Patent Application
20190327840
This description pertains to glass surfaces and articles having improved adhesion to copper.
Glass and glass ceramic substrates with vias are desirable for many applications, including for use as in interposers used as an electrical interface, RF filters, and RF switches. Glass substrates have become an attractive alternative to silicon and fiber reinforced polymers for such applications. But, it is desirable to fill such vias with copper, and copper does not adhere well to glass. In addition, a hermetic seal between copper and glass is desired for some applications, and such a seal is difficult to obtain because copper does not adhere well to glass.
Accordingly, a need exists for methods of better adhering copper to glass and glass ceramic materials.
In a first embodiment, a method comprises leaching a surface of a glass or glass ceramic substrate to form a leached layer. The glass or glass ceramic substrate comprises a multi-component material. The material has a bulk composition, in mol % on an oxide basis: 51% to 90% SiO2; 10% to 49% total of minority components ROx. Leaching comprises selectively removing components ROx of the glass or glass ceramic substrate preferentially to removal of SiO2. In the leached layer, the ROx concentration is 50% or less than the ROx concentration of the bulk composition.
In a second embodiment, the first embodiment further comprises etching the surface. Etching comprises selectively removing SiO2 from the substrate preferentially to removal of minority components ROx.
In a third embodiment, the second embodiment further comprises leaching the surface before etching the surface.
In a fourth embodiment, the second embodiment further comprises leaching the surface after etching the surface.
In a fifth embodiment, for the method of the first embodiment, after leaching, the surface has a surface roughness Ra of 0.3 nm or more, and the leached layer has a thickness of 100 nm or more.
In a sixth embodiment, for the method of any of the second through fourth embodiments, after leaching and etching, the surface has a surface roughness Ra of 0.4 nm or more, and the leached layer has a thickness of 20 nm or more.
In a seventh embodiment, for the method of any of the second through fourth embodiments, after leaching and etching, the surface has a surface roughness Ra of 0.5 nm or more, and the leached layer has a thickness of 20 nm or more.
In an eighth embodiment, for the method of any of the second through fourth embodiments, after leaching and etching, the surface has a surface roughness Ra of 1 nm or more, and the leached layer has a thickness of 50 nm or more.
In a ninth embodiment, for the method of any of the first through fourth embodiments, the leached layer has a thickness of 20 nm or more.
In a tenth embodiment, for the method of any of the first through fourth embodiments, the leached layer has a thickness of 50 nm or more.
In an eleventh embodiment, for the method of any of the first through fourth embodiments, the leached layer is nanoporous layer.
In a twelfth embodiment, for the method of the eleventh embodiments, the nanoporous layer comprises pores having a size of 2-8 nm.
In a thirteenth embodiment, for the method of any of the first through twelfth embodiments, the leached layer has a re-entrant geometry.
In a fourteenth embodiment, for the method of any of the first through thirteenth embodiments the surface is an interior surface of a via formed in the glass or glass ceramic substrate.
In a fifteenth embodiment, for the method of the fourteenth embodiment, the via is a through via.
In a sixteenth embodiment, for the method of the fourteenth embodiment, the via is a blind via.
In a seventeenth embodiment, the method of any of the first through sixteenth embodiments further comprises depositing electroless copper onto the surface, and depositing electroplated copper over the electroless copper.
In an eighteenth embodiment, the method of the seventeenth embodiment further comprises charging the leached layer by treating with aminosilanes or nitrogen-containing polycations. After charging, palladium complexes are adsorbed into the leached layer by treatment with a palladium-containing solution. Depositing electroless copper into the leached layer and onto the surface occurs after adsorbing.
In a nineteenth embodiment, for the method of any of the seventeenth and eighteenth embodiments, the electroplated copper is capable of passing a 3N/cm tape test after being annealed at 350° C. for 30 minutes.
In a twentieth embodiment, for the method of any of the first through nineteenth embodiments, ROx is selected from Al2O3, B2O3, MgO, CaO, SrO, BaO, and combinations thereof.
In a twenty first embodiment, for the method of any of the first through twentieth embodiments, the material has a bulk composition, in mol % on an oxide basis:
In a twenty second embodiment, for the method of any of the first through twenty first embodiments, leaching comprises exposing the surface to a solution consisting essentially of hydrochloric acid, sulfuric acid, nitric acid and combinations thereof.
In a twenty third embodiment, for the method of any of the second through twenty second embodiments, etching comprises exposing the surface to an etchant selected from: a solution comprising hydrofluoric acid and hydrochloric acid, and a solution comprising tetramethylammonium hydroxide (TMAH).
In a twenty fourth embodiment, an article comprises a glass or glass ceramic substrate having a plurality of vias formed therein, each via having an interior surface. The glass or glass ceramic substrate comprises a multi-component material, the material having a bulk composition, in mol % on an oxide basis: 51% to 90% SiO2, and 10% to 49% total of minority components ROx. A leached layer is formed under the interior surfaces of the vias. In the leached layer, the ROx concentration is 50% or less than the ROx concentration of the bulk composition. The leached layer has a thickness of 1 nm or more.
In a twenty fifth embodiment, for the article of the twenty fourth embodiment, the via is empty.
In a twenty sixth embodiment, the article of the twenty fourth embodiment further comprises copper filling the via.
In a twenty seventh embodiment, for the article of the twenty sixth embodiment, the copper filling the via is capable of passing a 3N/cm tape test after being annealed at 350° C. for 30 minutes.
In a twenty eighth embodiment, for the article of the twenty fourth through twenty seventh embodiments, the interior surface is an etched surface.
In a twenty ninth embodiment, for the article of the twenty fourth embodiment, the interior surface has a surface roughness Ra of 0.3 nm or more, and the leached layer has a thickness of 100 nm or more.
In a thirtieth embodiment, for the article of the twenty ninth embodiment, the interior surface has a surface roughness Ra of 0.4 nm or more, and the leached layer has a thickness of 20 nm or more.
In a thirty first embodiment, for the article of the thirtieth, the interior surface has a surface roughness Ra of 0.5 nm or more, and the leached layer has a thickness of 20 nm or more.
In a thirty second embodiment, for the article of the thirtieth embodiment, the interior surface has a surface roughness Ra of 1 nm or more, and the leached layer has a thickness of 50 nm or more.
In a thirty third embodiment, for the article of any of the twenty fourth through twenty eighth embodiments, the leached layer has a thickness of 20 nm or more.
In a thirty fourth embodiment, for the article of the thirty third embodiment, the leached layer has a thickness of 50 nm or more.
In a thirty fifth embodiment, for the article of the twenty fourth through thirty fourth embodiments, the leached layer is nanoporous layer.
In a thirty sixth embodiment, for the article of the thirty fifth embodiment, the nanoporous layer comprises pores having a size of 2-8 nm.
In a thirty seventh embodiment, for the article of the twenty fourth through thirty sixth embodiments, the leached layer has a re-entrant geometry.
In a thirty eighth embodiment, for the article of the twenty fourth through thirty seventh embodiments, the via is a through via.
In a thirty ninth embodiment, for the article of the twenty fourth through thirty seventh embodiments, wherein the via is a blind via.
In a fortieth embodiment, for the article of the twenty fourth through thirty ninth embodiments, ROx is selected from Al2O3, B2O3, MgO, CaO, SrO, BaO, and combinations thereof.
In a forty first embodiment, for the article of the twenty fourth through forty first embodiments, the material has a bulk composition, in mol % on an oxide basis:
Glass and glass ceramic substrates with vias are desirable for a number of applications. For example, 3D interposers with through package via (TPV) interconnects that connect the logic device on one side and memory on the other side are desirable for high bandwidth devices. The current substrate of choice is organic or silicon. Organic interposers suffer from poor dimensional stability while silicon wafers are expensive and suffer from high dielectric loss due to semiconducting property. Glass may be a superior substrate material due to its low dielectric constant, thermal stability, and low cost. There are applications for glass or glass ceramic substrates with through vias or blind vias. These vias typically need to be fully or conformally filled by conducting metals such as copper to provide an electrical pathway. Copper is a particularly desirable conducting metal. The chemical inertness and low intrinsic roughness of glass and glass ceramic materials, however, pose a problem related to adhesion of the copper to the glass wall inside the vias. Lack of adhesion between copper and glass could lead to reliability issues such as cracking, delamination, and a path for moisture and other contaminants along the glass-copper interface. Described herein are approaches to increase the effective adhesion between copper and glass or glass ceramic materials on any glass or glass ceramic surface, including the interior surface of vias as well as other surfaces.
In some embodiments, the effective adhesion between copper and glass or glass ceramic may be increased through glass surface treatment such as leaching, or a combination of leaching and etching. It has been discovered that acid leaching can generate a nanoporous layer on the surface both inside the vias and on the planar surface, which has interconnected porosity and thus allows better mechanical interlock. It has been discovered that a combination of leaching and etching leads to higher surface roughness than leaching alone, while still preserving the nanoporous layer created by leaching. It has also been discovered that a combination of leaching followed by etching is surprisingly effective at forming of nanoporous layer with an open surface microstructure and rougher surface. Both the nanoporous layer and higher surface roughness are believed to increase copper adhesion due to mechanical interlocking between copper and the glass or glass ceramic.
In some embodiments, copper is deposited using electroless deposition, or electroless deposition followed by electroplating. Electroless deposition often involves the use of a catalyst, such as Pd. For this type of electroless deposition of copper onto glass, the copper typically does not form a chemical bond to the glass, and instead relies on mechanical interlocking and surface roughness for adhesion. In some embodiments, this mechanical interlocking is achieved by creating rough structure in the glass or glass ceramic substrate with re-entrant geometries. Penetration of catalyst into the re-entrant geometry promotes deposition of electroless copper throughout the re-entrant geometry, which leads to good mechanical interlocking. One example of a re-entrant geometry is an interconnected nanoporous structure.
Substrates with Vias
As used herein, a “via” is an opening in a substrate. A via may extent all the way through the substrate, in which case it is a “through via.” A via may extend only partially through the substrate, in which case it is a “blind via.”
While
First surface 112 and second surface 114 have a pre-etch surface roughness (Ra). As used herein, “surface roughness” refers to arithmetic mean surface roughness. The literature often uses the notation “Ra” to arithmetic mean surface roughness. Surface roughness Ra is defined as the arithmetic average of the differences between the local surface heights and the average surface height, and can be described by the following equation:
where yi is the local surface height relative to the average surface height. Surface roughness (Ra) may be measured and/or calculated from measurements using a variety of techniques. Unless otherwise specified, surface roughness as described herein is measured using a Veeco Dimension Icon atomic force microscope (AFM) with the following parameters: 1 Hz, 512 scans/line, and 2 micron image size.
As used herein, a “nanoporous layer” has a porous structure, where the size of the pores is 100 nm or less. A nanoporous structure as used herein comprises a plurality of interconnected tunnels or “nanopores.” The nanoporous structures described herein are generally open structures, in that there is a path of travel from anywhere within a nanopore to the surface of the material. The nanoporous structures are open because of the manner in which they are formed—the leachant penetrates deeper into the material through the nanoporous layer as it is formed. While the nanoporous layers described herein are generally interconnected, it is possible that portions of the nanoporous network may be isolated from each other. Nano-pore 712 of
The “size” of a nanopore is the average dimension of a cross-section of the pore in a plane normal to the direction of the pore. So, if a cylindrical nanopore intersects a surface, the “size” of the nanopore is the diameter of the circle. For non-circular cross sections, the “size” of the cross-section is the diameter of a circle having the same area as the cross-section. Nanopore size is measured by obtaining a high-resolution SEM image, measuring the area of all visible nanopores in a 100×100 nm area, calculating the diameter of a circle with equivalent area, and calculating the average of these diameters. Where the nanopores are circular in shape, the same result may be obtained by directly measuring the diameter. In some embodiments, the size of the nano-pores are 2 nm to 10 nm, or 2 nm to 8 nm. In some embodiments, the size of the nanopores is 0.5, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 or 12 nm, or any range having any two of these values as endpoints.
In the most general sense, any glass or glass-ceramic composition having 51% or more SiOx may be used, i.e., the original (prior to leaching) bulk composition is:
As used herein, “bulk composition” refers to the composition of a material prior to any leaching or etching. Where leaching or etching preferentially removes some components of a material relative to others, there is a deviation from bulk composition in the leached or etched area. In a SIMS plot, such as that of
In some embodiments, Al2O3 is one of the minority components ROx, and Al2O3 is the component having the highest mol % on an oxide basis after SiO2.
In some embodiments, minority components ROx are selected from from Al2O3, B2O3, MgO, CaO, SrO, BaO, and combinations thereof. The leachants described herein remove each of these components at a rate significantly higher than the rate at which they remove SiO2.
In some embodiments, the material has a bulk composition, in mole percent on an oxide basis:
(Composition 1).
For the compositions described above, the etchants described herein remove SiO2 at a rate higher than that at which they remove the other components. And, the leachants described herein remove each of the ROx components (components other than SiO2) at about the same rate, which is significantly higher than the rate at which the leachants remove SiO2. The amount of SiO2 remaining after the other components have been leached is sufficient to form a robust framework. And, the amount of ROx components is sufficient to form a nanoporous layer when leached.
“Leaching” as used herein means selectively removing minority components ROx of the glass substrate preferentially to removal of SiO2. Leaching occurs when a leaching agent, such as an acid, removes the minority components ROx at a faster rate than SiO2. As a result, the percentage of ROx removed, compared to the amount of SiO2, is greater than would be expected if all components were removed at a rate proportionate to the amount of component in the composition.
As used herein, a “leached layer” refers to a layer in which the ROx concentration is 50% or less than the ROx concentration of the bulk composition due to preferential removal with a leaching agent of the ROx component from the leached layer compared to removal of SiO2. Due to the way it is formed, a leached layer has unique structural characteristics when compared, for example, to a layer having the same composition as the leached layer, but formed by a different method. Compared to the bulk composition, ROx has been removed from the leached layer. The SiO2 and reduced amount ROx components that remain retain the microstructure from the bulk composition, with spaces or pores where the leached ROx was removed. For the compositions described herein, such as Composition 1, leaching generally results in a leached layer having a nanoporous structure with a re-entrant geometry.
Directly measuring the ROx concentration to see whether it is 50% or less than the ROx concentration of the bulk composition by SIMS analysis involves measuring each ROx component by SIMS. Unless otherwise specified, this is how ROx concentration is measured. But, the inventors have determined that, for the compositions and leachants described herein, each of the ROx components leach at about the same rate. This is illustrated, for example, in
As used herein, a “reentrant geometry” refers to a surface geometry where there is at least one line perpendicular to a major surface that crosses the surface of the material more than once. A “major surface” of a material is the surface on a macroscopic scale—the surface defines by a plane that rests on, but does not intersect, the material. For a reentrant geometry, there is at least one line that enters the material, exits the material (into an open nanopore, for example), and reenters the material. Where the reentrant geometry is filled, for example, with copper, even if the copper is not bonded to the material, mechanical interlocking prevents pulling the copper straight out without deforming the copper or the material. A rough surface may or may not be reentrant. A nanoporous surface will almost always be reentrant, although the unlikely case of cylindrical pores, not interconnecting and all aligned perpendicular to the surface, is not reentrant.
In some embodiments, a substrate is subject to leaching but not etching before being metallized. Such a process is illustrated, for example, in
In some embodiments, where a substrate is subject to leaching but not etching before being metallized, after leaching the substrate has a surface roughness Ra of 0.1 nm, 0.2 nm, 0.3 nm, 0.4 nm, 0.5 nm or any range having any two of these values as endpoints, or any open-ended range having any of these values as a lower bound. In some embodiments, after leaching the substrate has a surface roughness of 0.3 nm or more, or 0.3 nm to 0.5 nm.
In some embodiments, where a substrate is subject to leaching but not etching before being metallized, after leaching the substrate has a leached layer with a thickness of 1 nm, 5 nm, 10 nm, 20 nm, 40 nm, 50 nm, 60 nm, 80 nm, 100 nm, 150 nm, 200 nm, or any range having any two of these values as endpoints, or any open-ended range having any of these values as a lower bound. In some embodiments, the leached layer has a thickness of 100 nm or more, or 100 nm to 200 nm.
In some embodiments, where a substrate is subject to leaching but not etching before being metallized, after leaching the substrate has any of the ranges described above for surface roughness Ra combined with any of the ranges described above for leached layer thickness. In some embodiments, the substrate has a surface roughness of 0.3 nm or more, or 0.3 nm to 0.5 nm, combined with a leached layer of 100 nm or more, or 100 nm to 200 nm.
In some embodiments as illustrated herein, all surfaces of a substrate are exposed to leachant. But, in some embodiments, selected surfaces the substrate may be protected from exposure to leachant, for example by photoresist or other protective layer, in which case the selected surfaces would not be leached.
“Etching” as used herein means selectively removing majority component A of the glass substrate preferentially to the removal of minority components B. The etchants used to preferentially remove majority component A can and often do also remove minority components B, but at a rate slower than they remove majority component A. Minority components B are generally removed along with majority component A during etching, as minority components B are quite exposed to etchant and have limited structural integrity once majority component A is removed.
In some embodiments as illustrated herein, all surfaces of a substrate are exposed to etchant. But, in some embodiments, selected surfaces the substrate may be protected from exposure to etchant, for example by photoresist or other protective layer, in which case the selected surfaces would not be etched.
A glass surface that has been etched has distinctive structural characteristics, and one of skill in the art can tell from inspecting a glass surface whether that surface has been etched. Etching often changes the surface roughness of the glass. So, if one knows the source of the glass and the roughness of that source, a measurement of surface roughness can be used to determine whether the glass has been etched. In addition, etching generally results in differential removal of different materials in the glass. This differential removal can be detected by techniques such as electron probe microanalysis (EPMA). Moreover, in the case of previously leached surfaces, etching may remove a portion of the leached layer, as described herein, which is another structural difference between etched and un-etched layers.
Process flow 310 shows steps for preparing substrate 110 for metallization. Schematic 410 shows substrate 110 prior to leaching and/or etching. Region 422, which is the whole substrate in schematic 410, has the bulk composition of substrate 110.
In step 320, substrate 110 is leached. As illustrated in
In step 340, substrate 110 is etched after having been leached in step 320. The etchant and etching parameters are selected to remove some, but not all, of leached layer 424. Schematic 430 shows substrate 110 after etching. Region 422 remains similar to how it appeared after step 320. A part of leached layer 424 has been removed by etching. Dotted line 426 shows the extent of substrate 110 (and leached layer 424) prior to etching. Substrate 110 is illustrated as being smaller in schematic 430 than schematic 420, because etching primarily results in the removal of a layer as opposed to modifying the composition of substrate 110.
In some embodiments, where a substrate is subject to leaching followed by etching before being metallized, after leaching and etching the substrate has a surface roughness Ra of 0.1 nm, 0.2 nm, 0.3 nm, 0.4 nm, 0.5 nm, 0.6 nm, 0.7 nm, 0.8 nm, 0.9 nm, 1.0 nm, or any range having any two of these values as endpoints, or any open-ended range having any of these values as a lower bound. In some embodiments, after leaching and etching the substrate has a surface roughness of 0.4 nm or more, 0.4 nm to 1.0 nm, 0.5 nm or more, 0.5 nm to 1.0 nm, or 1 nm or more.
In some embodiments, where a substrate is subject to leaching followed by etching before being metallized, after leaching and etching the substrate has a leached layer with a thickness of 1 nm, 5 nm, 10 nm, 20 nm, 40 nm, 50 nm, 60 nm, 80 nm, 100 nm, 150 nm, 200 nm, or any range having any two of these values as endpoints, or any open-ended range having any of these values as a lower bound. In some embodiments, the leached layer has a thickness of 20 nm or more, 20 nm to 200 nm, 50 nm or more, or 50 nm to 200 nm. The thickness of the leached layer in this case is that of the leached layer remaining after etching.
In some embodiments, where a substrate is subject to leaching followed by etching before being metallized, after leaching and etching the substrate has any of the ranges described above for surface roughness Ra combined with any of the ranges described above for leached layer thickness. In some embodiments, after leaching and etching the substrate has a surface roughness of 0.4 nm or more, 0.4 nm to 1.0 nm, 0.5 nm or more, 0.5 nm to 1.0 nm, or 1 nm or more, combined with a thickness of 20 nm or more, 20 nm to 200 nm, 50 nm or more, or 50 nm to 200 nm.
After leaching followed by etching, substrate 110 may optionally be metallized by any suitable method. One such method is illustrated in
Process flow 510 shows steps for preparing substrate 110 for metallization. Schematic 610 shows substrate 110 prior to leaching and/or etching. Region 422, which is the whole substrate in schematic 410, has the bulk composition of substrate 110.
In step 520, substrate 110 is etched. As illustrated in
In step 540, substrate 110 is leached after having been etched in step 320. Leaching is expected to form a nanoporous leached layer in an etched surface, just as it does in an un-etched surface. Schematic 630 shows substrate 110 after leaching. A leached layer 624 has been formed due to leaching. Region 422, which has the bulk composition of substrate 110, has correspondingly shrunk. As will become clear from the examples, there is a small part of region 422 next to leached layer 624 that has been subject to some leaching, but not enough to qualify as a “leached layer” as defined herein. Substrate 110 is illustrated as having about the same size in schematic 410 and schematic 420, because leaching primarily removes material from within leached layer 424 to modify the substrate composition, while leaving the shape and size of substrate 110 relatively the same.
In some embodiments, where a substrate is subject to etching followed by leaching before being metallized, after etching and leaching the substrate has a surface roughness Ra of 0.1 nm, 0.2 nm, 0.3 nm, 0.4 nm, 0.5 nm, 0.6 nm, 0.7 nm, 0.8 nm, 0.9 nm, 1.0 nm, or any range having any two of these values as endpoints, or any open-ended range having any of these values as a lower bound. In some embodiments, after etching and leaching the substrate has a surface roughness of 0.4 nm or more, 0.4 nm to 1.0 nm, 0.5 nm or more, 0.5 nm to 1.0 nm, or 1 nm or more.
In some embodiments, where a substrate is subject to etching followed by leaching before being metallized, after etching and leaching the substrate has a leached layer with a thickness of 1 nm, 5 nm, 10 nm, 20 nm, 40 nm, 50 nm, 60 nm, 80 nm, 100 nm, 150 nm, 200 nm, or any range having any two of these values as endpoints, or any open-ended range having any of these values as a lower bound. In some embodiments, the leached layer has a thickness of 20 nm or more, 20 nm to 200 nm, 50 nm or more, or 50 nm to 200 nm.
In some embodiments, where a substrate is subject to etching followed by leaching before being metallized, after etching and leaching the substrate has any of the ranges described above for surface roughness Ra combined with any of the ranges described above for leached layer thickness. In some embodiments, after etching and leaching the substrate has a surface roughness of 0.4 nm or more, 0.4 nm to 1.0 nm, 0.5 nm or more, 0.5 nm to 1.0 nm, or 1 nm or more, combined with a thickness of 20 nm or more, 20 nm to 200 nm, 50 nm or more, or 50 nm to 200 nm.
After etching followed by leaching, substrate 110 may optionally be metallized by any suitable method. One such method is illustrated in
After being leached, leached and then etched, or etched and then leached, substrate 110 may optionally be metallized. Any suitable metallization process may be used. Solution or gas based deposition methods that allow copper to penetrate into the leached layer are preferred.
In some embodiments, electroless deposition is used to deposit copper. Before depositing metal by electroless deposition, the substrate is treated with aminosilanes or nitrogen containing polycations, and a catalyst is deposited. The treatment with aminosilanes or nitrogen containing polycations produces a cationic charge state of the glass surface, which enhances catalyst adsorption. The catalyst adsorption step entails treatment of the glass surface with K2PdCl4 or ionic palladium or Sn/Pd colloidal solutions. The palladium complexes usually exist in anionic form and, therefore, their adsorption on the glass surface is enhanced by the cationic surface groups such as protonated amines. If K2PdCl4 or ionic palladium chemistries are used, the next step involved reduction of the palladium complex into metallic palladium, Pd(0), preferably (but not limited to) in the form of colloids of dimension ˜2-10 nm. If Sn/Pd colloidal solution is used, the palladium is already in Pd(0) form with a Sn shell around it which is removed by acid etching.
Copper deposits by electroless deposition at a much faster rate where a catalyst is present. Adsorbing catalyst inside the nanoporous structure as well as on the rough surface allows electroless deposition of copper inside the nanoporous structure. Such deposition allows for a much higher degree of mechanical interlocking than would be obtained, for example, with copper deposition on a rough surface without a nanoporous layer, or copper deposition on a rough surface with a nanoporous layer where catalyst was not adsorbed throughout the nanoporous layer.
Process flow 350 of
Step 360: charge the nanoporous layer by treating with aminosilanes or nitrogen-containing polycations;
Step 380: after charging, adsorb palladium complexes into the nanoporous layer by treatment with a palladium-containing solution;
Step 390: after adsorbing, deposit electroless copper into the nanoporous layer, for example, a nanoporous layer formed on interior surface 126 of via 124.
In some embodiments, if a thicker copper layer is desired, electroless deposition may optionally be followed by electroplating. Electroless deposition has certain advantages, such as the ability to deposit onto an initially non-conductive surface. But, electroless plating can be slow where thick layers are desired. Once an initial layer of electroless copper is deposited to form the conductive surface used in electroplating, electroplating may be used to more quickly deposit a thicker layer of copper.
After electroless deposition of copper, the samples were annealed at 350° C. for 30 minutes. As described below, the samples were tested for adhesion both before and after annealing. Some samples exhibited superior adhesion prior to annealing. But, avoiding exposure to temperatures similar to the annealing temperature may not be practical, as many applications for copper adhered to glass involve processing at elevated temperatures after the copper is deposited. In addition, annealing relieves stress in the copper, which might, if not relieved, lead to cracking and/or delamination.
Adhesion tests were performed on copper layers deposited as described herein. A tape test may be used to assess the strength of the bond between the conductive metal and first surface 112 of the metal oxide substrate 110. The tape test may be conducted according to ASTM 3359 using a tape having a specific adhesion strength when bonded to the conductive metal. In some embodiments, the tape test may be conducted on a conductive metal that is copper, and the tape used may have a bond strength to copper of 3 N/cm.
Samples were tested after electroless deposition of copper without annealing. For those samples that passed the pre-anneal adhesion test, a similar sample was annealed at 350° C. for 30 minutes and tested again for adhesion. While the samples tested for adhesion were planar, and the copper was not deposited on the interior surface of a via, the tests are indicative of copper adhesion to the interior surface of a via.
Leaching of high-silicon content multi-component glass, such as alkaline earth boro-aluminosilicate glasses with 51% or more SiO2 content, by HCl selectively removes the non-SiO2 components (such as alumina, magnesia etc.) leaving behind a nanoporous surface layer with interconnected geometry. However, the roughness after leaching alone is usually low. On the other hand, etching with HF+HCl selectively removes silicon leaving behind the other metal oxides. While the etchant itself does not necessarily remove the other metal oxides, they do not have sufficient structural integrity to remain once the SiO2 is etched, so the etching process effectively removes these other metal oxides in addition to SiO2. So, etching alone usually leads to high surface roughness but no nanoporous layer. A combination of leaching followed by etching achieves both high surface roughness and interconnected porosity. Surprisingly, the surface roughness observed with leaching followed by etching is significantly higher than the surface roughness observed with etching alone. And, with proper control of etching parameters, a leached layer still remains after leaching followed by etching.
Leaching of glass creates nanoporous surface layer with rougher surface, but the increase in roughness is limited. The nano-porosity could provide extra mechanical interlock in addition to surface roughness. On the other hand etching-only of glass could roughen the surface but couldn't generate nanoporous layer.
Leaching, etching, leaching followed by etching, and etching followed by leaching each lead to different and unique microstructures. Leaching alone results in a nanoporous leached layer and a relatively low surface roughness. Etching alone results in a relatively high surface roughness, but no nanoporous leached layer. Etching followed by leaching leads to a surface roughness comparable to that of etching alone, combined with a nanoporous layer. Leaching followed by etching leads to a surface roughness higher than that obtained by etching alone, combined with a nanoporous layer. Without being limited by theory, it is believed that the presence of the leached layer during etching changes the way the etchant interacts with the glass substrate, leading to a higher surface roughness due to etching. The microstructure generated by leaching followed by etching provides better mechanical interlocking and thus increased adhesion between copper and glass surface.
Different chemistries were explored for etching and leaching the glass composition tested herein. The chemistries are described in Table 1:
As described in the examples below, a variety of glass samples were subject to different etching and/or leaching treatments. The glass described and tested in the examples herein was Corning® Eagle XG glass, which met the criteria of composition 1. Basic testing for leaching was also performed on Corning® Gorilla® Glass, and Corning® Lotus™ Glass, which were observed to form nanoporous layers upon exposure to leaching chemistry.
After etching and/or leaching, the glass samples described below were treated with 1.0 vol % APTES (aminopropyltriethoxysilane) solution (95 mL methanol, 4 mL H2O and 1 mL APTES) for 15 minutes followed by baking in a 120° C. oven for 30 minutes. Afterwards, K2PdCl4 or ionic palladium chemistries were used followed by reduction of the palladium complex into metallic palladium by DMAB (dimethylaminoborane) to create catalyzed substrates. Unless otherwise specified, an ionic palladium chemistry was used in the examples described herein. Then, the catalyzed substrates were coated with a thin copper layer with a thickness of 100-200 nm by electroless plating, followed by a thick copper layer with a thickness of >1 um by electrolytic plating.
The samples were then annealed at 350° C. for 30 minutes. Depending on the sample, tape tests with an adhesion force of 3 N/cm were conducted before and/or after annealing.
Separate 6″ glass wafers (a type of substrate) were treated with (a) no etchant (sample 2a); (b) a weak echant, 5% TMAH (tetramethylammonium hydroxide) solution at 60 C for 10 min (sample 2b); and (c) a strong etchant, 0.1M HF solution mixed with 2M HCl at room temperature for 30 min (sample 2c).
Samples 2a, 2b and 2c were then catalyzed with K2PdCl4 chemistry, and copper was deposited, as described in Example 1.
After electroless plating, a full coverage copper was formed on control sample 2a and TMAH-etched sample 2b. HF—HCl etched sample 2c showed some copper delamination issues. After electroplating 2.5 um copper films, TMAH-etched sample 2b failed the 3N/cm tape test before annealing. Control sample 2a passed the tape test prior to annealing, but failed the tape test (3N/cm) after annealing at 350° C. for 30 minutes.
6″ glass wafers were leached with a 0.15M HCl solution at 75° C. for periods ranging from 2 to 1080 minutes. Dynamic SIMS analysis indicated an aluminum depleted surface layer in each of the leached samples. The thickness of this aluminum depleted surface layer increased with leaching time.
BET surface analysis indicates that the leach layer is nanoporous. For example, the sample 3e that was leached for 1080 minutes had a nanoporous layer with the BJH average pore diameter of 7.16 nm.
Table 2 also compares the roughness of the control and leached samples, which was measured by AFM at a resolution of 500 nm×500 nm. It can be seen that leaching can roughen the glass surface to some extent. The surface roughness was increased from 0.33 to 0.45 nm as leaching time increased from 2 to 240 minutes. Further extending leaching time did not increase roughness.
These SEM images were indistinguishable for samples 3a, 3d and 3e.
For electroless plating and copper-to-glass adhesion evaluation, 6″ glass wafers were leached with 0.15M HCl solution at 75° C. for 2 hours.
As shown in
A layer described as “silica enriched” does not necessarily mean that silica has been added to the layer. Rather, the “silica enriched” layer has a silica content higher than that of the bulk composition. This higher silica content may be due to preferential removal of components other than silica.
The 6″ glass wafers leached with 0.15M HCl solution at 75° C. for 2 hours were then catalyzed with K2PdCl4 chemistry, and copper was deposited as described in Example 1. After electroless plating, a full coverage uniform copper was formed. After electroplating to form the 2.5 um copper film, these samples passed the 3N/cm tape test, but failed after annealing at 350° C. for 30 minutes.
Another leaching experiment was done on 2″×2″ glass coupons with 0.15M HCl solution at 95° C. for 6 hours.
Image 1320 shows only leached layer 1312. Based on image 1320, the pore size is in the range of 2-8 nm. The surface SEM with ×10,000 didn't distinguish the difference of the surface morphology of the control sample without leaching and the leached sample. The surface roughness Ra of the leached samples was measured to be 0.36 nm by AFM.
For electroless plating and copper-to-glass adhesion evaluation, the 2″×2″ glass coupons of Example 4 were leached with 0.15M HCl solution at 95° C. for 6 hours, followed by electroless plating of a thin copper film (100-200 nm in thickness) and electroplating of thick copper film (2.5-5 um in thickness). The leaching solution used in Example 4 during evaluation of electroless plating and copper-to-glass adhesion was slightly different from that used for the SIMS profile, but is expected to work the same. The sample passed 3 N/cm tape test both before annealing and after annealing at 350° C. for 30 minutes.
In Example 5, the glass was first leached. Then, the glass was etched with a weak etchant with temperature and time controlled such that only a part of the leached layer was removed. This treatment surprisingly enables glass having a nanoporous layer with more open surface microstructure and a rougher surface compared to the leaching-only treatment and the etching-only treatment. This surface microstructure and roughness provides better mechanical interlocking and thus increased adhesion between copper and glass surface.
6 sets of 2″×2″ glass coupons were leached with 0.15 M HCl solution at 95° C. for 6 hrs, followed by etching with 5% TMAH solution for varied temperatures and etching times. The etching conditions and the resulting leached layer thicknesses and surface roughness for each sample are listed in Table 3. All of the samples had nanoporous leached layers remaining after etching, with leached layer thickness decreasing at higher etching temperatures and longer etching times. Surprisingly, it was found that the surface roughness of samples that were leached then etched was significantly higher than samples that were etched under the same conditions, but without any leaching. This difference was particularly marked for higher etching temperatures, e.g. 60° C.
The combination of leaching and etching led to the roughest surfaces with more open microstructure. Without being limited by any theories, it is believed that the nanoporous structure is prone to non-uniform etching, which increases post-etch surface roughness of nano-porous structures as compared to non-porous structures. In addition, it is believed that leaching results in an internal leach layer having a relatively open porous microstructure, covered by an external leach layer having a less open porous microstructure due to solution collapse of the porous network caused by the drying. Etching can remove this external leach layer, exposing the internal leach layer with its more open porous microstructure.
Electroless plating and copper-to-glass adhesion evaluation was conducted on 4 samples that were leached with 0.15M HCl at 95° C. for 6 hours followed by 5% TMAH etching at 60° C. for 2, 10, 20, and 30 minutes, respectively. The same procedure was applied to all four samples: electroless plating of a thin copper film (100-200 nm in thickness) followed by electroplating of thick copper film (2.5-5 um in thickness). All four samples passed 3N/cm tape test before annealing. After annealing at 350° C. for 30 minutes, the sample with 2 min-TMAH etching failed, and the other three samples passed 3 N/cm tape test.
Those skilled in the relevant art will recognize and appreciate that many changes can be made to the various embodiments described herein, while still obtaining the beneficial results. It will also be apparent that some of the desired benefits of the present embodiments can be obtained by selecting some of the features without utilizing other features. Accordingly, those who work in the art will recognize that many modifications and adaptations are possible and can even be desirable in certain circumstances and are a part of the present disclosure. Therefore, it is to be understood that this disclosure is not limited to the specific compositions, articles, devices, and methods disclosed unless otherwise specified. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting. Features shown in the drawing are illustrative of selected embodiments of the present description and are not necessarily depicted in proper scale. These drawing features are exemplary, and are not intended to be limiting.
Unless otherwise expressly stated, it is in no way intended that any method set forth herein be construed as requiring that its steps be performed in a specific order. Accordingly, where a method claim does not actually recite an order to be followed by its steps or it is not otherwise specifically stated in the claims or description that the steps are to be limited to a specific order, it is no way intended that any particular order be inferred.
Unless otherwise expressly stated, percentages of glass components described herein are in mol % on an oxide basis.
When referring to a percentage of a percentage herein, the percentages should be multiplied and not added or subtracted. For example, if a quantity is “50% or less than X,” where X is 80%, the quantity is 40% or less. The “50%” of “80%” results in 40% (80%×50%), not 30% (80%-50%).
It will be apparent to those skilled in the art that various modifications and variations can be made without departing from the spirit or scope of the illustrated embodiments. Since modifications, combinations, sub-combinations and variations of the disclosed embodiments that incorporate the spirit and substance of the illustrated embodiments may occur to persons skilled in the art, the description should be construed to include everything within the scope of the appended claims and their equivalents.
This application claims the benefit of priority to U.S. Provisional Application Ser. No. 62/660,677 filed on Apr. 20, 2018, the content of which is relied upon and incorporated herein by reference in its entirety.
| Number | Date | Country | |
|---|---|---|---|
| 62660677 | Apr 2018 | US |
