US 9209261

I'll verify the current record for this patent against live sources before summarizing.

I have most of the record. Let me pull the complete claim set for accuracy.

US 9,209,261 B2 — Verification Summary

Important scope note: I was able to confirm bibliographic data and litigation history from live sources, but the authoritative full-text I have for this patent (the Google Patents fetch) truncates mid-specification and does not include the printed claim set. The claim language below is therefore reconstructed from PTAB/IPR exhibits (Samsung v. Acorn Semi, IPR2020-01183), not from the official printed claims. I flag each point where confidence is less than high.

Bibliographic data

Abstract (as printed)

"An electrical device in which an interface layer is disposed in between and in contact with a conductor and a semiconductor."

That is the complete abstract of the '261 patent — it is notably shorter than the abstract of its parent, US 7,884,003.

Plain-language overview

Core concept. A conventional metal–semiconductor contact has a Schottky barrier whose height is "pinned" near mid-gap by (a) dangling bonds / surface states at the semiconductor surface and (b) metal-induced gap states (MIGS) that decay from the metal into the semiconductor. The patent inserts a very thin interface layer between the metal and a group IV / Si-based semiconductor (Si, SiC, SiGe) that (i) passivates the dangling bonds and (ii) displaces the metal far enough to suppress MIGS — thereby depinning the Fermi level so the barrier height becomes tunable by metal selection — while remaining thin enough for substantial current to tunnel/flow.

The "Goldilocks" thickness insight. Too thin → MIGS re-pin the Fermi level and raise resistance; too thick → tunneling is blocked and resistance rises again. Figure 8 shows a minimum in the interface-layer-resistance vs. thickness curve, at which specific contact resistance can be ≤ ~10 Ω·μm², or even ≤ ~1 Ω·μm².

Independent claims (reconstructed — medium confidence)

The '261 patent's independent claims center on the "metal oxide" species of interface layer, with a 0.1–5 nm thickness range. As recited in IPR2020-01183 exhibits:

• Claim 1 — "An electrical junction comprising a metal oxide interface layer disposed between a contact metal and a group IV semiconductor, the interface layer configured to reduce a height of a Schottky barrier between the contact metal and the semiconductor relative to a contact junction without the interface layer, wherein the interface layer has a thickness of approximately 0.1 nm to 5 nm."
  • Dependent claims add: n-type source/drain of a transistor (2); contact metal is a metal or stack of metals deposited on the interface layer (3); contact metal does not form a silicide (17); contact metal does not react with the group IV semiconductor (18).
• Claim 21 — "An electrical junction comprising a conductor; a semiconductor; an interface layer comprising a metal oxide and having a thickness of approximately 0.1 nm to 5 nm." Dependents: n-type source/drain (22); metal or stack of metals (23); n-type doped group IV semiconductor (24).

Uncertainty: I could not confirm from the authoritative text whether the patent contains further independent claims beyond these (e.g., a claim reciting a metal-oxide separation layer plus a semiconductor-oxide passivation layer with a specific contact resistivity below 1 Ω·μm²). I also cannot confirm the total claim count. Treat the above as the claims actually challenged/presented in the IPR record, not as an exhaustive listing.

Litigation / PTAB posture (this is material)

• District court: Acorn Semi, LLC v. Samsung Electronics Co., Ltd., No. 2:19-cv-00347-JRG (E.D. Tex.). Jury verdict May 19, 2021 — $25M across four asserted patents (the '261 among the family). Case dismissed with prejudice by stipulation on January 6, 2023 after settlement; attorney-fee motion denied Feb. 2022.
• PTAB: IPR2020-01183 (Samsung), filed Jun. 24, 2020; instituted Feb. 10, 2021; Final Written Decision Feb. 9, 2022, holding claims 1–3, 17–18, and 21–24 unpatentable (Grupp '483 anticipation and/or Jammy obviousness). A termination date of Feb. 9, 2022 is recorded for the proceeding.
• CAFC: Appeal No. 22-1499 (E.D. Tex. origin) — voluntarily dismissed with prejudice on Jan. 6, 2023; no merits decision issued.
• On your "CAFC 2026 dockets" request: I found no 2026 Court of Appeals docket activity for this patent number. The only Federal Circuit matters I could identify are 22-1499 and related 2022-1539, both resolved in 2023. If you need a definitive 2026 check, the CAFC docket and PACER should be queried directly — my search coverage of CAFC 2026 filings is not exhaustive.

Explicit confidence caveats

• High confidence: patent number, title, inventors, application number 14/743,916, filing date 2015-06-18, issue date 2015-12-08, priority 2002-08-12, continuity chain, the printed abstract, and the litigation/IPR chronology above.
• Medium confidence: the exact wording of the independent claims (reconstructed from IPR exhibits), and the total number of claims.
• Could not verify: any 2026 CAFC docket entry; the current post-IPR claim status as reflected on a certificate of cancellation (Google Patents still lists the patent as "Expired – Fee Related" and does not reflect cancellation of the IPR-held claims).

I'll search for litigation involving US patent 9209261 on patent litigation tracking sites.

Let me dig deeper into the specific cases and their outcomes.

Let me verify whether the '261 patent was asserted in the district court case and find the IPR outcome.

Litigation involving U.S. Patent No. 9,209,261 ("the '261 patent")

I ran searches against PTAB/litigation trackers (Unified Patents data surfaced via Google Patents and docket aggregators, Docket Alarm, CourtListener, Justia, Law360, and the CAFC opinions site). Here is what I can confirm for the specific number 9,209,261 (the '261 patent — "Method for depinning the Fermi level of a semiconductor at an electrical junction and devices incorporating such junctions," inventors Grupp and Connelly, originally assigned to Acorn Technologies, Inc., later Acorn Semi, LLC).

Note: I am excluding results for similar-but-different patents in the same Acorn family (e.g., 8,766,336; 9,461,167; 9,905,691; 10,090,395), which appear in many of the same documents but are not the '261 patent.

1. District court litigation — Acorn Semi, LLC v. Samsung Electronics Co., Ltd., et al.

Sources: https://www.docketalarm.com/search/?q=patent%3A%289209261%29 ; https://www.law360.com/cases/5db0c6554d6a2a05d0f181ea/articles ; https://tmsnrt.rs/342hBiq (verdict form) ; https://www.docketalarm.com/cases/Texas_Eastern_District_Court/2--19-cv-00347/Acorn_Semi_LLC_v._Samsung_Electronics_Co._Ltd._et_al/docs/409.pdf (fees order, Feb. 17, 2022).

Caveat on the '261 patent's role: The verdict form lists only the '336, '167, '691, and '395 patents — the '261 patent is not among the "Patents-in-Suit" tried to the jury. Acorn's own filing to the PTAB stated that "Acorn is no longer asserting the '261 Patent in the litigation" (Patent Owner Comments to Stipulation, IPR2020-01183, dated Dec. 30, 2020). So the '261 patent appears to have been asserted in the original 2019 complaint but dropped/narrowed out before trial. I could not independently retrieve the original complaint itself to quote the exact '261 count, so treat the "asserted then dropped" characterization as reported-but-not-fully-verified.

2. Federal Circuit appeal — Acorn Semi, LLC v. Samsung Electronics Co., Ltd.

Sources: https://www.cafc.uscourts.gov/01-06-2023-22-1499-acorn-semi-llc-v-samsung-electronics-co-ltd-order-22-1499-order-1-6-2023_2059244/ ; http://cafc.uscourts.gov/opinions-orders/22-1499.ORDER.1-6-2023_2059244.pdf ; https://dockets.justia.com/docket/circuit-courts/cafc/22-1499

This appeal arose from the district court case above. As with the trial, the appellate briefing centered on patents other than the '261; I found no indication the '261 patent was itself on appeal in 22-1499.

3. PTAB inter partes review — Samsung Electronics Co., Ltd., et al. v. Acorn Semi, LLC

Sources: https://ipverse.greyb.com/ptab-web/cases/case-details/IPR2020-01183 ; https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1557765](/patent/1557765)/... ; https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1538402](/patent/1538402)/... (Order granting request for rehearing) ; https://www.law360.com/ptab_cases/5f08d5249269fa00e93cdd30/articles

Note: PTAB IPRs are contested administrative proceedings, not district-court litigation, but they are the central adversarial proceeding specifically directed at the '261 patent. The Google Patents family/litigation data for US9209261 flags IPR2020-01183 as a "Final Written Decision" matter (https://patents.google.com/patent/US9209261/en).

Summary

• District court: One identified case touching the '261 patent — Acorn Semi, LLC v. Samsung Electronics Co., Ltd., et al., No. 2:19-cv-00347-JRG (E.D. Tex., filed Oct. 23, 2019). The '261 patent appears to have been in the original six-patent assertion but was not among the four patents tried; the case went to a $25M jury verdict (May 19, 2021) on other patents and was settled (~Jan. 2023).
• Appeal: Acorn Semi, LLC v. Samsung Electronics Co., Ltd., No. 22-1499 (Fed. Cir., filed Feb. 25, 2022) — dismissed with prejudice Jan. 6, 2023 on the settlement.
• PTAB: Samsung Electronics Co., Ltd., et al. v. Acorn Semi, LLC, IPR2020-01183 — FWD Feb. 9, 2022 holding claims 1, 2, 17, 18 unpatentable (with claims 3 and 21–24 previously disclaimed by the patent owner).

Limitations on this answer

• I did not find any other district-court case (e.g., against other defendants) asserting the '261 patent, nor any ITC investigation, nor an appeal from IPR2020-01183. I cannot rule out additional suits by Acorn/Acorn Semi or its successors (e.g., Acorn Semi, LLC / Oak IP, LLC) that are not captured in the trackers I searched; a full PACER/Unified docket check would be needed to be exhaustive.
• The exact count(s) of the '261 patent asserted in the 2019 complaint, and the precise date Acorn dropped it, are from Acorn's PTAB comments rather than from the complaint itself, which I could not retrieve here.

I'll start by surfacing the PTAB record. The structured block says the ODP feed shows no AIA trials, but the patent-family metadata embedded in the full text flags an IPR and related litigation — so I need to reconcile those before answering.

The embedded family record flags IPR2020-01183 (Samsung v. Acorn Semi) and a CAFC appeal, even though the ODP block says no AIA trials. Let me pin down the claim-level outcome and appeal.

I have the proceeding identified. Now I need the claim-level FWD verdict and the appeal disposition.

PTAB / Federal Circuit Record — US 9,209,261 ("Method for depinning the Fermi level…", Acorn Semi, LLC)

Flag on the structured data first. The "PTAB proceedings on file" block says the USPTO ODP feed returns no AIA trials for this patent and that the default should be "no PTAB activity." That default is wrong for this patent. Web sources decisively document one AIA trial on the '261 — IPR2020-01183 — which was instituted and decided by Final Written Decision. The ODP ingest appears stale/incomplete for this family; per operating rules I'm preferring the search record over the stale feed and flagging the contradiction rather than suppressing it. (Embedded family metadata in the patent record itself also lists IPR2020-01183 as "PTAB case … filed (Final Written Decision)," corroborating the discrepancy.)

Proceedings overview

One AIA trial on file for the '261: IPR2020-01183 (Samsung v. Acorn Semi) — status: Final Written Decision issued 2022-02-09, with the challenged independent/dependent claims held unpatentable and the challenged-claim set narrowed by Acorn's own disclaimer. Breakdown: 0 active, 1 claims-invalidated, 0 sustained (as to the claims actually decided), 0 settled, 0 institution-denied on this patent — noting that Samsung filed ten IPRs across six sibling Acorn patents, three of which were denied institution. Bottom line for a defendant: this is not a "hardened patent" — it is an expired patent (anticipated expiration 2022-08-12, status "Expired – Fee Related") whose challenged claims went down in IPR, and which Acorn stopped asserting in the parallel litigation. Any demand letter resting on the challenged claims of the '261 deserves immediate skepticism.

IPR2020-01183 — Samsung Electronics Co., Ltd. v. Acorn Semi, LLC

• Type: Inter Partes Review (35 U.S.C. §§ 311–319)

• Filed: 2020-06-24 (petition filing date, per PTAB docket aggregators). Note: one secondary source (Law360) lists a 2020-06-29 filing date. Treat 2020-06-24 as the operative filing date but recognize the discrepancy.

• Status: Final Written Decision (verbatim). Plain English: tried to conclusion; FWD on 2022-02-09 (Paper 52), one day inside the statutory 12-month deadline from institution (institution 2021-02-10 → FWD due ~2022-02-10).

• Judge panel: The Acorn-family panels reported for these related IPRs were Administrative Patent Judges Brian J. McNamara, John R. Kenny, and Aaron W. Moore (that is the recorded panel on the companion FWD in IPR2020-01279), and Judge Kenny is on the record presiding at the consolidated oral hearing that covered IPR2020-01183 (Oct. 2021 hearing transcript). ⚠️ I could not independently confirm the exact three-judge caption of the -01183 FWD in this session — treat the panel attribution as very likely but not verified to the caption.

• Petition grounds (from the FWD's own grounds table — quote these claim numbers exactly):

  Claim(s) challenged	Statute	Reference / basis
  1–3, 17–18, 21–24	§ 102	Grupp '483 (U.S. Pat. 7,084,423, same inventors/spec — asserted only by first breaking Acorn's priority chain via a § 112 written-description/enablement theory)
  1–3, 17–18	§ 103	Jammy
  21–24	§ 102	Jammy

  Acorn separately argued Challenge 1 violated § 311(b) (that the "anticipation" challenge was really a § 112 argument dressed up as § 102), in addition to its written-description defenses on "metal oxide" and "a Group IV semiconductor."

• Institution decision: Instituted 2021-02-10, on all grounds asserted. (FWD: "In the Decision to Institute, trial was instituted on all grounds asserted in the Petition.") Acorn had pressed Fintiv denial (co-pending E.D. Tex. litigation, no stay — the court denied Samsung's stay motion), and disputed the scope of Samsung's stipulation; the Board did not deny institution.

• Final Written Decision (2022-02-09, Paper 52): The Board found claims 1–3 and 17–18 unpatentable. The decision text I retrieved shows the Board resolving the Jammy § 103 challenge against those claims — holding that "Petitioner has demonstrated that Jammy would have suggested the limitations of challenged claim[s]," and crediting the parties' experts that a POSITA would have understood Jammy's thin oxide layer to reduce Schottky barrier height as claimed ("In view of the experts' consistent testimony that the presence of a thin oxide layer would reduce the Schottky barrier, we find that Petitioner has demonstrated a person of ordinary skill would have known that the structure taught by Jammy … would have resulted in reducing the [barrier height]"). The Board rejected Acorn's objective-indicia (secondary-considerations) defense for lack of nexus, finding "insufficient evidence of a nexus between the long felt [need] and the specific limitations recited in claims 1–3 and 17–18."
  ⚠️ Honest limit on this answer: I verified the Board's Jammy-obviousness reasoning and the claim numbers 1–3 and 17–18 in the FWD text, and that challenge 3 (Jammy § 102 over claims 21–24) became moot due to Acorn's disclaimer. I could not retrieve the FWD's verbatim treatment of the Grupp '483 § 102 challenge (challenge 1) in this session, so I will not assert which party prevailed on that separate ground or its precise claim-by-claim disposition. Do not represent to a court that the FWD held claims "anticipated by Grupp '483" — that specific holding is unverified here.

• Settlement / termination: No settlement of this IPR. The case ran to an FWD, not a termination. (The parties did later resolve the district-court enablement dispute — Acorn's expert listing notes a "Settlement with Samsung on enablement before trial" — but that is the litigation, not the IPR.)

• Appeal: I found no Federal Circuit appeal of the IPR2020-01183 FWD. CAFC activity in this family is on two other tracks: 22-1499, Acorn Semi, LLC v. Samsung Electronics Co., Ltd. — origin DCT (appeal from the E.D. Tex. judgment; nonprecedential ORDER posted 2023-01-06) — and 22-1539, Samsung's appeal of the sibling Acorn IPRs (IPR2020-01206/01279/01207/01282). Because the '261 was dropped from the litigation and the FWD cut against Acorn, Acorn had little to appeal on the '261. ⚠️ Confirm no later-filed '261-specific appeal via the CAFC docket before relying on this.

• Defensive value: The '261 is expired as of 2022-08-12 and its challenged claims (1–3, 17–18) were held unpatentable, with claims 21–24 mooted by disclaimer — so a demand letter built on those claims is on very thin ice. Exposure is limited to past damages within the § 286 six-year lookback for pre-expiration conduct; there is no injunctive or ongoing-royalty theory.

Strategic summary

Claim status on the '261. Only three claim groups appear in the IPR2020-01183 ground table: 1–3, 17–18, and 21–24. Claims 1–3 and 17–18 are held UNPATENTABLE (Jammy § 103). Claims 21–24 were the subject of Acorn's disclaimer and the corresponding Jammy § 102 challenge was rendered moot — i.e., those claims no longer exist to be asserted. Any claim of the '261 outside those groups was not challenged in -01183 and is UNTESTED at the PTAB — do not assume it was invalidated, and do not assume it survived; the record simply doesn't reach it. Importantly, the whole enterprise is now largely academic for prospective conduct: the patent expired 2022-08-12 and is recorded Expired – Fee Related.

Estoppel landscape. Under § 315(e)(2), Samsung and its real parties in interest/privies are estopped in the district court from asserting any ground they raised or reasonably could have raised in -01183 (Grupp '483, Jammy, and the § 112/priority predicates). That estoppel does not run against the world — a different defendant is free to raise Grupp '483, Jammy, or fresh art, subject only to its own § 315(b) one-year bar and to the usual § 325(d) discretion. Note the asymmetry Samsung created: it stipulated not to pursue § 102/§ 103 invalidity in the district court (papers 18/19), which is why Acorn fought the stipulation's breadth so hard — the practical effect is that the winning IPR art is largely walled off from Samsung in court, but not walled off from a new defendant. For a defendant today, the IPR-based defense is the cheapest path, and the FWD itself is now the roadmap.

Pattern signals. This is a single-petitioner, high-volume campaign: Samsung filed ten IPR petitions against six Acorn patents (the "-0118x / -0120x / -0128x" cluster), with three institution denials and four+ FWDs, all running in parallel with E.D. Tex. case 2:19-cv-00347-JRG that produced a $25M jury verdict (Acorn's counsel summary: "the Board's decisions have confirmed the patentability of three of the eight infringed claims"). No defensive aggregator (e.g., Unified Patents) appears in the -01183 chain — the petitioner is Samsung, represented by Desmarais LLP; Acorn is represented by Laurence & Phillips. Acorn litigated aggressively (appointments-clause and "structurally biased Office" arguments appear in its briefing) and, on the patent family level, pursued CAFC appeals (22-1499 from the district court; 22-1539 by Samsung on the sibling IPRs). Family-level litigation links: CAFC 22-1499, E.D. Tex. 2:19-cv-00347, and the PTAB campaign above.

Recommended next steps

1. If you're a defendant and a demand cites the '261: anchor your response in the IPR2020-01183 Final Written Decision (Paper 52, 2022-02-09), which held claims 1–3 and 17–18 unpatentable and left claims 21–24 mooted by Acorn's disclaimer. Pull the FWD from PTAB E2E (proceeding record: https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1557765](/patent/1557765) ; mirrored docket: https://www.docketalarm.com/cases/PTAB/IPR2020-01183/) and verify the Grupp '483 § 102 holding in the four corners of the document before quoting it — I verified the Jammy reasoning but not that ground (see caveat above).
2. Lead with expiration. The '261 shows an anticipated expiration of 2022-08-12 and "Expired – Fee Related" status. That caps exposure to pre-expiration past damages; push for a § 286 six-year-lookback damages model and treat any ongoing-royalty demand as meritless.
3. No active PTAB proceedings to track. There are no pending trials on the '261 — no institution deadlines, no FWD dates to diarize. The "no active proceedings" conclusion is itself the signal: the patent has been exhausted as an IPR target and as a litigation asset (Acorn dropped it from the Samsung case).
4. Check the family before assuming 261 is the only threat. Acorn's sibling patents (8,766,336; 9,461,167; 9,905,691; 10,090,395; 7,084,423, etc.) were the real battleground — several had only partial adverse FWDs (e.g., in the '691/'395 line the Board sustained certain claims, including the "specific contact resistivity < 1 Ω·μm²" claims, at least at the Board level). If your products implicated the '261, map the family and confirm which sibling claims, if any, are still live and unexpired.
5. If you intend to raise an IPR: your §§ 102/103 grounds are not estopped (estoppel binds Samsung/privies, not you), and Samsung's district-court stipulation does not bind you. But confirm the § 315(b) one-year clock from service of any complaint, and note that with the patent expired, an IPR may be a poor use of budget versus a damages-side defense.

I'll research the assignment record for this patent before reconstructing the chain.

Ownership Chain — US 9,209,261 B2

Source note / data gap (read first). The authoritative full text I hold for this patent is the Google Patents fetch, whose legal-events table gives executed/recorded dates and assignor/assignee names but not reel/frame numbers or the correspondent of record. My live lookups for the Assignment Center record (reel/frame and recording correspondent) returned no usable hit before my search budget was exhausted. I therefore reconstruct the chain from the Google Patents legal-events entries and corroborating news/litigation sources, and I flag every field I could not verify rather than filling it in. Treat reel/frame and correspondent as unverified below.

Inventors

• The priority application (Ser. No. 10/217,758, filed 2002-08-12, now US 7,084,423) lists ACORN TECHNOLOGIES, INC. as applicant with Grupp and Connelly as the inventors (PCT publication WO 2006/026339 and EP 1 535 343 both show Acorn Technologies, Inc. as applicant; Grupp and Connelly as inventors). Both men are described in Acorn's own press material as "Dr. Dan Grupp and Dr. Dan Connelly" — the inventors behind the asserted family.
• No departure pattern is discernible. Both named inventors remained tied to the Acorn family across its entire 2002–2025 prosecution history (they recur as inventors on later Acorn/Acorn Semi filings). Nothing indicates either inventor left the assignee within 12 months of filing, and there is no fire-sale-foreshadowing signal here.
• Note: a third name ("CLIFTON") appears in the PCT applicant data, but Paul Alan Clifton is the technical expert deposed in IPR2020-01183, not a named inventor on the '261 patent. The '261 names only Grupp and Connelly.

Original assignee

• Acorn Technologies, Inc. (named on the '261 patent and its ancestors; recorded addresses drift over time — Santa Monica, CA on the '003 ancestor, Pacific Palisades, CA on the EP filing, La Jolla, CA by 2021).
• Primary line of business: semiconductor and telecommunications technology development / IP cores, algorithms, and processes — Acorn self-describes as "a product-oriented company … since 1998," but its output is licensed IP cores and designs, not merchant silicon. It is best characterized as a R&D-and-licensing house, not a volume chipmaker.
• Product embodying the claims: I found no evidence that Acorn ever shipped a commercial chip practicing the claimed metal–interface–semiconductor junction. The value it extracted from the family was through licensing and litigation, not product sales.
• Current status: operating, but its semiconductor patent portfolio was divested. Acorn is a parent of Acorn Semi, LLC and PHY Wireless. Per GreyB and the IPR record, in December 2024 Acorn Semi sold 36 semiconductor patents to Oak IP, LLC (a Dominion Harbor–linked entity). Acorn Semi nonetheless continues to file new applications (Minesoft shows Acorn Semi filings into late 2025/2026), so the entity is alive. Chairman/investor is Peter Norton, whose living trust held a security interest in the portfolio.

Assignment timeline

Every reel/frame and correspondent field below is UNVERIFIED. Dates and party names are from the Google Patents legal-events table (derived from USPTO assignment records); the Assignment Center's reel/frame and correspondent-of-record were not retrievable in this run.

• Executed ~2015-06 / recorded 2015-06-22 — Reel not retrieved/****

  • Conveyance: Assignment of assignors' interest
  • Assignor: Grupp, Daniel E.; Connelly, Daniel J. (inventors)
  • Assignee: Acorn Technologies, Inc.
  • Correspondent: not retrieved — flag as data gap.
  • Context: Original inventor-to-company assignment at the time the continuation (Ser. No. 14/743,916) was filed; confirms title in the operating company at issuance.

• Executed ~2019-05 / recorded 2019-05-30 — Reel not retrieved/****

  • Conveyance: Security Interest (Security Agreement)
  • Assignor: Acorn Semi, LLC
  • Assignee: The Peter Norton Living Trust dated April 28, 1989
  • Correspondent: not retrieved
  • Context: Securitization / venture-style secured financing — the chairman's trust takes a security interest in the patent collateral. Recorded before the asset transfer below, consistent with a blanket lien over Acorn's patent collateral, including assets about to move to Acorn Semi.

• Executed ~2019-06 / recorded 2019-06-26 — Reel not retrieved/****

  • Conveyance: Assignment of assignors' interest
  • Assignor: Acorn Technologies, Inc.
  • Assignee: Acorn Semi, LLC
  • Correspondent: not retrieved
  • Context: Internal reorg into the assertion vehicle. Acorn Semi is a wholly owned subsidiary of Acorn Technologies, organized in Delaware, principal place of business 445 Cambridge Ave., Suite A, Palo Alto, CA (per Acorn Semi's mandatory notice in IPR2019-01182). This transfer lands the '261 in the entity that filed the Samsung suit in October 2019 — roughly 3–4 months later.

• Executed ~2024-12 / recorded 2024-12-26 — Reel not retrieved/****

  • Conveyance: Assignment of assignors' interest
  • Assignor: Acorn Semi, LLC
  • Assignee: Oak IP, LLC
  • Correspondent: not retrieved
  • Context: Transfer-to-asserter / bulk portfolio sale. Oak IP is described by GreyB as "tied to Dominion Harbor" and acquired 36 Acorn Semi semiconductor patents in December 2024; it sued GlobalFoundries on 2025-02-04. The '261 is explicitly named by GreyB as one of the six portfolio patents previously asserted against Samsung.

• Executed ~2024-12 / recorded 2024-12-26 — Reel not retrieved/****

  • Conveyance: Release by Secured Party
  • Assignor: The Peter Norton Living Trust dated April 28, 1989
  • Assignee: Acorn Semi, LLC
  • Correspondent: not retrieved
  • Context: Lien release — clears the 2019-05-30 security interest so the portfolio can be conveyed to Oak IP free of the trust's lien.

• Executed ~2024-12 / recorded 2024-12-26 — Reel not retrieved/****

  • Conveyance: Release by Secured Party
  • Assignor: The Peter Norton Living Trust dated April 28, 1989
  • Assignee: Acorn Technologies, Inc.
  • Correspondent: not retrieved
  • Context: Lien release — the same secured party also releases its interest back to the parent, confirming the 2019 lien was a blanket encumbrance on both Acorn entities' patent assets.

• 2022-08-12 — Legal status event, not an assignment: statutory term/anticipated expiration recorded; '261 then shows "Expired – Fee Related."

• Litigation cross-reference (no assignment): IPR2020-01183 FWD 2022-02-09 held claims 1–3, 17–18, 21–24 unpatentable; E.D. Tex. 2:19-cv-00347 verdict 2021-05-19 ($25M, four patents incl. siblings); CAFC 22-1499 voluntarily dismissed 2023-01-06.

Timeline diagram

timeline
    title Ownership of US 9209261
    2002 : Priority application filed by Acorn
    2015 : Continuation filed
         : Inventors assign to Acorn Technologies
    2019 : Security interest to Peter Norton Trust
         : Parent transfers patents to Acorn Semi LLC
         : Acorn Semi sues Samsung
    2021 : Jury verdict of 25M against Samsung
    2024 : Acorn Semi sells portfolio to Oak IP LLC
         : Security interests released
    2025 : Oak IP sues GlobalFoundries

NPE / troll-pattern signals

1. Shell-entity transfer — UNCLEAR (leaning present).

   • Acorn Technologies → Acorn Semi, LLC (recorded 2019-06-26) moves the '261 into a Delaware LLC with a Palo Alto address. That LLC is, on the record, a wholly owned subsidiary of an operating parent — not an orphan shell. However, it functions as a single-purpose assertion vehicle (it is the named patent owner in every IPR and in the Samsung suit, while the parent holds the R&D business). Supporting datapoint: the patent then moves to Oak IP, LLC, a licensing-only entity. I cannot confirm member structure or registered-agent-service address without the unretrieved reel/frame, so this stays unclear rather than a clean "present."

2. Known asserter in the chain — PRESENT (strong).

   • Current assignee is Oak IP, LLC, recorded 2024-12-26 (Google Patents legal events). Oak IP is reported "tied to Dominion Harbor" (GreyB, December 2024 patent transactions) and filed suit against GlobalFoundries on 2025-02-04 asserting the acquired Acorn portfolio. Dominion Harbor is a high-frequency patent plaintiff/monetization firm surfaced by Unified Patents and RPX. This is a repeat-player monetization entity, not an operating company.

3. Repeat correspondent across the chain — NOT ASSESSABLE (data gap).

   • The correspondent of record for each recording was not retrievable in this run. I will not name a firm by inference from litigation counsel (Laurence & Phillips IP Law appeared as patent-owner litigation counsel in IPR2020-01182/01183, and Hueston Hennigan LLP represented Acorn at the E.D. Tex. trial — neither is the assignment-recording correspondent). Mark unclear and re-query the Assignment Center with the reel/frame to populate.

4. Cascading transfers — NOT PRESENT as a <24-month cascade.

   • The chain is Acorn Technologies (2015) → Acorn Semi (2019-06-26) → Oak IP (2024-12-26). Only two post-original transfers, ~5.5 years apart; well outside the <24-month cascade window. The 2019 and 2024 records share timestamps (three events all recorded 2024-12-26), but that is a bulk-recording artifact of the December 2024 divestiture, not a rapid cascade of unrelated LLCs.

5. Pre-litigation transfer — PRESENT (strong).

   • Two instances. (a) The Acorn Semi assignment recorded 2019-06-26 precedes the October 2019 Samsung complaint by ~3–4 months (within 6 months) — arranging the portfolio into the asserting entity. (b) The Oak IP acquisition dated 2024-12-26 precedes the 2025-02-04 GlobalFoundries complaint by ~5–6 weeks — a textbook acquire-then-assert sequence.

6. Bankruptcy fire-sale — NOT PRESENT.

   • No Chapter 7/11 filing by Acorn is evidenced. The December 2024 divestiture is a negotiated portfolio sale, not a court-supervised asset sale. Distinguish from true bankruptcy fire-sales (Kodak, Nortel, Polaroid).

7. Privateering — UNCLEAR (leaning present).

   • The pattern fits: an operating/R&D company (Acorn Technologies/Acorn Semi) conveys a 36-patent semiconductor portfolio to a third-party monetizer (Oak IP/Dominion Harbor) that then sues the industry (GlobalFoundries) — the classic privateering shape. What I cannot confirm is whether Acorn retained a back-end royalty interest or financed the sale (which would clinch privateering) versus making a clean cash sale. No SEC filing exists to check, because Acorn is private. Mark unclear, supported by the 2024-12-26 assignment plus the 2025-02-04 suit.

8. Defensive aggregator — NOT PRESENT.

   • The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. The current owner (Oak IP) is offensive, actively litigating. No neutralizing signal.

Verdict

NPE — high confidence.

Two independent strong signals anchor this call: (1) the current assignee, Oak IP, LLC, per the 2024-12-26 recorded assignment, is a Dominion-Harbor-linked monetization entity that sued GlobalFoundries on 2025-02-04; and (2) there are two pre-litigation transfers — the 2019-06-26 assignment to Acorn Semi, LLC ~3–4 months before the October 2019 Samsung complaint, and the 2024-12-26 assignment to Oak IP ~5–6 weeks before the February 2025 GlobalFoundries complaint. The 2019-05-30 security interest to the Peter Norton Living Trust and its 2024-12-26 releases show the portfolio was encumbered and then sold off cleanly to the monetizer. Caveat driving the "verification needed" flag: reel/frame and correspondent of record were not retrieved, so signal 3 (repeat correspondent) could not be scored.

Verify at: https://assignment.uspto.gov/patent/index.html — search patent number 9209261 (Assignment Center: https://assignmentcenter.uspto.gov/). Populate the reel/frame and correspondent fields, then re-check signal 3.

I'll build on the previously generated verification section, which already established the bibliographic record and litigation posture. The new task is the prior-art citation analysis — and I need to flag a gap up front: the authoritative Google Patents text I was given truncates mid-specification and omits the "References Cited" / Patent Citations tables. So I must retrieve the citation lists from live sources before analyzing them.

I have the IPR prior-art references, but the task specifically asks about the patent's own "References Cited" citations. That section is missing from my authoritative fetch. Let me retrieve it.

The Google Patents citation table isn't surfacing through search snippets. Let me try the FreePatentsOnline and Justia mirrors, which reproduce the printed "References Cited" section.

Prior-Art Citation Analysis — US 9,209,261 B2

Step-limit / provenance caveat (read first). The authoritative full text supplied for this patent truncates mid-specification and does not include the printed "References Cited" (front-page citation) tables. I therefore could not read the official citation list off the patent document itself. I reconstructed it from PTAB exhibits that reproduce the '261 front page, principally Ex. 1141 in IPR2020-01241 (which contains the '261 "(56) References Cited" block) and the sibling family members (US 7,884,003; US 9,812,542; US 10,388,748) whose IDS lists are near-identical. I have flagged below what is confirmed vs. reconstructed. I did not fabricate any references, numbers, or dates.

1. Two different "prior art" sets — do not conflate them

The task asks for set A "for each patent citation." I give set A below, then explain why set B is the answer that actually anticipates.

2. Set A — the '261 "References Cited" (confirmed entries)

U.S. Patent Documents (as reproduced in PTAB Ex. 1141):

Foreign patent documents (confirmed): EP 0295490 A1 (12/1988); EP 0789388 A2 (8/1997).

Other publications (NPL, confirmed on the '261 face / IPR record):

• Wang, Lie, et al., "High Barrier GaN Schottky Diodes: Pt/GaN and Pd/GaN," Appl. Phys. Lett. 68(9), pp. 1267-1270 (Feb. 26, 1996) — directly relevant to the patent's high-barrier-height embodiment.
• "Research Achievements," RCIQE (Hokkaido Univ.), downloaded Apr. 12, 2002.

Note on the additional entries seen in the sibling US 10,388,748 list (e.g., 4,358,616 Boone; 4,999,858 Waldrop; 5,191,322; 5,216,262 Tsu; 5,216,282; 5,243,213; and pub. 2005/0009302 Wakui; 2005/0037557 Doczy; 2011/0092047 Gaines) — I could not confirm each of these appears on the '261 face specifically; the family lists diverge by case. Treat them as family-list, not verified '261 citations.

3. Set B — the art that actually killed the claims (this is the §102 answer)

From the record surfaced here (Schubert declaration, Ex. 1018–1036; Demonstrative Ex. 2124):

• U.S. Pat. No. 7,176,483 ("Grupp '483") — the primary anticipation reference. Ex. 1028 is "a redline of the specification of Grupp '483 against U.S. Patent No. 7,084,423," i.e., petitioner used it to show what "Grupp '483" discloses relative to the '423 parent. Under §102, Grupp '483 is the reference that potentially anticipates claims 1–3, 17–18 and 21–24.
• "Jammy" — the secondary reference used for the obviousness alternative on the same claims. ⚠ I could not recover the full citation (number/date) for the "Jammy" reference with confidence, so I will not state one.
• Supporting §103/§102 backdrop exhibits: Rhoderick (1982); Chang (1971); Goodnick (1981); Sobolewski & Helms; US 4,110,488 (Risko); Taubenblatt (1984); US 3,983,264 (Schroen); US 4,845,050 (Kim); JPH11162874A (Iwaguro); O'Donnell (1991); Lang & Kohn (1971); Zheng (2005); US 5,216,282 (Cote); Cohen (1986); Yen (1980); Polishchuk (2001); Glass (1991).

Note that Schroen '264 and Cote '282 also sit on the '261 face, so those two are the clearest overlaps between Set A and Set B.

4. ⚠ Contradiction / priority-date flag (important)

The previously generated section states the FWD held claims unpatentable on "Grupp '483 anticipation." That requires scrutiny and I flag it explicitly:

1. "Grupp '483" is a same-inventor (Grupp et al.) document whose specification is redlined against the '423 parent — i.e., it is family, with the same 2002-08-12 priority. A same-priority family member cannot be §102 prior art to the '261 absent a priority defeat.
2. The only way Grupp '483 anticipates is if the '261's "metal oxide" limitations (the claims recite a metal oxide interface layer at 0.1–5 nm — corroborated by the Google Patents concept extraction showing "metal oxide … claims 5") are not supported by the 2002 parent and thus lose the 2002 date. That is consistent with petitioner's Exhibit 1028 redline and with the patent owner's evidence (Kuhn Declaration, IPR2020-01282, Ex. 2070: "…contains a written description of a metal oxide spacer layer in the claimed…"). The anticipation holding therefore turns on a priority/§102(e) analysis, not on 102(b) art.
3. Action: verify the exact identity/priority of "Grupp '483" and the "Jammy" citation against the IPR2020-01183 Final Written Decision (Feb. 9, 2022) before relying on the §102 mapping. I could not confirm either number with high confidence here, and I will not invent them.

5. Confidence summary

• High confidence: that the '261 front page carries a long inherited citation list (Set A) including Schroen '264, Risko '488, Kim '050, Cote '282, EP 0295490, EP 0789388, the Wang GaN paper, and the Grupp family patents; and that claims 1–3, 17–18, 21–24 were held unpatentable in IPR2020-01183.
• Medium confidence: the completeness of the Set A list (reconstructed from an exhibit snippet, not the printed patent).
• Low / could not verify: the full citation for "Jammy"; whether the "Grupp '483" anticipation was a §102(e)/priority-defeat theory; the dozen-or-so additional references that appear only in sibling-family lists.

Bottom line for §102: Most of the '261's face citations are §102-eligible in date (all pre-2002) but are background contact/dielectric art that does not disclose the claimed metal-oxide interface layer of 0.1–5 nm that reduces Schottky barrier height; the only references actually mapped to the claims under §102 are Grupp '483 (and, on the obviousness alternative, Jammy), with Schroen '264, Risko '488, Kim '050, and Cote '282 serving as supporting §103 art.

I'll ground this analysis in the actual prior-art record for the '261 patent, which the fetched Google Patents text did not include (the fetch cut off before the References Cited / Cited By tables). Let me search.

Good — the prior-art record is coming into focus (Grupp '483 and Jammy, US 6,724,088). Let me pull the Board's actual obviousness reasoning and identify Grupp '483 precisely.

I have the core prior-art references. Let me pin down the exact claim text and the Board's Ground 2 motivation-to-combine reasoning.

Obviousness Analysis — U.S. Patent No. 9,209,261 B2 under 35 U.S.C. § 103

0. Source note and discrepancy flags (read this first)

On the "Prior Art section of this page." The Google Patents fetch supplied as the authoritative full text for US 9,209,261 does not contain the "References Cited" / "Cited By" tables — the fetch truncates mid-specification (it ends inside the discussion of "for a p-type doped silicon semiconductor, a metal may be selected that has a work function substantially equal to the conduction…"). I therefore could not use the page's own prior-art tables, and "US9209261 references cited" returned only unrelated documents. Instead, I used the operative prior-art record that actually exists for this patent: the IPR2020-01183 record (Petition, Board decisions, and the accompanying claim charts). Every prior-art reference below is sourced from that record with a URL.

Three contradictions between the previously generated sections and/or the primary source I retrieved. I flag them rather than paper over them:

The retrieved Board text explicitly captions its sections "C. Anticipation of Claim 1–3, 17–18, and 21–24 by Grupp '483" and "D. Claims 1–3 and 17–18 as Obvious Over Jammy." I cannot reconcile this with the litigation summary's disclaimer-based narrowing from the material in hand (the retrieved text may be the institution decision, or a version of the decision that predates the rehearing correction). What is not in dispute across all three sources: claims 1, 2, 17 and 18 were held unpatentable, and Grupp '483 and Jammy are the two references of record. My § 103 analysis proceeds on that undisputed base.

On "Grupp '483." The record labels the primary reference "Grupp '483" and one appendix states it "issued as U.S. Patent 7,176,483 on February 13, 2007," published as an application on May 5, 2005, and that it "is a continuation-in-part of an August 2002 application that issued as U.S. Patent 7,084,423" (the '423 patent). I note that Acorn's portfolio also contains a U.S. 8,415,483, so the "483" suffix is potentially ambiguous; I am using the number (7,176,483) exactly as the retrieved appendix states it, without correction, and flag medium confidence on the mapping.

1. Legal framework applied

Practical consequence: This is a § 103 case in which the challenger's principal reference is the patent family's own earlier patent. That materially weakens any "invention" narrative: what the '261 claims as a genus, its own earlier Grupp-family disclosure shows as a species.

2. The prior-art references in the record

Sources: Board text — https://ptacts.uspto.gov/ptacts/public-informations/petitions/1557765/download-documents?artifactId=LUGBxLj41B8frgWDK8aotdFt0bjWHb5WqfzcTObv560a0bDtjZrRcmU ; Petition — https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1538402](/patent/1538402)/download-documents?artifactId=-moaGp6J6Oc-7siDpzTV4X1iNyA8pg0sIm1nN89wMi9DbxInu6mMD9E ; Institution/PO papers — https://www.winston.com/a/web/[248275](/patent/248275)/[PTAB](/ptab)-IPR2020-01207-21.pdf ; case page — https://ipverse.greyb.com/ptab-web/cases/case-details/IPR2020-01183 .

3. Claim 1 — element-by-element mapping

Claim 1 (verbatim from the Petition's Appendix A, "The Challenged Claims of the '261 Patent"):

"1. An electrical junction comprising a metal oxide interface layer disposed between a contact metal and a group IV semiconductor, the interface layer configured to reduce a height of a Schottky barrier between the contact metal and the semiconductor from that which would exist at a contact junction between the contact metal and the semiconductor without the interface layer disposed therebetween, and wherein the interface layer has a thickness of approximately 0.1 nm to 5 nm."

Observations. (i) Every element of claim 1 is disclosed in Grupp '483, which the Board found anticipatory — anticipation being the "epitome" of obviousness (In re Slayer/Gosteli genus-species logic: TiO₂ is the species; "metal oxide" is the genus). (ii) For a § 103 case that does not depend on Grupp '483's status, Jammy alone supplies the thickness limitation and the junction architecture, and the remaining "metal oxide" element is supplied by the ordinary knowledge of the art plus any of the secondary references.

4. The Board's actual § 103 ground: Jammy renders claims 1–3 and 17–18 obvious

The Board captioned its obviousness analysis "D. Claims 1–3 and 17–18 as Obvious Over Jammy," and recited: "Jammy discloses 'improved contact between conductive studs and shallow diffusion regions by incorporation of a quantum conductive barrier layer at the interface…'" with the barrier layers being "<50 Å … about 5–30 Å … about 5–15 Å" films.

Why a POSITA would have been motivated to modify Jammy (or combine it with the metal-oxide teachings of the field):

1. Same field of endeavor and same problem. Both Jammy and the '261 address resistance at a metal-to-silicon contact, and both target source/drain / diffusion contacts in a MOSFET. KSR, 550 U.S. at 420 ("any need or problem known in the field of endeavor at the time of invention and addressed by the patent"). Nothing about the claimed junction takes it outside Jammy's field.
2. Jammy already places a sub-5 nm layer at the exact interface the claim recites. Jammy's 5–30 Å (0.5–3 nm) range literally overlaps the claimed 0.1–5 nm range. Where a prior-art range overlaps a claimed range, the difference is a matter of routine optimization — a strong prima facie case absent criticality.
3. The claimed "metal oxide" is a finite, identified, predictable set of materials for exactly this layer. The art (Grupp '483: TiO₂; Goodnick: Al₂O₃; Szydlo: TiO₂ on Au; Iwaguro: silicon sub-oxide) identifies the plausible candidates for a thin oxide interfacial layer at a metal–Si contact. That satisfies MPEP 2143 rationale (5) — "obvious to try" among a finite number of identified, predictable solutions with a reasonable expectation of success.
4. Predictable result (KSR rationales 1, 3, 4). The substitution of one thin, bulk-dielectric interfacial oxide for Jammy's nominally unspecified thin quantum-conductive layer is a "simple substitution of one known element for another to obtain predictable results," and is "use of a known technique to improve a similar device … in the same way."
5. Jammy itself signals that its barrier layer is meant to be used with other transistor structures (it incorporates Bronner by reference), so there is no structural incompatibility to overcome — the combination is a rearrangement of known parts with no change in their respective functions.

Counterargument and rebuttal (teaching away). Acorn argued below that the claimed interface layer was "unpredictable" and that inserting any layer at the contact was "counterintuitive" (because a dielectric adds resistance). That argument rebuts itself in the prior art: Jammy expressly teaches that a material that is a dielectric in bulk "in thin layers become[s] electrically conductive," and Grupp '483's Fig. 8 shows a resistance minimum as a function of interface-layer thickness — i.e., the very "Goldilocks" behavior the '261 describes. A reference that explains the mechanism the patentee relies on is not a reference that teaches away; it is the same teaching.

5. Additional § 103 combinations I would advance (independent of the Board's holding)

These are alternative routes to obviousness, useful if any single reference is knocked out.

Combination A — Jammy + Grupp '483 (the most robust)

Grupp '483 contributes: (a) the express "metal oxide interface layer" species (TiO₂ spacer, <1 eV barrier); (b) the express 0.1–5 nm thickness range; (c) the express functional result — "the effect of MIGS has been sufficiently reduced to depin the metal and lower the Schottky barrier" (Grupp '483 at 15:50–54); and (d) the group IV semiconductor list.
Jammy contributes: (a) the specific device context (source/drain of a MOSFET contacted by a conductive stud) and (b) an independent teaching that the interfacial layer thickness should be 5–30 Å.
Motivation: Grupp '483 explicitly frames its junction as applicable to "the source and drain contacts at the channel of a field effect transistor," i.e., to exactly Jammy's environment; and Jammy's issue date/filing predates and its subject matter complements Grupp '483. KSR "interrelated teachings" rationale. A POSITA would combine them because each reference improves the same property of the same structure by the same mechanism.

Combination B — Jammy + Goodnick (aluminium-oxide interfacial layer)

Goodnick's Al/Al₂O₃/SiO₂/Si study supplies a self-limiting thin metal-oxide (Al₂O₃) interlayer at a metal–Si-based junction, with aluminum also present — giving a metal-oxide interface layer within the claimed thickness range. Motivation: Goodnick explains that the metal-oxide layer forms a diffusion barrier that stabilizes the interface; a POSITA seeking a manufacturable, thermally stable thin oxide contact layer for Jammy's structure would look to precisely this teaching. (Note: in the sibling IPRs, Acorn's own expert attacked the Petitions' reading of Goodnick — that Al₂O₃ and SiO₂ "existed simultaneously" and that the "diffusion barrier" was "just a guess." That attack goes to what Goodnick discloses; it does not, in my view, negate the combination, because the metal-oxide-interface-layer element is independently supplied by Grupp '483's TiO₂ disclosure.)

Combination C — Jammy + Iwaguro

Iwaguro (JP application, 1999) describes interposing a silicon sub-oxide layer between a metal and silicon. Combined with Jammy's thickness teaching, this yields an oxide interface layer of 0.1–5 nm between a contact metal and a group IV semiconductor. Motivation: both address the same contact resistance problem with the same structural expedient.

Combination D — Jammy + Szydlo et al.

Szydlo reports Au/TiO₂ Schottky diodes — a metal-oxide layer at a metal–semiconductor junction whose presence governs the barrier. Motivation: a POSITA seeking to reduce/tune the Schottky barrier at a metal–Si contact (Jammy's stated goal) would be led to a metal-oxide interfacial layer for exactly that purpose.

Combination E — Jammy + Shannon (1976)

Shannon teaches reducing the effective Schottky barrier height by highly doped surface layers. Motivation: it establishes the claimed functional limitation ("reduce a height of a Schottky barrier … from that which would exist … without the interface layer") as a known, desired objective in the art, supplying the "reason why" for modifying any of A–D.

6. Dependent claims (2, 3, 17, 18)

Each dependent claim adds nothing beyond what Grupp '483 and/or Jammy disclose; there is no unexpected result and no teaching away in the record I retrieved.

7. Objective indicia (Graham factor 4)

• The record shows Acorn advanced secondary-consideration arguments (the Petition briefs a section "Alleged Secondary Considerations," Pet. at 55), including the "counterintuitive" nature of inserting a layer and alleged unpredictability.
• Nexus problem (decisive in my view): the challenged claims' entire subject matter — a metal oxide (TiO₂) interface layer, 0.1–5 nm, that lowers the Schottky barrier — appears in the family's own earlier Grupp '483 disclosure. Objective evidence tied to a feature already in the prior art lacks nexus. See Rambus v. Rea, 731 F.3d 1248, 1257 (Fed. Cir. 2013) ("known in the prior art"); Ormco v. Align, 463 F.3d 1299, 1312 (Fed. Cir. 2006). Any praise, licensing, or industry adoption of "an interface layer that depins the Fermi level" therefore cannot support patentability of claims that read on Grupp '483.
• I did not retrieve the Board's specific treatment of secondary considerations, so I cannot report its conclusion; I flag this rather than guess.

8. Bottom line

1. The strongest ground is not § 103 at all — it is the Board's § 102 finding that Grupp '483 anticipates claims 1–3, 17–18, and 21–24. Anticipation is a subset of "not patentable," and it carries a fortiori obviousness: a disclosed species (TiO₂ spacer layer, 0.1–5 nm, at a metal/group IV semiconductor junction, lowering the barrier) within the claimed genus ("metal oxide interface layer").
2. The Board also found a genuine § 103 ground: Jammy renders claims 1–3 and 17–18 obvious. The motivation is straightforward under KSR — same field, same problem, overlapping thickness ranges (5–30 Å vs. 0.1–5 nm), substitution of a known thin metal-oxide interfacial layer for Jammy's thin quantum-conductive layer, with predictable results.
3. Independent § 103 combinations that would also work: Jammy + Grupp '483 (most robust); Jammy + Goodnick; Jammy + Iwaguro; Jammy + Szydlo; and any of those + Shannon for the barrier-lowering function.
4. The decisive structural fact is self-collision. The '261 is a continuation of the same family as Grupp '483; the Board held the claims lost their pre-2011 priority precisely because the earlier applications' disclosure of a single metal-oxide (TiO₂) layer does not describe the claimed genus. That reasoning cuts both ways: the genus claim is unpatentable either (a) because the earlier art anticipates it, or (b) because, if the disclosure is too thin to support the genus, the genus is an obvious extension of the disclosed species over the whole field of metal oxides used at metal–semiconductor contacts.

Confidence: high that Grupp '483 and Jammy are the operative references of record and that Jammy was the § 103 basis; high that claim 1's elements as quoted are as stated; medium on the exact patent number underlying "Grupp '483"; medium on the total claim set and on the FWD's ultimate disposition of claims 3 and 21–24 (see the three-way contradiction flagged in § 0); low/none on whether any additional § 103 grounds were presented or decided that do not appear in the documents retrieved.