|Publication number||US20060175202 A1|
|Application number||US 11/291,679|
|Publication date||Aug 10, 2006|
|Filing date||Nov 30, 2005|
|Priority date||Nov 30, 2004|
|Also published as||CN101065520A, EP1817443A2, WO2006060520A2, WO2006060520A3|
|Publication number||11291679, 291679, US 2006/0175202 A1, US 2006/175202 A1, US 20060175202 A1, US 20060175202A1, US 2006175202 A1, US 2006175202A1, US-A1-20060175202, US-A1-2006175202, US2006/0175202A1, US2006/175202A1, US20060175202 A1, US20060175202A1, US2006175202 A1, US2006175202A1|
|Original Assignee||Stephen Mazur|
|Export Citation||BiBTeX, EndNote, RefMan|
|Referenced by (5), Classifications (21), Legal Events (1)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This invention relates to processes and apparati for selectively electroplating a metal layer or layers into recessed topographic features on a conductive surface. The processes and apparati of the invention are useful for fabricating metal circuit patterns, for example for creating copper interconnects between integrated circuit elements embedded in a thin layer of dielectric material on the surface of a semiconductor wafer.
In the damascene process for fabricating integrated circuits, the electrical interconnections are created as patterns of lines and holes etched through a dielectric layer on the surface of the wafer. Such patterns are then filled with metallic copper, and electroplating is commonly used. An ideal deposition process would completely fill the recesses in the dielectric layer with copper to a level that is flush with the surrounding plateau surfaces and not deposit any copper on the plateau surfaces.
Although conventional electroplating technology can provide control over thickness and uniformity of the plated layer, no practical method has been disclosed that selectively deposits a metal layer into the holes and trenches or the recessed areas in the dielectric layer and simultaneously precludes depositing a metallic layer of comparable thickness on top of the plateaus separating the circuit features. More often, under conventional electroplating conditions, a thick layer of copper is deposited on the plateaus and must be removed by a highly specialized polishing process to planarize the surface, and smooth it to within an extremely fine tolerance, and simultaneously avoid loss of, or damage to, the circuit features.
One aspect of the present invention is a process of electroplating metal onto a conductive surface, wherein the conductive surface comprises plateaus and trenches, the method comprising:
Another aspect of the present invention is an apparatus for electroplating metal onto a conductive surface, the conductive surface comprising plateaus and trenches, the apparatus comprising:
These and other aspects of the present invention will be apparent to those skilled in the art in view of the present disclosure and the appended claims.
The description of the invention is aided by use of the following figures, which are not intended to be drawn to scale:
One embodiment of this invention is an apparatus for electroplating metal onto a conductive surface, the conductive surface comprising plateaus and trenches, the apparatus comprising:
Unless otherwise stated, the following terms when used herein have the meanings set forth below.
By “trenches” is meant recessed features on the substrate or conductive surface. In the present specification, “trenches,” “recessed features,” “holes,” “recessed trenches,” “topographic recesses,” and “vias,” may be used alternatively, in conjunction, selectively, or interchangeably. Unless otherwise specified, usage of any of these terms includes every type of recessed feature that is not a plateau and the meaning is construed to comprehensively include all types of features.
By “plateau,” is meant the generally flat area of the substrate or conductive surface that is at the level of the top of the trenches and/or vias.
Unless otherwise specified, a “conductive” fluid or solution has conductivity greater than about 5 mS/cm, preferably equal to or greater than about 30 mS/cm, more preferably equal to or greater than about 100 mS/cm.
A “conductive surface” has a sheet resistance no greater than about 10 milli-ohms per square.
A “low-conductivity” fluid or solution has conductivity below about 1000 μS/cm. “Non-platable metal ions” are known to those skilled in the art and include, for example, Na and K.
The electroplating solution may contain, in addition to the platable metal ions, other electrolytes, surfactants, and/or other additives well known in the art and variously designated as “brighteners”, “levelers”, or “accelerators”.
Any apparatus or device suitable for supplying electroplating solution to an area between a membrane and a substrate is useful, including, for example, baths and sprayers. The electroplating solution can be supplied at any pressure, and the supply can be intermittent or continuous. The electroplating solution can be of one composition, or can change composition during the plating process.
To drive the electroplating process, a source of DC electrical power is connected between the conductive surface (which functions as the cathode) and the anode. The source of DC power can be steady or can advantageously provide pulses and/or variable DC power. A typical damascene wafer comprises a barrier layer that may not provide sufficient electrical connection to a power source and thus cannot by itself serve as a cathode. For this reason a seed layer of metal, for example copper, covers the barrier layer to provide the electrical connection to the power source.
The anode is an electrically conductive material such as a metal or alloy (e.g., stainless steel, platinum, palladium or the dimensionally stable anodes commonly used in the chlor-alkali process) or carbon. Electrical contact between the anode and the first surface of the membrane can beneficially be through an anolyte contacting the anode and the first surface of the membrane, wherein preferably the anolyte is a conductive solution, fluid, or composition.
The anolyte also acts as a source and/or sink for ions passing through the membrane. The anolyte solution may comprise water, a polar organic solvent, or a combination of such solvents and beneficially also includes solutes such as acids, bases or salts. Higher conductivity anolyte compositions, for example having a conductivity of at least 20 mS/cm, are generally preferred, as they can reduce voltage loss of current passing from the anode and through the anolyte. The anolyte can contain one or more non-plating metal ions, e.g., Na, K, or such, For use with anion-conducting membranes, the anolyte should not contain any readily reducible negatively charged anions. For use with cation-conducting membranes, the anolyte should not contain any readily reducible positively charged cations.
The membrane-limited selective electroplating process can be used for deposition of a wide variety of platable metals and metal alloys. Suitable metals include silver, nickel, cobalt, tin, aluminum, copper, lead, tantalum, titanium, iron, chromium, vanadium, manganese, zinc, zirconium, niobium, molybdenum, ruthenium, rhodium, hafnium, tungsten, rhenium, osmium, iridium, and combinations thereof. Preferred metals include silver, nickel, cobalt, tin, aluminum, copper, lead and combinations thereof. The method is particularly suitable for electroplating copper and/or copper-containing alloys on damascene wafers.
Membrane-limited selective electroplating may employ conventional electroplating solutions comprising salts of metal ions or complex metal ions and other ingredients, for example acids or bases, buffers, surfactants and/or other additives known in the electroplating art. Any or all adjuvants known for use in electroplating solutions can be used in the processes herein. The platable metal in the electroplating solution either has a positive charge or a negative charge. Common commercial aqueous electroplating solutions fall into two general categories, depending upon whether the dissolved metal ions are positively charged cations or negatively charged anions. Membrane-limited selective electroplating may use either type of plating solution depending on the type of membrane that is employed. It is desirable that substantially all (meaning at least 80%) of the platable metal ions in electroplating solutions for use with an anion-conducting membrane are present in the form of positively charged cations. On the other hand, substantially all of platable metal ions in electroplating solutions for use with a cation-conducting membrane are desirably in the form of negatively charged anions.
The electroplating solution for copper plating is commonly either an acid solution containing, for example CuSO4 in aqueous H2SO4, or a solution containing basic cyanide or other nitrogen-containing-ligand, for example CuCN and NaCN in aqueous NaOH or Na2CO3. In the former example, the platable copper species is the hydrated cupric cation Cu(H2O)n +2, whereas in the latter example the platable copper species is the complex cyanocuprate anion Cu(CN)3 −2.
The ion-conducting membrane serves two functions. The first function of the ion-conducting membrane is to displace plating solution from the plateau areas of the conductive surface while trapping plating solution within the recessed areas. The second function of the membrane is to serve as a gate that allows certain ions to carry electrical current through the membrane, but specifically prevents electrochemically active metal ions from contacting or plating onto the plateau areas.
Suitable charge-selective ion-conducting membranes include film-forming ionic polymers that are stable under the conditions of the electroplating process. Ionic polymer membranes useful in electro-coating, electro-dialysis, the chloralkali process and fuel-cells may also be useful in the electroplating process herein. The ion-conducting membrane can be of any thickness, but advantageously the membrane thickness is greater than the width of trenches to be filled with electroplated metal. In practice, the membrane thickness is typically at least 2 times the largest width of trenches to be filled with electroplated metal. Exemplary thickness of the membrane is in the range of from about 40 microns to about 500 microns, or alternatively about 3 to about 120 mils. The reason for having appreciable thickness is that the thickness will resist bending and, coupled with the stiffness of the membrane, should be sufficient so that the membrane does not conform to the topography of the conductive surface. The distribution of charged moieties in the pores need not be uniform, and the membrane can comprise one or more separate membranes laminated one to another.
Similarly, the membrane is advantageously sufficiently stiff and incompressible so that the active portion of the membrane does not conform to the topography of the conductive surface. The stiffness and compressibility of the membrane can vary with degree of saturation as well as the ion content in the membrane, but generally, most commercially available NafionŽ and FlemionŽ cation-exchange membranes, or Fumatech FAP and PCA60 anion-exchange membranes have the requisite stiffness and incompressibility when the saturation level is near 100%.
Suitable ion-conducting membranes are substantially impervious (impermeable) to the platable metal ions in the electroplating solution. By substantially impermeable to the metal ions in the electroplating solution we mean firstly, that for cation-exchange membranes, the transference number for cations is at least 0.9, and secondly, that at least 80% of the platable metal ions are anions. Similarly for anion-exchange membranes, the transference number for anions is at least 0.9, and at least 80% of the platable metal ions are cations. Under conditions where electroplating requires transfer of a cation, a cation-conducting membrane is used. An exemplary cation-conducting membrane comprises a polymeric ionomer functionalized with at least one type of acidic moiety. Cation-selective ion-conducting membranes (also called cation-exchange membranes), generally comprise organic polymer films with acidic functional groups (e.g., —CO2H or —SO3H), bound covalently to the polymeric backbone. In one embodiment, the cation-conducting membranes are formed from polymeric ionomers functionalized with strong acid groups that have a pKa of less than about 3. Sulfonic acid groups are preferred strong acid groups. Preferred polymeric ionomers are copolymers of fluorinated and/or perfluorinated olefins and monomers containing strong acid groups. An exemplary cation-selective membrane may have 1.0 to 4.0 milli-equivalents of strong acid groups per cubic centimeter of membrane. Suitable membranes include polytetrafluorethylene polymer-based membranes, perfluorocarboxylic acid/PTFE copolymers, polymeric ionomers functionalized with both sulfonic acid groups and carboxylic acid groups, and perfluorosulfonic acid/ polytetrafluorethylene copolymer membranes. Other acid moieties can be attached to the membrane as an alternative to, or in addition to, the carboxylic acid moieties and/or the sulfonic acid moieties, including for example a sulfanilamide moiety, a phosphonate moiety, a sulfonyl moiety, or any combination thereof, wherein the acidic moieties can independently be substituted with, for example, a C1 to C4 alkyl group.
Commercially available cation-conducting membranes useful in the processes of this invention include FlemionŽ perfluorocarboxylate ionomer membranes (Asahi Glass Co., Ltd, Yokahama, Japan) and/or NafionŽ perfluorosulfonate ionomer membranes (E.l. du Pont de Nemours, Inc., Wilmington, Del.), which are composed of fluorocarbon chains bearing highly acidic carboxylic and sulfonic acid groups, respectively. On exposure to water, the acid groups of NafionŽ ionize, leaving fixed sulfonate anions and mobile hydrated protons. The protons may be readily exchanged with various metal cations. NafionŽ is particularly well-suited for use in membrane-limited selective electroplating due to its strong common-ion exclusion, high conductivity, strong acidity, chemical stability and robust mechanical properties.
In one embodiment the membrane is layered, and comprises a fluoropolymer membrane comprising at least two integrally laminated layers including a first layer made of a perfluorocarbon polymer having carboxylic acid groups as its ion exchange groups, and a second layer comprising perfluorocarbon polymer having sulfonic acid groups as its ion exchange groups. Alternatively, the layers can be separated by a fluid layer. A suitable membrane can be a single layer having both sulfonic and carboxylic groups made, for example, by the copolymerization of a carboxylic acid type monomer with a sulfonic acid type monomer, or by the copolymerization of a carboxylic acid type monomer with a sulfonic acid type monomer, or by impregnating a sulfonic acid type fluoropolymer membrane with a carboxylic acid type monomer, followed by polymerization. Suitable membranes include those formed from a blend comprising a sulfonic acid group-containing polymer and a carboxylic acid group-containing polymer, which is laminated on a sulfonic acid group membrane, as described in U.S. Pat. No. 4,176,215, and herein incorporated by reference.
Under conditions where electroplating requires transfer of an anion, an anion-conducting membrane is used. An anion-conducting membrane comprises a polymeric ionomer functionalized with at least one type of basic moiety, for example quaternary ammonium groups. Tertiary or lower amino groups are also suitable functional groups. Anion-selective ion-conducting membranes (also called anion-exchange membranes) generally comprise organic polymer films with positively charged covalently bound functional groups such as ammonium ions —NH3 +, NH2R+, —NHR2 +, or —NR3 +, or basic salts such as —NRH2OH, NR2HOH, or NR3OH, where R is an organic radical. When saturated with water, these functional groups hydrate and dissociate. The resulting cations —NH3 +, —NRH2 +, —NR2H+, and NR3 + remain confined within the membrane while the hydroxide ions —OH, are free to diffuse, migrate and exchange with other anions in adjacent solutions. An exemplary anion-conducting membrane may have 5 to 200 microequivalents of basic moieties per cm2 of membrane area. Examples of anion-selective ion-conducting membranes (anion conducting membranes) include the PC amine-functionalized epoxide polymers (PCA—Polymerchemie Altmeier GmbH, Heusweiler, Germany). Strongly basic styrenic anion-conductive membranes can be formed from a cross-linked poly-styrene-divinylbenzene that is chloromethylated using a Lewis acid and further functionalized by addition of a tertiary amine. Methods for making anion-conducting membranes can be adapted from methods to make anion-exchange membranes described in U.S. Pat. No. 6,646,083, which is incorporated by reference herein.
Also provided are processes for electroplating metal onto conductive surfaces.
According to one embodiment there is provided a process for electroplating metal onto a conductive surface, wherein the conductive surface comprises plateaus and trenches, the method comprising:
Electroplating occurs when a portion of the external surface of the ion-conducting membrane is brought into sensible contact with a portion of the conductive surface that is covered by the electroplating solution, with the substrate held at a voltage more negative than the open circuit voltage. The term “sensibly contact” means there may be a thin layer of fluid disposed between the membrane and the wetted plateaus; preferably there is none. Some solution remains trapped within the recessed features. The trapped electroplating solution within the recesses serves as a source of metal ions in an electroplating step. The continuous or intermittent exchange of the plating solution within the recessed features with fresh plating solution is usually achieved by moving the membrane with respect to the conductive surface, where the exposed portions of the surface not contacting the membrane are subject to a rinse of fresh electroplating solution. A membrane moving along a surface may not displace all the electroplating solution from the plateaus, and controlling the velocity of the membrane and the pressure exerted on the membrane can influence the thickness of any layer of electroplating solution disposed between the plateaus and the membrane.
The conductive surface typically comprises a metal such as copper, but other metals may be used as long as they provides suitable electrical contact to the power source and adhesion to the plated copper.
The distance between the membrane and the plateaus desirably provides an electroplating solution layer between the plateaus and the membrane that is less than twice the depth of the trenches preferably much less than twice the depth of the trenches. The “depth of the trenches” is the difference in height between the floor of trenches and vias to be filled and the top of the surrounding plateaus. For example, if the membrane contacts and electroplates material onto a damascene wafer where the depth of the trench to be filled is about 1 micron, then the average height of the electroplating solution disposed between a membrane and a nearby plateau is less than 0.5 microns, and may preferably be less than 10 nanometers. When the membrane is being moved relative to the conductive surface, this layer may provide lubrication between the membrane and the conductive surface.
In a process of the current invention, the pressure exerted by the membrane on the conductive surface is sufficient to reduce the thickness of the layer of electroplating solution disposed between the top of plateaus and the membrane to the desired thickness. For example, the pressure can range from about 0.03 to about 30 psi.
The velocity of the membrane relative to the conductive surface may range from about 0 to about 200 cm/sec or more, but is typically from about 1 cm/sec to about 30 cm/sec.
Then the membrane is held in sensible contact with the conductive surface and a suitable voltage is applied between the anode and cathode under these conditions, metal ions of the electroplating solution are reduced to elemental metal and are deposited into the trenches of the conductive surface. Unlike conventional electroplating, a disproportionately large portion of the electrical current flows through small volumes of the electroplating solution trapped within the topographic recesses (trenches) of the surface. The entrapment of the electroplating solution is achieved by holding the conductive surface in intimate contact with a first surface of a charge-selective ion-conducting membrane. The conductive surface can be held stationary and the membrane moved, or the membrane can be held stationary and the surface moved, or both the surface and the membrane can be in motion. The relative motion may be parallel or perpendicular to the conductive surface or some combination of the two.
The rate of depletion of platable metal ions in the electroplating solution is typically not constant, and the rate of plating from a trapped volume of electroplating solution can slow over time as the concentration of platable ions in the electroplating solution is depleted. In some embodiments, the amount of time for which a given portion of electroplating solution is trapped without being refreshed is at least sufficient to lower the average concentration of platable metal ions by at least 30% in the layer of fluid disposed between the membrane and the recessed areas. As the recessed areas become filled with copper, they retain progressively less electroplating solution, so that metal ion depletion occurs more rapidly. There is no advantage in allowing the concentration to fall to less than 90% of its original value. The amount of time to electroplate before replenishing or replacing the electroplating solution disposed in a trench can in some cases beneficially be changed as an endpoint is approached.
If the membrane is moving relative to the surface, then fresh electroplating solution is supplied to the area between the membrane and the surface. Since electroplating solution is typically disposed on the conductive surface prior to the membrane passing over that surface, it is important to prevent current flowing to the plateau areas outside the area where the membrane sensibly contacts the surface. Once the membrane contacts the surface, then electroplating solution is displaced from the plateaus, and the electroplating process can beneficially proceed. In one embodiment of the invention, in order to avoid plating metal onto the plateau areas the electrical circuit is temporarily opened or the voltage is set to the open-circuit voltage during disengagement, movement, and re-engagement of the membrane.
In other embodiments, such as those represented in
During the electroplating process, there can be a substantial flux of solvent across the membrane and the surface during the electroplating process, and an accompanying substantial change in concentrations of constituents in the electroplating solution. In addition, the composition of the anolyte changes as various reagents are either consumed or generated by anodic reactions and other reagents may accumulate or be lost through the membrane. In order to maintain uniform processing conditions, it is advantageous to maintain substantially stable compositions for the anolyte and electroplating solutions. Therefore, it may be advantageous to remove and replace used solutions from the anode compartment and from the conductive surface, or to otherwise affect the composition to maintain stable concentrations in the anolyte and electroplating solution.
As metal is deposited into the recessed areas, the recessed volumes become progressively filled with metal and retain progressively less plating solution when pressed against the membrane surface. Consequently, the metal ions in the recess volumes become more rapidly depleted. Near the end of the process, when the recessed volumes are nearly filled with metal and approach the level of the plateau areas, the rate of deposition will eventually decrease to a negligible value. Accordingly, the process of the invention is self-limiting in the sense that the plating process automatically slows as the recessed volumes have been filled with metal to a level approaching or comparable with the plateau areas. The corresponding decrease in plating current may be used as a diagnostic indication of the process end-point. Near the end of the process, it may be advantageous to intentionally create a small gap of less than 1 micron between the membrane and the plateaus, which will result in plating a small quantity of copper or other metal on plateaus, to insure the trenches are completely filled and to make sure there is sufficient electroplating solution between the bottom of trenches and vias to result in an minimum rate of metal deposition. Increasing velocity, lowering the hydrostatic pressure, or a combination thereof, may be used to increase the thickness of the layer of electroplating solution disposed on plateaus as the endpoint of the polishing is neared.
It shall be understood that the examples illustrated in
The apparatus for membrane-limited electroplating is advantageously designed so that limited or no electrolytic current can flow to plateau areas of the substrate not sealed by the membrane.
Another embodiment of the invention is illustrated in
In yet another embodiment of the invention (not shown), similar to that shown in
The area of contact between the membrane 13 and the substrate 10 may be continuously moved over the surface of the substrate 10 in such way that the area under the anode 18 always remains in contact. In this manner, fresh plating solution can be continuously replaced in the recessed features 9 and plating current can be maintained without interruption until the desired amount of metal has been deposited.
In the embodiment represented in
Embodiments of this invention are not limited to a single area of contact between the membrane and the substrate. An apparatus of this invention can comprise a multiplicity of contact areas involving a single large membrane, a multiplicity of membranes and/or a multiplicity of insulating masks, and may further comprise a multiplicity of anodes and a multiplicity of anolyte solutions. Such embodiments can provide advantages for increasing the productivity of the process and/or improving the macroscopic uniformity of the process.
Although the present invention is described with reference to certain preferred embodiments, it is apparent that modification and variations thereof may be made by those skilled in the art without departing from the spirit and scope of this invention as defined by the appended claims. In particular, it will be clear to those skilled in the art that the present invention may be embodied in other specific forms, structures, arrangements, proportions, and with other elements, materials, and components, without departing from the spirit or essential characteristics thereof. One skilled in the art will appreciate that the invention may be used with many modifications of materials, methods, and components otherwise used in the practice of the invention, which are particularly adapted to specific substrates and operative requirements without departing from the principles of the present invention. The presently disclosed embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims, and not limited to the foregoing description.
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US8158532||Nov 20, 2006||Apr 17, 2012||Novellus Systems, Inc.||Topography reduction and control by selective accelerator removal|
|US8168540||Dec 29, 2009||May 1, 2012||Novellus Systems, Inc.||Methods and apparatus for depositing copper on tungsten|
|US8377824||Apr 2, 2012||Feb 19, 2013||Novellus Systems, Inc.||Methods and apparatus for depositing copper on tungsten|
|US20120298518 *||Jul 14, 2011||Nov 29, 2012||Seung Ki Joo||Electroplating apparatus and method|
|WO2008112086A2 *||Feb 29, 2008||Sep 18, 2008||Du Pont||Electroplating process|
|U.S. Classification||205/149, 204/252|
|International Classification||C25B9/00, C25D7/00|
|Cooperative Classification||C25D17/008, H05K3/423, C25D5/02, H05K2201/09563, H05K2203/0557, C25D5/34, H01L21/76877, H01L21/2885, C25D7/123, C25D17/001|
|European Classification||C25D7/12, H01L21/768C4, H01L21/288E, H05K3/42D, C25D5/34, C25D5/02, C25D17/00|
|Apr 12, 2006||AS||Assignment|
Owner name: E. I. DU PONT DE NEMOURS AND COMPANY, DELAWARE
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:MAZUR, STEPHEN;REEL/FRAME:017460/0637
Effective date: 20060309