Ex Parte Bashyam et alDownload PDFPatent Trials and Appeals BoardMar 21, 201915162867 - (D) (P.T.A.B. Mar. 21, 2019) Copy Citation UNITED STA TES p A TENT AND TRADEMARK OFFICE APPLICATION NO. FILING DATE 15/162,867 05/24/2016 23494 7590 03/25/2019 TEXAS INSTRUMENTS IN CORPORA TED PO BOX 655474, MIS 3999 DALLAS, TX 75265 FIRST NAMED INVENTOR Murlidhar Bashyam UNITED STATES DEPARTMENT OF COMMERCE United States Patent and Trademark Office Address: COMMISSIONER FOR PATENTS P.O. Box 1450 Alexandria, Virginia 22313-1450 www .uspto.gov ATTORNEY DOCKET NO. CONFIRMATION NO. TI-76303 4110 EXAMINER FOURSON III, GEORGE R ART UNIT PAPER NUMBER 2899 NOTIFICATION DATE DELIVERY MODE 03/25/2019 ELECTRONIC Please find below and/or attached an Office communication concerning this application or proceeding. The time period for reply, if any, is set in the attached communication. Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the following e-mail address(es): uspto@ti.com PTOL-90A (Rev. 04/07) UNITED STATES PATENT AND TRADEMARK OFFICE BEFORE THE PATENT TRIAL AND APPEAL BOARD Ex parte MURLIDHAR BASHY AM and RICHARD ALLEN FAUST 1 Appeal2018-006250 Application 15/162,867 Technology Center 2800 Before MARK NAGUMO, JEFFREY R. SNAY, and DEBRA L. DENNETT, Administrative Patent Judges. NAGUMO, Administrative Patent Judge. DECISION ON APPEAL Texas Instruments Incorporated ("Bashyam") timely appeals under 35 U.S.C. § 134(a) from the Final Rejection2 of all pending claims 1- 22. We have jurisdiction. 35 U.S.C. § 6(b). We affirm. 1 The applicant under 37 C.F.R. § 1.46, and hence the appellant under 35 U.S.C. § 134, is the real party in interest, identified as Texas Instruments Incorporated. (Appeal Brief, filed 30 November 2017 ("Br."), 3.) 2 Office Action mailed 6 April 2017 ("Final Rejection"; cited as "FR"). Appeal2018-006250 Application 15/162,867 A. Introduction 3 OPINION The subject matter on appeal relates to fabricating bonding pads on integrated circuits ("IC"). Aluminum is said to have been the standard metal for bonding pads for decades, but as device sizes shrink, manufacturers are looking to copper or copper alloys (hereinafter collectively "copper"), which are said to have "improved RC characteristics." (Spec. 11 [0029].) The '867 Specification teaches, however, that copper, unlike aluminum, does not form a self-passivating oxide. (Id.) Moreover, when aluminum or gold wires are bonded to copper, "intermetallics are formed which are more resistive and which expand volumetrically, causing cracks and lowering reliability." (Id.) To address this problem, it is known to provide a barrier metal composition layer, also known as a diffusion barrier or metal passivation layer (id.), such as Ni-Pd (id. at 2 [0005]), over the copper bonding pad. As shown in Figure 5, reproduced with annotations on the following page, copper bonding pad 5024 is formed, typically in interlevel dielectric ("ILD") (id. at 18, 1. 1) 504, and covered by a protective overcoat ("PO") 510 (id. at 3 [0006]), which is then patterned, e.g., by photolithographic processes, to form recess 512. (Id. at 19 [0043].) The entire top surface of the substrate, including recess 512, is coated with 3 Application 15/162,867, Surface Conditioning and Material Modification in a Semiconductor Device, filed 24 May 2016. We refer to the '"867 Specification," which we cite as "Spec." 4 Throughout this Opinion, for clarity, labels to elements are presented in bold font, regardless of their presentation in the original document. 2 Appeal2018-006250 Application 15/162,867 Ni-Pd. See barrier metal layers 514 on the bottom and 516 on the sides of recess 512. (Id. at 2 [0005].) The barrier metal layers are then treated by chemical-mechanical polishing ("CMP") to obtain the desired thickness. (Id. at 3 [0007].) Following CMP, residual contaminants may remain, especially in recesses on the substrate ( e.g., debris 518 in recess 512 for bond pad 502, and packed-in contaminants 522 in the slot over scribe seal 522) (id. at 20 [0043]). 500 ~, {Figure 5 is shown below} 512 \ 1 518 ) 516· ! ' .. ,.:~:-.,.t,_.r524 510 522 520 \ ) 506 506 506 FIG. 5 ~504 {Figure 5 shows device 500 under 02 plasma ash treatment 524 to remove residue 518 and 522} In the words of the Specification, "successful wire bonding is critically dependent upon the surface finish of both the component's bond pads and the substrate." (Id. at 1 [0002].) To address this problem, the 3 Appeal2018-006250 Application 15/162,867 Specification discloses that "the diffusion barrier layer is treated by a plasma ashing process [524] for improved residue removal, whereby CMP's residual particulate matter or other byproducts are removed by microincineration." (Id. at 3 [0007].) Wire bonds may then be formed to the bonding pad. Claim 1 is representative and reads: A method in fabricating a semiconductor device, the method compnsmg: creating a plurality of recessed features in a process layer of the semiconductor device, wherein the recessed features include a nickel-palladium (Ni-Pd) surface that can contain residual materials generated in creating the recessed features; and removing the residual materials from the Ni-Pd surfaces of the recessed features by subjecting the semiconductor device to a plasma of reactive ion species for a specific duration. (Claims App., Br. 7; some formatting, and emphasis added.) The Examiner maintains the following ground of rejection: s, 6 Claims 1-22 stand rejected under 35 U.S.C. § 103 in view of the combined teachings of Lehr 7 and Yu. 8 5 Examiner's Answer mailed 04 April 2018 ("Ans."). 6 Because this application was filed after 16 March 2013, the effective date of the America Invents Act, we refer to the AIA version of the statute. 7 Matthias Lehr and Frank Kuechenmeister, Wire bonding on reactive metal surfaces of a metallization of a semiconductor device by providing a protective layer, U.S. Patent Application Publication 2009/0243105 Al (2009). 8 Tsung-Yuan Yu et al., Method of forming bump structure, U.S. Patent Application Publication 2014/0242791 Al (2014). 4 Appeal2018-006250 Application 15/162,867 B. Discussion The Board's findings of fact throughout this Opinion are supported by a preponderance of the evidence of record. Bashyam urges the Examiner erred harmfully in finding that the combined teachings of Lehr and Yu suggest removing residual materials from the Ni-Pd surfaces by a plasma of reactive ion species. (Br. 5, 11. 9- 12.) In particular, Bashyam (Br. 4, last para., through 6, first full para.) urges that the Examiner misapprehends the passages in Lehr cited by the Examiner as "disclos[ing] protective layer 240 being useful in preventing increased creation of contaminants on surface 212S resulting from an oxygen/fluorine plasma process [0031---0032]" (FR 2, 11. 12-14.) Bashyam urges that, in this passage, "Lehr alludes to a plasma process that creates the contaminants." (Br. 5, 1. 7.) This, in Bashyam's view, does not disclose or suggest removing residual materials from the Ni-Pd surface purportedly suggested by Yu. (Id. at 11. 7-12.) Bashyam appears to misapprehend or overlook the teachings of Lehr. Lehr, like Bashyam, is concerned with difficulties associated with wire bonding to copper pads, which is said to be "very difficult to achieve due to an inhomogeneous self-oxidization of the copper surface in combination with extensive corrosion, which may result in highly non- reliable bond connections." (Lehr [0006].) More particularly, Lehr teaches that the surface 212S of copper metal region 212 (that will become a bonding pad) "represents a copper-containing surface, which may suffer from increased creation of contaminants upon exposure to a reactive ambient, such as ambient air, a plasma ambient containing reactive 5 Appeal2018-006250 Application 15/162,867 components, such as fluorine, oxygen and the like." (Id. at 5 [0031], 11. 18- 23.) See Lehr Figure 2a, reproduced below. ./ct:. ! ' ··. ·'',;i':•t-\' , .... · I : .. \ ! t I {Lehr Figure 2a shows device 200 undergoing formation of protective layer 240 by exposure to deposition ambient 241.} Lehr teaches that the problems with top surface 212S of metal region 212 may be overcome by providing protective layer 240. Protective layer 240 may be a dielectric (id. at 5, col. 2, 1st full sentence), or a conductive metal (id. at 2d full sentence), and is "selected such that interaction of the exposed portion of the surface 212S with the ambient may be significantly suppressed, while a bond process to be performed in a later stage may not be substantially negatively affected by the presence of the protective layer 240." (Id. at 5 [0032] 11. 23-27.) In particular, Lehr teaches that protective layer 240 "may be provided in the form of nickel, palladium and the like." (Id., penultimate sentence.) 6 Appeal2018-006250 Application 15/162,867 In this regard, the Examiner's finding, that Yu teaches that Ni-Pd can serve as a cap on a copper pad (FR 2, last sentence, citing Yu [0011]), provides adequate support for the conclusion that it would have been obvious to use Ni-Pd as protective layer 240 in processes disclosed and suggested by Lehr. Basham's unelaborated criticisms (Br. 5---6) that the teachings of Lehr would not have suggested removing the residual materials from metal protective layers 240 in recesses of devices described by Lehr are not well- taken. The ordinary worker would have recognized that exposure of device wafer 200 to ambient 241 results in the non-selective deposition of protective layer 240-that is, everything gets coated. The ordinary worker would have recognized further that removal of protective layer 240 from where it isn't wanted, e.g., the top surface of dielectric layer 223, is necessary and would proceed by standard removal processes such as CMP. 9 The routineer would also have recognized plasma etching as a standard method of removing dielectric materials, such as polymers, used in association with CMP processes. 10, 11 In light of the well-known sensitivity 9 Lehr discloses such processes with a conductive protective layer at 6 [0040]-7 [0044] and Figures 2d-2f. 10 Lehr also teaches that following CMP removal, "a further etch process for removing any residues of the protective material 242 [a polymer material to enhance the integrity of the protective layer 240 within the opening 220A during the removal process 241 (id. at [0043])]. (Id.) 11 Lehr also discusses at some length plasma-etch processes to reduce the polymer protective layer 240 to a thickness through which a wire bond connection can be made. (Lehr 8 [0052]-[0055], Figures 21 and 2m.) These 7 Appeal2018-006250 Application 15/162,867 of wire bonding to the surface presence of surface contaminants, 12 we have no difficulty determining that the routineer would have considered using techniques such as plasma etching as one of these standard processes, particularly when protective layer 240 is a metal such as Ni-Pd, selected to be insensitive to conditions that would degrade a copper-containing surface. Because we conclude that harmful error has not been demonstrated in the appealed rejection, we affirm. C. Order It is ORDERED that the rejection of claims 1-22 is affirmed. AFFIRMED processes leave a thin layer of protective layer, but the routineer, even without the promptings of Lehr, would have recognized that if the protective layer 240 were metal, all the dielectric material could be removed without damaging the underlying copper layer. 12 Lehr 1 [0006], cited supra; see also id. at 2 [0008], col. 2, 11. 19-25, elaborating on [0006] and discussing the problems of contaminants 112A on the exposed Cu surface shown in Figure la (not reproduced here). 8 Copy with citationCopy as parenthetical citation