Patent Application
20190136397
The present invention is directed to electroplated copper and methods of electroplating the copper where copper metal has high tensile strength. More specifically, the present invention is directed to electroplated copper and methods of electroplating the copper where copper metal has high tensile strength and has a certain percentage of specific angle misorientations between adjacent grains of the copper metal with respect to a crystal plane direction of <111> to provide the high tensile strength copper metal in addition to other improved material properties.
In the drive toward future applications in the electronics industry, such as artificial intelligence and autonomous vehicles, advanced semiconductor packaging products and processes that enable high density circuits, reduced form factor (size, configuration or physical arrangement of a device) and increased functionality in electronic devices are highly desired. In addition to the required smaller package size, more reliable chip-to-chip, chip-to-circuit board, and terminal to terminal interconnects are increasingly in demand. For example, copper has been used for interconnect applications for the past few decades, such as in copper redistribution Layers (RDL) which are used to reroute conduction paths within chip packages. As the package size decreases, the demand for reliable fine-line RDLs is increasing. During thermal cycling tests (TCT) copper cracking has been reported for fine-line RDLs due to thermal stresses induced from the differences in the coefficient of thermal expansion (CTE) of copper and adjacent materials when cycling between hot and cold chambers. The pathway to resolve the cracking issue is still controversial. In conventional printed circuit board (PCB) applications, high elongation copper is commonly known as the preferred way to resolve the cracking issue. However, as feature sizes of electronic components on PCBs decrease and the components become more integrated to microsized and even to nanosized dimensions, such as is currently taking place in advanced packaging applications, this approach may not be suitable. There is another school of thought which argues that high material strength or high tensile strength, not high elongation, is essential to avoid copper cracking upon multiple thermal cyclings. A disadvantage of copper with high tensile strength is that such copper can become brittle. One approach to addressing the problem of cracking is to use nanotwinned copper which is characterized by having both high tensile strength and high elongation. Although nanotwinned copper may be suitable to address the problem of cracking of copper in RDLs, nanotwinned copper has been found to be unsuitable in many applications for plating vias, such as in advanced packaging applications where 3D stacking is needed to increase circuit densities. In many such applications via fill has not been found to be acceptable and the surface of the copper deposits have been found to be unacceptably rough and nonuniform.
Therefore, there is a need for a copper metal which can withstand thermal stresses induced from differences in the CTE between copper and adjacent materials when cycling between hot and cold chambers without the copper cracking and enables smooth and uniform copper deposits in features, even in high circuit density application.
The present invention is directed to a copper metal including twin fractions of 30% or greater of grain boundaries between adjacent copper grains having angles of misorientation of 55° to 65° with respect to a crystal plane direction of <111>.
The present invention is also directed to a method of electroplating copper including:
The copper metal of the present invention has improved tensile strength over many conventional copper metals deposited from copper plating baths or by physical or chemical vapor deposition. In addition, the copper metal of the present invention has good elongation and low thermal stress. The properties of the copper metal of the present invention inhibit the copper metal from cracking when the copper metal is exposed to heat in high temperature environments, such as TCT and as are found in annealing processes. The copper metal electroplating compositions of the present invention can be used to electroplate the copper metal of the present invention at high current density applications with non-conformal plating where the copper is simultaneously deposited on a surface of a substrate at a faster rate than in apertures, such as vias, within the substrate to provide smooth and uniform copper deposits in the aperture and on the surface of the substrate. The method of the present invention can also electroplate the copper metal of the present invention directly on or adjacent to metal seed layers which are adjacent to or joined to dielectric or semiconductor materials used in electronic devices where the CTE of the materials differ without the concern for cracking. The copper metal of the present invention is highly suitable for fine line redistribution layer technology used in advanced packaging where redistribution line pitch is decreasing and circuit density is increasing.
As used throughout this specification the following abbreviations shall have the following meanings unless the context clearly indicates otherwise: A=amperes; A/dm2=amperes per square decimeter=ASD; DC=direct current; ° C.=degrees Centigrade; mmol=millimoles; mg=milligrams; g=gram; L=liter; mL=milliliter; ppm=parts per million=mg/L; m=meters; μm =micron=micrometer=10−6 meters; mm=millimeters; cm=centimeters; nm=nanometers=10−9 meters; Å=angstroms=1×10−10 meters; 2.54 cm=inch; MPa=megapascal=N/m2; N=Newtons; kV=kilovolts; V=volts=joule/coulomb; mA=milliamperes; DI=deionized; mJ=millijoules; Joule=kg(m)/s2; kg=kilograms; s=seconds; Mw=weight average molecular weight; Mn=number average molecular weight; wt %=weight percent; XRD=X-ray diffraction; EBSD=electron backscatter diffraction; FE-SEM=field emission scanning electron microscope; EO/PO=ethylene oxide/propylene oxide copolymer; IPF=inverse pole figure; RDL=redistribution layer; N2=nitrogen gas; vs.=versus; e.g.=example; ohm-cm resistance; and L/S=line spacing or distance between two features or structures, such as for an RDL.
As used throughout this specification, the term “plating” refers to metal electroplating. “Deposition” and “plating” are used interchangeably throughout this specification. The terms “composition” and “bath” are used interchangeably throughout the specification. “Accelerator” refers to an organic additive that increases the plating rate of the electroplating composition and is also used to improve brightness of copper deposits. “Suppressor” refers to an organic additive that suppresses the plating rate of copper during electroplating. The term “electrolyte” means a chemical compound which dissociates into ions and hence capable of transporting electric charge, e.g. an acid. The term “moiety” means a part of a molecule or polymer that may include either whole functional groups or parts of functional groups as substructures. The terms “moiety” and “group” are used interchangeably throughout the specification. The term “aperture” means opening, hole, gap or via. The term “aspect ratio” means thickness of the substrate divided by the diameter of the aperture of a feature in the substrate. The term “grain boundary” means the interface between two grains, or crystallites, in copper metal, wherein grain boundaries are 2 dimensional defects (2D) in the crystal structure of the copper. The term “grain”, “crystal” and “crystallite” are used interchangeably throughout this specification. The term “misorientation” means the difference in crystallographic orientation between two grains, or crystallites, with a common interface, wherein crystallographic orientation between the two grains, or crystallites, can range from 0-180° angles, wherein 0° indicates a perfect crystal without any misorientations. The term “tensile strength” means the resistance of a material to breaking under tension. The term “thermal stress” means stress that occurs as a result of thermal expansion of copper structural members when its temperature changes. The term “annealing” means a heat treatment that alters the physical and sometimes chemical properties of a material. The term “Miller Indices: (hkl), [hkl], {hkl} and <hkl>” mean the orientation of a surface of a crystal plane defined by considering how the plane (or any parallel plane) intersects the main crystallographic axis of a solid (i.e., the reference coordinates—x, y, and z axis as defined in a crystal, wherein x=h, y=k and z=I), wherein a set of numbers (hkl), [hkl], {hkl} and <hkl> quantify the intercepts and are used to identify the plane. The expression “(hkl)” defines a specific crystal plane in a lattice. The expression “[hkl]” defines the specific direction of a crystal plane in a lattice. The expression “{hkl}” defines the set of all planes that are equivalent to (hkl) by the symmetry of the lattice. The expression “<hkl>” defines the set of all directions that are equivalent to [hkl] by the symmetry of the lattice. The term “plane” means a two-dimensional surface (having length and width) where a straight line joining any two points in the plane would wholly lie. The term “lattice” means an arrangement in space of isolated points in a regular pattern, showing the position of atoms, molecules or ions in a structure of a crystal. The term “grain boundary energy” means the energy at an interface between two grains due to interface formation. The term “crystallographic domains” means atoms inside a specific space have the same atomic arrangements, crystallinity, orientations and symmetry. The term “texture (crystalline)” means distribution of crystallographic orientations of a copper sample, wherein the sample in which these orientations are fully random is said to have no distinct texture and, if the crystallographic orientations are not random, but have some preferred orientation, then the sample has a weak, moderate or strong texture, wherein the degree is dependent on the percentage of crystals having the preferred orientation. The term “slip system” means a set of symmetrically identical slip planes and associated family of slip directions for which dislocation motion can easily occur and lead to plastic deformation, wherein an external force makes parts of the crystal lattice glide along each other, changing the material's geometry. The term “pitch” means a frequency of feature positions from each other on a substrate. The term “amino group”=—NHR, where R is —H (hydrogen) or a linear or branched hydrocarbyl group. The term “aminoalkyl”=-(C1-C4)—NH—R, where R—H (hydrogen) or a linear or branched hydrocarbyl group. The term “hydrocarbyl group” means a hydrogen and carbon functional group. The term “halide” means chloride, fluoride, bromide and iodide. The term “adjacent” means directly on or next to such that two structures or materials have a common interface. The articles “a” and “an” refer to the singular and the plural.
As used throughout the specification the average of a parameter means the sum of the individual measurements of a parameter divided by the number of measurements taken for the parameter. Grain size (spherical equivalent diameter) is based on the calculation that all grains are spherical, wherein grain size area=π(d/2)2, wherein d=grain diameter. Copper has a cubic structure of six sides and is the same in all directions through symmetry. Twin fractions by grain length (μm), and texture (crystalline) are based on EBSD analysis techniques which is synonymous With FE-SEM. As used throughout the specification the mechanical pull test parameters are based on test procedure IPC-TM-650 available from IPC® Association Connecting Electronics Industries using an INSTRON™ pull tester. As used throughout the specification the area under the curves ratio of diffraction peak (111) plane orientation and diffraction peak (200) plane orientation is based on XRD analysis of diffraction intensity (I) vs. diffraction angle 2θ (°) as done by Jade 2010 MDI software available from KSA Analytical Systems, Aubrey, Tex.
All numerical ranges are inclusive and combinable in any order, except where it is clear such numerical ranges are constrained to add up to 100%.
The present invention is directed to copper metal including twin fractions of 30% or greater of grain boundaries between adjacent copper grains having angles of misorientation of 55° to 65° with respect to a crystal plane direction axis of <111>. Twin fraction is defined as the ratio of the sum of the lengths of grain boundaries in μm with misorientations of 55° to 65° divided by the sum of the lengths of all grain boundaries in pm with misorientations of 0° to 1.80° with respect to a crystal plane direction of <111> observed for a given measureable sample area, such as, for example, 60 μm×3 μm.
Preferably, the twin fractions are 35% or greater of grain boundaries between adjacent copper grains having angles of misorientation of 55° to 65° with respect to a crystal plane direction of <111>; more preferably, the twin fractions are 35% to 55% of grain boundaries between adjacent copper grains having angles of misorientation of 55° to 65° with respect to a crystal plane direction of <111>; further preferably, the twin fractions are 35% to 52% (e.g. 35%, 37% or 52%) of grain boundaries between adjacent copper grains having angles of misorientation of 55° to 65° with respect to a crystal plane direction of <111>. Preferably, the grain boundaries between adjacent copper grains have angles of misorientation of 60° with respect to a crystal plane direction of <111>. Such angles of misorientation are thermodynamically stable such that the misorientations do not change over time.
The grain boundaries between adjacent copper grains having angles of misorientation of 55° to 65° with respect to a crystal plane direction of <111> are high-angle grain boundaries, wherein high-angle grain boundaries are defined as misorientations of angles greater than 10° with respect to a crystal plane direction of <111>. Low angle grain boundaries have misorientations from 2-10° with respect to a crystal plane direction of <111>. In addition to the high-angle grain boundary misorientations of 55° to 65° with respect to a crystal plane direction of <111>, the copper metal of the present invention can include twin fractions of less than 30% of grain boundaries between adjacent copper grains having angles of misorientation of less than 55° and greater than 65° with respect to a crystal plane direction of <111>. Such angles of misorientation can range from 0° to less than 55° and greater than 65° to 180° with respect to a crystal plane direction of <111>.
The copper metal of the present invention has a texture index (crystalline), also referred to as multiples of random distribution (MRD), of equal to or greater than 2, preferably, equal to or greater than 5 at (111) plane orientation, such as from 5 to 10.5 (e.g. 5.7 to 10.2). High (111) texture indicates that the present invention has more (111) planes that are available for slip to occur because the slip system in copper includes a slip plane of {111} and a slip direction of <110>({111}<110>) and therefore results in improved mechanical performance of tensile strength and elongation. A texture index of I indicates random orientation of (111) plane and less copper crystals in a sample having (111) plane orientation, thus less slip which results in inferior mechanical performance of tensile strength and elongation. A texture index greater than 1 indicates that there are more crystals in a sample having (111) plane orientation. Accordingly, a texture index of 2 means the number of crystals in a sample with (111) plane orientation is 2x more than in a sample having a texture index of 1, and a texture index of 5 means the number of crystals with (111) plane orientation in a sample is 5× the number of crystal in a sample with a texture index of 1 enabling improved slip and improved mechanical properties. Additional orientations detected at MRD greater than 2 are, for example, (001), (101), (201), (212), (311) and (511), but can have a texture index, such as less than 5, or such as from 0-5, or such as 1-4 (e.g. 1 to 3.5).
The copper metal of the present invention has an XRD area ratio of (111) plane orientation/(200) plane orientation at diffraction angle 2θ (°) which is equal to or greater than 1 on a graph of diffraction intensity (I) vs. diffraction angle 2θ (°). Preferably, the XRD area ratio of (111) plane orientation/(200) plane orientation at the diffraction angle 2θ (°) is greater than or equal to 5. More preferably, the XRD area ratio of (111) plane orientation/(200) plane orientation at the diffraction angle 2θ (0) is 5-31 (e.g. 5.3, 21 or 31) which indicates copper crystals having a high number of (111) planes.
Average grain size (spherical equivalent diameter) of the copper metal of the present invention does not substantially increase upon exposure to heat at high temperatures of 200° C. and greater, such as are found in annealing processes, thus reducing the potential of cracking of the copper metal. For example, the average grain size (spherical equivalent diameter) of all the grains in a copper sample having angles of misorientation of 55° to 65° at a crystal plane direction of <111> can be from 100 nm and greater, preferably, from 500 nm and greater, more preferably from 1-2 μm before annealing. The average diameter of copper grains having angles of misorientation of 55° to 65° at a crystal plane direction of <111> have a diameter (spherical equivalent diameter) of 100 nm or greater after thermal annealing, preferably 500 nm or greater. More preferably, the average diameter of copper grains having angles of misorientation of 55° to 65° at a crystal plane direction of <111> have a diameter (spherical equivalent diameter) of 0.1 μm to 3μm (e.g. 1 μm to 2.5 μm, or such as from 1.4 μm to 2.3 μm) after thermal annealing; even more preferably, the average diameter of copper grains having angles of misorientation of 55° to 65° at a crystal plane direction of <111> have a diameter (spherical equivalent diameter) of 1 μm to 2.5 μm, most preferably, from 1.5-2.3 μm after thermal annealing. Small grain size diameter (spherical equivalent diameter) of the copper metal of the present invention can strengthen the materials such that the tensile strength is improved as evidenced by the Hall-Petch relationship: σy (yield stress)=σ0+k1D−1/2
where σy is the material strength at yield in MPa.
The copper metal of the present invention is electroplated from aqueous acid copper electroplating compositions (baths) of the present invention. The aqueous acid copper electroplating compositions (baths) of the present invention contain (preferably consisting of) a source of copper ions and counter anions; an electrolyte; a leveler including (preferably consisting of) a reaction product of one or more imidazole compounds, or one or more 2-aminopyridine compounds with one or more bisepoxides; an accelerator; a suppressor;
optionally, but preferably, a source of halide ions; and water.
Preferably, the imidazole compounds have the following general formula:
where R1, R2 and R3 are independently chosen from a hydrogen atom, linear or branched (C1-C10) alkyl, hydroxyl, linear or branched alkoxy, linear or branched hydroxy(C1-C10)alkyl, linear or branched alkoxy(C1-C10)alkyl, linear or branched, carboxy(C1-C10)alkyl, linear or branched amino(C1-C10)alkyl, or substituted or unsubstituted phenyl where the substituents are chosen from hydroxyl, hydroxy(C1-C3)alkyl, or (C1-C3)alkyl. Preferably, R1, R2 and R3 are independently chosen from a hydrogen atom; linear or branched (C1-C5)alkyl, hydroxyl, linear or branched hydroxy(C1-C5)alkyl, or linear or branched amino(C1-C5)alkyl. More preferably R1, R2 and R3 are independently chosen from a hydrogen atom or (C1-C3)alky, such as methyl, ethyl or propyl moieties. Examples of such compounds are 1H-imidazole, 2,5-dimethyl-1H-imidazole and 4-phenylimidazole.
2-aminopyridine compounds of the present invention are pyridine compounds where carbon-2 of the pyridine ring is substituted with an amino group or an aminoalkyl group.
Preferably, 2-aminopyridine compounds of the present invention have a formula:
wherein R8 is —H or linear of branched (C1-C4)alkyl, R9 is —H, linear or branched (C1-C4)alkyl, halide, linear or branched amino(C1-C4)alkyl, or phenyl, and p is an integer of 0-4, wherein when p=0, the nitrogen of NHRs forms a covalent bond with carbon-2 of the pyridine ring. Preferably, R8 is —H or (C1-C2)alkyl, R9 is —H of C1-C2)alkyl, amino(C1-C2)alkyl, chloride and p is an integer of 1-2. More preferably, R8 is —H or methyl, R9 is —H or methyl and p is an integer of 1-2. Most preferably, R8 is —H, R9 is —H and p=2. Exemplary compounds of formula (II) are 2-amino-4-methyl pyridine, 2-amino-5-methyl pyridine, 2-amino-5-chloropyridine, 2-aminopyridine, 2-(2-aminoethyl)pyridine and 4-(2-aminoethyl)pyridine, wherein 2-(2-aminoethyl)pyridine is preferred.
Preferably, bisepoxides have a formula:
where R4 and R5 are independently chosen from hydrogen or (C1-C4)alkyl; R6 and R7 can be the same of different and are independently chosen from hydrogen, methyl or hydroxyl; m=1-6 and n=1-20. Preferably, R4 and R5 are hydrogen. Preferably R6 and R7 are independently chosen from hydrogen, methyl or hydroxyl. More preferably R6 is hydrogen, and R7 is hydrogen or hydroxyl. Preferably m=2-4 and n=1-2. More preferably m=3-4 and n=1.
Compounds of formula (II) include, but are not limited to, 1,4-butanediol diglycidyl ether, ethylene glycol diglycidyl ether, di(ethylene glycol) diglycidyl ether, glycerol diglycidyl ether, neopentyl glycol diglycidyl ether, propylene glycol diglycidyl ether, di(propylene glycol) diglycidyl ether, poly(ethylene glycol) diglycidyl ether compounds and polypropylene glycol) diglycidyl ether compounds.
The reaction products (levelers) of the present invention can be prepared by various processes known in the art. Typically, one or more imidazole compounds, or one or more 2-aminopyridine compounds are dissolved in DI water at room temperature followed by dropwise addition of one or more bisepoxide compounds. The temperature of the bath is then increased from room temperature to around 100 ° C. Heating with stirring is done for 2-5 hours. The temperature of the heating bath is then reduced to room temperature with stirring for an additional 8-12 hours. The amounts for each component may vary but, in general, sufficient amount of each reactant is added to provide a product where the molar ratio of the moiety from the imidazole compound, or the 2-aminopyridine compound to the moiety from the bisepoxide ranges from 1:1 to 100:70. The reaction products or copolymers of the present invention are positively charged (cationic) in the acid copper electroplating compositions of the present invention.
In general, the reaction products have a number average molecular weight (Mn) of 200 to 100,000, preferably from 300 to 50,000, more preferably from 500 to 30,000, although reaction products having other Mn values can be used. Such reaction products can have a weight average molecular weight (Mw) value in the range of 1000 to 50.000, preferably from 5000 to 30,000, although other Mw values can be used.
The amount of the reaction product included in the copper electroplating baths for plating copper metal of the present invention can range from 2 ppm to 15 ppm, preferably, from 2 ppm to 10 ppm, more preferably, from 2 ppm to 5 ppm, most preferably from 3 ppm to 4 ppm based on the total weight of the plating bath.
Copper ion sources are copper salts (preferably water soluble) and include without limitation: copper sulfate, such as copper sulfate pentahydrate; copper halides such as copper chloride; copper acetate; copper nitrate; copper tetrafluoroborate; copper alkylsulfonates; copper aryl sulfonates; copper sulfamate: copper perchlorate and copper gluconate. Exemplary copper alkane sulfonates include copper (CI-C6)alkane sulfonate and more preferably copper (C1-C3) akane sulfonate. Preferred copper alkane sulfonates are copper methanesulfonate, copper ethanesulfonate and copper propanesulfonate. Exemplary copper arylsulfonates include, without limitation, copper benzenesulfonate and copper p-toluenesulfonate. Mixtures of copper ion sources may be used. Preferably, the copper salt is present in an amount sufficient to provide an amount of copper ions of 30 to 60 g/L of plating solution. More preferably, the amount of copper ions is from 35 to 50 g/L; most preferably, the amount of copper ions is from 35 to 45 g/L.
The electrolyte of the present invention is acidic. Preferably, the pH of the electrolyte is less than or equal to 2; more preferably, the pH is less than or equal to 1. Acidic electrolytes include, but are not limited to, sulfuric acid, acetic acid, fluoroboric acid, alkanesulfonic acids such as methanesulfonic acid, ethanesulfonic acid, propanesulfonic acid and trifluoromethane sulfonic acid, aryl sulfonic acids such as benzenesulfonic acid, p-toluenesulfonic acid, sulfamic acid, hydrochloric acid, hydrobromic acid, perchloric acid, nitric acid, chromic acid and phosphoric acid. Mixtures of acids can be used in the copper electroplating compositions of the present. Preferred acids include sulfuric acid, methanesulfonic acid, ethanesulfonic acid, propanesulfonic acid, hydrochloric acid and mixtures thereof. Sulfuric acid is a most preferred acid. The acids can be present in amounts of 1 to 400 g/L; preferably from 10 g/L to 300 g/L: more preferably from 25 g/L to 250 g/L; most preferably from 30 g/L to 100 g/L. When sulfuric acid is included in the copper electroplating composition, a preferred concentration range is 40 g/L to 80 g/L, most preferably, from 40 g/L to 60 g/L. Electrolytes are generally commercially available from a variety of sources and can be used without further purification.
Such electrolytes can, optionally, but preferably, contain a source of halide ions. Preferably, chloride ions and bromide ions are used. Exemplary chloride ion sources include copper chloride, sodium chloride, potassium chloride and hydrochloric acid. Examples of sources of bromide ions are bromine chloride and bromine water. A wide range of halide ion concentrations may be used in the present invention. Preferably, the halide ion concentration is in a range of 0.5 ppm to 100 ppm based on the plating bath. More preferably, halide ions are included in amounts of 50 ppm to 80 ppm, most preferably from 65 ppm to 75 ppm. Such halide ion sources are generally commercially available and may be used without further purification.
The aqueous acid copper electroplating baths contain an accelerator. Accelerators (also referred to as brightening agents) include, but are not limited to, N,N-dimethyl-dithiocarbamic acid-(3-sulfopropyl) ester; 3-mercapto-propylsulfonic acid-(3-sulfopropyl)ester; 3-mercapto-propylsulfonic acid sodium salt; carbonic acid,dithio-O-ethylester-S-ester with 3-mercapto-1-propane sulfonic acid potassium salt; bis-sulfopropyl disulfide; bis-(sodium sulfopropyl)-disulfide; 3-(benzothiazolyl-S-thio)propyl sulfonic acid sodium salt; pyridinium propyl sulfobetaine; 1-sodium-3-mercaptopropane-1-sulfonate; N,N-dimethyl-dithiocarbamic acid-(3-sulfoethyl)ester; 3-mercapto-ethyl propylsulfonic acid-(3-sulfoethyl)ester; 3-mercapto-acid sodium salt; carbonic acid-dithio-O-ethylester-S-ester with 3-mercapto-1-ethane sulfonic acid potassium salt; bis-sulfoethyl disulfide; 3-(benzothiazolyl-S-thio)ethyl sulfonic acid sodium salt; pyridinium ethyl sulfobetaine; and 1-sodium-3-mercaptoethane-1-sulfonate. A preferred accelerator of the present invention is N,N-dimethyl-dithiocarbamic acid-(3-sulfopropyl) ester. Preferably, accelerators are included in amounts of 0.1 ppm to 1000 ppm. More preferably, the accelerators are included in amounts of 10 ppm to 50 ppm, most preferably, from 40 ppm to 50 ppm.
Suppressors include, but are not limited to, polypropylene glycol copolymers and polyethylene glycol copolymers, including ethylene oxide-propylene oxide (“EO/PO”) copolymers and butyl alcohol-ethylene oxide-propylene oxide copolymers. The weight average molecular weight of the suppressors can range from 800-15000, preferably 900-12,000. Preferably, suppressors are present at a range of 0.5 g/L to 15 g/L based on the weight of the composition; more preferably, from 1 g/L to 5 g/L.
The electroplating baths can be prepared by combining the components in any order. It is preferred that the inorganic components such as source of copper ions, water, electrolyte and optional halide ion source are first added to the bath vessel, followed by the organic components such as reaction product (leveler), accelerator, suppressor, and any other optional organic components.
The aqueous copper electroplating baths of the present invention can, optionally, contain a conventional leveling agent provided such the leveling agent does not substantially compromise the structure and function of the copper features. Such leveling agents may include those disclosed in U.S. Pat. No. 6,610,192 to Step et al., 7,128,822 to Wang et al., U.S. Pat. No. 7,374,652 to Hayashi et al. and U.S. Pat. No. 6,800,188 to Hagiwara et al. However, it is preferred that such leveling agents are excluded from the baths.
Electroplating is preferably done from 15-65 ° C.; more preferably, electroplating is from room temperature to 50 ° C.; even more preferably, from room temperature to 40 ° C.; and, most preferably, from room temperature to 30 ° C., wherein room temperature is optimum.
Preferably, the copper electroplating baths of the present invention are agitated during plating. Agitation method include, but are not limited to: air sparging, work piece agitation, and impingement. Preferably agitation is done from 10 cm/second to 25 cm/second, more preferably from 15 cm/second to 20 cm/second.
A substrate is electroplated by contacting the substrate with the plating bath by immersing the substrate in the bath or by spraying the substrate with the bath. The substrate functions as a cathode. The plating bath contains an anode, which may be a soluble anode or an insoluble anode. Potential is applied to the electrodes. Current densities preferably range from 2 ASD to 8 ASD; more preferably, from 4 ASD to 8 ASD; and most preferably, from 5 ASD to 7 ASD (e.g., such as 5 ASD to 6 ASD, or 5 ASD to 7ASD, or 6 ASD to 7ASD).
After the substrate is electroplated with copper from the aqueous based acid copper electroplating composition of the present invention, the copper along with the substrate is annealed to complete the method of preparing the copper metal of the present invention. Preferably, the annealing is done at 200° C. or greater; more preferably. from 200° C. to 260 ° C.; most preferably from 230° C. to 250° C. Preferably, annealing is done from 2 hours to 10 hours; more preferably, from 5 hours to 8 hours; most preferably, from 5.5 hours to 6.5 hours. Preferably, annealing is done in an inert atmosphere, such as a gaseous N2 atmosphere. The annealing processes does not substantially increase copper grain size.
In addition to the properties described above, the copper metal of the present invention has good mechanical properties of tensile strength (at break) and elongation% (at break). Preferably, the tensile strength at break of the copper metal of the present invention is equal to or greater than 330 MPa; more preferably, from 330 MPa to 360 MPa. Elongation% at break is greater than or equal to 20% (e.g. 20% to 25%); preferably, from 21% to 23%.
While the copper metal of the present invention and the method of electroplating copper metal of the present invention can be used for copper metallizing various substrates, preferably, the copper metal of the present invention is electroplated by the method of the present invention in the formation of fine line copper RDLs used as a way to reroute conduction paths within chip packages, such as in chip packages where fine line RDL has an LIS less than or equal to 10 μm×10 μm: preferably, 5 μm×5 μm; more preferably, less than or equal to 2 μm×2 μm; most preferably, less than or equal to 1 μm×1 μm.
In addition to copper plating RDLs, the copper electroplating method of the present invention can be used to electroplate copper metal of the present invention on dielectric substrates and semiconductors with metal seed layers, such as copper seed layers. Dielectric materials include, but are not limited to, thermoplastic resins and thermosetting resins. A particularly preferred dielectric material is polyimide. Semiconductor materials include, but are not limited to, silicon.
The copper metal electroplating method can be used to non-conformally electroplate copper metal of the present invention on surfaces of the substrates as well as in apertures, such as vias. Preferably, the apertures, including vias, have high aspect ratios of 2:1 or greater; more preferably from 4:1 or greater; even more preferably from 6:1 or greater, such as 10:1 to 20:1.
Apertures, such as vias, preferably, have diameters of 0.5 μm to 200 μm, more preferably, from 1 μm to 50 μm. The depth of the apertures can range, preferably, from 0.5 μm to 500 μm; more preferably, from 1 μm to 100 μm.
The following examples are included to further illustrate the invention but are not intended to limit its scope.
Glycerol diglycidyl ether (60 mmols) and 1H-imidazole (100 mmols) were added at room temperature to a round-bottom reaction flask set in a heating bath. Then 40 mL of DI water were added to the flask. The temperature of the heating bath was set to 98° C. The reaction mixture was heated for 5 hours and left stirring at room temperature for another 8 hours. The reaction product (reaction product 1) was used without purification. The molar ratio of the moieties from the 1H-imidazole to the molar ratio of the ether moieties was 100:63.
Glycerol diglycidyl ether (30 mmols) and an imidazole compound mixture (30 mmols) of 1H-imidazole (25% by mole)+4-phenylimidazole (75% by mole) were added at room temperature to a round-bottom reaction flask set in a heating bath. Then 40 mL of DI water were added to the flask. The temperature of the heating bath was set to 98° C. The reaction mixture was heated for 5 hours and left stirring at room temperature for another 8 hours. The reaction product (reaction product 2) was used without purification. The molar ratio of the moieties from the imidazole mixture to the molar ratio of the ether moieties was 1:1.
2-(2-aminoethyl)pyridine (100 mmols) was added at room temperature to a round-bottom reaction flask set in a heating bath. Then 40 mL of DI water were added to the flask. The temperature of the heating bath was heated to a jacket temperature set at 90° C. Once the bath reached an internal temperature of 76-78° C., glycerol diglycidyl ether (100 mmols) was slowly fed into the round-bottom reaction flask to moderate any exotherm. The reaction mixture was heated with the jacket temperature set at 90° C. for 4 hours with stirring. The reaction mixture was then cooled down to 50-55° C. and a sulfuric acid solution was added to dilute the mixture to 40 wt %. The final reaction product (reaction product 3) was cooled to 25° C. then gravity drained. Reaction product 3 was used without purification. The molar ratio of the moieties from the imidazole mixture to the molar ratio of the ether moieties was 1:1.
In a 250 mL round-bottom, three-neck flask equipped with a condenser and a thermometer, 100 mmol of 1H-imidazole and 12 mL DI water were added followed by addition of 200 mmol of epichlorohydrine. The resulting mixture was heated for 5 hours using an oil bath set to 95 ° C. and then left to stir at room temperature for an additional 8 hours. The reaction product was transferred to a 200 mL volumetric flask, rinsed and adjusted with DI water to the 200 mL mark. The reaction product (comparative reaction product) solution was used without further purification.
The following aqueous copper electroplating baths were prepared at room temperature by mixing and stirring the components of the bath in water.
The following aqueous copper electroplating bath was prepared at room temperature by mixing and stirring the components of the bath in water.
A copper blank silicon wafer (size=4 cm×4 cm) with a copper seed layer of 1500 Å thick was placed into a plating cell which included the aqueous copper electroplating composition of Bath 1 from Example 5. The pH of the bath during plating was less than 1 and the plating composition was paddle agitated with a linear speed of 20 cm/second during plating. A soluble copper electrode served as an anode. DC plating was done at room temperature using a current density of 6 ASD. Copper electroplating was done until a copper deposit having a thickness of 20 μm was plated on the wafer. The copper deposit was annealed in an inert N2 atmosphere filled oven at 230° C. for 6 hours. After annealing, the copper metal plated wafer was cooled to room temperature.
A copper blank silicon wafer (size=4 cm×4 cm) with a copper seed layer of 1500 Å thick was placed into a plating cell which included the aqueous copper electroplating composition of the comparative bath from Example 6. The pH of the bath during plating was less than 1 and the plating composition was paddle agitated with a linear speed of 20 cm/second during plating. A soluble copper electrode served as an anode. DC plating was done at room temperature using a current density of 6 ASD. Copper electroplating was done until a copper deposit having a thickness of 20 μm was plated on the wafer. The copper deposit was annealed in an inert N2 atmosphere filled oven at 230° C. for 6 hours. After annealing, the copper metal plated wafer was cooled to room temperature.
EBSD was used to quantitatively measure properties of the copper deposits from Examples 7 and 8. The EBSD revealed grain size, grain orientation, texture, and grain boundary angles.
4 mm×8 mm pieces of the plated copper wafer (300 mm polycrystalline silicon, P/boron, <100>, 0-100 ohm-cm from Pure Wafer, 2240 Ringwood Ave. San Jose Calif. 951311) was cut and mounted on a sample holder. Argon milling cross section polisher Model JEOL IB09010CP from JEOL USA, Inc. was used to polish the surface of each piece, and the surfaces were analyzed. FE-SEM (FEI model Helios G3) coupled with EBSD detector (EDAX Inc., model Hikari Super and data was analyzed by OIM™ Analysis software) was used to collect diffracted signals from the samples. For grain size analysis. step size was 0.025 μm (measured at every 0.025 μm intervals) with 10 scans at different random sample locations were collected to obtain statistically significant data. For texture analysis, the step size is 0.075 μm (measured at every 0.075 μM intervals) with 5 different location scans (statistically significant).
EBSD was used to determine misorientations between two adjacent grains as shown in
Copper wafers from Examples 7 and 8 were mechanically broken to approximately 1 cm×2 cm pieces. Each piece was then mounted with copper facing up on a plastic sample holder using double-sided tape. A Bruker D8 Advance θ-θ X-ray diffractometer (XRD) equipped with a copper sealed-source tube and Vantec-1 linear position sensitive detector was used to collect diffraction patterns (Bruker AXS Inc. 5465 East Cheryl Parkway, Madison Wis. 53711). The tube was operated at 35 kV and 45 mA and the samples were illuminated with copper Kα radiation (1=1.541 Å). XRD data were collected with a 3° detector window from 15° to 84°2θ, with a step size of 0.0256° and 1 s/step collection time. Analysis was performed with Jade 2010 MDI software application available from KSA Analytical Systems, Aubrey, Tex.
Mechanical Property Tests were performed using INSTRON™ Pull Tester 33RR64. Test specimens were first plated on a stainless steel substrates (dimensions 12 cm×12 cm) using the formulations in Examples 7 and 8 under the same plating conditions and plated at a current density of 6ASD. The plated copper was then peeled off the stainless steel plates and cut to strips with dimension of 1.3 cm×10 cm. The thickness of the standalone copper films was 50 μm. The test procedure (IPC-TM-650) was followed in this test using an INSTRON™ pull tester 33R4465. The copper strips were annealed in a furnace (Blue M industrial laboratory oven Model 01440A) at 230° C. for 6 hours. After allowing the samples to cool to room temperature, the samples were tested in the pull tester. The pull rate applied was 0.002 inches/minute until the samples broke. The data was recorded with Bluehill-3 software, available from INSTRON®. Table 3 shows the results of the elongation test. The mechanical pull test showed that the sample of the present invention had improved tensile strength vs. the comparative sample while not sacrificing significant elongation performance.
Table 3 shows the comparisons of the copper deposits of the invention vs. the comparative or conventional bath. The grain size for the copper deposits of the present invention were approximately 43% less than the copper from the comparative example after thermal annealing.
The EBSD was also used to determine the twin fraction and the texture index for the copper deposits of the present invention and the comparative copper. The copper deposits of the present invention showed twin fractions of 35% of grain boundaries with 60°±5° misorientations between adjacent grains at a crystal plane direction of <111>. The texture index for the copper metal of the present invention was 5.7 at (111) plane orientation. High (111) plane orientation is preferred as the slip system for copper is {111}<110>. High (111) plane fraction can facilitate the slip to occur easily which results in better mechanical performances. On the other hand, in the comparative copper metal, (001) plane orientation texture dominated with a texture index ratio of 5.1, which is not favorable for the system to slip. The (111) plane and the (001) plane are the two most significant planes in terms or MRD>2 in copper. Therefore, the two planes were compared where the higher number of (111) planes vs. (001) planes was preferred for the copper metal of the invention.
XRD also revealed that the copper metal of the present invention had significantly higher (111)/(200) ratio than that of the comparative copper metal. Samples for the XRD test were about 5 μm thick copper film plated by the copper baths of Examples 7 and 8 on silicon wafers. The ratio was determined using diffraction peak at (200) plane orientation because the (200) plane orientation was the second strongest diffraction peak after (111). Other diffraction peaks were too weak or were unable to be detected. Diffraction intensity (I) vs. diffraction angle 2θ (°) was recorded and plotted for each sample as shown in
Copper metal was plated from Bath 2 on the same type of substrate as disclosed in Example 7 above. The plating conditions were substantially the same as disclosed in Example 7. Copper plated from the copper electroplating composition of Bath 2 disclosed in Table 2 in Example 5 was analyzed for its properties according to the methods described in Example 9 above. The results of the EBSD, XRD and mechanical pull test results are disclosed in Table 4 below.
Copper metal plated from Bath 3 was plated on the same type of substrate as disclosed in Example 7. Plating conditions were substantially the same as in Example 7. Copper plated from the composition of Bath 3 disclosed in Table 2 in Example 5 was analyzed for its properties according to the methods described in Example 9 above. The results of the EBSD, XRD and mechanical pull test results are disclosed in Table 5 below.
| Number | Date | Country | |
|---|---|---|---|
| 62583251 | Nov 2017 | US |
