EP4352277A1

EP4352277A1 - A composite and a method for manufacturing a composite of a copper layer and an organic layer

Info

Publication number: EP4352277A1
Application number: EP22733030.5A
Authority: EP
Inventors: Ralf Schmidt; Ingomar WOLTER; Gerhard Steinberger; Philipp HAARMANN; Norbert LÜTZOW; Thomas HÜLSMANN
Original assignee: Atotech Deutschland GmbH and Co KG
Current assignee: Atotech Deutschland GmbH and Co KG
Priority date: 2021-06-09
Filing date: 2022-06-09
Publication date: 2024-04-17
Also published as: KR20240018632A; CN117597469A; TW202307270A; WO2022258726A1

Abstract

The present invention relates to a composite and a method for manufacturing a composite and a composite, the method comprises in this order the steps: (i) providing a first pure copper structure having a first copper surface; (ii) contacting the first copper surface with an aqueous alkaline conditioner solution and nano-oxidizing the first copper surface by converting Cu-(0) into Cu-(I) and Cu-(II) oxide species obtaining a uniformly oxidized first copper surface layer having a uniformly oxidized first copper surface; (iii) contacting the uniformly oxidized first copper surface with an aqueous alkaline enhancer solution, comprising at least one oxidizing agent, and oxidizing the uniformly oxidized first copper layer by converting the Cu-(I) oxide species into Cu-(II) oxide species and Cu-(0) oxide species into Cu-(I) oxide species obtaining a controlled oxidized first copper surface layer having a controlled oxidized first copper surface; (iv) providing a first non-conductive organic material; and (v) applying the first non-conductive organic material as a first non-conductive organic material layer onto the controlled oxidized first copper surface to obtain the composite.

Description

A composite and a method for manufacturing a composite of a copper layer and an organic layer

Field of the Invention

The present invention generally relates to a composite and a method for manufacturing a composite of a copper structure and a non-conductive organic material, in particular for wa fer/panel level packaging in the production of electronic articles. The method is especially suitable for fan out wafer/panel level packaging for the production of electronic articles. The composite shows excellent adhesion and reliability performance.

Background of the Invention

Recent developments in the semiconductor manufacturing technology request microelec tronic components to become smaller and circuitry within such components to become in creasingly dense. To reduce the dimensions of such components and the structures by which these components are packages and assembled with circuit boards must become more compact. Heterogeneous integration is considered as the key technology to meet cur rent and upcoming requirements. It involves the integration of separately produced compo nents into a package, which provides enhanced functionality and improved operating char acteristics. In this context, system-in-package (SiP) assemblies allow or increasing intercon nect density, decreasing form factor, and increasing efficiency.

Examples for such packages include embedded wafer ball grid array (eWLB) or fan-out wa fer-level packaging (FOWLP) that constitutes a packaging process in which contacts of a semiconductor die are redistributed over a larger area through a redistribution layer (RDL) that is typically formed on a substrate such as a TSV interposer.

While the FOWLP process sequence varies from one manufacturer and packaging variant to the next, the baseline processes are generally comparable. An adhesive material is ap plied to a carrier wafer and one or multiple die are then placed face down onto the adhesive layer. This is followed by a wafer-level over-molding, which essentially embeds the die(s) into the molding layer. Debonding is next in the process, during which the carrier wafer is removed from the newly reconstituted over-molded wafer, thus exposing the active area of the die. A redistribution layer (RDL) is subsequently formed, which occurs across the in creased area of the over-mold, followed by soldering, and finally die singulation.

Such a RDL structure is typically defined by the repeated addition of metal as a copper layer and non-conductive layers as a build-up layer onto the surface of the wafer to re-route the I/O layout into a looser pitch footprint. Such redistribution requires thin film polymers such as BOB, PI or other organic polymers and metallization such as Cu to reroute the peripheral pads to an area array configuration.

Heterogeneous integration usually requires the combination of many different materials such as organic films, metallic Cu, and silicon. Due to the different thermal expansions of these materials, the packages may be prone to reliability issues.

It is essential that no delamination occurs during the lifetime of such a multi-layer structure. Therefore, it is desired that the adhesion between a copper circuitry and a non-conductive layer laminated thereon is as high as possible and remains strong during the lifetime of the respective structure as part of an electronic article.

However, non-conductive layers sometimes suffer the disadvantage that insufficient adhe sion with copper surface of the copper layer is obtained, which results in delamination at the interface between copper layer and non-conductive layer.

To improve the adhesion between the copper layer and the non-conductive layer in the pro duction of printed circuit boards (RGBs) it is known to roughen the copper surface.

In WO96/19097, a copper surface is micro-roughened to improve the adhesion of polymeric materials, by the use of an adhesion-promoting composition comprising hydrogen peroxide an inorganic acid, a corrosion-inhibitor which is for example a tri azole, tetrazole or imidaz ole, and a quaternary ammonium surfactant.

Unfortunately, the use of strong etchant, even with corrosion-inhibitor, leads to a strongly roughened copper surface, usually known as a micro-roughened copper surface. Such a treatment typically removes copper harshly from the circuitries’ surface leaving behind a strongly roughened copper surface, including broad and deep cavities, even in the microm eter range. As a result, such a method is less suitable for fine line circuitries, in particular high frequency applications or wafer/panel level packaging technologies in the production of electronic articles, because the loss of copper is too dramatic and the surface becomes too rough, and the obtained adhesion is often too weak.

EP3310137B1 discloses a method for increasing the adhesion strength between a copper surface and an organic layer wherein in a first step, the copper surface is oxidized to copper oxide and reduced back to metallic copper, and in a second step, the surface is treated with an acidic aqueous non-etching protector solution comprising amino azole and peroxide, and finally laminating the organic layer onto the substrate obtained after the second step.

US 2018/0223412 A1 discloses a roughened copper foil which can significantly improve adhesion to an insulating resin and reliability. The roughened copper foil has a roughened surface by applying an oxidation-reduction treatment, wherein the entire copper surface is composed of a mixed phase of Cu metal and Cu₂0.

US 4,816,086 A discloses a composition which can be used to give copper an oxide coating in order to improve adhesion between the dielectric material and the copper and a method to provide the oxide surface.

US 4,717,439 A relates to a composition and process for improving the leach resistance of the copper oxide coating on the copper circuitry of printed circuit boards to solutions used in their preparation which comprises contacting the copper oxide of the circuit board with a solution containing an amphoteric element which forms an acidic oxide, such as selenium dioxide.

US 2011/186221 A1 relates to a process for improving the bond between a copper/copper alloy and resin, such as in multilayer printed circuit boards.

While the methods are of particular value in production of multi-layer RGBs to promote inter layer adhesion, the methods are not suitable for smaller sizes e.g. in processing/connecting embedded dies having much smaller size and much higher demands in production and re liability. One example of embedding dies is fan out wafer/panel level packaging in which the die is placed on a wafer carrier and the package is built up around it. Fan out is a preferred packaging approach as it is designed to considerably increase I/O density with a reduced footprint and profile, partly due to the fact that it’s thinner than flip chip, as it does not require a package substrate. In fan out processing, the redistribution layers are plated partially on the die and the molding resin. Using a metal or glass wafer carrier which is removed after formation of the RDLs, the RDLs and second layer interconnect (SLI) pads are left open to connect with solder balls to the RGB. Warpage is a critical processing challenge in fan out due to the use of molding resins, thinner substrates, and thicker Cu depositions.

Due to further miniaturization in particular for applications as wafer/panel level packaging and molding of chips in the production of electronic articles, it is generally demanded to further improve existing methods such that adhesion strength is further increased and relia bility of the obtained composites is improved.

Objective of the present Invention

It was therefore the objective of the present invention to provide an improved method for manufacturing a composite, preferably for wafer/panel level packaging for the production of electronic articles, more preferably for fan out wafer/panel level packaging which leads to increased adhesion strength between a surface of a copper layer and a non-conductive organic material layer and improved reliability of the obtained composites compared to prior art methods. It was further the objective of the present invention to provide a method which is more cost efficient environmentally friendly.

Description of the Invention

The aforementioned objectives are solved by a method for manufacturing a composite, pref- erably for wafer/panel level packaging for the production of electronic articles, more preferably for fan out wafer/panel level packaging, the method comprises in this order the steps:

(i) providing a first pure copper structure having a first copper surface;

(ii) contacting the first copper surface with an aqueous alkaline conditioner solution and nano-oxidizing the first copper surface by converting Cu-(0) into Cu-(l) and Cu-(ll) oxide species obtaining a uniformly oxidized first copper surface layer having a uniformly oxidized first copper surface;

(iii) contacting the uniformly oxidized first copper surface with an aqueous alkaline enhancer solution comprising at least one oxidizing agent, and oxidizing the uni formly oxidized first copper layer by converting the Cu-(l) oxide species into Cu- (II) oxide species and Cu-(0) oxide species into Cu-(l) oxide species obtaining a controlled oxidized first copper surface layer having a controlled oxidized first copper surface;

(iv) providing a first non-conductive organic material, preferable a polyimide-contain- ing resin, polybenzoxazole-containing resin or an epoxide-containing resin or mixtures thereof; and

(v) applying the first non-conductive organic material as a first non-conductive or ganic material layer onto the oxidized first copper surface to obtain the compo site, wherein step (ii) is carried out electroless and the aqueous alkaline conditioner solution comprises a mineral base, preferably sodium hydroxide or potassium hydroxide, in a concentration range from 2 to 5 g/L.

The objective is furthermore solved by a composite comprising

(a) a first pure copper structure having a controlled oxidized first copper surface layer comprising Cu-(l) and Cu-(ll) oxide species, which are forming a controlled oxidized first copper surface layer comprising Cu-(l) oxide species and needle like Cu-(ll) oxide species, wherein said Cu-(l) oxide species are interspersed with the needle-like Cu-(ll) oxide species,

- wherein the Cu-(l) oxide species are forming a layer thickness from 5 nm to 100 nm and the needle-like Cu-(ll) oxide species having alength up to 500 nm, more preferably from 100 nm to 500 nm;

(b) a first non-conductive organic material layer, preferable a polyimide-containing resin, polybenzoxazole-containing resin or an epoxide-containing resin or mix- tures thereof;

(c) wherein the first non-conductive organic material layer is attached to the oxidized first copper surface.

It is clear to the skilled person, that the layer thickness is also understood as “average layer thickness”, because the layer thickness can be varied locally, but will be overall in the de- nominated ranges. If sometimes “average layer thickness” is used in the following, it means that the locally layer thickness of a considered total layer thickness can be varied, but count ing all local layer thickness values together and dividing them through the number of the local thickness values, an average layer thickness for the considered layer thickness can be calculated and is within the given ranges. The layer thickness can be seen and determined in FIB pictures of said layers. With other words, e.g. a local layer thickness of less than 5 nm or more than 100 nm of Cu-(l) oxide species can be seen and determined in FIB pictures of a layer or layer assembly, but if the layer thickness is determined over the total layer thickness, shown in the FIB picture, the calculated average layer thickness value is within the given ranges. If “average length” of the needle-like Cu-(ll) oxide species is used in the following, it means according to the explanation of “average layer thickness” above, that a length can be deter mined in FIB pictures for the considered needle-like Cu-(ll) oxide species by taking together all considered needle-like Cu-(ll) oxide species within the FIB picture. Brief Description of the Figures

Fig. 1 shows a FIB picture according to the present invention;

Fig. 2 shows adhesion test results and FIB pictures of inventive examples and comparative example;

Fig. 3 shows the thickness of Cu-(l) oxide species and thickness of the total controlled ox- idized first copper surface layer depending on the temperature in step (iii) according to the invention;

Fig. 4 shows the self-limiting behavior of the formation of the needle-like Cu-(ll) oxide species over time according to the invention; Fig. 5 shows normalized ductility and normalized peel strength for composites of copper layer and polyimide layer in absence and presence of the invention.

Detailed Description of the Invention

According to the present invention, preferably the first pure copper structure having the first copper surface is part of a substrate which is selected from the group consisting of a redis tribution layer (RDL) structure (more specific, the copper structure of redistribution layer structure is the first copper surface), a copper pillar-like structure and/or a copper contact structure of an integrated circuit (IC) or of a die (e.g. onto a wafer). That means that said substrate comprises the first pure copper structure having the first copper surface to be provided for the method and is to be treated. The present invention is in particular suited to form a strong composite between the first pure copper structure having the first copper sur face as part of said substrate and the non-conductive organic material layer to enable ex cellent mechanical properties, especially for shrinking dimensions. Further, the present in vention is in particular suited to improve the ductility of the copper layer of the inventive composite which is considerably increased over composites of the prior art which were not produced by the present invention.

The method of the present invention and the composite of the present invention provide increasing adhesion strength between the copper structure having the controlled oxidized first copper surface and the non-conductive organic material layer without strong etching or a noticeable removal of copper from the copper structure.

Further the present invention does not need and does not include an additional reduction step of the copper oxides species back to metallic copper which reduces the number of method steps, safe costs and time, and avoid the use of a reducing agent as formaldehyde, hydrazine or other environmental critical reducing agents. The present invention in particular does not provide a surface with mixed phases of Cu metal and CU2O over the whole surface derived from a reduction step.

Also, the present invention does not need applying an additional coupling layer comprising e.g. triazole, tetrazole or imidazole, before applying the first non-conductive organic mate rial.

Preferred is a method of the present invention, wherein the first pure copper structure having a copper content of copper of 99 weight- % or higher, preferably 99,9 weight % or higher. In this case pure copper is preferably free of, preferably does not comprise, other elements than copper; more preferably is substantially free of, preferably does not comprise, one, more than one or all elements selected from the group consisting of molybdenum, cobalt, nickel, tungsten, and titanium. Without wishing to be bound by theory, the improved adhesion, ductility and reliability of the composite bases on the combination of steps (ii) and (iii) which allows a first very slow cop per oxidation, creating the uniformly oxidized first copper surface by uniform and smooth reaction, followed by the controlled oxidation during step (iii) of said uniformly oxidized first copper surface obtained during step (ii). Preferably, step (iii) is provided directly after step (ii) without any additional steps as rinsing or cleaning steps, in particular without an acidic dipping step comprising organic or inorganic acids as formic or sulfuric acid.

That means, in step (ii), the first copper surface is nano-oxidized by converting the Cu-(0) surface of the first pure copper structure into Cu-(l) oxide species and Cu-(ll) oxide species, wherein probably Cu-(l) oxide species are slowly converted into Cu-(ll) oxide species wherein the copper-(l) and Cu-(ll) oxide still having the same dimension. Finally, a uniformly oxidized first copper surface layer having a uniformly oxidized first copper surface is formed wherein both, Cu-(l) and Cu-(ll) oxide species, are present. The obtained Cu-(l) and Cu-(ll) oxide species have a nanocrystalline structure. The Cu-(l) and Cu-(ll) oxide species are uniformly distributed over the first pure copper structure wherein Cu-(0) of the first copper surface is converted. The obtained Cu-(l) and Cu-(ll) oxide species are preferably CU2O, CuO, and corresponding hydroxides as Cu(OH)2., more preferably CU2O and CuO. This uni form converting is not limited to certain areas of the first copper surface, e.g. area of copper grain borders of the first pure copper structure, but takes place over the whole surface of the first copper surface, An uneven converting is not desired and will be prevented by applying step (ii). That means the original first copper surface is covered by a uniform copper oxide layer consisting of Cu-(l) and Cu-(ll) oxide species, preferably the Cu-(l) and Cu-(ll) oxide species consisting of Cu₂0, CuO, and corresponding hydroxides as Cu(OH)₂, more prefer ably CU2O and CuO.

The used aqueous alkaline conditioner solution provides very mild oxidation conditions to achieve the slow nano-oxidation in step (ii). Preferably the aqueous alkaline conditioner so lution does not comprise any intentionally added oxidizing agent except of normally (under normal atmospheric pressure) air oxygen dissolved within the temperature range during preparation and use of the aqueous alkaline conditioner solution, preferably between 15°C and 60°C.

Preferably the Cu-(l) and Cu-(ll) oxide species of the uniformly oxidized first copper surface of step (ii) form a uniformly oxidized first copper surface layer having an average layer thick ness from 5 nm to 30 nm or having a layer thickness from 5 nm to 30 nm. That means after step (ii) the uniformly oxidized first copper surface layer formed in step (ii) has a layer thick ness from 5 nm to 30 nm. The Cu-(l) and Cu-(ll) oxide species have a nanocrystalline struc ture wherein the uniformly oxidized first copper surface layer surface is very even and the Cu-(l) and Cu-(ll) oxide species have the same length or dimension, respectively. In step (iii), the uniformly oxidized first copper surface is oxidized by further converting Cu-(l) oxide species into Cu-(ll) oxide species and Cu-(0) into Cu-(l) oxide species with an aqueous al kaline enhancer solution, comprising at least one oxidizing agent, obtaining a controlled ox- idized first copper surface layer having a controlled oxidized first copper surface. Controlled in this context means, that the inventors found, that due to a dense and uniformly oxidized first copper surface layer formed in step (ii), the oxidizing agent oxidizes mainly the Cu-(l) oxide species into Cu-(ll) oxide species since very little amounts of Cu-(0) are available and can be converted into Cu-(l) oxide species. The obtained Cu-(l) species have again a nano- crystalline structure and the obtained Cu-(ll) oxide species have now a larger crystalline structure compared to the dense and small Cu-(l) oxide species. Thus, a converting of Cu- (I) oxide species into needle-like Cu-(ll) oxide species takes place while at the same time small amounts of Cu(0) will be still converted into Cu(l) oxide species. The obtained Cu-(l) and Cu-(ll) oxide species are preferably CU2O, CuO and corresponding hydroxides as CU(OH)2, more preferably CU2O and CuO.

Although the uniformly oxidized first copper surface is well covered (by uniformly oxidation) by the obtained Cu-(l) and Cu-(ll) oxide species, it is believed that the aqueous alkaline enhancer solution can still reach very small areas of the original first copper surface or with other words can be in contact with small amounts of metallic copper of the first copper sur- face. Therefore, also in step (iii) Cu-(l) oxide species are formed, while mainly Cu-(l) oxide species are converted into needle-like Cu-(ll) oxide species.

In any case the first pure copper structure remains without strong etching or a noticeable removal of copper.

Preferably, after step (iii) of the inventive method the converted Cu-(l) oxide species have a nanocrystalline structure even distributed of the whole surface and the converted Cu-(ll) oxide species are needle-like Cu-(ll) oxide species, wherein said Cu-(l) oxide species are interspersed with the needle-like Cu-(ll) oxide species. The Cu-(l) oxide species are forming within the controlled oxidized first copper surface layer a layer thickness from 5 nm to 100 nm and the needle-like Cu-(ll) oxide species having an average length up to 500 nm, more preferably from 100 nm to 500 nm or having a length up to 500 nm, more preferably from 100 nm to 500 nm. That means after step (iii) the controlled oxidized first copper surface layer formed in step (iii) has a layer thickness of Cu-(l) oxide species from 5 nm to 100 nm and the needle-like Cu-(ll) oxide species having a length up to 500 nm, more preferably from 100 nm to 500 nm. With other word, the controlled oxidized first copper surface layer can be considered as having an overall thickness from about 5 to about 500 nm wherein the Cu-(l) oxide species are forming a layer thickness from 5 nm to 100 nm and the needle-like Cu-(ll) oxide species having a length up to 500 nm, more preferably from 100 nm to 500 nm, most preferably from 120 nm to 350 nm. The overall (or total) layer thickness of up to 500 nm of the controlled oxidized first copper surface layer is mainly determined by the length of the needle-like Cu- (II) oxide species which is preferably up to 500 nm after step (iii) and the layer thickness of the Cu-(l) oxide species which is preferably from 5 nm to 100 nm after step (iii). Fig. 1 illus trates the controlled oxidized first copper surface layer after step (iii) wherein dense and uniform Cu-(l) oxide species interspersed with needle-like Cu-(ll) oxide species can be seen.

Advantageously, the method of the present invention provides a very smooth, nano-rough- ened, uniformly oxidized first copper surface layer after step (ii) (and also step iii) compared to conventional etching methods, which usually result in a very pronounced surface rough ness or even surface destruction; usually including altering the surface topography. This is not desired, in particular not for fine line circuitries and high frequency applications. After step (iii) of the method of the present invention, the controlled oxidized first copper surface is desired “uneven” by remaining the Cu-(l) oxide species which are forming within the con trolled oxidized first copper surface layer a layer thickness from 5 nm to 100 nm and the needle-like Cu-(ll) oxide species (needle type morphology) having a length up to 500 nm, more preferably from 100 nm to 500 nm, most preferably from 120 nm to 350 nm.

Preferably the layer thickness of the Cu-(l) oxide species formed in step (iii) is from 10 to 90 nm, preferably from 20 nm to 80 nm.

This mixed structure of controlled oxidized Cu-(l) and Cu-(ll) oxide species after step (iii) is responsible for the improved adhesion and reliability of the composite. This controlled oxi dized first copper surface layer having this uneven surface in height according to the formed layer thickness of Cu-(ll) oxide species and the interposing needle-like Cu-(ll) oxide species is well-suited for the following steps (iv) and (v).

Interestingly in step (iii) the converting of Cu-(l) into Cu-(ll) oxide species terminates by itself after all available Cu-(0) is converted. “Available” in this context means, that converting from Cu-(0) to Cu(l) stops, if the aqueous alkaline enhancer solution cannot reach the original first copper surface through the dense nanocrystalline Cu-(l) oxide species or with other words conversion takes place as long the aqueous alkaline enhancer solution is in contact with metallic copper of the first copper surface.

After step (iii) the surface morphology is significantly modified and typically results in said needle-like surface morphology of the copper-(ll) oxide species and the nano-crystalline structure of the Copper-(l) oxide species, with a very low surface roughness. However, the total surface area is significantly increased compared to the total surface area before and after steps (ii) or . The obtained controlled oxidized first copper surface layer (as the result of this conversion) creates the advanced adhesion strength of the composite.

Preferably step (iii) is performed by a self-limiting formation of the Cu-(ll) oxide species as CuO and/or hydroxides thereof, preferred CuO. It was found that the formation of the Cu-(ll) oxide species, preferably the needle-like Cu-(ll) oxide species also automatically stops if all available Cu-(0) is converted into Cu-(l) oxide species. Self-limiting in this context means that even if oxidizing agent is still available no further converting from Cu-(l) into Cu-(ll) oxide species takes place under the given conditions because the aqueous alkaline enhancer so lution is no longer in contact with metallic copper of the first copper surface or with other words, it is believed that no Cu-(0) can diffuse from this first copper surface to the controlled oxidized first copper surface.

Generally, the surface layer according to steps (ii) and (iii), which are also considered as nano-roughened (because of its nanometer scale), can be investigated, analyzed and meas ured by Atomic Force Microscopy (AFM), Fourier Transform Infrared Spectroscopy (FT-IR), Focused Ion Beam high resolution Scanning Electron Microscopy (FIB high resolution SEM), X-Ray Photoelectron Spectroscopy (XPS) and Transmission Electron Microscopy (TEM). Preferably, the analysis is carried out in vertical cross sections of respective samples. More preferably, the nano-roughened surface layer thickness is observed and determined by means of FIB high resolution SEM and AFM, e.g. to determine the maximum layer thickness. A very preferred method if AFM.

The method according to step (ii) and (iii) is preferably carried out electroless. The first cop per surface in step (ii) can be immersed into the aqueous alkaline conditioner solution or the aqueous alkaline conditioner solution can be sprayed onto said surface.

The aqueous alkaline conditioner solution is nano-oxidizing the first copper surface and con verting it in to the uniformly oxidized first copper surface and is adjusting pH and electro chemical potential of said surface.

The aqueous alkaline conditioner solution used in step (ii) comprises a mineral base, pref erably sodium hydroxide or potassium hydroxide, in a concentration range from 2 to 5 g/L, preferably from 2 to 4 g/L. In a preferred embodiment, the aqueous alkaline conditioner so lution used in step (ii) does not comprise phosphate salts, an organic nitrogen-containing compound or sulfur-containing compound. More preferred, the aqueous alkaline conditioner solution used in step (ii) consists of a mineral base, preferably sodium hydroxide or potassium hydroxide, in a concentration range from 2 to 5 g/L, preferably from 2 to 4 g/L, and water.

The method according to step (ii) wherein the nano-oxidizing is preferably carried out at a temperature from 30 to 60°C, preferably from 45°C to 55°C, and for a time from 0.25 min to 6 min. Preferably the nano-oxidizing is carried out for 0.25 min to 3 min by immersing pro cess or for 3 min to 6 min by spraying process.

The uniformly oxidized first copper surface treated in step (iii) can be immersed into the aqueous alkaline enhancer solution or the aqueous alkaline enhancer solution can be sprayed onto said surface. The aqueous alkaline enhancer solution in step (iii) comprises preferably a mineral base, more preferably sodium hydroxide or potassium hydroxide, in a concentration range from 2 to 5 g/L, preferably from 2 to 4 g/L, and the oxidizing agent is in a concentration range from 10 to 100 g/L, preferably from 15 g/L to 40 g/L if immersed, or preferably from 40 g/L to 100 g/L if sprayed. In a preferred embodiment, the aqueous alkaline enhancer solution used in step (iii) does not comprise phosphate salts, organic nitrogen-containing compound or sulfur-containing compound. More preferred, the aqueous alkaline enhancer solution used in step (iii) consists of a mineral base, preferably sodium hydroxide or potassium hydroxide, in a concentration range from 2 to 5 g/L, preferably from 2 to 4 g/L, the oxidizing agent in a concentration range from 10 to 100 g/L and water.

The method according to step (iii) wherein the oxidizing in step is carried out at a tempera ture from 50 to 80°C, preferably from 55°C to 75°C and for a time from 0.5 min to 20 min, preferably from 2 min to 10 min.

Preferably the temperature in step (ii) is lower than in step (iii), more preferred the tempera- ture in step (ii) is from 45°C to 55°C and in step (iii) from 55°C to 75°C.

Step (iii) preferably follows directly step (ii) without applying other steps as rinsing steps in between.

The oxidizing agent is preferably selected from the group consisting of chlorites including chlorous acid, hypochlorites including hypochloric acid, peroxides including hydrogen per- oxide, permanganates, perchlorates including perchloric acid, persulphates including perox- omonosulfates, peroxydisulfates and its related acids. A preferred chlorite is chlorous acid and alkaline chlorite, most preferably sodium chlorite. A preferred hypochlorite is hypochloric acid and salts thereof. A preferred peroxide is hydrogen peroxide. A preferred perchlorate is perchloric acid and salts thereof. Preferred persulfates are selected from the group consisting of peroxomonosulfates, peroxydisulfates and its related acids. Said oxidizing agents are present in a total concentration sufficient to accomplish the desired oxidation. After step (ia), copper is predominantly present in the oxidation number +2.

In one embodiment of the inventive method before step (ii) a cleaning step (ia) is applied wherein the first copper surface is treated with an acidic solution to remove copper oxide and other residues as grease.

The acidic solution preferably comprises a mineral acid, more preferably sulfuric acid. The concentration of mineral acid is preferably from 3 - 7 weight-% of the total solution.

The acidic solution preferably does not comprise an oxidizing agent as H2O2.

The acidic solution preferably has a pH < 2, preferably < 1.

The cleaning step (ia) is carried out at a temperature from 15 to 25 °C and for a time from 0.25 min to 1 min.

A method of the present invention is preferred, wherein the first non-conductive organic ma terial is a polyimide-containing resin, polybenzoxazole-containing resin or an epoxide-con taining resin or mixtures thereof. Also, liquid crystal polymer (LCP) can be used. In one embodiment of the invention the first non-conductive organic material layer preferably com prises material selected from the group consisting of epoxy compounds including epoxy es ter and fluorine containing epoxy, polyimides including photo-imageable polyimides, cyanate esters, polybenzoxazole, bismaleimide-triazine compounds, polypropylene ethers, and pol yolefins, more preferably the first non-conductive organic material layer is a build-up layer or a mold comprising polyimide-containing resin.

In one embodiment of the invention, the first non-conductive organic material layer is sub stantially free of, preferably does not comprise, filler fibers, preferably is substantially free of, preferably does not comprise, glass fibers.

According to step (v) of the inventive method the first non-conductive organic material as a first non-conductive organic material layer is applied onto the controlled oxidized first copper surface to obtain the composite. Preferably step (v) is directly applied onto the controlled oxidized first copper surface to obtain the composite without any additional treatment steps to treat the controlled oxidized first copper surface as, cleaning, rinsing, applying an adhe sion film, e.g. basing on a silane compound, or applying a solution to prevent leaching as a solution containing an amphoteric element which forms an acidic oxide, such as selenium dioxide.

A method of the present invention is preferred, wherein the first non-conductive organic ma terial is a foil, a dry film, as a dry film build-up layer, or a liquid. Preferably, the first non- conductive organic material is a liquid which can be spin coated or casted by known tech nologies. The skilled person knows that a liquid organic layer in this context is not fully fluid but exhibits a certain viscosity typical in this technical field. In the same manner the skilled person knows that a dry film is not completely dry but rather contains a certain amount of a typical solvent.

Preferably, if the first pure copper structure having the first copper surface is part of a sub strate which is selected from the group consisting of a copper pillar-like structure and/or a copper contact structure of an integrated circuit (IC) or of a die, the first non-conductive organic material can be applied in liquid form by e.g. spin-coating or casting onto the con trolled oxidized first copper surface. Preferably, if the first pure copper structure having the first copper surface is part of a substrate which is selected from the group consisting of a redistribution layer (RDL) structure, the first non-conductive organic material can be applied as foil or dry film and laminated onto the controlled oxidized first copper surface. The appli cation by spin-coating, casting or foil lamination are known to the skilled person.

During step (v) the first non-conductive organic material layer is in direct contact with the controlled oxidized first copper surface. Since said copper surface is modified according to the method of the present invention, the adhesion strength between the non-conductive or ganic material layer and said surface is increased. Preferably the first non-conductive or ganic material layer does not comprise silane-containing compounds as silane-coupling agents.

A method of the present invention is preferred, wherein in step (v) the first non-conductive organic material layer can be in form of a coated layer or in form of a casted mold. The layer has a layer thickness of 10 pm or less, preferably 5 pm or less, more preferably 3 pm or less, even more preferably 15 pm or less, most preferably 10 pm or less. The mold has a thickness of 100 pm to 300 pm, preferably 200 pm to 250 pm.

Own experiments have shown benefits by using a first non-conductive organic material layer, e.g. a build-up layer, exhibiting a layer thickness of 10 pm or less, in particular of 5 pm or less are highly susceptible towards oxygen permeation, in particular when subjected to increased temperatures. Such permeation is highly undesired because it leads to re-oxida- tion of copper. Therefore, the method of the present invention is in particular beneficial for very thin non-conductive organic material layer, in particular below 3 pm, in order to prevent undesired re- oxidation of copper into copper oxides.

Preferably, the composite as a result of the method of the present invention can be used as a precursor for further processing in the manufacturing of an electronic article as a smartphone or computer wherein the further processed composite can connect different electronic components e.g. a microchip with a sensor or with a RAM. Thus, the method of the present invention can be used in the hheterogeneous integration by involving the inte gration of separately produced components into a package as system-in-package (SiP) as semblies, which provides enhanced functionality and improved operating characteristics.

Preferably the method of the present invention comprises additional steps, wherein the non- conductive organic material layer of the composite of step (v), more specific the non-con- ductive organic material layer of the composite, comprises an outer non-conductive surface and wherein the method additionally comprising after step (v) the following steps:

(vi) applying a plating resist layer onto the outer non-conductive surface of the com posite;

(vii) inserting at least one opening going through the plating resist layer, the non- conductive organic material layer and ending up at the controlled oxidized first copper surface and optionally inserting traces going through the plating resist layer and ending up at the outer non-conductive surface of the composite;

(viii) electrochemical depositing of a metal or metal alloy layer into the at least one opening and optional traces to build a conductive via and optional conductive lines.

The metal can be preferably copper, nickel, gold, silver and metal alloys of these metals.

Thus, a method of the present invention is preferred, wherein the first non-conductive or ganic material layer in step (v) is a build-up layer.

After step (viii) is performed, the method according to steps (ii) to (v) can be repeated e.g. 3 to 4 times. In this case, the metal or metal alloy layer, e.g. copper layer, of step (viii) is provided analog as the first pure copper structure having a first copper surface of step (i). Such a sequence of cycles is typical for semi-additive processes and own experiments have shown that the method of the present invention is in particular beneficial for such a process.

Therefore, the method of the present invention is preferably utilized in a semi-additive pro cess in order to build up a redistribution layer structure onto a contact structure of a die.

The benefits of the inventive method are already described in the text above. The aforemen tioned regarding the method of the present invention applies likewise to the composite of the present invention.

The present invention also belongs to a composite comprising:

(a) a first pure copper structure having a controlled oxidized first copper surface layer comprising Cu-(l) and Cu-(ll) oxide species, wherein the converted Cu-(l) oxide species have a nanocrystalline structure and the converted Cu-(ll) oxide species are needle-like Cu-(ll) oxide species, wherein said Cu-(l) oxide species are interspersed with the needle-like Cu-(ll) oxide species,

- wherein the Cu-(l) oxide species are forming within the controlled oxidized first copper surface layer a layer thickness from 5 nm to 100 nm and the needle-like Cu-(ll) oxide species having a length up to 500 nm, more preferably from 100 nm to 500 nm;

(b) a first non-conductive organic material layer, preferable a polyimide-containing resin, polybenzoxazole-containing resin or an epoxide-containing resin or mix tures thereof;

(c) wherein the first non-conductive organic material layer is attached to the con trolled oxidized first copper surface.

Preferably the first pure copper structure having the first copper surface is part of a substrate which is selected from the group consisting of a redistribution layer (RDL) structure, a copper pillar-like structure and/or a copper contact structure of an integrated circuit (IC) or of a die.

The invention is further explained by the following figures and non-limiting examples.

Examples

Example set 1:

A. Sample preparation

Step (i): providing first pure copper structure having a first copper surface:

11 copper foils (5 pm) were plated on top of a Stainless-Steel Wafer by using a state of the art ECD Cu plating electrolyte. The foils were peeled-off from the Carrier and annealed for 1 h at 230°C under Nitrogen atmosphere. Nine foils were treated with the invented method (Inventive Example 1 to 9) and 2 foils (Comparative Example 1 and 2) without the method steps ((ii) and (iii) of the present invention was prepared for comparison. The comparative examples were processed according to standard process flow in the industry.

In step (ia), the surfaces of copper of all foils were cleaned by using an aqueous solution containing 5 weight- % sulfuric acid in Dl water, to obtain cleaned copper foils. The cleaning removed oxides and other compounds, such as anti-tarnishes and/or surfactants. After the cleaning the cleaned copper foils were rinsed with cold water for approximately two minutes. As a result, cleaned and rinsed copper foils were obtained.

Subsequently, step (ii) was performed: In step (ii), said foils were immersed for 45 seconds at 50°C into the aqueous alkaline conditioner solution comprising 3.2 g/L NaOH in Dl water. As a result, conditioned copper foils having uniformly oxidized first copper surface were obtained.

Next, the copper foils obtained after the step (ii) were treated by performing step (iii). In step (iii), the conditioned copper foils were treated with an aqueous alkaline enhancer solution comprising 2,5 g/L NaOH and 250g/l NaCI02 (25 weight- % solution) in Dl water. The treatment was carried out at 70°C for different times from 0.5 min to 8 min leading to a controlled and self-limiting conversion of Cu (0) into Cu (l)-oxides and Cu (l)-oxides into Cu (ll)-oxides, respectively, forming a needle-type layer of converted copper with a maximum layer thickness of more than 350 nm. After the oxidation in step (iii) the copper foils were rinsed with cold water.

In Fig. 1 a FIB picture according to the IE 4 is shown. The picture represents 6 min process time at 70°C in step (iii). The picture shows the layer thickness of the layer of Cu-(l) oxide species, needle-like Cu-(ll) oxide species, and the total layer thickness of the controlled oxidized first copper surface layer.

Step (iv): providing a first non-conductive organic material after step (iii):

In order to provide the Pull-Strength-Test, which needs a stable base layer, an FR4-like material was used, basing on epoxy resin.

Step (v): applying the first non-conductive organic material as a first non-conductive organic material layer onto the controlled oxidized first copper surface to obtain the composite:

The foils of Inventive Examples 1 to 9 were pressed with the controlled oxidized side onto organic dielectric material (FR4-like material) using 20kp/cm². The same procedure was car ried out for the Comparative Examples 1 and 2 without the additional treatment according to the invention. Six foils (Inventive Examples 1 to 5 and Comparative Examples 1) were ex- posed to a temperature of 180°C for 3 hours without N2 Protection and five foils (Inventive Examples 6 to 9 and Comparative Examples 2) were exposed to a 5times Standard Leadfree Reflow Profile.

This test scenario is a known standard test scenario in order to show the quality of the com posite formation e.g. in case of mold material. B. Adhesion strength evaluation:

After the thermal treatment all foils were tested by using the Pull-Strength-Test, were the foil is removed from the dielectric material under 90° direction and the associated force was determined. The test procedure is know to the skilled person. A summary of the results after thermal treatment of 3h/180°C without nitrogen protection is given in Table 1.

A summary of the results after thermal treatment of 5times Standard Leadfree Reflow Profile is given in Table 2.

The values in Table 1 show that Inventive Example 1 and 9 provides significantly improved adhesion strength over Comparative Example 1 and 2.

Fig. 2 shows adhesion test results of inventive example IE 2 and IE 4 as well as comparative example Cl 1. Furthermore, FIB pictures according to IE 2 and IE 4 after step (iii) of the present invention and Cl 1 without inventive treatment are shown. The controlled oxidized first copper surface layer has a total layer thickness of about 350 nm after 8 min. While the FIB picture of Cl 1 does not show a uniformly oxidized copper surface or a controlled oxi dized copper surface, the FIB pictures of the inventive examples show the nanocrystalline Cu-(l) oxide species interspersed with the needle-like Cu-(ll) oxide species.

Fig. 3 shows inventive examples which were prepared as described above wherein the pro cess temperature for the different inventive examples was 60°C and 70°C. The process time was 6 min in every case. FIB pictures (a) and (b) show the total controlled oxidized first copper surface layer depending on the temperature in step (iii) according to the invention (FIB pictures (a) at 60°C and (b) at 70°C). Fig. 3c shows the total thickness of the controlled oxidized first copper surface layer. Fig. 3d shows the thickness of Cu-(l) oxide species of the controlled oxidized first copper surface layer.

Fig. 4 shows the self-limiting behavior in step (iii) of the present invention according to the formation of the needle-like Cu-(ll) oxide species over time. It can be seen that the layer thickness of the layer of Cu-(l) oxide species and the total layer thickness of the controlled oxidized first copper surface layer is growing within the first 3 min very fast to a total layer thickness of about 300 nm and remains quite stable at this point during step (iii). At 4 and 6 min also the FIB pictures of the controlled oxidized first copper surface layer are shown. Example set 2:

A. Sample preparation

Wafer substrates with copper layers were prepared for further processing with the inventive method (inventive example - Cu w AP). The comparative example was not treated according to the inventive method. In particular steps (ii) and (iii) of the inventive method were not conducted (comparative example - Cu).

Step (i): providing first pure copper structure having a first copper surface:

A copper layer (20 pm deposit thickness) was applied on top of a gold wafer substrate by employing a state of the art ECD Cu plating electrolyte (Spherolyte Cu UF 3 process, Atotech). Copper layer was provided by using a standard electrolyte plating tool (Rena) at 25 °C and 2 ASD. The resulting copper layer was self-annealed at room temperature prior to further processing.

Initially in step (ia), the surface was cleaned with an acidic solution (5% sulfuric acid in water) for 60 s at 35 °C and rinsed with water for 35 s at ambient temperature. The cleaning re- moved oxides and other compounds, such as anti-tarnishes and/or surfactants.

Subsequently, step (ii) was performed wherein the aqueous alkaline conditioner solution was applied on the previously prepared copper layer. In this step, the copper-covered wafer substrates were immersed for 30 s at 50 °C in the solution of NaOH (3.2 g/L) in water. As a result, conditioned copper surfaces with uniformly oxidized first copper surface were ob- tained.

In step (iii), the conditioned copper foil surface was treated with an aqueous alkaline en hancer solution comprising NaOH (2.5 g/L) and 80 g/L sodium chlorite (25weight-% solution) for 6 min at 70 °C. The samples were subsequently rinsed with water for 0.5 min at ambient temperature. Step (iv) and (v):

A polyimide material as a first non-conductive organic material was provided and applied onto the controlled oxidized first copper surface of step (iii). The comparative example was not treated with step (ii) and (iii) and the bare copper foil was used. The polyimide material (LTC 9320 E07, Fuji Film) was applied by spin coating at 1000 rpm using industrial standard processes. Prebaking was conducted twice for 6 min at 105 °C, followed by UV curing with 300 mJ. Further curing for performed under nitrogen atmosphere for 1 h at 230 °C.

B. Tensile test and peel strength test: Both examples were further subjected to 2 h at 230 °C thermal treatment after applying the polyimide material and the resulting composites of copper and polyimide were removed from the gold substrate.

The ductility of the composite of copper layer and polyimide layer in absence and presence of the use of inventive method were determined by tensile tests. Test stripes (12.7 mm width), 100 mm length) were cut first using a JDC Precision Sample Cutter and afterwards annealed for 1 h at 120 °C in inert atmosphere. The ductility was determined with a tensile test device (Zwick Z1.0, force sensor 500 N pneumatic clamps with Vulkollan coating). Fig ure 5 compares the normalized ductility values of the composite of copper layer and polyi- mide layer with and without the inventive method (comparative example “Cu” vs. inventive example “Cu w AP” (AP - adhesion promotion)). For better comparison, the values were normalized to the comparative example in absence of inventive method. Composite for mation upon application of inventive method resulted in a considerable increase of the duc tility. For measuring the peel strength, double-sided polyimide tape was placed onto the spin- coated polyimide of the composite. After application of the double-sided tape, a part of the composite was cut out to the dimensions of the tape. Therefore, the composite of polyimide layer and copper layer was transferred to a back panel for mechanical reinforcement and cut into strips of 10 mm width. At the beginning of each peel stripe, the part had to be sepa- rated from the polyimide layer using a knife blade to avoid separation of copper and back panel. The polyimide layer of the composite was pulled in perpendicular direction of the sample orientation and the required force to remove the layer from the subjacent copper layer in absence and presence of inventive method was measured. The corresponding re sults are shown in figure 5 and revealed significantly stronger adhesion in the presence of inventive method (comparative example “Cu” vs. inventive example “Cu w AP” (AP - adhe sion promotion)). While negligible adhesion was observed for the comparative example of the composite of the copper layer and the polyimide layer, large values were obtained upon application of inventive method. For better comparison, the values were normalized to the comparative example and the inventive method provided an improvement by a factor of ca. 4.5.

Claims

1. A method for manufacturing a composite, preferably for wafer/panel level packaging for the production of electronic articles, more preferably for fan out wafer/panel level packaging, the method comprises in this order the steps:

(i) providing a first pure copper structure having a first copper surface;

(ii) contacting the first copper surface with an aqueous alkaline conditioner solution and nano-oxidizing the first copper surface by converting Cu-(0) into Cu-(l) and Cu-(ll) oxide species obtaining a uniformly oxidized first copper surface layer having a uni formly oxidized first copper surface;

(iii) contacting the uniformly oxidized first copper surface with an aqueous alkaline en hancer solution comprising at least one oxidizing agent, and oxidizing the uniformly oxidized first copper layer by converting the Cu-(l) oxide species into Cu-(ll) oxide species and Cu-(0) oxide species into Cu-(l) oxide species obtaining a controlled ox idized first copper surface layer having a controlled oxidized first copper surface;

(iv) providing a first non-conductive organic material, preferable a polyimide-containing resin, polybenzoxazole-containing resin or an epoxide-containing resin or mixtures thereof; and

(v) applying the first non-conductive organic material as a first non-conductive organic material layer onto the controlled oxidized first copper surface to obtain the composite, wherein step (ii) is carried out electroless and the aqueous alkaline conditioner solution comprises a mineral base, preferably sodium hydroxide or potassium hydroxide, in a concentration range from 2 to 5 g/L.

2. Method according to claim 1, wherein the uniformly oxidized first copper surface layer formed in step (ii) has a layer thickness from 5 nm to 30 nm.

3. Method according to claim 1 or 2,

- wherein the controlled oxidized first copper surface layer formed in step (iii) has a layer thickness of Cu-(l) oxide species from 5 nm to 100 nm and the needle-like Cu-(ll) oxide species having a length up to 500 nm, more preferably from 100 nm to 500 nm.

4. Method according to claim 3, wherein the layer thickness of the Cu-(l) oxide species formed in step (iii) is from 10 to 90 nm, preferably from 20 nm to 80 nm.

5. Method according to any of the preceding claims wherein in step (iii) the converting of Cu- (I) into Cu-(ll) oxide species terminates by itself after all Cu-(0) is converted.

6. Method according to any of the preceding claims, wherein the first pure copper structure having the first copper surface is part of a substrate which is selected from the group con sisting of a redistribution layer (RDL) structure, a copper pillar-like structure and/or a copper contact structure of an integrated circuit (IC) or of a die.

7. Method according to any of the preceding claims wherein step (ii) is carried out without adding an oxidizing agent to the aqueous alkaline conditioner solution and that the aqueous alkaline conditioner solution does not contain said oxidizing agent, preferably without add ing an oxidizing agent to the aqueous alkaline conditioner solution selected from the group consisting of chlorites including chlorous acid, hypochlorites including hypochloricacid, per oxides including hydrogen peroxide, permanganates, perchlorates including perchloric acid, persulphates including peroxomonosulfates, peroxydisulfates and its related acids, and that the aqueous alkaline conditioner solution does not contain said oxidizing agent.

8. Method according to any of the preceding claims wherein step (iii) is carried out electroless and the aqueous alkaline enhancer solution comprises a mineral base, preferably sodium hydroxide or potassium hydroxide, in a concentration range from 2 to 5 g/L and the oxidizing agent is in a concentration range from 10 to 100 g/L.

9. Method according to any of the preceding claims wherein the oxidizing agent is selected from the group consisting of chlorites including chlorous acid, hypochlorites including hypo chloricacid, peroxides including hydrogen peroxide, permanganates, perchlorates including perchloric acid, persulphates including peroxomonosulfates, peroxydisulfates and its re lated acids.

10. Method according to any of the preceding claims wherein the nano-oxidizing in step (ii) is carried out at a temperature from 30 to 60°C and for a time from 0.25 min to 6 min.

11. Method according to any of the preceding claims wherein the oxidizing in step (iii) is carried out at a temperature from 50 to 80°C and for a time from 0.5 to 20 min.

12. Method according to any of the preceding claims wherein before step (ii) a cleaning step (ia) is applied wherein the first copper surface is treated with an acidic solution to remove copper oxide.

13. Method according to claim 12 wherein the cleaning step (ia) is carried out with a tempera ture from 15 to 25 °C and for a time from 0.25 min to 1 min.

14. Method according to any of the preceding claims wherein the non-conductive organic ma terial layer of the composite of step (v) comprises an outer non-conductive surface and wherein the method additionally comprising after step (v) the following steps:

(vi) applying a plating resist layer onto the outer non-conductive surface of the composite;

(vii) inserting at least one opening going through the plating resist layer, the non-conduc tive organic material layer and ending up at the controlled oxidized first copper surface and optionally inserting traces going through the plating resist layer and ending up at the outer non-conductive surface of the composite;

15. A composite comprising:

(a) a first pure copper structure having a controlled oxidized first copper surface layer comprising Cu-(l) and Cu-(ll) oxide species, wherein the converted Cu-(l) oxide species have a nanocrystalline structure and the converted Cu-(ll) oxide species are needle-like Cu-(ll) oxide species, wherein said Cu-(l) oxide species are inter spersed with the needle-like Cu-(ll) oxide species,

(b) a first non-conductive organic material layer, preferable a polyimide-containing resin, polybenzoxazole-containing resin or an epoxide-containing resin or mixtures thereof;

(c) wherein the first non-conductive organic material layer is attached to the controlled oxidized first copper surface.