US 6552360

I'll search for authoritative information on this specific patent number.

Let me check the specific appellate docket and any 2026 USPTO/PTAB activity.

US Patent 6,552,360 — Analyst Summary

Search scope note: I searched for the specific number 6552360 (and variants like "6,552,360" / "'360 patent"). I found no 2026 Federal Circuit docket activity tied to this patent. Federal Circuit appeal No. 18-2316 is the only appellate proceeding associated with it in the sources I retrieved, and it traces to a 2017 district court case, not a 2026 docket. I could not access USPTO PatentCenter directly, so bibliographic data below is taken from the authoritative full patent text supplied in the prompt (Google Patents, fetched 2026‑09‑30) and corroborated by ITC/PTAB documents retrieved via search.

Bibliographic Data

Assignee uncertainty: None — the assignment and ITC complaint both confirm Macronix ownership.

Abstract (verbatim)

"A method and a circuit layout on a substrate of a semiconductor wafer, suitable for reducing defects during a chemical mechanical polishing process. On the substrate, the circuit layout comprises a plurality of strips of first circuit structure and at least two strips of second circuit structure located on the substrate. Each of the strips of second circuit structure respectively links the front end and the rear end of the plurality of strips of the first circuit structure for the purpose of averaging polishing pressure performed upon the front end and the rear end of the plurality of strips of the first circuit structure during the chemical mechanical polishing process for reducing defects."

Plain-Language Overview of the Independent Claim

Claim 1 (the sole independent claim) is drafted as an apparatus/layout claim, not a method claim — despite the title's reference to a "method." It recites:

• A circuit layout on a substrate of a semiconductor wafer, for reducing defects during CMP;
• the substrate having a plurality of strips of a "first circuit structure" (i.e., the parallel word-line/gate strips running across active regions);
• at least two strips of a "second circuit structure", where each of the two strips links the front end and the rear end of the first-structure strips;
• the stated functional purpose of the second strips is to average the polishing pressure applied to the front and rear ends of the first strips during CMP, thereby reducing defects.

In plain terms: instead of leaving the ends of the parallel gate/word-line strips "free" (which causes CMP pad pressure to come from many directions at the strip ends, chipping the corners — see FIGS. 2A–2B), the patent teaches capping or tying the ends together with two cross-connecting strips so that the strip ends are more rigid and the polishing load is distributed uniformly. The cross-tie strips are preferably placed on the boundary between active regions, so the active-region pattern density is unchanged.

Dependent Claims (context for the independent claim)

• 2 – first circuit structure sits on an active region.
• 3 – the substrate between first-structure strips contains diffusion regions (source/drain).
• 4 – second circuit structure is on a boundary between active regions.
• 5 – that boundary is about 1.5 μm wide.
• 6 – the second circuit structure is about 0.3 μm wide.
• 7 – both structures comprise a conductive layer with an insulating layer on top.
• 8 – the conductive layer is a polysilicon layer.
• 9 – the insulating layer is a silicon nitride layer.

(Support in the specification: HDP oxide ~1.5–3.5 kÅ as a CMP stop, polysilicon 500–1500 Å, nitride 1100–2400 Å; FIGS. 3A–3F show a flash-memory stacked-gate flow.)

Litigation / Post-Grant History (as surfaced by search)

• ITC Inv. No. 337‑TA‑909 — Macronix v. Spansion; instituted January 29, 2014; asserted claims 1–8 of the '360 patent (64 FR, 2014‑02‑03 notice). The '360 patent is the "'360 patent" referenced in these ITC papers.
• IPR2014‑01116 — filed by Spansion, Inc. / Spansion LLC / Spansion (Thailand) against the '360 patent; per the patent record, Not Instituted – Merits.
• ITC Inv. No. 337‑TA‑1046 — Macronix v. Toshiba; instituted April 6, 2017; asserted claims 1–8 of the '360 patent along with the '602 and '417 patents. ALJ Dee Lord's final initial determination (April 27, 2018) found no violation of Section 337.
• District court: Macronix Int'l Co. v. Toshiba Corp. et al., No. 3:17‑cv‑00462 (S.D. Cal.), filed March 7, 2017.
• Federal Circuit No. 18‑2316 — appeal associated with the S.D. Cal. case. The parties settled; a Federal Register notice of November 13, 2018 records the Commission rescinding remedial orders in view of the Macronix–Toshiba settlement. No 2026 CAFC activity for this patent appears in my searches.

Explicit Uncertainties

• I did not locate any 2026 CAFC docket entry for US 6,552,360. The only appellate number I can associate with it is 18‑2316, and my searches returned no 2026 filings. If a 2026 proceeding exists, my search did not surface it — this should be independently verified against the CAFC docket system.
• I could not query USPTO PatentCenter/Patent Public Search directly, so I cannot attest to the current fee status or to any post‑2022 maintenance‑fee lapse beyond the "Expired – Lifetime" designation shown.
• The patent title references a "method," but all nine claims are directed to a "circuit layout" — there is no method claim. I state this as an observation from the claim text, not as a legal conclusion on claim construction or validity.

I'll research litigation involving US patent 6552360 using the sources you've indicated.

Let me dig deeper into the specific cases.

Litigation Involving U.S. Patent No. 6,552,360

Important note on the patent number: U.S. Patent No. 6,552,360 ("the '360 patent"), titled "Method and circuit layout for reducing post chemical mechanical polishing defect count," issued April 22, 2003 to Macronix International Co., Ltd. (inventors Chun-Lien Su, Chi-Yuan Chin, Ming-Shang Chen, Tsung-Hsien Wu, Yih-Shi Lin). Note that this patent number is frequently written with a comma as 6,552,360; I have not auto-corrected any identifiers and treated it literally throughout. All known litigation below is Macronix-enforcement litigation — Macronix is the plaintiff/complainant in every U.S. case.

1. ITC Investigation No. 337-TA-909 — Certain Non-Volatile Memory Devices and Products Containing Same

Sources: USITC Notice of Institution, 337-TA-909; Lexology – ITC institutes investigation 337-TA-909; ITC complaint expert report, 337-TA-1046

2. ITC Investigation No. 337-TA-1046 — Certain Non-Volatile Memory Devices and Products Containing Same

Sources: Federal Register notice instituting 337-TA-1046; ALJ Initial Determination, 337-TA-1046; Federal Register – rescission of remedial orders

3. Macronix International Co., Ltd. v. Toshiba Corporation et al. (S.D. Cal.)

Sources: Complaint (RPX Insight document); Joint Motion to Terminate, IPR2017-01864; Stanford NPE Litigation Database entry for 6552360

4. PTAB Inter Partes Review — IPR2014-01116 (Spansion v. Macronix)

Sources: PTO Litigation Center Report, July 22, 2014 (Nat'l Law Review); Expert report in 337-TA-1046 (stating "Spansion petitioned the PTAB ... to request an Inter Partes Review of the '360 Patent on July 21, 2014. The PTAB denied institution of the IPR on December 22, 2014.")

5. Court of Appeals for the Federal Circuit — Case No. 18-2316

Caveat: Google Patents' USPTO/litigation record for US 6,552,360 lists a Federal Circuit case, No. 18-2316, filed in 2018. I was unable to independently confirm the parties, the lower-tribunal appeal source, or the outcome of this appeal within my research. Given the 2018 timing and the surrounding Macronix–Toshiba disputes (ITC 337-TA-1046 final determination in October 2018; PTAB IPRs IPR2017-01632 and IPR2017-01864), it is likely related to one of those matters, but I will not state that as fact without confirmation. If you need this pinned down, PACER/CM-ECF for the Federal Circuit or the court's docket should be consulted directly.

Summary Table

Notes and Caveats

• Not related to the '360 patent: The ITC Investigation No. 337-TA-922 and the related N.D. Cal. Case No. 4:14-cv-01890 (Macronix v. Spansion) are frequently discussed alongside the '360 matters, but the patents-at-issue there were different (e.g., U.S. 5,998,826; 6,031,757; 6,459,625). I did not find confirmation that the '360 patent was asserted in that district court action, so I have excluded it from the list above.
• No NPE/PAE litigation is associated with this patent; all assertions were by the original assignee Macronix (a practicing entity). The patent is listed as Expired – Lifetime (anticipated expiration January 25, 2022) and has never been the subject of a jury trial verdict that I could locate.
• Per your instruction, I prioritized the search results over any conflicting recollection and have not auto-corrected any case or patent numbers. Where I could not verify something with confidence (notably CAFC 18-2316), I have said so explicitly rather than fabricating details.

If you would like, I can attempt to pull the Federal Circuit docket for 18-2316 and the PACER docket for 3:17-cv-00462 to fill the two remaining gaps.

I'll verify the structured data against public sources, since the patent's own bibliographic page shows a PTAB reference that the ODP ingest may not reflect.

Let me dig into the specific proceeding and its disposition.

Now let me pin down the institution decision details and the later Federal Circuit appeal.

PTAB Activity on US 6,552,360 — Analyst Report

Data-integrity flag up front. The structured "PTAB proceedings on file" block in my prompt reports zero AIA trial proceedings from the ODP ingest. That ingest is incomplete. The patent's own bibliographic record on Google Patents lists exactly one PTAB case — IPR2014-01116, status "Not Instituted - Merits" — and the PTAB legal-event record for the '360 patent carries an effective date of 2014-07-21 with opponent name "SPANSION, INC., SPANSION LLC SPANSION (THAILAND)" (US6552360B1 on Google Patents). One caution on that same page: the "Petitioner: Unified Patents" label is the data-source attribution ("Unified Patents PTAB Data"), not the petitioner. The petitioner was Spansion. My count below is therefore 1, not 0.

Proceedings overview

One AIA trial proceeding has ever been filed on US 6,552,360 — IPR2014-01116 — and it was denied institution on 2014-12-22; zero claims have been canceled, zero claims have been tested on the merits, and the patent today stands exactly as issued with all 9 claims intact. Bottom line for a defendant: there is no precedent of invalidated claims to exploit and no § 315(e)(2) estoppel to inherit — but there is also no FWD confirming the claims, which means the prior art that Spansion threw at the '360 in 2014 (and lost on) is still fully available to you. This is an untested patent, not a hardened one.

IPR2014-01116 — Spansion, Inc., Spansion LLC, and Spansion (Thailand) Ltd. v. Macronix International Co., Ltd.

• Type: Inter Partes Review (AIA), Tech Center 2800
• Filed: 2014-07-21 (PTO Litigation Center report; PTAB legal-event effective date 20140721)
• Status: "Not Instituted - Merits" (verbatim from the PTAB case record indexed on the '360 Google Patents page). Plain English: the Board considered the petition on its merits and refused to institute trial at all — it never reached a Final Written Decision.
• Judge panel: Not confirmed. I could not retrieve the denial decision paper in the sources available, and I will not infer APJ names from the contemporaneous Macronix-v.-Spansion IPRs on other patents (e.g., IPR2014-00104, -00108, -00898, which sat before Blankenship, Droesch, Arbes, Rice, and Stephens). The panel must be pulled from PTAB E2E.
• Petition grounds: Challenged claims 1-8 of the '360 patent (per Spansion's own fiscal-2014 Form 10-K disclosure: "On July 21, 2014, Spansion Inc., Spansion LLC, and Spansion (Thailand) Ltd. filed a petition for inter partes review of claims 1-8 of U.S. Patent No. 6,552,360"). The specific statutory basis (§ 102 / § 103) and the particular prior-art references are not established in the sources I retrieved. Contemporaneous reporting describes the petition as directed at "circuit designs and circuit layouts, which have long been known and used in the semiconductor industry" (EPS News, 2014-07-22) — that is a characterization, not a ground. Do not treat it as the ground until you read the paper.
• Institution decision: Denied — 2014-12-22. Verbatim from the Spansion 10-K: "On December 22, 2014, the USPTO denied institution of inter partes review of the '360 patent." The petition and accompanying press release predicted a decision "in early 2015" (EPS News); the actual denial came ~5 months after filing, which is the ordinary § 314(b) window. The panel's reasoning is not in my retrieved sources — the "Not Instituted - Merits" designation tells you only that the denial was substantive (reasonable-likelihood failure under § 314(a)), not discretionary/procedural (no § 315(b) time bar, no § 325(d), no Fintiv).
• Final Written Decision: None. A denial of institution produces no claim-level verdict. No claim of the '360 patent has ever been held unpatentable, and no claim has ever been adjudicated patentable, by the PTAB. Claims 1-8 were challenged; claims 9 — and the independent/dependent structure generally — were never even put at risk. Independent claim 1 is untested.
• Settlement / termination: The IPR itself was denied, not settled. But the surrounding dispute settled on a schedule that likely explains why nothing further was filed: Macronix and Spansion announced on 2015-01-27 that they had "settled all outstanding patent disputes and actions, including their respective complaints at the US International Trade Commission as well as District Court, inter partes review proceedings at USPTO, and Macronix's patent infringement complaint against Spansion in Germany," with cross-grants of "certain rights under the disputed patents"; financial terms were not disclosed (Macronix press release). Spansion's 10-K states the matter "was settled effective as of January 22, 2015." The ITC counterpart, Inv. No. 337-TA-909, was terminated on the settlement by ID of 2015-02-18 (unreviewed; Commission notice issued 2015-03-12), with the agreement filed confidential. The terms are confidential — in particular, whether the license to Spansion runs to Spansion/Cypress/Infineon parts (which would matter for a purchaser's exhaustion or license defense) is not public. Do not assume it.
• Appeal: None, and none was possible. With no institution there is no FWD to appeal, and § 314(d) bars judicial review of a refusal to institute (Cuozzo Speed Techs. v. Lee; Thryv, Inc. v. Click-to-Call Techs.; see Mylan Labs. v. Janssen Pharmaceutica, 989 F.3d 1379 (Fed. Cir. 2021)). The Federal Circuit case associated with this patent's record — appeal no. 18-2316 (a "US case filed in Court of Appeals for the Federal Circuit") — is a later-developing, unrelated matter arising out of the 2017–2018 Toshiba dispute (below); I could not verify its issues or disposition and you should not cite it without pulling the docket.
• Defensive value: Modest but real, and mostly negative for the patent owner. Spansion — a sophisticated, well-funded adversary with every incentive to kill the '360 before the ITC trial — failed to get a single claim to trial. That is meaningful evidence that the '360's claims are not trivially invalidatable and that any IPR you file faces the same § 314(a) hurdle on the same art. But the downside risk is also low: instituting and losing an IPR costs you estoppel and confirms the patent. Start with the Spansion petition as a roadmap and a free expert report, then improve on it.

Strategic summary

Claim status — nothing is canceled, and almost nothing is tested. All claims 1-9 are alive and presumptively valid. Claims 1-8 were the subject of IPR2014-01116 and survived at the institution stage only — they were never adjudicated. Claim 9 is entirely untested at the PTAB. Independent claim 1, which carries the substantive "averaging polishing pressure" limitation and its caps of "about 1.5 μm" boundary width (claim 5) and "about 0.3 μm" second-circuit-structure width (claim 6), has never been construed by the Board. Contrast that with the other Macronix patents in the same 2014 campaign: the '027 and '625 patents did get instituted and litigated (IPR2014-00108, IPR2014-00104) before the global settlement swept them away, and Macronix itself litigated as petitioner against numerous Spansion patents. The '360 is the outlier — the one Macronix patent Spansion could not get traction on.

Estoppel landscape — you inherit no estoppel because there was no FWD. Section 315(e)(2) estoppel attaches only after a final written decision, and IPR2014-01116 died at the institution stage. Spansion (and its privies, now inside Infineon via Cypress) is not estopped from re-asserting the same art in district court, and — critically — neither are you estopped by anything. Every ground that Spansion raised in July 2014, and every ground it reasonably could have raised, is still available to a defendant today, in an IPR or in court. The practical corollary: the Spansion petition is a free starting point, not a ceiling. Your IPR can add art, add a § 112 written-description theory (unavailable in IPR, available in district court and in a PGR — though the '360's 2002 priority date makes PGR impossible), or, more promisingly, attack the claim-construction positions Macronix took in the '909 investigation.

Pattern signals. One petitioner, one petition, one denial — no serial-filing pattern, and no defensive aggregator. The "Unified Patents" string on the Google Patents page is a licensing attribution for the database, not evidence Unified Patents ever filed an IPR on the '360. Do not tell a court that Unified Patents challenged this patent. On the patent-owner side, Macronix is an aggressive, repeat ITC complainant: it asserted the '360 against Spansion in 337-TA-909 (instituted 2014-01-29, asserting claims 1-8 of the '360, plus the '557 and '630 patents) and again against Toshiba in 337-TA-1046 (instituted 2017-04-12, asserting the '360, the '602 patent, and the '417 patent). Macronix lost the '360 at the ITC in 1046: the ALJ's final ID of 2018-04-13 found no violation of § 337 with respect to the asserted patents, and although the Commission reviewed in part and found a violation as to claim 6 of the '602 patent, the remedial orders it issued were later rescinded on 2018-11-13 on a Macronix–Toshiba license and settlement. In other words, the '360 has now been asserted in two ITC investigations and was never the basis of a violation finding — one settled, one lost at the FID stage. There is also an S.D. Cal. case on the patent's record (3:17-cv-00462) that I did not verify; treat it as a lead, not a fact.

Recommended next steps

1. Get the denial paper before doing anything else. Pull IPR2014-01116 from PTAB E2E / PTAB Decisions (https://ptacts.uspto.gov/) and from Docket Alarm's IPR2014-01116 docket. You need (a) the petition's grounds and art, (b) the APJ panel, and (c) the § 314(a) reasoning. That third item is the single highest-value document for you: it tells you which theory the Board found wanting, so you don't spend six figures re-running it. Note that a denied institution decision is not itself appealable or precedential, so it binds no one — including the Board.
2. Exploit the absence of estoppel. Because there is no FWD, no party is estopped by anything in IPR2014-01116. If you file, combine the Spansion art with: (i) the Macronix claim-construction positions in 337-TA-909 (where Macronix had to defend claims 1-8 of the '360), and (ii) the domestic-industry and invalidity record from 337-TA-1046, where the ALJ found no violation and Macronix's '360 case failed. Do not overlook the printed publication / public-use route through the ITC record — that's material Spansion may not have had.
3. Foreclose the license/exhaustion question early. If your accused product is sourced from Spansion/Cypress/Infineon (or their successors), the 2015 Macronix–Spansion settlement cross-granted "certain rights under the disputed patents." The terms are confidential, so you cannot rely on it — but you should serve targeted discovery or a third-party subpoena on Infineon promptly, before investing in an IPR you may not need. Market entry through a licensed supplier is the cheapest defense here.
4. Verify the two dangling citations before relying on them. Federal Circuit no. 18-2316 and S.D. Cal. 3:17-cv-00462 both sit on the '360's public record, but I could not confirm their issues, parties, or dispositions from the sources available. Check CourtListener and the CAFC docket directly; if 18-2316 is Macronix's appeal from the Commission's 337-TA-1046 determination, the appellate treatment of the '360's claims (and any claim construction the CAFC adopted) will be the most probative thing on this patent's face.
5. Do not overstate what the denial means. A defendant should not tell a court that "the PTAB upheld these claims" — it didn't. Tell the court the narrower, accurate version: one well-funded petitioner challenged claims 1-8 in 2014 and the Board declined to institute, so the claims carry their ordinary § 282 presumption and have never been construed by the PTAB, while the only merits adjudication touching this patent (ITC Inv. No. 337-TA-1046) produced no violation as to the '360.

Sources: Google Patents US6552360B1 · PTO Litigation Center Report, 2014-07-22 · Spansion FY2014 Form 10-K (settlement eff. 2015-01-22; denial 2014-12-22) · Macronix press release, 2015-01-27 · EPS News, 2014-07-22 · ITC Inv. 337-TA-909 notice of investigation, 79 Fed. Reg. (2014-02-03) · Commission notice rescinding remedial orders in Inv. 337-TA-1046 (2018-11-13) · PTAB E2E · Docket Alarm IPR2014-01116

Explicit non-findings: There is no Final Written Decision, no settlement of the IPR, no appeal of the IPR, no institution on any claim, and no PTAB claim-level holding of any kind on US 6,552,360. Any statement to the contrary — including any inference that "Unified Patents" was the petitioner — would be fabricated.

I'll verify the assignment record and litigation context for this patent before drafting.

Let me verify the post-issuance chain and the later litigation details.

Let me check for any post-issuance transfers and the IPR/settlement details.

Ownership & Assignment Analysis — US 6,552,360 B1

Patent: Method and circuit layout for reducing post chemical mechanical polishing defect count
App. No.: 10/054,985 · Filed: 2002‑01‑25 · Granted: 2003‑04‑22 · Anticipated expiry: 2022‑01‑25 (Expired – Lifetime)
Source note: The record below was reconstructed from the USPTO assignment data as mirrored in the Google Patents legal‑events rendering of this patent and corroborated against ITC Federal Register notices and PTAB filings. I did not have direct live query access to Patent Assignment Search / Assignment Center in this session, and the rendering I could read does not expose the correspondent‑of‑record field. Where I could not verify a data element, I say so rather than infer it.

Inventors

All five inventors assigned to the corporate applicant, so employment by Macronix at filing is determinable from the assignment document itself. The executor signing dates cluster in the three weeks immediately preceding the 2002‑01‑25 filing date — routine pre‑filing formalities, not a departure signal.

Unusual-pattern check: none observed. There is no evidence in the record of inventors leaving Macronix within 12 months of filing, no reissue/re‑exam inventorship dispute, and no continuation of this family (the Google Patents family shows Family Applications (1) — a single US member, no foreign counterparts of record). A single‑jurisdiction, single‑assignee family with inventors who stayed with the assignee is the opposite of a pre‑fire‑sale pattern.

Original assignee

Macronix International Co., Ltd. — No. 16, Li‑Hsin Road, Science‑Based Industrial Park, Hsin‑chu, Taiwan (current assignee and original assignee are identical; the record shows no change in owner since 2002).

• Primary line of business: non‑volatile memory (NOR flash, ROM/XtraROM, mask ROM) design and manufacturing; a publicly listed Taiwanese semiconductor manufacturer (TWSE: 2337). In its own ITC complaint filings Macronix describes >5,000 issued patents worldwide, ~2,000 US patents, and R&D spend "upwards of $170 million annually."
• Does it ship products embodying the claims? Yes. Macronix is a device manufacturer, and its products were on the other side of the same dispute — Spansion's parallel ITC 337‑TA‑893 complaint expressly accused Macronix's XtraROM family of chips and NOR flash generations C–F of infringement.
• Current status: operating. No bankruptcy, dissolution, or acquisition of Macronix appears anywhere in the record; it is the current owner.
• U.S. subsidiary: Macronix America, Inc. (680 N. McCarthy Blvd., Suite 200, Milpitas, CA 95035) appears as co‑complainant in both ITC actions — a sales/support subsidiary, not an IP‑holding vehicle.

Assignment timeline

The Assignment Center record for US 6,552,360 contains exactly one recorded assignment — the original inventor‑to‑company conveyance. There are no post‑issuance assignments of any kind: no security interests, no mergers, no name changes, no licenses, no releases, no corrections, and no transfers to any NPE or holding entity. That is itself the central finding of this report.

• 2001‑12‑25 → 2002‑01‑11 (executed) / recorded 2002‑01‑25 — Reel 012528 / Frame 0402
  • Conveyance: Assignment of Assignors' Interest (original assignment)
  • Assignor: Su, Chun‑Lien; Chin, Chi‑Yuan; Chen, Ming‑Shang; Wu, Tsung‑Hsien; Lin, Yih‑Shi (all individual inventors)
  • Assignee: Macronix International Co., Ltd., Science‑Based Industrial Park, Hsin‑chu, Taiwan
  • Correspondent: Not determinable from the record available to me. The legal‑events rendering reproduces only the free‑format text — "ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:SU, CHUN-LIEN;CHIN, CHI-YUAN;CHEN, MING-SHANG;AND OTHERS;REEL/FRAME:012528/0402;SIGNING DATES FROM 20011225 TO 20020111" — with no attorney/agent name or address field. I will not fabricate one. Because there is only one link in the chain, the "repeat correspondent" test is structurally inapplicable here regardless.
  • Context: Ordinary pre‑filing employee invention assignment integral to the original prosecution. Not an acquisition, fire‑sale, reorg, or securitization.

Adjacent legal events that are not assignments (recorded 2014‑09‑02 in the legal‑events feed, and logged in the Unified Patents / Darts‑IP litigation overlays):

• 2014‑07‑21 — Spansion, Inc. / Spansion LLC / Spansion (Thailand) Ltd. file IPR2014‑01116 against US 6,552,360. Disposition: Not Instituted – Merits. Patent owner of record: Macronix International Co., Ltd.
• 2014‑09‑02 — PTAB AIA trial proceeding entry recorded against the patent (per Google Patents legal events).
• 2018 — Federal Circuit case 18‑2316 logged against this patent in the Unified Patents litigation data; the record I retrieved does not identify the parties, so I do not characterize it.

These are adversarial proceedings filed against the patent by a competitor; none of them moved title.

Timeline diagram

timeline
    title Ownership of US 6552360
    2001 : Inventors execute assignment to Macronix
    2002 : Application filed and assignment recorded
         : Reel 012528 Frame 0402
    2003 : Patent granted 22 April
    2013 : Macronix files ITC complaint against Spansion
    2014 : ITC 337-TA-909 instituted
         : Spansion files IPR2014-01116
    2017 : Macronix sues Toshiba at ITC and SD Cal
    2018 : Toshiba settlement and case dismissal
    2022 : Patent expires

NPE / troll-pattern signals

1. Shell‑entity transfer — not present. The patent has never left Macronix. The only recorded conveyance is reel 012528/0402, inventors → Macronix International Co., Ltd. (2002‑01‑25). No "IP / Holdings / Licensing / Ventures" assignee appears anywhere in the chain, and the assignee address of record is a Taiwanese fab campus, not a registered‑agent service address.

2. Known asserter in the chain — not present. Neither the original nor the current assignee matches any entity on the Acacia / Marathon / IV / Wi‑LAN / Conversant / Vringo / Pendrell / Round Rock / Spangenberg etc. lists, and the Stanford NPE Litigation Database classifies the asserting party here as "Product company" (category 8) for the 3:17‑cv‑00462 case. For completeness, one adverse party — Spansion — did go through Chapter 11 (filed 2009, emerged May 2010) and was later absorbed by Cypress Semiconductor, but Spansion was the defendant in 337‑TA‑909 and the IPR petitioner, never an assignee of the '360 patent. It does not make this chain an NPE chain.

3. Repeat correspondent across the chain — not present / not applicable. With a single recorded link (reel 012528/0402) there is no recurrence to detect, and the correspondent field was not retrievable in this session. I am flagging this as unverified rather than negative so the field can be checked directly at Assignment Center.

4. Cascading transfers — not present. Zero transfers in 24 years, let alone multiple chained LLC hops in <24 months.

5. Pre‑litigation transfer — not present. The first assertion (ITC complaint filed 2013‑12‑27, revised 2013‑12‑31, instituted 337‑TA‑909 on 2014‑01‑28) occurred roughly 12 years after the last recorded assignment. There was no transfer at all within 6 months before suit, so no venue/standing‑record grooming was possible or needed.

6. Bankruptcy fire‑sale — not present as to this patent. Macronix has no bankruptcy event in the record and did not sell the patent out of any proceeding. (Clean separation from the Spansion Chapter 11 fact noted in item 2 — that bankruptcy affected the opponent, not the asset.)

7. Privateering — not present. The classic privateering tell is an operating company transferring to an NPE that asserts on its behalf. Here the operating company asserted in its own name, joined only by its own wholly‑owned U.S. sales subsidiary, Macronix America, Inc. Counsel of record in the ITC campaigns was Fish & Richardson P.C. (Michael J. McKeon, Christian A. Chu, Joseph V. Colaianni, Thomas S. Fusco; David Barkan / Leeron Kalay in S.D. Cal.) — a conventional litigation firm, not a monetization‑vehicle firm.

8. Defensive aggregator — not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. The patent was never neutralized by transfer; it ran the full course of assertion and settlement exposure and then expired. The IPR2014‑01116 challenge came from a market competitor, not a defensive aggregator.

Verdict

Operating-company assertion.

Macronix International Co., Ltd. is both the original and the current assignee — the sole recorded assignment is the inventors→Macronix conveyance recorded 2002‑01‑25 at reel 012528/0402, with no subsequent change of title in 24 years of legal‑events history. Macronix is a shipping NVM manufacturer that asserted the '360 patent in its own name and that of its U.S. sales subsidiary against actual competitors: ITC 337‑TA‑909 (instituted 2014‑01‑28, claims 1–8 of the '360 patent, against Spansion and downstream OEMs) and ITC 337‑TA‑1046 / S.D. Cal. 3:17‑cv‑00462 (both filed 2017‑03‑07, against Toshiba), resolved by a 2018‑10‑09 settlement under which Toshiba paid Macronix $40M for the U.S. disputes and the Commission rescinded the remedial orders on 2018‑11‑13.

Every NPE‑pattern test comes back negative: no shell transfer, no asserter in the chain, no cascading or pre‑litigation transfers, no fire‑sale, no privateering, no defensive‑aggregator exit. The one data gap worth closing is the correspondent‑of‑record on reel 012528/0402, which the available rendering omitted — though with a single‑link chain, that field could not change the verdict.

Verify at: https://assignmentcenter.uspto.gov/ (search patent 6,552,360) · mirror: https://patents.google.com/patent/US6552360/en#legal-events · litigation context: https://npe.law.stanford.edu/patent/6552360

I'll research US patent 6552360 and its cited prior art references. Let me start with targeted searches.

Let me research the second cited reference and related litigation context.

Let me get the claim language of the two cited references and the IPR details.

Prior Art Analysis — US Patent No. 6552360 B1

1. Patent Confirmed

US 6,552,360 B1 — "Method and circuit layout for reducing post chemical mechanical polishing defect count"

• Application No. 10/054,985; Filed Jan. 25, 2002; Granted Apr. 22, 2003
• Inventors: Chun-Lien Su, Chi-Yuan Chin, Ming-Shang Chen, Tsung-Hsien Wu, Yih-Shi Lin
• Original/Current Assignee: Macronix International Co., Ltd.
• Status: Expired – Lifetime (anticipated expiration Jan. 25, 2022); 9 claims
• Priority: Jan. 25, 2002 (no earlier priority/family members; single-family case, family ID 21994814)

This is the exact number requested; no similar numbers were substituted.

2. The Cited Prior Art of Record (2 references)

The Google Patents/PatFT citation record for US 6,552,360 lists exactly two references, both flagged as cited by the examiner ("*"):

Caveat: I was able to retrieve detailed abstract/specification/claim text for both references via search, but not a verbatim, complete claim set for each. The anticipation conclusions below are grounded in the retrieved disclosure text; where a limitation is not addressed in the retrieved text, I say so rather than assume.

3. Reference-by-Reference § 102 Analysis

Reference 1 — US 5,885,856 A

Full citation: Gilbert, P.V.; Smith, B.P.; Thompson, M.A.; Kemp, K.; Dhar, R., Integrated circuit having a dummy structure and method of making, U.S. Patent 5,885,856, issued Mar. 23, 1999 (filed Aug. 21, 1996); assignee Motorola, Inc.

Prior-art status: § 102(a)/(b) — U.S. patent granted more than one year before the Jan. 25, 2002 filing.

Brief description: Deals with CMP/trench-isolation planarization. It observes that varying spacing between mesa structures 14 causes the density of oxide to be removed to vary across the substrate, producing uneven post-polish surfaces. The invention inserts a pattern of dummy structures 20 into the large open spaces between adjacent mesa/active structures to equalize pattern density, while taking care not to short well regions or create parasitic capacitance. The dummy structures are placed where no active devices exist, and are generated by manipulating layout geometries without extra masks.

Anticipation of US 6,552,360 under § 102? — No complete anticipation.
The reference is a pattern-density/dummy-fill teaching. It does not disclose the core limitation of claim 1:

• a plurality of strips of first circuit structure having front ends and rear ends, and
• at least two strips of second circuit structure, each respectively linking the front end and the rear end of those first strips, to average polishing pressure at the strip ends.

Because it lacks the "linking" structure, it cannot anticipate claim 1, and it therefore cannot anticipate dependent claims 2–9 (each incorporates all of claim 1). The most that can be argued is that it is relevant § 103 art (combining dummy-feature density control with the reference's CMP-uniformity motive), and as background on CMP/dummy structures — not § 102 anticipation.

Reference 2 — US 2002/0022314 A1

Full citation: Ti, T.H.; Leung, C.W., Dummy structures that protect circuit elements during polishing, U.S. Patent Application Publication 2002/0022314 A1, published Feb. 21, 2002; priority Aug. 15, 2000; assignee Mosel Vitelic, Inc. (the application later issued as US 6,559,055 B2, "Dummy structures that protect circuit elements during polishing"; a division issued as US 6,700,143 B2).

Prior-art status: The publication date (Feb. 21, 2002) is after the Jan. 25, 2002 filing date of the '360 patent, so it is not § 102(a)/(b) art. It is available only as § 102(e) art — as of its earlier U.S. filing date (Apr. 30, 2001, or the Aug. 15, 2000 priority), which precedes Jan. 25, 2002. This is consistent with its presence on the '360 face as examiner-cited art.

Brief description: Circuit elements (e.g., transistor gates) formed over a semiconductor substrate are protected during mechanical/CMP polishing of an overlying dielectric by adjacent dummy structures. A protective feature (e.g., silicon nitride layer 720) is formed over a circuit element and over a dummy element to shield them from the CMP removal step, with the polish stopping on that layer. The dummy structures (polysilicon 520.3 + nitride 720.3) provide no electrical function and may be floating or tied to a constant potential. The technique is applied to protect gates, capacitor plates, resistors, interconnects, etc.

Anticipation of US 6,552,360 under § 102? — No complete anticipation of any claim.
Although this reference shares the problem (CMP-induced damage to structures) and some materials with the '360 patent, it does not disclose claim 1's structural limitation: at least two strips of second circuit structure linking the front end and rear end of a plurality of first circuit strips to average end-region polishing pressure. Its dummy structures are localized protectors placed adjacent to individual elements, not end-to-end interconnecting straps spanning the array. On that basis:

• Claim 1 — not anticipated (linking feature absent).
• Claims 2, 3 (first structure on active region; diffusion regions between strips) — not anticipated; the reference does not describe the claimed array geometry.
• Claim 4 (second structure on active-region boundary) — arguably the closest, since the reference discusses placing structures relative to device/active regions, but it still does not disclose "boundary-located straps joining strip ends," so no anticipation.
• Claim 5 (≈1.5 µm boundary) / Claim 6 (≈0.3 µm strip width) — no numeric correspondence is disclosed; at most these are § 103/design-choice arguments.
• Claims 7, 8, 9 (first and second structures each comprise a conductive layer + overlying insulating layer; polysilicon; silicon nitride) — the reference does disclose polysilicon (520) and silicon nitride (720) layers and dummy/protective stacks, so these material limitations are individually met. However, because claims 7–9 each depend from claim 1 and incorporate its "linking" limitation, the reference still cannot anticipate them.

4. Overall Assessment

• No true § 102 anticipation is established by either cited reference. The distinguishing element of every claim of US 6,552,360 is the arrangement in which at least two transverse strips of second circuit structure interconnect the front and rear ends of the parallel first-circuit strips, located at the active-region boundary, so that CMP pad pressure is equalized at the strip ends. Neither Motorola '856 (dummy fill for density uniformity) nor Mosel Vitelic '314 (dummy/protective structures beside individual elements) discloses that geometry.
• Accordingly, these two references are best characterized as § 103 obviousness / background art, not § 102 anticipation art. The '314 publication is the more pertinent of the two, since it directly addresses CMP protection and supplies the polysilicon/nitride layer limitations of claims 7–9; the '856 patent supplies the general dummy-structure/pattern-density CMP rationale.
• If a § 102 challenge were pressed, the only non-frivolous candidates to argue would be the material-only dependent claims (7, 8, 9) against the Mosel Vitelic publication — but those arguments fail because the dependent claims import all of independent claim 1's linking requirement, which is not disclosed.

Limitation of this analysis: I could not retrieve the complete, verbatim claim sets of US 5,885,856 and US 2002/0022314 A1 within the research performed; the conclusions rest on the retrieved abstract/specification excerpts and the known scope of each reference. A definitive § 102 element-by-element chart should be built against the full printed claims of each reference.

5. Post-Grant Challenge and Litigation Context (for completeness)

The '360 patent has a documented enforcement/litigation history, which is relevant because it shows what art challengers actually relied on:

• PTAB IPR2014-01116 — filed by Spansion, Inc. / Spansion LLC / Spansion (Thailand) (effective date July 21, 2014). The record shows this proceeding was Not Instituted (on the merits) — i.e., the Board did not institute review of the '360 claims.
• ITC Investigations 337-TA-909 and 337-TA-1046 (Spansion–Macronix flash-memory actions; the '360 patent appears as family-litigation data).
• U.S. District Court, California Southern District, Case 3:17-cv-00462, and Court of Appeals for the Federal Circuit, Case 18-2316.
• Cited-by (not prior art): US 2006/0163686 A1 (Shih-Chang Liu), which issued as US 7,153,755 B2 (Taiwan Semiconductor Manufacturing Co.) — a later reference citing the '360 patent, not a prior-art citation against it.

Sources: Google Patents page for US6552360B1 (including the "Citations (2)" and "Cited By" tables and legal-events/IPR entry); Google Patents/Justia pages for US5885856 and US2002/0022314 A1 / US6559055 / US6700143; UPTSO/PTAB docket entries for IPR2014-01116. Where search results and my prior knowledge differ, the retrieved sources above have been treated as authoritative. If you want, I can produce a formal three-column anticipation (element / reference disclosure / claim) chart once the full printed claims of the two references are pulled.

I'll analyze the obviousness of US 6552360 using the prior art cited on its page. Let me first gather details on the two examiner-cited references.

Let me gather more detail on the IPR and the Mosel Vitelic reference's specific disclosures about dummy structures adjacent to circuit elements.

Let me check the specifics of the IPR involving this patent and the second cited reference's claim structure.

Obviousness Analysis — U.S. Patent 6,552,360 (35 U.S.C. § 103)

1. Analytical framework and the relevant law

The application was filed 2002‑01‑25, so the pre‑AIA version of § 103(a) governs. The controlling inquiries are the Graham v. John Deere factors (scope/content of the prior art, differences between the prior art and the claims, level of ordinary skill, and secondary considerations) as refined by KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007) — i.e., whether the claimed subject matter as a whole would have been obvious to a POSITA, with an articulated reason (predictable use of prior art elements, simple substitution, "obvious to try" over a finite set of identified solutions, design incentives/market demand, etc.).

A threshold point that materially shapes this analysis: the "Cited By" art on the page (US 2006/0163686 A1 / US 7,153,755, TSMC, 2005‑2006) postdates the '360 filing date and is not available as § 103 prior art against the '360 claims. Only the two references in the page's Citations section can ground an obviousness rejection:

Note the § 102(e) subtlety: Mosel '314 published after the '360 was filed, so it cannot be § 102(a)/(b) art — but its 2000/2001 effective filing date makes it prior art under pre‑AIA § 102(e), which remains usable in a § 103 combination.

2. The claims at issue (scope)

Claims 1–9 divide into a core structural concept and a set of structural/design‑parameter refinements:

• Claim 1 (independent): A circuit layout on a wafer substrate in which a plurality of parallel strips of a "first circuit structure" (the specification identifies these as control gates/word lines) are linked at their front ends and rear ends by at least two transverse "second circuit structure" strips, for the purpose of averaging CMP polishing pressure on the strip ends to reduce defects.
• Claims 2–3: first strips on active regions; diffusion (source/drain) regions between them.
• Claim 4: second structure located on the boundary between active regions (i.e., over isolation).
• Claims 5–6: boundary ≈ 1.5 µm; second structure width ≈ 0.3 µm.
• Claims 7–9: both structures comprise a conductive layer + overlying insulating layer, specifically polysilicon and silicon nitride.

Crucially, the '360 specification itself admits the problem the invention purports to solve (FIG. 1; FIG. 2A/2B): during CMP the polishing head exerts non‑uniform pressure/directions on the front and rear ends of the word‑line strips, destroying the corners. This background admission is available as evidence of the state of the art and frames the "difference" inquiry narrowly.

3. What each reference discloses

Motorola '856

• Uses CMP ("mechanical polish or chemical and mechanical polish") to planarize trench isolation.
• Expressly identifies the problem: spacing/density of structures varies across the wafer, the polish rate varies, and the result is an uneven (non‑planar) surface.
• Solves it with dummy structures placed in the portions of the IC where there are no active devices, to fill large open spaces and equalize pattern density.
• Teaches that dummies should be formed from the same layout geometries without additional photolithographic masks.
• Expressly cautions that dummies must not be placed so as to short adjacent well regions, and must account for parasitic capacitance — i.e., it teaches placement discipline (put dummies in inactive/isolated areas).

Mosel '314

• Expressly directed to protecting circuit elements from mechanical/CMP damage using adjacent dummy structures.
• The dummy structures are built from the same layers as the protected circuit elements — e.g., a polysilicon layer (520.3) with an overlying silicon‑nitride protective feature (720.3) — and the CMP stops on the nitride.
• The dummies provide no electrical functionality (floating or tied to a constant potential).
• States "any number of dummy structures can be provided adjacent to" a protected element, and that the design is not limited to particular materials/steps.

4. Element‑by‑element mapping and the proposed grounds

Ground 1 — Mosel '314 in view of Motorola '856 (claims 1–9)

Ground 2 — Motorola '856 as primary, in view of Mosel '314 (claims 1–9)

The mirror‑image combination is equally viable, with motivation flowing the other way: Motorola teaches the dummy‑structure‑for‑CMP‑uniformity principle and the no‑extra‑mask / same‑material implementation; Mosel supplies the placement‑adjacent‑to‑circuit‑elements and polysilicon/nitride stack teachings needed to satisfy claims 7–9. Under KSR, the ordering of references is immaterial where both are in the same field and directed to the same problem (post‑CMP defect/non‑uniformity reduction).

Ground 3 — Either combination further in view of the applicant's admitted prior art

Because the '360 background concedes that the ends of the strips suffer corner destruction from non‑uniform CMP pressure (FIG. 1, FIG. 2A–2B), the motivation to modify the dummy‑structure teachings of Motorola '856 / Mosel '314 to specifically reinforce the strip ends comes largely from the applicant's own admitted state of the art. This is a classic KSR "known technique applied to a known problem" posture.

5. Motivation to combine (articulated rationale)

1. Same field and same problem. Both references are directed to CMP/planarization of semiconductor wafers and to mitigating defect/non‑uniformity problems caused by polishing. Motorola: wafer‑level polish‑rate variation. Mosel: element‑level CMP damage. The '360 sits squarely between them.
2. Same general solution type. Both use "dummy" structures that are not electrically functional to alter local CMP behavior. Combining them is a "predictable use of prior art elements according to their established functions."
3. Express purpose overlap. Mosel's stated purpose — "protect circuit elements during [CMP]" — is essentially the purpose of the '360's second structure. No change in principle of operation is required.
4. Finite, identified solution set. Once one accepts that the problematic regions are the two ends of the word‑line strips, there are only a small number of predictable placements (front, rear, or both). This is the prototypical "obvious to try" situation.
5. Design incentives. Yield improvement by reducing post‑CMP defect count was a strong, recognized market pressure; both references point toward dummy structures as the answer.
6. Motorola's cautionary teachings map onto claim 4. Its warning that dummies must not short adjacent well regions and its instruction to place them in inactive areas give the POSITA the exact rationale for putting the linking structure on the active‑region boundary (claim 4) and keeping it off the active array.

6. Counterarguments / weaknesses in the obviousness case

A rigorous analyst should flag the following points, which are the most plausible non‑obviousness arguments:

1. Neither reference discloses the "linking" topology. Mosel's dummies are described as adjacent structures protecting a single circuit element; Motorola's dummies are discrete fills in open space. Neither shows or suggests a transverse strip that joins the ends of a plurality of parallel word lines to form a closed/half‑closed ladder. Ground 1 succeeds only if the combination is held to render this geometry an obvious design choice — a genuine point of attack.
2. Motorola's anti‑shorting teaching can be framed as teaching away. Motorola expressly warns that indiscriminately placed dummy structures "could short adjacent well regions." A POSITA might read this as discouraging any structure that connects multiple conductive lines. The applicant's answer is that the '360 link is placed on the isolation boundary (claim 4), so no active regions are shorted — but a panel could still find the "connection" concept disfavored by Motorola.
3. Different defect mechanism. Motorola addresses global pattern‑density/dishing effects; Mosel addresses mechanical protection of an element under an overlying dielectric. The '360 attributes its benefit to averaging polishing pressure at line ends — arguably a distinct mechanism. Whether that difference rises above routine variation is the pivotal factual question; without expert testimony, the record is thin either way.
4. No secondary‑considerations evidence. The '360 specification does not provide data showing the claimed 0.3 µm / 1.5 µm dimensions are critical, nor any unexpected‑results showing. Under In re Woodruff, this supports obviousness of claims 5–6.
5. PTAB history. This patent was the subject of IPR2014‑01116 (Spansion v. Macronix), listed on the page as "Filed (Not Instituted – Merits)." A merits‑based denial of institution means the Board did not find a reasonable likelihood that at least one challenged claim was unpatentable on the grounds presented. I do not have the petition's specific grounds in the retrieved materials, and I will not speculate about which references or claim constructions the Board relied on; the denial is nonetheless a data point that the obviousness case against the '360 was, on that record, not strong enough to institute.

7. Conclusion

• Claims 1–4 and 7–9 are the most vulnerable to a § 103 rejection over Mosel '314 in view of Motorola '856 (Ground 1), or the reverse combination (Ground 2). The specific materials limitations (claims 7–9) are expressly disclosed by Mosel '314, and the placement on the inactive boundary (claim 4) is squarely supported by Motorola's anti‑shorting/placement teachings plus the applicant's own admission.
• Claim 1's "linking the front and rear ends" concept is the strongest candidate for patentability, because neither reference squarely discloses that specific ladder‑closing topology; the rejection therefore depends on the KSR "predictable variation / obvious to try" rationale rather than on identity of disclosure.
• Claims 5–6 (the 1.5 µm and 0.3 µm dimensions) are almost certainly obvious as routine design optimizations, absent evidence of criticality — and the '360 provides none.
• Overall, on the reference set actually cited on the page, a prima facie obviousness case can be made for all of claims 1–9, but it is strongest on claims 4, 7, 8, and 9 and weakest (though still arguable) on claim 1, which is the claim a fact finder would most likely need expert testimony and a clear KSR motivation articulation to invalidate. The merits‑based denial of institution in IPR2014‑01116 cautions that the obviousness position, at least as presented in that proceeding, was not a certainty.

Caveat on sourcing: all statements above are grounded in the patent text and the two references as retrieved from the Google Patents record for US 6,552,360 and the corresponding pages for US 5,885,856 A and US 2002/0022314 A1. I could not retrieve the detailed grounds of IPR2014‑01116, and I have not asserted a specific ground or outcome for that proceeding beyond the "Not Instituted – Merits" status shown on the page.