US 8766336

I'll search for authoritative information on US patent 8766336, including any 2026 CAFC docket activity.

Let me check for specific CAFC 2026 docket activity and confirm the claims.

US Patent 8,766,336 — Analyst Summary

Search scope note: I queried for this exact number (8,766,336 / US8766336B2). No CAFC 2026 docket activity involving this patent surfaced. The only Federal Circuit proceeding I can tie to this family is No. 22-1499 (listed on Google Patents' litigation timeline, arising from the Texas litigation). That is a 2022 appeal, not 2026. I did not find evidence of any 2026 CAFC case for '336, and the patent's nominal term has expired — treat any "2026 litigation" premise as unverified.

Bibliographic Data

Continuity: Continuation of Ser. No. 13/552,556 (filed 2012-07-18) → continuation of Ser. No. 13/022,522 (2011-02-07) → divisional of Ser. No. 12/197,966 (2008-08-25, now US 7,884,003) → divisional of Ser. No. 11/181,217 (2005-07-13, now US 7,462,860) → continuation of Ser. No. 10/217,758 (2002-08-12, now US 7,084,423). Related to Ser. No. 10/342,576 (now US 6,833,556).

Prosecution note (relevant to claim scope): On 2020-10-14, Acorn filed a statutory disclaimer under 37 CFR 1.321(a) disclaiming claims 9–12 and 65–76 of '336. Claims 1–8, 13–16, and 77–80 remained.

Abstract (verbatim)

"An electrical device in which an interface layer comprising arsenic is disposed between and in contact with a conductor and a semiconductor. In some cases, the interface layer may be a monolayer of arsenic."

Note: the granted abstract narrows to arsenic specifically, even though the specification describes a much broader class of interface materials (nitride, oxide, oxynitride, arsenide, hydride, fluoride, and metal oxides).

Plain-Language Overview of the Invention

The patent addresses "Fermi-level pinning" at metal–semiconductor contacts. Ordinarily, dangling bonds at the semiconductor surface plus metal-induced gap states (MIGS) lock the semiconductor's Fermi level near mid-gap, making the Schottky barrier height largely independent of which metal is used. The inventors insert a very thin interface layer between the metal and the silicon-based semiconductor that (i) passivates dangling bonds and (ii) physically displaces the metal far enough to suppress MIGS — thereby "depinning" the Fermi level so barrier height becomes tunable by metal choice. The layer must be thin enough to still pass current; FIG. 8 shows a resistance-vs-thickness curve with a minimum, i.e., a "Goldilocks" thickness.

Independent Claims — Plain-Language Overview

⚠️ Uncertainty flag: The claim text was not present in the full patent text supplied to me, and my searches did not return verbatim claim language. The following is reconstructed from the specification's stated embodiments and PTAB filings (IPR2020-01204 / IPR2020-01264), which identify the challenged independent claims as including at least claims 1 and 5, and describe the challenged limitations as reciting (1) specific contact resistance ranges, (2) "metal oxide" interface layers, (3) a stack of metals, and (4) a "group IV semiconductor." Verify exact wording against the USPTO PatentCenter copy.

• Claim 1 (device/junction — believed to be the principal independent claim): An electrical device/junction having a semiconductor, a conductor (metal), and an interface layer disposed between and in contact with both, where the interface layer is configured to depin the Fermi level of the semiconductor while still permitting current flow when the device is biased — with the device exhibiting a specific contact resistance below a recited threshold (the specification's thresholds range from ≤2500 Ω·μm² down to ≤1 Ω·μm²; the summary highlights <10 Ω·μm² and <1 Ω·μm² as preferred). The interface layer is characterized by selected materials, thickness (e.g., 0.1–5 nm, or a monolayer), and/or presence of a separation/spacer layer.

• Claim 5 (independent, related configuration): A junction/contact in which the conductor's Fermi level is aligned with (or independent of) the semiconductor's conduction or valence band edge, with current flow enabled because the interface layer thickness corresponds to a minimum or near-minimum specific contact resistance. Likely depends on/echoes the "thickness sufficient to reduce MIGS + passivates the surface" formulation from the summary.

• Other independents (probable, among claims 13, 65, 77): Later independent claims appear to recite metal/semiconductor material selections (group IV semiconductor, SiC, SiGe), interface layer stack variants (passivation layer + separation/spacer layer, e.g., metal oxide spacers), and in some cases Fermi-level alignment limitations. Claims 65–80 form a later claim set; claims 65–76 were disclaimed.

Non-claimed / disclaimed subject matter: claims 9–12 and 65–76 are statutorily disclaimed and cannot be enforced.

Litigation / Post-Grant History (for context)

• IPR2020-01204 — Samsung Electronics Co., Ltd. et al. v. Acorn Semi, LLC, filed 2020-06-29; instituted 2021-01-13; Final Written Decision, terminated 2022-01-12. Petitioner counsel: Yung-Hoon Ha; Patent Owner counsel: Matthew Phillips.
• IPR2020-01264 — filed by Unified Patents / Samsung; Not Instituted – Merits.
• District court — Acorn Semi, LLC v. Samsung Electronics Co., No. 2:19-cv-00347 (E.D. Tex.); reported $25M settlement. The same Acorn patent family (including US 8,766,336; US 7,084,423; US 9,461,167; US 9,209,261; US 9,905,691; US 10,090,395) was later transferred to Oak IP LLC in Dec. 2024.
• Federal Circuit — No. 22-1499 is the only CAFC matter listed for this family on Google Patents' timeline.

No 2026 CAFC docket entry for US 8,766,336 was found. Given the patent's nominal expiration (2022-08-12) and the 2024 transfer to Oak IP, I cannot confirm any 2026 appellate activity. Do not rely on the premise that a 2026 CAFC proceeding exists.

Sources

• Google Patents, US8766336B2: https://patents.google.com/patent/US8766336/en
• FreePatentsOnline, US8766336: https://www.freepatentsonline.com/8766336.html
• PTAB IPR2020-01264 record / '336 disclaimer (2020-10-14), EFS ID 40839495, App. 13/687,907: https://www.docketalarm.com/cases/PTAB/IPR2020-01264/Samsung_Electronics_Co._Ltd._v._Acorn_Semi_LLC/
• GreyB, "December's Patent Transfers Signal Higher NPE Risks for Semiconductors": https://greyb.com/blog/december-2024-patent-transactions/
• IPverse case summary, IPR2020-01204: https://ipverse.greyb.com/ptab-web/cases/case-details/IPR2020-01204

I'll research the litigation history of US patent 8766336 across the sources you mentioned.

Let me gather more detail on the appeal, the IPR outcomes, and any other defendants.

Let me verify the IPR outcomes and check for any additional proceedings or defendants.

Litigation Involving U.S. Patent No. 8,766,336

Based on searches of Unified Patents Portal, CourtListener, Docket Alarm, Justia, the CAFC opinions/orders database, and Law360/PTAB reporting, the following proceedings involving U.S. Patent No. 8,766,336 (the "'336 patent") are known. All substantive proceedings trace to a single dispute between Acorn (patent owner) and Samsung. The '336 patent was one of several Acorn semiconductor-contact patents asserted together.

1. District Court Litigation

Case: Acorn Semi, LLC v. Samsung Electronics Co., Ltd., et al.

• Plaintiff: Acorn Semi, LLC (subsidiary of Acorn Technologies, Inc.)
• Defendants (collectively "Samsung"): Samsung Electronics Co., Ltd.; Samsung Electronics America, Inc.; Samsung Semiconductor, Inc.; Samsung Austin Semiconductor, LLC
• Jurisdiction: U.S. District Court for the Eastern District of Texas, Marshall Division
• Case Number: 2:19-cv-00347-JRG (Chief Judge J. Rodney Gilstrap)
• Filing Date: October 23, 2019
• Asserted patents: Originally six (U.S. Pat. Nos. 7,084,423; 8,766,336; 9,209,261; 9,461,167; 9,905,691; 10,090,395). After narrowing, four remained at trial: the '336, '167, '691, and '395 patents.
• Outcome / Status:
  • Jury verdict May 19, 2021 in favor of Acorn — jury found infringement of all asserted claims and awarded $25 million in damages. (Acorn's press release, June 3, 2021; jury verdict at Dkt. 368.)
  • Acorn's motion for attorney's fees was denied (Feb. 17, 2022, Dkt. 409 — court found neither party without blame and did not deem the case exceptional).
  • The case was ultimately dismissed with prejudice by order dated January 6, 2023, following a stipulation by the parties resolving all disputes.
• Source: https://www.courtlistener.com/opinion/[9978246](/patent/9978246)/acorn-semi-llc-v-samsung-electronics-co-ltd/ ; https://cases.justia.com/federal/district-courts/texas/txedce/2:2019cv00347/[193059/409](https://assignmentcenter.uspto.gov/search/patent/reelFrameDetail?reelFrame=193059-0409)/0.pdf ; https://www.marketscreener.com/news/latest/... (Acorn Technologies press release)

2. PTAB — Inter Partes Review (Instituted)

Proceeding: Samsung Electronics Co., Ltd. v. Acorn Semi, LLC

• Petitioner: Samsung Electronics Co., Ltd.
• Patent Owner: Acorn Semi, LLC
• Case Number: IPR2020-01204
• Patent challenged: U.S. Patent No. 8,766,336
• Filing Date: June 29, 2020
• Institution Decision: January 13, 2021 (instituted on claims 1–8, 13–16, 77–80; petition had challenged claims 1–16 and 65–80)
• Final Written Decision: January 12, 2022
• Status / Outcome: Final Written Decision entered. Reporting indicates the Board found claims unpatentable (Law360: "Samsung Strikes Down Some Patents Tied To $25M Trial Loss"; IPWatchdog reported that across the related Acorn/Samsung IPRs, all asserted claims of two of the four patents were held unpatentable, while the patent owner prevailed on some asserted claims of another). Note: I could not, within this search, confirm the exact claim-by-claim outcome of the '336 FWD with high confidence — the precise set of claims held unpatentable should be verified against the FWD (Paper 48) itself.
• Sources: https://ipverse.greyb.com/ptab-web/cases/case-details/IPR2020-01204 ; https://www.law360.com/ptab_cases/5efb0b759d43ef0f35106839 ; https://ipwatchdog.com/2022/01/20/patent-filings-round-claims-25-million-verdict-held-unpatentable-stay-denied-panel-denies-ipr-lengthy-reexam-history/

3. PTAB — Inter Partes Review (Not Instituted)

Proceeding: Samsung Electronics Co., Ltd. v. Acorn Semi, LLC

• Case Number: IPR2020-01264
• Patent: U.S. Patent No. 8,766,336
• Status: Not Instituted — Merits (denied institution).
• Note: A statutory disclaimer under 37 C.F.R. § 1.321(a) by Acorn appears in this record (Exhibit 2001).
• Sources: Google Patents litigation entry; https://www.docketalarm.com/cases/PTAB/IPR2020-01264/Inter_Partes_Review_of_U.S._Pat._8766336/

4. Court of Appeals for the Federal Circuit

Appeal: Acorn Semi, LLC v. Samsung Electronics Co., Ltd., et al.

• Appellants: Samsung Electronics Co., Ltd.; Samsung Electronics America, Inc.; Samsung Semiconductor, Inc.; Samsung Austin Semiconductor, LLC
• Appellee: Acorn Semi, LLC
• Case Number: 2022-1499
• Filed: February 25, 2022 (appeal from E.D. Tex. No. 2:19-cv-00347-JRG)
• Outcome / Status: Appeal voluntarily dismissed with prejudice pursuant to Fed. Cir. R. 42(b) by order dated January 6, 2023 (parties to bear their own costs/fees); mandate issued. This dismissal followed the parties' settlement/resolution of the district court action.
• Sources: http://cafc.uscourts.gov/opinions-orders/22-1499.ORDER.1-6-2023_2059244.pdf ; 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/ ; https://dockets.justia.com/docket/circuit-courts/cafc/22-1499

Summary

No litigation involving the '336 patent against any party other than Samsung is known to me. The '336 patent was litigated as part of a broader Acorn–Samsung dispute over Acorn's metal-interface-layer-semiconductor contact patents (the '423, '336, '261, '167, '691, and '395 patents), and the PTAB proceedings against the '336 patent were companion filings to Samsung's IPRs against those other family members (e.g., IPR2020-01205, -01206, -01207, -01279, -01282).

Caveats:

• I was unable to confirm the exact claim-by-claim disposition in the IPR2020-01204 Final Written Decision within this session; the precise list of claims held unpatentable should be verified against the FWD document itself (PTAB Paper 48) and the subsequent appellate treatment (if any) of that FWD.
• Post-trial motions and any license/settlement terms were not fully visible in the searchable records; the January 2023 dismissal was by stipulation, so settlement details are not public in the docket.
• Per your instruction, I have treated all identifiers literally; one secondary source (IPWatchdog) contains an apparent typo referring to "IPR2022-01204," which I have read as referring to IPR2020-01204 for the '336 patent based on the surrounding context.

I'll work from the structured block, but note up front that it appears to conflict with the patent text in this prompt — so I'll verify via search before writing anything.

Good — the searches confirm PTAB activity, contradicting the structured block. Let me drill into the FWD outcome and the appeal.

Let me pin down the FWD claim-level holdings, the panel, and any later petitions.

⚠️ Data-integrity flag before anything else

The structured "PTAB proceedings on file" block in this prompt says the USPTO ODP API "returns no AIA trial proceedings for this patent." That is wrong, and I am overriding it. The patent's own Google Patents record, the PTAB docket, and the prior sections of this analysis all confirm two Samsung IPR petitions against US 8,766,336 (IPR2020-01204 and IPR2020-01264). I did not take the ODP "no proceedings" default, because the corroborating docket evidence is direct and specific (Paper numbers, institution dates, FWD date, judge panel). Treat the ODP block as a stale/incomplete ingest, not as ground truth.

Two additional corrections to the previously generated summary:

• The earlier section described IPR2020-01264 as filed by "Unified Patents / Samsung." The petitioner of record is Samsung Electronics Co., Ltd. et al.; "Unified Patents" is merely the source of the metadata column Google Patents republishes, not the filer.
• The earlier section said "No CAFC 2026 docket activity." Confirmed — and note the only Federal Circuit matter tied to this family, 22-1499, is the district-court appeal (Acorn Semi v. Samsung), not an appeal of any IPR on '336.

Proceedings overview

Two AIA trial petitions were filed against US 8,766,336 — both by [Samsung Electronics Co., Ltd. et al.](/litigations/by-plaintiff/Samsung%20Electronics%20Co.%2C%20Ltd.%20et%20al.) on the same day in mid-2020. Breakdown: 0 active, 1 went to Final Written Decision (2022-01-12), 1 institution denied (2021-01-13), 1 claims-invalidated proceeding, 0 settled, 0 pending appeals.

Bottom line for a defendant today: the patent is not "hardened by surviving two IPRs" — it is a 2022-expired, partially knocked-out patent now owned by Oak IP LLC, and one of its two challenges produced claim cancellations. But it also is not a case of "claims 1–8 are canceled, ignore the demand letter." Acorn took claims 9–12 and 65–76 off the table by statutory disclaimer (2020-10-14) before institution, and the instituted set (1–8, 13–16, 77–80) is the set that matters. Get the FWD text (Paper 48) before you build any invalidity or non-infringement theory on claim numbers.

IPR2020-01204 — Samsung Electronics Co., Ltd. et al. v. Acorn Semi, LLC

• Type: Inter Partes Review
• Filed: 2020-06-29
• Status: Final Written Decision — terminated 2022-01-12
• Judge panel: Aaron W. Moore, Brian J. McNamara, John R. Kenny; Moore authored the Final Written Decision. (Kenny's questioning at the 2021-10-13 oral hearing is preserved in the transcript — he pressed both sides hard on whether the district-court infringement verdict was "final" for secondary-considerations purposes.)
• Petition grounds:
  • § 102 anticipation by Grupp '483 (i.e., US 7,084,423 — a member of the same family) against claims 1–8, 13–16, and 77–80. This ground is priority-dependent: Samsung argued the challenged claims were not entitled to any priority date before 2011-02-07, which would make the family's own earlier disclosure prior art.
  • § 112 (written description and enablement) as the vehicle for the priority break — specifically lack of written-description support in the pre-2011 priority applications for the "metal oxide" genus, for "group IV semiconductor," and enablement of the recited specific contact resistivity values (1000 / 100 / 10 / 1 Ω·μm²).
  • Note the § 311(b) wrinkle: Acorn argued on the merits that Samsung's reliance on Grupp '483 (a family member) violated § 311(b).
• Claim set: Petition challenged 1–16 and 65–80. On 2020-10-14 Acorn filed a statutory disclaimer under 37 C.F.R. § 1.321(a) disclaiming claims 9–12 and 65–76, mooting those. The Board instituted on 1, 2, 3, 4, 5, 6, 7, 8, 13, 14, 15, 16, 77, 78, 79, 80 — i.e., the entire post-disclaimer surviving claim set.
• Institution decision: Instituted 2021-01-13. The panel expressly declined to discretionarily deny under § 314(a) / Fintiv: "Petitioner's stipulation has minimized any overlap with the parallel district court litigation… the minimization of overlap and the strength of the merits outweigh the upcoming trial date." It also rejected § 325(d) and the General Plastic serial-petition argument.
• Final Written Decision: Issued 2022-01-12 (Paper 48). I could not verify the verbatim claim-level disposition from the sources retrieved, and I will not invent it. What is documented: (i) the FWD issued on the statutory deadline; (ii) contemporaneous reporting (Law360, 2022-01-13, "Samsung Strikes Down Some Patents Tied To $25M Trial Loss") characterized the combined result across the Acorn IPR set as a partial win for Samsung; (iii) no Federal Circuit appeal of this FWD appears in the docket record (contrast the four Samsung appeals from the other Acorn IPRs, below), which is inconsistent with a total wipeout of the instituted claims and is at least suggestive that Acorn prevailed on at least some of claims 1–8 / 13–16 / 77–80. Go to Paper 48 for the operative claim numbers. Link: https://www.docketalarm.com/cases/PTAB/IPR2020-01204/Samsung_Electronics_Co._Ltd._v._Acorn_Semi_LLC/
• Settlement / termination: None. Trial ran to a merits decision. (The district court case settled for a reported $25M; the IPR did not.)
• Appeal: No CAFC appeal of the IPR2020-01204 FWD identified. Do not confuse this with Acorn Semi, LLC v. Samsung Electronics Co., Ltd., No. 22-1499 (Fed. Cir.), which is the appeal from the district court judgment (2:19-cv-00347-JRG) and was voluntarily dismissed with prejudice under FRAP 42(b) on 2023-01-06, each side bearing its own costs — a procedural dismissal with no merits ruling. https://www.courtlistener.com/opinion/[9357041](/patent/9357041)/acorn-semi-llc-v-samsung-electronics-co-ltd/
• Defensive value: This FWD is the reason you must check claim numbers before responding to any Oak IP assertion. Acorn itself already abandoned two claim blocks (9–12, 65–76) rather than defend them — those claims cannot be asserted by anyone, ever. And Samsung's § 315(e)(2) estoppel attached on 2022-01-12 as to grounds raised or reasonably could have been raised, and Samsung separately stipulated in the district court not to raise prior-art-based invalidity — so Samsung's own toolbox is closed off, but a different defendant is not bound by Samsung's estoppel.

IPR2020-01264 — Samsung Electronics Co., Ltd. et al. v. Acorn Semi, LLC

• Type: Inter Partes Review
• Filed: 2020-07-08 (nine days after -01204)
• Status: Institution Denied — Merits ("Not Instituted – Merits" per Google Patents; ipverse status "Institution Denied"). Minor docket-label inconsistency flagged: the docket entry generated for the 2021-01-13 decision is labeled "Trial Instituted Document," but it is followed by Petitioner's Request for Refund of Post-Institution Fees (2021-03-01) and Notice of Refund (2021-03-05) — a refund of the post-institution fee is the signature of a denial, and both Google Patents and ipverse classify the case as not instituted. I treat it as denied.
• Judge panel: Not confirmed from the sources retrieved (the denial panel is not public in what I could access). Panel overlap with -01204 (Moore / McNamara / Kenny) is likely but unverified.
• Petition grounds: This was a parallel/paired petition to -01204. The Board grouped the pairs across the Acorn set: (01204 / 01264) for '336, (01205 / 01241), (01206 / 01279), (01207 / 01282). Acorn's comments-to-stipulation brief confirms the pairing and that the Board invited a third, revised stipulation from Samsung in these two cases — a procedural fight Acorn characterized as "unprecedented… abusive gamesmanship." The substantive ground set is the same § 112 priority-disentitlement attack (written description for "metal oxide" / "group IV semiconductor") that the -01204 petition pressed; the Board opted to try the challenge in one of the pair.
• Institution decision: Denied 2021-01-13 — the same date the companion was instituted. Where parallel petitions challenge the same patent, the Board routinely institutes one and denies the other to conserve resources; that is the pattern here.
• Final Written Decision: N/A (never instituted).
• Settlement / termination: Terminated by denial; post-institution fee refunded 2021-03-05.
• Appeal: None possible — § 314(d) makes the institution determination final and non-appealable.
• Defensive value: Low incremental value on its own — a non-institution is not a merits win and creates no estoppel. But it matters procedurally: it shows the Board declined to run a second, parallel '336 trial, which is why there is only one merits record on this patent to work from.

Strategic summary

Claim status on 8,766,336. The claim picture is layered and you must not over-read it. Claims 9–12 and 65–76 are statutorily disclaimed (2020-10-14) — unenforceable, full stop, not just "invalid in this proceeding." Claims 1–8, 13–16, and 77–80 were the only claims left standing at institution, and they were the entire instituted set. The FWD of 2022-01-12 resolved that set; I cannot responsibly state which of those claims were canceled without the Paper 48 text, and I am not going to guess. There are no untested claims left on this patent — every surviving claim was put through IPR2020-01204.

Estoppel landscape. Samsung's § 315(e)(2) estoppel attached 2022-01-12 and bars Samsung and its privies from raising in litigation any ground raised or that reasonably could have been raised in IPR2020-01204 — which sweeps in the Grupp '483 anticipation theory and the § 112 priority-disentitlement theory. Samsung also stipulated in E.D. Tex. not to assert prior-art-based invalidity at all. Critically, none of that binds a new defendant. A defendant sued today by Oak IP LLC can re-run Grupp '483, can re-run the "metal oxide" written-description attack, and — because IPR2020-01264 was never instituted — has no General Plastic bar on a fresh petition either. The estoppel runs one way.

Pattern signals. This was a coordinated six-patent, ten-petition campaign: Samsung filed ten IPR petitions against six Acorn patents, with parallel paired petitions per patent. The Board instituted one of each pair and denied the other. Acorn was an aggressive, expensive respondent — it litigated Appointments Clause challenges, structural-bias arguments, § 311(b), a motion-to-exclude fight over Dr. Clifton's fact-vs-expert testimony, and heavy secondary-considerations evidence including a judicially found infringement and a stipulated $25M jury verdict against Samsung's Exynos chips. The defensive aggregator chain is real but indirect: Unified Patents appears only as the metadata provider Google Patents cites; the actual petitioner throughout was Samsung. The portfolio was later transferred Acorn Semi, LLC → Oak IP, LLC (recorded 2024-12-26), after a Norton Living Trust security interest (2019-05-30) was released the same day — a classic monetization-vehicle handoff that typically precedes fresh assertion campaigns.

Term posture. Google Patents lists the patent as Expired – Lifetime, anticipated expiration 2022-08-12. That materially caps exposure: no injunctive relief, and any damages are past-only, confined to infringement occurring before 2022-08-12 and within the § 286 six-year lookback from the complaint date.

Recommended next steps

1. Pull Paper 48 now — it is the single document that controls your response. Do not state a claim-level validity position until you have it. The FWD is indexed as IPR2020-01204, Paper 48 (P.T.A.B. Jan. 12, 2022) at https://www.docketalarm.com/cases/PTAB/IPR2020-01204/Samsung_Electronics_Co._Ltd._v._Acorn_Semi_LLC/ and via USPTO PTAB E2E under Application 13/687,907.
2. If a demand letter cites claims 9–12 or 65–76, it is citing disclaimed claims — respond by attaching Exhibit 2001 from the IPR2020-01264 record (the 2020-10-14 statutory disclaimer) and the § 1.321(a) disclaimer in the file. That is a near-dispositive response on those numbers.
3. Confirm whether any instituted claim survived the FWD. If the FWD canceled all of claims 1–8 / 13–16 / 77–80, your answer is that the patent is fully neutralized. If some survived, scope your invalidity work to exactly those claims and re-run Grupp '483 — Samsung's estoppel does not travel to you.
4. Mine the parallel records for free prior art and admission ammunition. The Goodnick/Taubenblatt/Iwaguro/Kim/Jammy/Chang record is in the IPR2020-01205 / -01279 / -01282 files (including Dr. Goodnick's own declaration rebutting Samsung's reading of his paper) and is directly reusable against any surviving '336 claim.
5. Exploit expiration. For any product shipping today, the damages window closed 2022-08-12. Pair the § 286 lookback with the expiration date and Oak IP's exposure theory narrows sharply.
6. Watch the Oak IP docket. Oak IP LLC took this portfolio in 2024-12-26; a 2025–2026 assertion wave is the expected next move, and any new IPR would be the first on this patent since 2020 — with no General Plastic or § 315 estoppel obstacle.

Do not confuse the numbers: a search for "8,766,336" surfaced an unrelated US 8,756,336 owned by Cellspin Soft (e.g., IPR2024-00757, TikTok v. Cellspin). It is a different patent, different owner, different technology. Per the operating rule to read patent numbers literally, I did not treat it as this patent.

I'll research the assignment record for US 8,766,336. Let me start with several targeted searches.

Assignment & Ownership Analysis — US 8,766,336 (Grupp / Connelly)

Source-access caveat (read first): I could not render the USPTO Assignment Center's per-record detail fields (reel/frame, conveyance text, assignee address, correspondent of record) for this patent in this session. The underlying USPTO Patent Assignment Search (https://assignmentcenter.uspto.gov/ and its legacy mirror https://assignment.uspto.gov/patent/index.html) is the authoritative place to pull reel/frame and correspondent, and I was unable to extract those specific fields. I will not fabricate reel/frame numbers or correspondent names. Everything below comes from (a) the authoritative full patent text supplied (Google Patents legal-events block), and (b) search results cited inline. Where I mark "not retrieved," treat it as an explicit gap to close at the Assignment Center, not as absence of a record.

Inventors

Pattern check: No "inventors departed within 12 months" signal. Both inventors were still conveying to Acorn Technologies on the 2012 continuation, ~10 years after the 2002 priority filing, and remained the named inventors through the family. This is the profile of a small founder-engineer team holding its own IP, not a fire-sale precursor. Inventor-to-assignee correspondence address is not separately determinable beyond the assignment record itself.

Original assignee

Acorn Technologies, Inc. (original assignee; also the entity that received the inventors' 2012-11-28 conveyance).

• Primary line of business: fabless semiconductor technology development and IP licensing — the company commercialized "Fermi-level depinning / passivated Schottky barrier" contact technology rather than selling chips. Per the Semiconductor Digest report on the jury win, "Acorn Semi LLC (together with Acorn Technologies, 'Acorn'), a wholly owned subsidiary of Acorn Technologies Inc." (https://www.semiconductor-digest.com/acorn-technologies-prevails-at-trial-in-patent-infringement-lawsuit-against-samsung/). That confirms Acorn Semi LLC is a captive subsidiary, not an arm's-length acquirer.
• Did it ship a product embodying the claims? No evidence of a commercial semiconductor product; Acorn's revenue model in the record is licensing/litigation (the $25M Samsung verdict). Treat "operating company" loosely for Acorn — it is an R&D/IP house, not a merchant chipmaker.
• Current status: Still nominally in existence, but it has divested the asserted portfolio: the '336 patent and 35 others moved to Oak IP LLC on/around 2024-12-26 (Google Patents legal events; GreyB, https://greyb.com/blog/december-2024-patent-transactions/). No bankruptcy found.

Cross-reference / contradiction flag: The supplied patent text lists the current assignee as Oak IP LLC and the status as Expired – Lifetime, anticipated expiration 2022-08-12. That means Oak IP acquired an already-expired patent as part of the 36-asset portfolio. Any '336 enforcement value is limited to pre-2022-08-12 back damages (and the family's unexpired members). This is consistent with the prior summary but worth stating plainly.

Assignment timeline

Dates below are the recorded legal-event dates from the authoritative patent page (https://patents.google.com/patent/US8766336/en). Reel/frame and correspondent fields were not retrievable in this session and are flagged as such.

1. 2012-11-28 (recorded) — Reel/frame not retrieved

   • Conveyance: Assignment (inventor→company)
   • Assignor: Grupp, Daniel E.; Connelly, Daniel J.
   • Assignee: Acorn Technologies, Inc.
   • Correspondent: not retrieved
   • Context: Original inventor assignment perfecting title in the company for this continuation (13/687,907).

2. 2019-05-30 (recorded) — Reel/frame not retrieved

   • Conveyance: Security Interest ("Grant of Security Interest")
   • Assignor: Acorn Semi LLC
   • Assignee: The Peter Norton Living Trust dated April 28, 1989
   • Correspondent: not retrieved
   • Context: Securitization — Acorn pledged the portfolio as collateral to a private investor trust (Peter Norton). Financing, not a transfer of beneficial ownership.
   • (Anomaly to verify: the security-interest record names Acorn Semi LLC as assignor on 2019-05-30, i.e., ~4 weeks before the Acorn Technologies→Acorn Semi assignment in step 3. Either the Acorn Semi transfer was executed earlier with a later recordation, or Google's assignor field reflects the collateral grantor rather than the then-record owner. Worth confirming against the actual reel/frame at the Assignment Center.)

3. 2019-06-26 (recorded) — Reel/frame not retrieved

   • Conveyance: Assignment
   • Assignor: Acorn Technologies, Inc.
   • Assignee: Acorn Semi, LLC
   • Correspondent: not retrieved
   • Context: Internal reorg / transfer-to-assertion-vehicle — portfolio moved into the wholly owned subsidiary that filed the Samsung suit in October 2019.

4. 2024-12-26 (recorded) — Reel/frame not retrieved

   • Conveyance: Assignment
   • Assignor: Acorn Semi, LLC
   • Assignee: Oak IP, LLC
   • Correspondent: not retrieved
   • Context: Transfer-to-asserter / portfolio sale — 36 patents sold; GreyB ties Oak IP to Dominion Harbor. Oak IP sued GlobalFoundries on 2025-02-04.

5. 2024-12-26 (recorded) — Reel/frame not retrieved

   • Conveyance: Release by Secured Party
   • Assignor: The Peter Norton Living Trust dated April 28, 1989
   • Assignee: Acorn Semi, LLC
   • Context: Collateral release at closing (clearing the step-2 security interest so title could pass clean to Oak IP).

6. 2024-12-26 (recorded) — Reel/frame not retrieved

   • Conveyance: Release by Secured Party
   • Assignor: The Peter Norton Living Trust dated April 28, 1989
   • Assignee: Acorn Technologies, Inc.
   • Context: Collateral release — same closing, releasing the trust's interest as to Acorn Technologies' related assets.

Correspondent note : The one attorney-of-record datapoint I obtained is Tarek N. Fahmi (Reg. No. 41,402), 2150 N First Street, Suite 420, San Jose, CA 95131 (USPTO Customer No. 27571), who filed the 2020-10-14 statutory disclaimer for '336, co-signed by Sarah Zizzo, docket 3771-0003-DC4 (PTAB Ex. 2001, https://www.docketalarm.com/cases/PTAB/IPR2020-01264/...). This is prosecution-side correspondence, not confirmed as the assignment-recordation correspondent. Whether Fahmi also handled the 2019 or 2024 reel/frame recordings must be verified at the Assignment Center before it can support a "repeat correspondent" finding.

Timeline diagram

timeline
    title Ownership of US 8766336
    2002 : Priority application filed
    2012 : Continuation filed by Acorn Technologies
    2014 : Patent issued
    2019 : Moved to Acorn Semi LLC
         : Security interest to Norton Trust
         : Samsung suit filed Oct 2019
    2021 : 25M dollar jury verdict vs Samsung
    2022 : Patent term expires
    2024 : Sold to Oak IP LLC
    2025 : Oak IP sues GlobalFoundries

NPE / troll-pattern signals

1. Shell-entity transfer — PRESENT (moderate). The asserted patents migrated from Acorn Technologies, Inc. to Acorn Semi, LLC on 2019-06-26, a nominal licensing/litigation LLC whose only public role in the record is as plaintiff in Acorn Semi, LLC v. Samsung, 2:19-cv-00347 (E.D. Tex.). No products in commerce; the entity exists to hold and assert. The name suffix "Semi" rather than "IP/Holdings" is cosmetic — the functional test (no products, litigation vehicle) is met. Note it is a wholly owned subsidiary, which softens this signal relative to a true third-party shell.

2. Known asserter in the chain — PRESENT (strong). Current assignee Oak IP, LLC is documented as tied to Dominion Harbor (a high-frequency patent monetization/NPE firm) and filed a new infringement complaint against GlobalFoundries on 2025-02-04, weeks after the 2024-12-26 acquisition (GreyB, https://greyb.com/blog/december-2024-patent-transactions/). Dominion Harbor is exactly the kind of repeat-player monetization entity the asserter directories (RPX / Unified) track. This is the chain's clearest NPE marker.

3. Repeat correspondent across the chain — UNCLEAR (records not retrieved). Google's legal-events block does not expose correspondents, and my searches did not surface the recording attorneys for the 2019/2024 reel/frames. The only attorney datapoint is Tarek N. Fahmi (prosecution/PTAB, 2020), which is not an assignment-recording correspondent. This signal cannot be scored without the Assignment Center's correspondent fields.

4. Cascading transfers — PRESENT (weak-to-moderate). Three recordings on a single day, 2024-12-26: (i) Acorn Semi → Oak IP assignment, plus (ii) and (iii) two Norton-Trust releases of security interest. Same-day stacking of a sale plus collateral releases is the signature of a single orchestrated closing, not independent transactions. The earlier 2019-05-30 security interest followed within ~4 weeks by the 2019-06-26 assignment is a second tightly-clustered pairing.

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

   • Acorn Technologies → Acorn Semi, LLC (2019-06-26), then the Samsung complaint in October 2019 (~3–4 months): title was parked in the assertion vehicle immediately before suit.
   • Acorn Semi → Oak IP (2024-12-26), then the GlobalFoundries complaint 2025-02-04 (~5–6 weeks). GreyB literally frames it as "a patent changes hands today, and six months later, it's at the center of a lawsuit."
     This is the textbook pre-assertion clean-up of the record.

6. Bankruptcy fire-sale — NOT PRESENT. No Chapter 7/11 filing by Acorn Technologies or Acorn Semi appears anywhere in the record (contrast Kodak/Nortel). The 2024 transfer is a negotiated portfolio sale to a monetizer, not a distressed asset sale.

7. Privateering — NOT PRESENT (or unclear). Acorn asserted on its own behalf (via its own subsidiary), not on behalf of a separate operating manufacturer against competitors. Oak IP's post-acquisition suits are classic third-party monetization rather than privateering for a sponsor; no SEC filing ties GlobalFoundries' assertion to another operating company's sponsorship.

8. Defensive aggregator — NOT PRESENT. The chain terminates at an asserting NPE (Oak IP, LLC), the inverse of neutralization. No RPX / AST / LOT / OIN / Unified assignment appears.

Additional finding not in the checklist — expired-asset purchase: Oak IP acquired '336 after its 2022-08-12 expiry. The purchase value for '336 specifically is backward-looking (pre-expiry damages) or defensive/portfolio-flanking (e.g., § 285 / family leverage); the live assertion value sits in unexpired family members such as US 9,905,691 (the patent actually named in the GlobalFoundries IPR petitions, https://ptacts.uspto.gov/.../1557953). Confirm whether '336 itself was pleaded in the GF complaint before treating it as an active assertion.

Verdict

NPE — high confidence.

Two independent strong signals: (1) a known asserter in the chain — current assignee Oak IP, LLC (Dominion Harbor–linked) acquired the portfolio on 2024-12-26 and sued GlobalFoundries on 2025-02-04; and (2) pre-litigation transfers — the patents were moved into the litigation entity Acorn Semi, LLC on 2019-06-26, roughly three months before the October 2019 Samsung complaint, and again into Oak IP roughly five weeks before the February 2025 GlobalFoundries complaint. These are reinforced by shell-entity characterization (Acorn Semi, LLC and Oak IP, LLC are non-product assertion vehicles) and same-day cascading recordings on 2024-12-26 (sale plus two secured-party releases). The "repeat correspondent" signal remains unscored because I could not retrieve the reel/frame correspondent fields — verify at the USPTO Assignment Center before finalizing.

Verify / close-out links:

• USPTO Patent Assignment Search: https://assignmentcenter.uspto.gov/ (search "8766336") and https://assignment.uspto.gov/patent/index.html
• Google Patents legal-events block (authoritative dates used above): https://patents.google.com/patent/US8766336/en
• PTAB record (IPR2020-01204 / IPR2020-01264) and the 2020-10-14 disclaimer: https://www.docketalarm.com/cases/PTAB/IPR2020-01264/
• Oak IP → GlobalFoundries context: https://greyb.com/blog/december-2024-patent-transactions/

Open items I could not resolve (state these to the requester, do not fill by inference): (a) reel/frame for all six recordings; (b) the correspondent of record on each recording; (c) whether Tarek N. Fahmi or his firm is also the assignment correspondent; (d) the exact execution dates (vs. recording dates) of the 2019 and 2024 transfers; (e) whether '336 was among the three patents pleaded in Oak IP v. GlobalFoundries (2025).

I'll search USPTO sources and related databases for the specific patent number 8766336 and its cited prior art.

I found general information but not the front-page "References Cited" list. Let me search specifically for the cited references and IPR grounds.

Let me retrieve the actual front-page cited references for US 8,766,336 from primary sources.

Scope and confidence caveat (important)

I was asked to enumerate the front-page cited references ("References Cited" / (56)) of US 8,766,336 and map each to claims under 35 U.S.C. § 102. I must flag a limitation up front: I could not retrieve the verbatim (56) front-page list from the USPTO PatentCenter record or Google Patents within my available searches. The Google Patents page for US8766336 returned its "Definitions" text rather than the citations block, and my queries for the FPO/PatentCenter "References Cited" table returned unrelated documents. I will not fabricate patent numbers, dates, or examiner citations I have not verified.

What I can provide, and have verified, falls into three distinct buckets — I label each so you don't conflate them:

Buckets B and C are the "most relevant prior art" in substance; bucket A is necessary context because those parent filings are themselves potential §102 art if the '336 claims lose the 2002 priority claim (the central dispute in the IPRs).

A. Intrinsic cross-references (from the '336 specification) — verified

The specification's "RELATED APPLICATIONS" paragraph names these (all incorporated by reference). They are same-family filings, not examiner citations, but they matter for §102 because Samsung argued the challenged claims may be entitled to no earlier than 2012-11-28:

"Grupp '483." The IPR2020-01204 petition (Samsung v. Acorn Semi) is captioned the "Grupp '483 Petition" and is described by Samsung as asserting "a patent related to the '336 Patent as anticipatory prior art" (Samsung's Statement on Parallel Petitions, E.D. Tex./PTAB record). I could not confirm the exact patent number behind the shorthand "Grupp '483," and I will not guess it. Note the naming pattern is consistent with an Acorn/Grupp patent whose number ends in 483. Treat this as the single most important §102 reference in the record — verify its number in the IPR2020-01204 petition itself.

B. Non-patent literature cited in the '336 specification — verified

These are the printed publications the applicants cited in the body of '336 (they appear in the Background/Discussion, quoting the retrieved text):

1. J. Tersoff, "Schottky Barrier Heights and the Continuum of Gap States," Phys. Rev. Lett. 52(6), Feb. 6, 1984. — MIGS/"gap center" pinning model. Cited as the theoretical basis for why the Fermi level pins. §102: Background art only; it is a theory paper and does not disclose the claimed metal–interface-layer–semiconductor structure. Not an anticipation reference.
2. Louie, Chelikowsky & Cohen, "Ionicity and the theory of Schottky barriers," Phys. Rev. B 15, 2154 (1977). — same role; background.
3. Heine (work cited); Flores & Tejedor (work cited) — MIGS antecedents; background.
4. Bardeen's model (surface-state pinning) — background.
5. Grondahl & Geiger (1926); Schottky (1938) — historical rectification background.

None of these alone anticipates any claim; they are cited by the applicants to frame the problem, not as §102 art.

C. Prior art actually asserted against '336 (PTAB + district court) — verified

This is the substantive answer to "what is the most relevant prior art." From the PTAB filings and E.D. Tex. invalidity contentions that surfaced:

C1. References in the PTAB IPRs

⚠️ The record shows these were marshaled as obviousness (§103) combinations, not pure §102 anticipation (with the possible exception of Goodnick and the "Grupp '483" patent). That is a material distinction for your question.

C2. References in the E.D. Tex. invalidity contentions (Acorn Semi, LLC v. Samsung, No. 2:19-cv-00347-JRG; Appendix F claim charts, served 2020-05-19). These are the technical printed publications Samsung charted:

Note these four (De Bosscher, Schmidt, Waldrop, Defives) appear in the same contentions list as the Goodnick paper — i.e., they are the core technical printed-publication art the defendants charted against the family.

§102 mapping — what I can and cannot responsibly assert

• Anticipation (§102) requires a single reference disclosing every claim limitation, arranged as claimed. On the record I retrieved, "Grupp '483" is the one reference Samsung framed as anticipatory — and its mapping target is the challenged independent claims (claims 1–16 and 65–80 per the petitions; claims 1, 5, 13, 65, and 77 being independent). I cannot responsibly assign specific claim numbers to it without the petition text and the exact '336 claims.
• Schmidt et al. (1988) is the printed publication most facially aligned with the "depin the Fermi level / tune barrier height with a thin layer" concept, and Goodnick et al. (1981) with the "thin interface layer between metal and Si" structure. These are your best §102 candidates among the publications, but in the PTAB they were run as §103 combinations (Goodnick + Taubenblatt, etc.).
• The Tersoff/Bardeen/Louie materials are not §102 art against the claims — they are cited to explain pinning theory.

Uncertainty flags (do not treat as verified):

1. The exact front-page (56) list of US 8,766,336 (U.S. and foreign patent documents, and the examiner-cited NPL) — not retrieved.
2. The exact number behind "Grupp '483."
3. Claim numbers for each reference — I could not confirm the verbatim claim text, so any claim-number mapping would be reconstruction, not quotation.

Recommendation to close the gaps

To produce a fully grounded §102 chart you should pull, from primary sources:

1. USPTO PatentCenter record for App. 13/687,907 — the (56) "References Cited" and the examiner's Form 892/PTOL-326.
2. IPR2020-01204 petition (Samsung, "Grupp '483") — supplies the anticipatory reference and its number.
3. IPR2020-01264 petition ("Jammy/Kim") — supplies the §103 combinations.
4. The E.D. Tex. Appendix F claim charts (already partially listed above) — supply the element-by-element printed-publication mapping.

Sources

• Google Patents, US8766336B2: https://patents.google.com/patent/US8766336/en
• FreePatentsOnline, US8766336: https://www.freepatentsonline.com/8766336.html
• Samsung Statement on Parallel Petitions, IPR2020-01204 (Docket Alarm): https://www.docketalarm.com/cases/PTAB/IPR2020-01204/...
• Samsung's Reply, IPR2020-01204 (PTACTS): https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1538602](/patent/1538602)/...
• Samsung Federal Circuit "Statement of Related Cases" (Goodnick/Taubenblatt/Kim/Chang references): https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1558034](/patent/1558034)/...
• Defendants' Preliminary Invalidity Contentions + Appendix F charts (De Bosscher, Defives, Goodnick, Schmidt, Waldrop), cited in Justia Docket for US 11,804,533: https://patents.justia.com/patent/[11804533](/patent/11804533)
• Patexia summary, IPR2020-01204 (claims challenged/instituted: 1–16, 65–80): https://services.patexia.com/lawsuits/Samsung-Electronics-Co-Ltd-v-Grupp-et-al-id-[139157](/patent/139157)

Bottom line: The two most relevant prior-art references for US 8,766,336 on substance are Goodnick et al. (1981) and Schmidt et al. (1988), with De Bosscher (1986) and Waldrop (1993) rounding out the metal/interface-layer barrier-height art, and the "Grupp '483" patent as the one reference Samsung pressed as anticipatory under §102. I was unable to verify the granted patent's own front-page (56) citation list, and I have deliberately not invented it.

I'll note upfront: no "Prior Art section" was actually included in the material provided to me — the previously generated sections are the patent summary and litigation summary only. I'll flag that as a gap and instead ground the § 103 analysis in the actual prior-art record I can retrieve (principally the PTAB grounds in IPR2020-01204, which are the authoritative prior-art combinations already litigated against this patent). Let me search.

Let me pull the specific grounds and the actual claim language for the '336 patent.

Let me pin down the IPR2020-01204 Final Written Decision outcome and the '336 claim text.

Scope note before I begin

No "Prior Art section" was provided to me. The two previously generated sections in this thread are the Patent summary and the Litigation summary; neither contains a prior-art analysis. I will not pretend otherwise. Instead, I reconstructed the actual prior-art universe that has been applied to US 8,766,336 by retrieving the PTAB record for IPR2020-01204 (Samsung v. Acorn) and the companion Acorn-family IPRs (‑01183, ‑01205, ‑01206, ‑01207, ‑01279, ‑01282), plus the underlying petition and expert declaration (Schubert). That is the authoritative § 103 record on this patent, and it is where any serious obviousness analysis of '336 must start.

Where the earlier sections and the retrieved record diverge, I flag it explicitly (see "Reconciliation" below).

§ 103 Obviousness Analysis — U.S. 8,766,336

1. Governing law / timing

The '336 patent issued from App. 13/687,907, filed 2012-11-28 — before March 16, 2013 — so pre‑AIA 35 U.S.C. §§ 102/103 govern. The priority dispute in the IPR is therefore dispositive of which references are available.

Threshold issue — effective filing date. Acorn contended the challenged claims are entitled to the 2002-08-12 priority date. Samsung contended (and the Board preliminarily agreed in related proceedings) that:

• the genus "metal oxide" interface layer was first arguably described only in the '522 application filed 2011-02-07 (the pre‑2011 applications disclosed only a single species, a TiO₂ spacer);
• the pre‑2011 applications do not enable the full breadth of the recitations "specific contact resistance/resistivity less than [1/10/100/1000] Ω·μm²" (no lower bound; the sole species said to reach just below 1 Ω·μm² is TiO₂); and
• the pre‑2011 applications do not adequately describe the "group IV semiconductor" genus.

If claims carry no priority earlier than 2011‑02‑07, then U.S. Pat. 7,176,483 ("Grupp '483," issued 2007‑02‑13) — a same-family, same-inventor patent sharing the '423 specification — becomes § 102(b) prior art. Critically for § 103: pre‑AIA § 103(c) does not shield Grupp '483 from combination, because § 103(c) disqualifies only § 102(e)/(f)/(g) art; Grupp '483 is § 102(b) art. So a same-inventor patent can be a § 103 primary reference here.

⚠️ Contradiction flag: The earlier Patent summary reported the granted abstract is limited to "arsenic" while my retrieved IPR record states "every challenged claim includes a generic 'metal oxide' interface layer." These are reconcilable: the challenged set (claims 1–16, 65–80) recites metal oxide; the arsenic/arsenide subject matter evidently sits in the unchallenged middle claims (17–64) reflected in the abstract. Treat the claim set as containing both a metal-oxide family and an arsenic family.

2. Claim scope carried into the analysis

From the petition's own element-by-element mapping of '336 claim 1, the challenged independent claims recite, in substance:

The arsenic family (unchallenged) would add "interface layer comprising arsenic / a monolayer of arsenic."

3. The prior-art references available

Retrieved from the Samsung petitions/declarations (Exhibit numbers as used in the IPRs):

4. Obviousness combinations and motivations to combine

Ground 1 — Grupp '483 as the § 103 primary reference (+ Jammy; + Iwaguro)

Combination: Grupp '483 in view of Jammy, and optionally Iwaguro.

• Grupp '483 is the strongest reference: same specification as the '336, same field, same inventors, and it expressly discloses the interface layer between metal and Si-based semiconductor, the depinning concept, the SCR below ~1 Ω·μm², and even a TiO₂ spacer (a metal oxide).
• Motivation to combine with Jammy: The only thing Grupp '483 arguably lacks is an express genus-level "metal oxide" recitation (it names TiO₂ as a species). Jammy supplies exactly that — it expressly teaches that the interface layer can include a metal oxide (alumina) in a metal–semiconductor contact, and that a SiO₂ layer on the Si substrate has a passivating effect. A POSITA seeking to generalize an interface layer across material choices would naturally consult Jammy, which is in the identical field and addresses the identical problem (contact resistance/passivation at a metal–Si interface). Rationale: In re Kerkhoven / In re Crockett — combining known elements taught as useful for the same purpose is prima facie obvious.
• Motivation to combine with Iwaguro: Iwaguro teaches the U-shaped resistance-vs-thickness curve, i.e., that a minimum resistance occurs at an intermediate interfacial-layer thickness. That is precisely the "thickness corresponding to a minimum specific contact resistance" element and supplies the reason to select a thickness that is thick enough to help but thin enough to conduct. This is the same optimization argument Samsung made on the record, using Iwaguro's U-curve against Acorn's FIG. 8.

Why obvious: Every element appears in the four corners of the references, and the references are in the same field addressing the same problem (minimizing metal–semiconductor contact resistance and controlling Schottky barrier height). The combination produces nothing more than the predictable sum of known properties.

Ground 2 — Goodnick + Taubenblatt 1984 (the "thin interfacial dielectric" combination)

Combination: Goodnick in view of Taubenblatt 1984; + Jammy to supply the metal oxide; + Chang for the SCR limitation.

• Goodnick teaches that a thin SiO₂ interlayer at a metal–Si contact and the self-limiting formation of Al₂O₃ reduce contact resistance — i.e., inserting a thin dielectric between metal and Si is known to be beneficial, not merely tolerable.
• Taubenblatt 1984 expressly teaches that an interfacial SiO₂ decreases the metal–Si Schottky barrier for Ti and Hf — supplying the "depin / lower the barrier" functional element.
• Motivation to combine: Same field (metal–silicon Schottky contacts), same problem (barrier height / contact resistance), same mechanism (interfacial oxide). A POSITA optimizing a Ti–Si contact would look to both. Reasonable expectation of success is supported because both references report the effect experimentally.
• Jammy then supplies the metal oxide genus (alumina) and the passivation teaching → satisfies the "interface layer comprising a metal oxide" element and, in claim sets reciting it, the "oxide of the semiconductor."
• Chang teaches the metric itself (specific contact resistance of metal–semiconductor barriers), supplying motivation to target a low SCR and the "specific contact resistance ≤ X Ω·μm²" recitations.

Ground 3 — Goodnick + Iwaguro (the "optimum thickness" combination)

• Goodnick shows the thin-dielectric benefit; Iwaguro shows the U-shaped optimum; together they render obvious the "thickness sufficient to reduce MIGS but thin enough to pass current → minimum specific contact resistance" element and the FIG. 8 relationship. Samsung relied on this exact mapping in the IPR.

Ground 4 — Nitride/other-chemistry variants

• Sobolewski (metal–silicon-nitride–silicon diodes with modified barrier height) in view of Taubenblatt/Rhoderick would render obvious the nitride-interface-layer claims (useful against the broader disclosure and against any claim not limited to metal oxide).

Ground 5 — "Stack of metals" and Markush claims

• Kim (metal stack / TiW alloy) supplies "a stack of metals deposited on the interface layer"; Jammy/Cote were offered for the metal selections. (Note: on the sibling record the Board found Samsung's reliance on Jammy+Cote for titanium inadequate, and its Goodnick+Taubenblatt motivations insufficient, in IPR2020-01282 — a caution that the "stack of metals"/metal-substitution sub-combinations are the weakest grounds.)

5. Why a POSITA would have combined them (the KSR-style rationale)

1. Common field and common problem. All primary references address the Schottky barrier and contact resistance at a metal–silicon interface — the exact problem the '336 patent addresses. Problem and field identity is a classic motivation to combine (KSR; In re Kerkhoven).
2. Known, finite, predictable options. Interposing a thin dielectric (SiO₂, silicon nitride, metal oxide) between metal and Si was a known technique with a known trade-off (tunneling vs. barrier reduction). Choosing among the small set of known oxides/nitrides is the "predictable variations of known elements" prong of KSR.
3. Iwaguro supplies the optimization direction. The U-shaped curve tells the artisan where to operate (an intermediate thickness), removing any "unexpected result" character from the claimed thickness/SCR relationship.
4. Reasonable expectation of success. Goodnick and Taubenblatt report measured, beneficial barrier/contact-resistance changes, so an artisan would reasonably expect a thin interfacial layer to lower the effective barrier and reduce contact resistance.
5. No teaching away. Nothing in the references teaches away from a thin interfacial layer; the record shows the opposite (multiple references observe reduced resistance with an interlayer).

6. Acorn's non-obviousness theories (and how the record treats them)

These are the arguments any § 103 conclusion must confront:

• Acorn's core argument: prior art (Goodnick, Taubenblatt, Sobolewski) observed reduced resistance with an interlayer but did not appreciate the MIGS mechanism and therefore did not recognize the "Goldilocks" optimization problem; on Acorn's telling, those were "anomalous results nobody noticed." Acorn also argued Goodnick does not disclose distinct stratified Al₂O₃/SiO₂ layers (reflected in Dr. Goodnick's own declaration that the Al₂O₃ "diffusion barrier" was only an untested, speculative hypothesis — he even disclaimed that the petition's layered diagram matches his paper).
• Objective indicia (Graham factors): commercial success, industry praise (the 2004/2006 Connelly papers), long-felt need (ITRS roadmap), and unexpected results (Dr. Kuhn's "counterintuitive" testimony about adding a resistive insulating layer to a contact).
• Samsung's rebuttals on the record: (i) no nexus — the praised Connelly work used silicon nitride, not a metal oxide, whereas the challenged claims recite metal oxide; (ii) the ITRS roadmap already provided low-contact-resistance solutions (silicide), negating long-felt need; (iii) secondary considerations cannot overcome a strong prima facie case; and (iv) Iwaguro's U-shaped curve already disclosed the optimum-thickness insight, defeating "unexpectedness."

Analytical consequence: Ground 2 (Goodnick+Taubenblatt(+Jammy)) is vulnerable to the sameness-of-invention / distinct-layer attack that succeeded in the sibling IPR2020-01282, where the Board held Samsung failed to show a motivation to combine Goodnick and Taubenblatt as proposed. The robust § 103 theory for '336 is therefore Ground 1, anchored on Grupp '483, because it shares the inventors' own specification and thus meets essentially every limitation on its face — leaving only the "metal oxide genus" generalization (Jammy) and the optimum-thickness point (Iwaguro) to be supplied.

7. Reconciliation with the earlier sections

• ✅ Consistent: claims 9–12 and 65–76 disclaimed (2020‑10‑14) and not instituted; instituted set = 1–8, 13–16, 77–80; IPR2020‑01204 instituted 2021‑01‑13, FWD 2022‑01‑12 (APJs Moore, McNamara, Kenny); IPR2020‑01264 not instituted.
• ⚠️ Correction to the earlier "Patent summary": that section described the independent claims as "reconstructed" and listed a "metal oxide" or "arsenide" interface layer as if interchangeable. The IPR record is more precise: all challenged claims recite a generic metal oxide interface layer; the arsenic family is the (unchallenged) subject matter reflected in the abstract. Use the metal-oxide characterization for § 103.
• ⚠️ Uncertainty retained: I could not verify, claim-by-claim, the disposition of claims 1–8, 13–16, 77–80 in the FWD (Paper 48). Public reporting ("Samsung Strikes Down Some Patents Tied To $25M Trial Loss") and the related '691 record (Board finding no anticipation for '691 claims 18 and 26–29) show the outcome was mixed across the family. Do not assume all '336 claims fell. The precise list must be read from the FWD itself.

8. Bottom line

A strong prima facie § 103 case exists against the challenged claims of '336, built primarily as:

Grupp '483 (primary; § 102(b) art if the claims lack pre‑2011 priority) + Jammy (metal-oxide interface layer + passivation) + Iwaguro (U-shaped optimum-thickness curve) + Chang (specific contact resistance),

with Goodnick + Taubenblatt 1984 (+ Jammy, + Chang) as the principal alternative that does not depend on the priority challenge. The motivations are field/problem identity, the finite and predictable set of known interfacial dielectrics, and the known tunneling-vs-barrier trade-off whose optimum Iwaguro expressly teaches.

Caveats: (1) The single biggest swing factor is the effective filing date — if Acorn's 2002 priority survives on the merits, Grupp '483 drops out and only Grounds 2–3 remain (and those drew adverse motivation-to-combine findings in the sibling IPR2020‑01282). (2) The "stack of metals" and metal-substitution sub-combinations are the weakest § 103 theories on this record (cf. the Board's rejection of Jammy+Cote for titanium). (3) Claims 9–12 and 65–76 are statutorily disclaimed and outside any enforcement analysis. (4) I have not verified the exact FWD claim-by-claim result — verify against PTAB Paper 48 before relying on any specific claim-level conclusion.

Sources: https://patents.google.com/patent/US8766336/en · https://www.docketalarm.com/cases/PTAB/IPR2020-01204/Inter_Partes_Review_of_U.S._Pat._8766336/ (Petition, ‑2, App. 13/687,907; Petitioner's Statement on Parallel Petitions) · https://ipverse.greyb.com/ptab-web/cases/case-details/IPR2020-01204 · Ex. 1003 (Declaration of Dr. E. Fred Schubert, listing Goodnick, Taubenblatt 1984, Chang, Rhoderick, Sobolewski, Iwaguro) · https://www.winston.com/a/web/[248292](/patent/248292)/PTAB-IPR2020-01282-20.pdf (motivation-to-combine findings) · https://www.winston.com/a/web/[248275](/patent/248275)/PTAB-IPR2020-01207-21.pdf (Grupp '483 § 102(b) prior-art status).