed to confirm the suitability of transfer of PBTA onto the wafer.
FIG. 2 shows the test results. The lower graph of FIG. 2 shows the reflection spectrum of the reference PBTA, and the upper graph shows the reflection spectrum of the wafer after printing.
According to this, the printed wafer spectrum includes CN stretching vibration (1240 cm ⁻¹ ) derived from PBTA, C—O—C symmetrical stretching vibration (840 cm ⁻¹ ), aromatic C—H bending vibration. Since each peak at (960 cm ⁻¹ ) was confirmed, it was found that PBTA was reliably transferred onto the wafer.

<< Measurement example of cathode polarization curve in copper plating bath by PBTA >>
Next, the cathode polarization curve in the copper plating bath of PBTA-SAMs based on the above Example 1 and the cathode polarization curve in the case where the PBTA based on the above Comparative Example 1 is not transferred are obtained by measurement, and the PBTA is compared by comparing them. The effect of copper on the deposition potential of copper was investigated.
FIG. 3 shows the test results. The cathode polarization curve (dotted line) of Example 1 is substantially parallel to the horizontal axis and largely changed in the negative direction as compared with the polarization curve (solid line) of Comparative Example 1. It was.
That is, in Example 1 in which a PBTA self-assembled film was formed on the wafer surface, the copper deposition potential was greatly shifted compared to Comparative Example 1, and it was difficult to reduce copper ions to metallic copper. As a result, the copper precipitation suppression effect was confirmed.
In particular, in the potential region of −100 to +100 mV (vs. Ag / AgCl / 3.3M KCL), the polarization curve of Example 1 shifts to a base direction larger than that of Comparative Example 1 at the same cathode current density, and strong copper It was found that the effect of suppressing precipitation was working.

《Example of spectroscopic analysis by glow discharge emission in the depth direction of copper film》
Subsequently, glow discharge was performed in the depth direction of the copper film of each wafer on which the electrolytic copper plating of Example 1 and Comparative Example 1 was performed, and a spectroscopic analysis test using the same discharge luminescence was performed.
FIG. 4 shows the test results, and FIG. 4c is an intensity curve of the wafer at the stage where the copper seed layer before the electrolytic copper plating is formed.
FIG. 4a is an intensity curve of the wafer of Example 1 in which PBTA was microcontact-printed. From this figure, a carbon peak due to PBTA was confirmed, and it was confirmed again that PBTA was transferred to the surface of the wafer. . Further, the intensity curve in FIG. 4a (Example 1) includes a copper peak caused by the copper seed layer formed on the wafer and a titanium peak caused by the titanium nitride layer coated as a base of the copper seed layer. It was appearing.
On the other hand, FIG. 4b is an intensity curve of Comparative Example 1 in which electrolytic copper plating was performed without transferring PBTA to the wafer, and a large peak of precipitated copper appeared. In FIG. 4a, FIG. No large peak of precipitated copper was observed.
Moreover, the film thickness of the deposited copper in FIG. 4a was almost the same as the film thickness of the copper seed layer before plating in FIG. 4c.
From the above, when FIG. 4a (Example 1) is compared with FIG. 4b (Comparative Example 1) and FIG. 4c (preceding stage of electrolytic copper plating), when PBTA is transferred to the wafer surface by microcontact printing, copper deposition is observed. It was greatly suppressed, and it became clear that the function as a resist for electrolytic copper plating was sufficiently achieved.

<< Test example of microscopic observation of the time-dependent change of copper deposition on the wafer to which PBTA was transferred >>
Next, with respect to Example 1 in which PBTA was transferred to the wafer and electroplated with copper, the time course of copper deposition was microscopically observed with a field emission scanning electron microscope (FE-SEM).
FIG. 5 shows the results, and almost no copper deposition was observed on the wafer surface until the deposition time passed 450 seconds, and it was confirmed that copper was deposited only in the via holes (FIGS. 5a to 5c). reference).
However, in the electrodeposition time of 600 seconds, some copper deposition was observed on the wafer surface, but copper could be completely filled inside the via hole having a hole diameter of 500 nm (see FIG. 5d).

It is process drawing which applied the copper filling method of this invention to the wafer. It is a reflection spectrum which shows the IR measurement test result with the wafer which carried out the micro contact printing of the copper precipitation inhibitor. It is a cathode polarization curve in the copper plating bath of the self-organization film | membrane of a copper precipitation inhibitor. It is an intensity | strength curve which shows the spectral-analysis test result by the glow discharge light emission to the depth direction of the copper film of the wafer which transferred the copper precipitation inhibitor and gave the copper plating. It is a field emission type scanning electron micrograph (magnification: 20,000 times) showing a change with time of copper deposition when electrolytic copper plating is applied to a wafer to which a copper deposition inhibitor is transferred.

Claims (2)

Polyethylene glycol-bis (1,2,3-benzotriazolyl ether), polyethylene glycol-1,2,3-benzotriazolyl ether, amino group, mercapto group or chloro group-containing polyethylene glycol derivative, thiol A step of applying an ink made of a copper precipitation inhibitor that is at least one compound selected from the group consisting of benzothiazoles ,
Forming a metal seed layer on the surface of the wafer or substrate in which fine recesses are formed;
A step of bringing the stamp into contact with the wafer or substrate to transfer the ink to the surface of the wafer or substrate, and selectively microcontact printing a copper deposition inhibitor film on the surface of the wafer or substrate excluding fine recesses;
The method comprises the steps of applying electrolytic copper plating to the wafer or substrate to suppress copper deposition on the wafer or substrate surface, and accelerating and filling copper into the fine recesses of the wafer or substrate. Copper filling method. 2. The copper filling method according to claim 1 , wherein the elastomeric stamp is a stamp made of polydimethylsiloxane, and the metal seed layer is a copper seed layer.

Images