US 9935165

I'll verify the patent details against live sources and check for any litigation activity.

Let me look for the actual claim set and any litigation records.

Let me try to pin down the actual granted claim language.

US Patent 9,935,165 — Summary

Important note on sources: The full patent text you supplied is authoritative and cuts off partway through the description — immediately after the start of the "Actions and Effects" discussion of Comparative Example 1 — so the verbatim granted claim set is not present in that text. My claim overview below is therefore drawn from the patent's own "Summary of the Invention" section (which recites the illustrative embodiments in claim-like form), plus my searches. I flag where this creates uncertainty.

Bibliographic data

Assignment chain (from the Google Patents record): JOLED Inc. → INCJ, Ltd. (security interest, 2023-04-20) → correction by affidavit (2023-06-12) → JDI Design and Development G.K. (2024-01-31) → Magnolia Blue Corporation (2025-07-18).

Related application: A continuation, US 15/880,990 (filed 2018-01-26), issued as US 10,283,579 B2 (2019-05-07), bearing the same title and inventorship.

Apparent metadata discrepancy — flagged, not resolved: The Google Patents record lists a "prior art date" of 2015-08-31 for this patent, which does not match the only priority claim stated on the face of the specification (JP 2016-033836, filed 2016-02-25). I could not reconcile these in this session and am not auto-correcting either date.

Abstract (verbatim, as retrieved)

"There is provided a semiconductor device that includes a substrate, an electric field shielding layer, and a semiconductor element. The electric field shielding layer is provided on the substrate. The semiconductor element includes an electrode, and is provided on the electric field shielding layer with an insulating film in between."

Plain-language overview of the independent claims

Based on the patent's Summary section, the disclosure defines four independent aspects. Grouped in plain terms:

1. Semiconductor device — a substrate (typically a flexible resin substrate); an electric field shielding layer formed on the substrate; and a semiconductor element (e.g., a thin-film transistor, TFT) containing at least one electrode, disposed on top of the shielding layer with an insulating film in between.
   Plain meaning: you put a conductive "shield" layer between the plastic substrate and the transistor, so that when voltages are applied to the transistor's gate/source/drain, the resulting electric field can't reach the substrate and induce charges or mobile species there.

2. Display unit — the same semiconductor device stack, plus a display element layer containing multiple pixels formed on top of the semiconductor element (the specification's example is an organic EL element with anode / organic EL layer / cathode, with the anode coupled to the TFT's source-drain electrode).
   Plain meaning: the shielded-transistor backplane plus the light-emitting pixel layer on top.

3. Method of manufacturing a display unit — three recited steps: (a) forming the electric field shielding layer on the substrate; (b) forming the semiconductor element (with an electrode) on the shielding layer with an insulating film in between; and (c) forming the display element layer including multiple pixels on the semiconductor element.

4. Electronic apparatus — an apparatus (television, PC, smartphone, tablet, mobile phone, digital camera, etc.) that includes the display unit of aspect 2.

Uncertainty — stated explicitly: I could not retrieve the issued claim numbers (e.g., whether these correspond to claims 1, 7, 13, etc.), the preamble wording as granted, or the full set of dependent claims in this session. The Summary language above mirrors but is not guaranteed to be verbatim-identical to the allowed claims. The specification does describe subject matter that may appear in dependent claims — e.g., the dual-layer organic/inorganic insulating film to suppress "back-channel effect"; the shielding layer having an "island part" that overlaps the semiconductor layer in plan view; the shielding layer being a light-transmissive conductive film (ITO, IZO, n⁺-type a-Si) to survive laser repair of metal wiring; a fixed/ground potential on the shielding layer; sheet resistance ranges; a metal thin film on the substrate's back side; and (in Example Embodiment 2) an opening through the substrate filled with a conductive layer that ties the shielding layer to a grounded metal film/housing. Whether each of these is claimed, and at what dependency, is not confirmed here.

Litigation / CAFC check

• I found no Federal Circuit 2026 docket entry and no district-court litigation naming US 9,935,165. Searches for "9935165" in Federal Circuit 2026 summaries and for Magnolia Blue / JOLED infringement actions returned nothing tied to this patent.
• Caution on search noise: several PTAB filings discuss a different "'993 Patent" (a method of forming a transistor gate with a Ti/WN/W metal stack) and various display-apparatus petitions reciting "shield layers." Those are unrelated patents that happen to share the digits "993"; they should not be conflated with US 9,935,165. Likewise, the "Magnolia Medical Technologies" Delaware case is a different entity from Magnolia Blue Corp.
• Caveat: the absence of litigation results may reflect search-coverage limits rather than a definitive absence of disputes; I'd verify against Docket Navigator / PACER / the USPTO PTAB record for a definitive answer.

Bottom line: US 9,935,165 is a JOLED-originated (now Magnolia Blue Corp) patent on placing a grounded, transparent electric-field shielding layer between a flexible substrate and the backplane TFTs (and under the OLED pixel layer) to suppress bias-stress-driven threshold-voltage drift, plus a substrate-embedded conductor variant for grounding the shield.

I'll research litigation involving US patent 9935165 across patent litigation databases.

Let me search litigation-specific databases for this patent.

Let me find the specific JOLED v. Samsung case details and which patents were asserted.

Bottom line

Based on my searches, I could not confirm any litigation that specifically asserts US Patent No. 9,935,165 ("the '165 patent") as of the current date. I found no PACER docket, Unified Patents litigation entry, RPX record, or news report tying this specific patent number to a filed case. I want to be explicit about this rather than manufacture a match, because several prominent results for "the '516 patent" refer to entirely different patents (see "False positives" below).

Caveat on my research: my tooling hit its step limit before I could exhaustively query PACER, the Unified Patents caselist, and RPX/Insight directly for this number. The conclusion above is what the available evidence supports, but it is a "not found," not a certified "none exists."

What the patent is (for identification, not auto-corrected)

• Patent: US 9,935,165 B2 — Semiconductor device, display unit, method of manufacturing display unit, and electronic apparatus
• Application: US 15/207,047; filed 2016-07-11; granted 2018-04-03
• Priority: JP 2016-033836 (JP 6654466 B2); the front-page "prior art date" is listed as 2015-08-31, but the claimed priority is 2016-02-25 — I have not reconciled that discrepancy.
• Original assignee: JOLED Inc. Current assignee (per Google Patents): Magnolia Blue Corp.
• Assignment chain: JOLED Inc. → JDI Design and Development G.K. (recorded 2024-01-31) → Magnolia Blue Corporation (recorded 2025-07-18).
• Related family member: US 10,283,579 B2 (continuation via US 15/880,990).

Confirmed findings

• Asserted in litigation: none located. I found no case in which the '165 patent number appears as an asserted patent.
• PTAB/IPR: none located for this number.
• The only hits that reproduced the number were the Google Patents page for US 9,935,165 itself, a patent-inventor aggregator (Takashige Fujimori / JOLED), and a 1999 Lehigh University dissertation (UMI Microform 9935165 — coincidental, unrelated).

False positives — do NOT cite these as litigation for US 9,935,165

The searches returned several items about a "'516 patent" that are different patents:

• Maxeon v. HSC (solar cells) — the "'516 patent" there is described as a Maxeon back-contact/solar-cell patent challenged in a DJ action (HSC v. Maxeon), not US 9,935,165.
• Ariad Pharmaceuticals v. Eli Lilly — the "'516 patent" is US 6,410,516 (NF-κB), not this patent.
• PDI / Danaher (CYBEREX BCM) — the "'516 Patent" is an unrelated branch-monitor patent.

None of these involve US 9,935,165.

Related JOLED litigation context (adjacent, not confirmed to include the '165 patent)

There is well-documented JOLED enforcement activity you may be thinking of, but my searches did not confirm the '165 patent was among the asserted patents:

• JOLED Inc. v. Samsung Electronics Co., Ltd., Samsung Display Co., Ltd., and [Samsung Electronics America, Inc.](/litigations/by-plaintiff/Samsung%20Electronics%20America%2C%20Inc.) — reported as filed June 22, 2020 in the U.S. District Court for the Western District of Texas (with a parallel action in the District Court of Mannheim, Germany). JOLED publicly announced it; multiple trade-press outlets (TechPowerUp, BusinessKorea, Legal Era, Chinese/Korean press) covered it. I could not verify the case number or the specific asserted patent numbers — the press coverage did not enumerate them, and I could not reach a docket listing before my search budget ran out. Reported accused products: Samsung Galaxy smartphones (OLED panels).
• Samsung Display Co., Ltd. v. JOLED Inc. et al. — No. 6:21-cv-00017 (W.D. Tex., Judge Alan D. Albright), filed January 8, 2021; defendants JOLED Inc. and ASUSTeK Computer Inc. This is a separate case in which Samsung Display is the plaintiff. A related complaint document references US 9,768,240 (a JOLED/Samsung Display OLED patent) — not the '165 patent.

Assessment and recommended next steps

Because the '165 patent recently moved to Magnolia Blue Corporation (recorded July 2025), it is the kind of asset that could be asserted by a non-practicing entity in the future; there is no public record I located showing that has happened yet.

To definitively confirm the "no litigation" conclusion, you should run these targeted queries (my step limit prevented completing them):

1. Unified Patents litigation caselist — search the patent number directly: https://portal.unifiedpatents.com/litigation/caselist
2. RPX Insight / docket databases (Darts-IP, UniCourt, Justia Dockets) — search "9,935,165."
3. PACER — query the patent number in W.D. Tex. (Albright) and other high-volume districts, and check the AO-120 ("Report on the Filing of an Action Regarding a Patent") records filed with the USPTO, which are keyed to the patent number.
4. USPTO Patent Center / Global Dossier "Litigation" tab for US 9,935,165, and the PTAB for any IPR/PGR naming it.

If you can share the JOLED v. Samsung 2020 complaint (W.D. Tex.) or its case number, I can verify whether US 9,935,165 appears among the asserted patents — that is the single most likely place this patent, if ever asserted, would appear.

I'll start by verifying the canonical "no proceedings" signal against web sources before drafting.

Let me run a few more targeted checks for any proceeding or litigation referencing this patent.

Let me check litigation/assertion history and the Unified Patents portal listing for this patent.

Proceedings overview

Zero AIA trial proceedings on file. The USPTO Open Data Portal returns no IPR, PGR, or CBM proceeding naming US 9,935,165 as the challenged patent — breakdown: 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denied. My independent web checks (PTAB petitions portal, Unified Patents PTAB database, JOLED-party IPR filings) corroborate the null result; I found no petition, institution decision, FWD, or CAFC appeal touching this patent. Defensive posture: this is a blank slate, not a hardened patent. Nothing here has been tested at the Board — no claim has been canceled, but none has been blessed either. A defendant gets no free invalidity judgment to lean on, and equally faces no § 315(e)(2) estoppel and no adverse claim-construction ruling binding anyone. The patent is simply unchallenged, which is a weaker-than-average signal for a patent that has been in force since 2018-04-03 and has changed hands three times since 2023.

Because there are no proceedings, the per-proceeding template below is empty. I have added a short "nearby proceedings" block so you do not mistake related JOLED/Samsung activity for coverage of this patent.

---

No proceedings to report

No ### IPR20XX-XXXXX sections follow because none exist for US 9,935,165. I am not going to manufacture proceeding numbers or infer outcomes from a different patent in the same family.

---

Nearby proceedings that do not cover US 9,935,165 (context only — do not cite as invalidating this patent)

Related continuity that matters for prior-art and claim-scope work: US 10,283,579 (filed 2018-01-26, granted 2019-05-07) is a same-spec JOLED continuation claiming priority through this family, and JP2016033836 / JP6654466B2 is the foreign priority. None of those shows PTAB activity either.

---

Strategic summary

Claim status: entirely UNTESTED. No claim of US 9,935,165 has been canceled, confirmed, or construed by the PTAB. The PGR window closed on 2019-01-03 (nine months after the 2018-04-03 grant), so any AIA challenge today must be an IPR under §§ 311–319, limited to § 102/§ 103 grounds on patents and printed publications, or a CBM (unavailable — this is a semiconductor/display-structure patent, not a covered business method). Ex parte reexamination under § 302 remains fully available and requires no petitioner standing, which is worth noting given the 2025 USPTO environment.

Estoppel landscape: nothing is barred. Section 315(e)(2) estoppel attaches only to a petitioner whose IPR was instituted and reached a final written decision. Because no IPR was instituted against this patent, no estoppel exists against any party. Every prior-art ground is on the table for a new petitioner. The only hard deadline is the § 315(b) one-year bar: if you have been served with a complaint alleging infringement of the '165 patent, your petition is due within one year of that service. If you have not been served and are contemplating a preemptive filing, there is no statutory clock — but see the discretionary-denial caution below, and note that a declaratory-judgment-driven preemptive IPR is the classic way to avoid being boxed in later.

Pattern signals. JOLED was an aggressive but conventional operating-company assertor: it sued Samsung in the U.S. and Germany in 2020 and Samsung struck back with IPRs — just not on this patent. The ownership chain is the more interesting signal for a defendant today. Per the assignment record: JOLED → INCJ security interest (2023-04-20) → correction by affidavit (2023-06-12) → JDI Design and Development G.K. (2024-01-31) → Magnolia Blue Corporation (2025-07-18), which is the current assignee and expiry is 2036-07-11. That is a liquidation-successor/NPE-style ownership posture for a patent whose terminal value is monetization. There is no evidence in my searches of a defensive aggregator (Unified Patents, RPX, LOT) filing against it — which cuts against a coordinated invalidity effort existing already, and means the first real challenge is probably still ahead.

---

Recommended next steps

If you are a defendant: there is no FWD to cite and no canceled claim to point at. Your validity story has to be built from scratch, and the good news is that you have the entire prior-art universe. Two practical anchors: (1) the specification's own admission of the prior art at JP-A 2013-235598 and JP-A 2014-103410, both of which describe an electrode formed below a circuit including a TFT — that is a strong § 103 starting point against a claim whose entire point is a shielding layer between substrate and semiconductor element; and (2) the breadth of independent claim 1 ("a substrate; an electric field shielding layer provided on the substrate; and a semiconductor element including an electrode, and provided on the electric field shielding layer with an insulating film in between") invites a § 112(b) indefiniteness or § 103 obviousness attack on "electric field shielding layer" as a functional label rather than structure.

Timing if you do file: check your § 315(b) clock first. On the merits timeline, the Board's statutory framework gives an institution decision at ~6 months from petition and a final written decision at 12 months from institution — so budget roughly 18 months petition-to-FWD. Link: PTAB E2E / USPTO PTAB Decisions.

Discretionary-denial caution — this is the live risk, not the merits. Under the 2025–2026 regime, the Director personally controls IPR institution and denies via terse or unreasoned summary notices; the "settled expectations" factor targets patents in force more than roughly six years, and this patent's priority is 2016-02-25 with a 2018 grant — squarely past that line. A Fintiv-style discretionary denial is a realistic outcome even on a strong petition, and Sotera-type stipulations are no longer dispositive. Plan accordingly: either stipulate aggressively and file early, or consider ex parte reexamination, which hit an all-time-high 726 requests in 2025 precisely because PTAB institution became unreliable. Sources: Irwin IP, PTAB 2025 Year in Review and Unified Patents, 2025 in Review.

Caveats. I could not query the USPTO ODP API or PTAB E2E directly from this session; my verification rests on the canonical zero-count block plus public search, and pre-2019 filings occasionally index poorly. If a proceeding was filed between roughly 2018 and 2020 and heavily designated confidential, a docket-level check on PTAB E2E by patent number and by the three assignee names (JOLED, JDI Design and Development, Magnolia Blue) would be the confirming step before you rely on the absence in a brief.

I'll research the assignment records, correspondents, and entity backgrounds before assembling the report.

Let me get more specifics on the assignment records and the entities involved.

Assignment & Ownership Analysis — US 9935165 B2

Patent: US 9935165 B2 — Semiconductor device, display unit, method of manufacturing display unit, and electronic apparatus
Application: 15/207,047 · Filed: 2016-07-11 · Granted: 2018-04-03 · Anticipated expiration: 2036-07-11 (per Google Patents legal events)
JP priority: JP2016-033836, filed 2016-02-25 (counterpart JP6654466B2)

Data-quality note up front: The authoritative patent text supplied for this patent surfaces the conveyance types, parties and recording dates of the post-issuance chain but does not surface reel/frame numbers (except one) or the recorded correspondent. I could not reach the USPTO Assignment Center record itself within this session. Where the correspondent or reel/frame is unknown I say so explicitly rather than guess — the correspondent field is the single highest-value missing datum for the troll-pattern test below.

---

Inventors

All six are named as assignors in the 2016-07-11 assignment of assignors' interest to JOLED Inc. — the standard inventor-to-employer filing assignment.

Pattern note. JOLED was incorporated in January 2015 out of the OLED display units of Sony and Panasonic, so these six were almost certainly inherited from one of those parents; the individual pre-JOLED employers are not determinable from the record before me. I found no evidence that any inventor departed within 12 months of filing — the "departure precedes fire-sale" heuristic is not triggered here on the data available (and note this portfolio's distress event came ~7 years after filing, not <12 months, so the heuristic would not apply regardless).

---

Original assignee

JOLED Inc. (Tokyo, Japan) — named assignee on the issued patent; a 2015 joint venture of Sony and Panasonic, capitalised in part by the government-backed INCJ (Innovation Network Corporation of Japan). JOLED's business was printed/OLED display technology — research and development plus limited production (industrial/medical monitors and signage panels). It is best characterised as a display-technology operating company, not a patent-holding vehicle.

Product embodying the claims: JOLED was an operating panel developer and did bring OLED display products to limited markets, but it never achieved mass-production scale and was chronically unprofitable. I would not call it a high-volume commercial shipper of the claimed TFT/backplane architecture.

Current status: Insolvent / dissolved. JOLED filed for civil rehabilitation (民事再生手続 — the Japanese analogue of Chapter 11) on 2023-03-27; the commencement decision issued 2023-04-03. Its OLED technology-development business and IP were carved out and transferred to JDI Design and Development G.K., a subsidiary of Japan Display Inc. (JDI), for roughly ¥1 billion, with 2023-06-30 as the target execution date. JOLED's manufacturing/sales business was wound down.

---

Assignment timeline

Chronological, from the legal-events record supplied (Google Patents, sourced from USPTO assignment data):

• 2016-07-11 (executed) / recorded 2016-07-11 — Reel/frame not surfaced

  • Conveyance: Assignment of assignors' interest
  • Assignor: Yuichiro Ishiyama; Yuichi Kato; Tomoatsu Kinoshita; Takashige Fujimori; Kenta Masuda; Keiichi Akamatsu
  • Assignee: JOLED Inc.
  • Correspondent: not surfaced in the authoritative text
  • Context: Customary inventor-to-employer assignment recorded contemporaneously with filing.

• 2023-04-20 (executed) / recorded 2023-04-20 — Reel/frame not surfaced (see 2023-06-12 below)

  • Conveyance: Security interest (security agreement / collateral record, not a transfer of title)
  • Assignor: Joled, Inc.
  • Assignee: INCJ, Ltd. (the Innovation Network Corporation of Japan, later renamed INCJ, Ltd.)
  • Correspondent: not surfaced
  • Context: Securitization — INCJ, a long-standing JOLED financier and shareholder, recording collateral over the patent estate. Note the timing: this is recorded 24 days after JOLED's civil-rehabilitation filing.

• 2023-06-12 (executed) / recorded 2023-06-12 — Reel/frame 063396/0671 (the record being corrected)

  • Conveyance: Correction by affidavit filed against reel/frame 063396/0671
  • Assignor: Joled, Inc.
  • Assignee: Joled, Inc.
  • Correspondent: not surfaced
  • Context: Change/correction of record only — no transfer of title. This is the only reel/frame number the authoritative text surfaces, and it is cited as the target of the correction; the underlying record it corrects is not labelled in the text, though on timing it aligns with the 2023-04-20 INCJ security interest.

• 2023-06-30 (target) / recorded 2024-01-31 — Reel/frame not surfaced

  • Conveyance: Assignment of assignors' interest
  • Assignor: Joled, Inc.
  • Assignee: JDI Design and Development G.K. (JDIDD), a wholly-owned subsidiary of Japan Display Inc.
  • Correspondent: not surfaced
  • Context: Distressed asset transfer out of JOLED's civil rehabilitation — a court-approved out-of-plan business transfer (事業譲渡) under Art. 42(1)(2) of the Civil Rehabilitation Act, covering JOLED's OLED tech-development business, ~100 employees, and its IP estate. This is the fire-sale link.

• 2025-07-18 (executed) / recorded 2025-07-18 — Reel/frame not surfaced

  • Conveyance: Assignment of assignors' interest
  • Assignor: JDI Design and Development G.K.
  • Assignee: Magnolia Blue Corporation
  • Correspondent: not surfaced
  • Context: Transfer-to-monetization-vehicle. Magnolia Blue was incorporated 2025-05-19 in Minato-ku, Tokyo, capital ¥10,000, business purpose IP management and licensing, representative Koichi Irie — a single-purpose IP-holding entity created alongside sister vehicles Magnolia Purple and Magnolia White. JDI contributed LCD/OLED patents into the three vehicles as in-kind capital, took a royalty-free licence back, then sold 100% of the shares of all three to Ichigo Trust, its largest shareholder (~78.2% of voting rights), completing 2025-07-30 as partial repayment of ~¥65 billion of JDI borrowing.

Correspondent flag: Because the correspondent field is not surfaced in the material available to me, I cannot run the repeat-correspondent test. Flagging that as a genuine gap rather than a null finding — it is the field most likely to reveal a single attorney running these recordings.

---

Timeline diagram

timeline
    title Ownership of US 9935165
    2016 : Filed by JOLED Inc
         : Inventors assign rights to JOLED
    2018 : Patent issued to JOLED
    2023 : INCJ records security interest
         : JOLED files civil rehabilitation
         : Correction by affidavit filed
    2024 : IP transferred to JDI Design and Development
    2025 : Assigned to Magnolia Blue Corporation
         : Magnolia Blue shares sold to Ichigo Trust

---

NPE / troll-pattern signals

1. Shell-entity transfer — PRESENT (with structural caveat).
The patent moved on 2025-07-18 from JDI Design and Development G.K. — an operating OLED-development subsidiary of an operating display manufacturer — to Magnolia Blue Corporation. Magnolia Blue hits the shell tells squarely: incorporated 2025-05-19 (weeks before the transfer), capital ¥10,000, stated business "management and licensing of intellectual property," no products in commerce, single-purpose, and its shares were immediately sold to a financial sponsor (Ichigo Trust) rather than retained by an operating company. The caveat: it is a Japanese 合同会社/corporation, not a Delaware or Texas LLC, and its owner is a large institutional asset manager rather than a one-person troll. The structure is nonetheless a textbook IP-monetisation vehicle.

2. Known asserter in the chain — NOT PRESENT.
I found no match for any of JOLED, INCJ, JDI Design and Development G.K. or Magnolia Blue against the named NPE lists (Acacia, Marathon, IV, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, Spangenberg entities). INCJ is a government-backed investment fund, not an asserter. I found no Unified Patents or RPX high-frequency-plaintiff listing for Magnolia Blue — unsurprising, since the entity is only months old.

3. Repeat correspondent across the chain — INSUFFICIENT DATA.
The recorded correspondent for each of the five events is not surfaced in the authoritative material. I will not infer a shared attorney from the pattern of conveyances alone. This test is unresolved, not negative.

4. Cascading transfers — PRESENT.
Four recorded events inside ~27 months: 2023-04-20 (INCJ security interest) → 2023-06-12 (correction) → 2024-01-31 (assignment to JDIDD) → 2025-07-18 (assignment to Magnolia Blue). Restricting to true title transfers, two consecutive hops (JOLED → JDIDD → Magnolia Blue) occurred in under 19 months. I could not confirm a common correspondent address because that field is not surfaced.

5. Pre-litigation transfer — NOT PRESENT (no suit found).
I found no infringement litigation naming US 9935165 in the material reviewed, so there is no transfer dated within 6 months of a first suit to point to. Note the 2025-07-18 transfer to Magnolia Blue is the kind of move that typically precedes assertion — flag it as a watch item, but as of today it is not a pre-litigation-transfer finding.

6. Bankruptcy fire-sale — PRESENT.
Original assignee JOLED Inc. filed civil rehabilitation on 2023-03-27 (commencement decision 2023-04-03). The patent estate was transferred to JDI Design and Development G.K. for approximately ¥1 billion, recorded 2024-01-31, in a court-supervised, out-of-plan business transfer. This is the Japanese-system analogue of the Kodak/Nortel pattern the test describes.

7. Privateering — UNCLEAR.
The 2025 structure — operating company (JDI) contributing patents to licensing vehicles, selling the vehicles' equity to its own largest shareholder, and retaining a royalty-free licence back — is monetisation, and JDI keeps the ability to practise. That is not the classic privateering shape (operating company arm's an NPE to sue competitors on its behalf), and I found no competitor-directed assertion. Marked unclear rather than present.

8. Defensive aggregator — NOT PRESENT.
The chain terminates at Magnolia Blue Corporation, a for-profit sponsor-owned licensing vehicle. No RPX, AST, LOT Network, Unified Patents or OIN participation anywhere in the chain — the patent is not neutralised.

---

Verdict

NPE — high confidence.

Three independent signals are present on the record: a shell-entity transfer to Magnolia Blue Corporation (IP-only vehicle, ¥10,000 capital, incorporated 2025-05-19, no products, recorded 2025-07-18); cascading transfers through two hops in under 19 months (JOLED → JDI Design and Development G.K., recorded 2024-01-31 → Magnolia Blue, recorded 2025-07-18); and a bankruptcy fire-sale out of JOLED's civil rehabilitation (filed 2023-03-27, estate conveyed for ~¥1 billion, recorded 2024-01-31). The chain terminates in a licensing-only entity whose equity was sold to a financial sponsor (Ichigo Trust), with the operating transferor retaining only a bare licence — the classic posture from which assertion is launched.

Two honest qualifications that keep this from being a slam-dunk: there is no litigation history on this patent yet, and the correspondent field — the strongest troll-pattern tell — could not be retrieved, so the repeat-correspondent test is unresolved rather than negative. If the Assignment Center record shows a single recurring recording attorney across the 2023-04-20, 2024-01-31 and 2025-07-18 links, that would move this decisively from "high" to "certain"; if it shows three unrelated correspondents, the monetisation read still stands but the organised-campaign read weakens.

Verify at: USPTO Assignment Center — search patent 9935165 (mirror: assignment.uspto.gov). Cross-check USPTO Patent Center for application 15/207,047 for the full chain-of-title documents and correspondents of record.

I'll search for the USPTO record and the cited references for US 9,935,165. Note up front: the authoritative full text supplied for this patent cuts off before the "References Cited" / patent-citation table, so I need to retrieve that from live sources rather than infer it.

The searches so far haven't surfaced the front-page "References Cited" table. Let me go directly at the patent's own PDF and citation listings.

Important provenance caveat — read first

I was not able to retrieve the front-page "References Cited" table for US 9,935,165 in this session, and I hit my tool-step limit before I could pull the issued-patent PDF (patentimages.../US009935165B2.pdf) or the FreePatentsOnline/PatentGuru citation lists directly.

Two things follow from that, and I want to be blunt about them:

1. The authoritative full text you supplied for this patent cuts off mid-description ("...the semiconductor layer 103 is provided in a selected region on the insulating film 102..." in the Comparative Example 1 discussion). It therefore does not contain the claim set, the "References Cited" table, or the examiner-cited art.
2. I will not invent patent numbers to fill that table. Per the operating rules, I invent nothing. Below I list only references I can actually ground, and I explicitly label the ones I could not verify.

---

Step 1 — USPTO record for 9935165 (verified, no auto-correction)

The record confirms this is the correct patent; nothing was substituted:

No similar number was substituted — the results returned were literally for US9935165B2. For completeness, the same searches also returned an unrelated 1999 Lehigh University dissertation bearing UMI microform number 9935165 (heavy-metal sorption). That is a coincidental numeric collision, not this patent, and I have excluded it.

---

Step 2 — Prior art actually tied to this patent

A. Applicant-admitted prior art (the only prior-art references the authoritative text surfaces)

The Background section expressly cites two references. These are the "Patent documents cited in the description" rather than examiner citations, but they are the strongest grounded art I have:

A1. JP-A No. 2013-235598 (Japanese Unexamined Patent Application Publication)

• Citation: Japanese Unexamined Patent Application Publication No. 2013-235598
• Publication date: 2013 (the JPO "2013-" series number; I could not confirm the exact publication day in this session — flag as unverified to the day)
• Content (per the '165 specification's own characterization): discloses "a semiconductor device that performs wireless communication and in which an electrode is provided below a circuit including an antenna and a circuit including a thin film transistor."
• §102 relevance: Highest relevance of anything located. Its disclosed structure — an electrode below a TFT circuit — reads directly onto the core independent-claim concept of "electric field shielding layer provided on the substrate; semiconductor element ... provided on the electric field shielding layer with an insulating film in between." If the recited electrode sits on/over the substrate with insulating material between it and the TFT, this anticipates the semiconductor-device independent claim (and, with added pixel language, the display-unit claim) under §102(a)(1) (publication pre-dates the 2016-02-25 priority) and independently under pre-AIA-style §102(b) reasoning if its publication preceded 2016-02-25 by more than a year (2013 does).
• Caveat: I could not read A1's actual disclosure; the mapping rests on JOLED's own admission in the '165 Background. The claim's functional label "electric field shielding layer" is the likely distinguisher — the reference may disclose the structure without the shielding purpose. That is a §103/§112 question, not a clean §102 kill.

A2. JP-A No. 2014-103410

• Citation: Japanese Unexamined Patent Application Publication No. 2014-103410
• Publication date: 2014 (exact day not confirmed in-session)
• Content: cited jointly with A1 as disclosing "a semiconductor device that performs wireless communication and in which an electrode is provided below a circuit including an antenna and a circuit including a thin film transistor." (The two are described as a pair, so likely the same assignee/family or companion filings.)
• §102 relevance: Same structural argument as A1 — an electrode beneath a TFT circuit. Cumulative with A1; useful as the §103 secondary reference or as an independent §102 reference if the examiner's mapping of A1 to a limitation fails.
• Caveat: identical — full disclosure not read.

Also cited on the face of the record (via Google Patents' "prior art keywords"/description): the specification's own statement that the disclosure "relates to..." — no further prior-art documents appear in the authoritative text.

B. Documents I found that CITE the '165 patent — these are NOT §102 prior art

These surfaced in search as forward citations (later documents listing US9935165B2). By definition they post-date the '165 priority and cannot anticipate or render obvious its claims. Listed so you don't mistake them for prior art:

• EP3097594 A4 — "Organic light-emitting diode display with bottom shields." Google Patents shows the US9935165B2 (en) * 2015-08-31 | 2018-04-03 | Joled, Inc. entry on this document's page.
• US 2021/0057417 A1 — "Semiconductor device including active region with variable atomic concentration of oxide semiconductor material..." — lists US9935165B2 (en) * in its citation block.

Caution / genuine lead worth chasing: a document titled "Organic light-emitting diode display with bottom shields" is conceptually on point for the shielding-under-the-TFT idea. Whether EP3097594 is a backward citation in the '165 front page (i.e., possible §102 art) or a forward citation (i.e., not art) is exactly the ambiguity I could not resolve. This is the single most important verification target — see Step 4.

C. Same-family documents (not prior art against this patent)

From the global family record, these share the '165 specification/priority and are not prior art to it:

• JP 2017-049568 A (app. JP2016033836; priority source, filed 2016-02-25) and its grant JP 6654466 B2
• US 2017/0062541 A1 (the '165 pre-grant publication)
• US 2018/0151659 A1
• US 10,283,579 B2 (continuation, app. 15/880,990)

---

Step 3 — § 102 mapping (as far as the record supports)

Because the granted claim set is not in the supplied text, this maps against the independent-claim concepts as recited in the patent's Summary section (substrate / electric-field shielding layer on the substrate / semiconductor element with an electrode on the shielding layer with an insulating film in between / + display element layer).

Bottom line on §102: the only anticipatory candidates I can ground are the two JOLED-admitted JP-A references, and both suffer the same weakness — the '165 claims are drafted around the function of "electric field shielding," which the admitted art is not described as possessing. That makes these references far stronger as § 103 obviousness art (substrate→shield→insulator→TFT is a concededly known stack; adding the shielding purpose is a predictable result) than as clean § 102 anticipation.

---

Step 4 — What must be verified, and how (I could not finish this)

To complete the citation list properly — and it is completable, I simply ran out of queries — do these three lookups:

1. Front page of the issued patent. Open https://patentimages.storage.googleapis.com/.../US009935165B2.pdf and read the "(56) References Cited" block (U.S. Patent Documents + Foreign Patent Documents) verbatim. That is the definitive answer and will list every examiner citation, including the JP-A numbers with their exact publication dates.
2. USPTO Patent Center, application 15/207,047 → Documents & Transactions → the Notice of Allowance / PTO-892 (List of References Cited by Examiner) and the IDS (PTO/SB-08). The IDS shows what the applicant cited; the PTO-892 shows what the examiner actually relied on. These often differ, and the IDS is where earlier JOLED/Sony OLED-shield art would appear.
3. Global Dossier / Espacenet citation tab for the family (US 15/207,047, JP2016033836, US 10,283,579) — the EPO search report there will show X/Y-category citations that can be cross-checked against the US record.

Explicit limits of this answer: (a) the issued "References Cited" table was not retrieved; (b) the publication dates of JP-A 2013-235598 (to the day) and JP-A 2014-103410 (to the day) are unverified; (c) the disclosure contents of both JP-A references are described here only as the '165 specification characterizes them, not from the references themselves; (d) I could not determine the direction of the EP3097594 citation. No patent number, date, or mapping above is auto-corrected or inferred — where the record is silent, I have said so.

I'll ground this analysis in the prior art actually identified on the patent's face and in what's searchable. Let me run several targeted searches.

Let me search for prior art addressing the core structural concepts.

Let me run a few more targeted searches on the more specific claimed features.

Obviousness Analysis — US 9,935,165 under 35 U.S.C. § 103

Note on the "Prior Art section of this page": The authoritative patent text supplied identifies exactly two references on the face of the document — JP-A 2013-235598 and JP-A 2014-103410 — both characterized by the patent itself in the Background as disclosing "a semiconductor device that performs wireless communication and in which an electrode is provided below a circuit including an antenna and a circuit including a thin film transistor." I could not retrieve the full texts of either within this session; I rely on the applicant's own characterization of them and flag that limitation. Everything else below is drawn from searchable art, with dates and any uncertainty stated.

Carried-forward caveat (from the prior sections): The supplied patent text truncates partway through the Comparative-Example-1 discussion, so the verbatim granted claim set is not verified. This analysis therefore proceeds against the claim-like language in the Summary of the Invention (substrate / electric-field-shielding layer on the substrate / semiconductor element including an electrode, on the shielding layer with an insulating film in between), and flags where the analysis would change if the issued claims add limitations. I also carry forward the unresolved date discrepancy (Google "prior art date" 2015-08-31 vs. claimed JP priority 2016-02-25). I am using 2016-02-25 as the effective filing date because it is the date the specification states; if the true date is 2015-08-31, the analysis is unchanged for the 2014-and-earlier references below and only affects the two marginal 2016 publications.

---

1. Legal standard

Post-AIA § 103 uses the Graham factors: scope and content of the prior art, differences between the prior art and the claims, the level of ordinary skill, and objective indicia. Under KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), a combination is obvious where an element was known in the field, there was a known problem with an identified, finite set of solutions, and the combination yields predictable results — or where the reference teaches that the claimed arrangement is "obvious to try." A claim is not saved merely because the inventors combined known elements; the question is whether the combination was within the ordinary creativity of the skilled artisan.

Critically for this patent, the claim language is almost entirely structural and generic: "a substrate," "an electric field shielding layer," "a semiconductor element including an electrode," "an insulating film in between." "Electric field shielding layer" is a functional label applied to any conductive body; the claim does not (as summarized) require grounding, a fixed potential, overlap with the channel, transparency, or a sheet-resistance value. That breadth is the central § 103 vulnerability.

---

2. Level of ordinary skill (PHOSITA)

A person having ordinary skill in the art here is a display backplane process/device engineer with ~3–5 years of experience, or an advanced degree with ~2 years, in oxide-semiconductor (IGZO/ITZO) TFT backplanes for flexible/plastic OLED displays. That artisan would know:

• oxide TFTs on polyimide (PI) substrates suffer threshold-voltage (Vth) drift under positive/negative bias stress driven by charge trapped at or migrating in the plastic substrate and its interface with the barrier/insulator stack;
• plastic substrates are prone to outgassing, moisture, and mobile-ion migration, and PI surface chemistry (e.g., fluorine-containing species) can charge under bias;
• the standard mitigations are barrier layers, thicker planarization stacks, and an underlying conductive/back-gate layer.

Post-dating literature corroborates that this was squarely the industry's known problem and its known solution set: Kim et al., ACS Appl. Mater. Interfaces (2024-02-28) describes PI-substrate surface charging causing abnormal positive Vth shifts and image sticking, cured by a charge-compensation layer in the backplane; Park et al., ACS Appl. Mater. Interfaces (2025) describes a "bottom shield metal" structure that "suppress[es] threshold-voltage shifts under positive bias stress" on unstable substrates. These are not § 102/§ 103 prior art (they post-date 2016), but they are probative that the problem and the conductive-layer solution were routine engineering, not inventive insight — the KSR "known problem / known solution" scenario.

---

3. Claim chart — the reconstructed independent claims

---

4. Combination A (strongest): KR 10-2014-0081487 A as the primary reference, in view of the cited JP art and flexible-display art

Primary reference — KR 10-2014-0081487 A ("Flexible display device and manufacturing method of the same"). (Publication numbering indicates a 2014 Korean publication — verify date, but 2014 is well before the 2016-02-25 priority even on the later date.) Its abstract and description disclose: an insulating, flexible substrate with an adhesive layer; a support layer; a buffer layer; a shielding layer formed on the buffer layer; a rear-surface insulating layer formed over the entire buffer layer so as to cover the shielding layer; a cell array formed on the rear-surface insulating layer, defining pixel regions and including a ground wire connected to an external ground power supply; and a light-emitting array and sealing layer over the cell array. The ground wire connects to the shielding layer through a contact hole penetrating the cell array and the rear-surface insulating layer.

This is, structurally, the claimed invention in miniature: substrate → shielding layer → insulating film → TFT-containing cell array → display element, with the shield grounded. Notably, its stated motivation is noise/charging mitigation without an external conductive film — the same problem-and-solution vector the '165 patent claims.

Secondary references:

• JP-A 2013-235598 and JP-A 2014-103410 (on the face of the '165 patent): "an electrode is provided below a circuit including… a thin film transistor." Whatever their purpose (they involve wireless communication/antenna coupling), they teach the structural arrangement of a conductor beneath a TFT circuit — negating any argument that placing a conductive layer under the TFT was itself the inventive contribution.
• US 7,829,394 B2 and US 2013/0119561 A1 (Semiconductor Energy Laboratory): flexible OLED/display devices formed on a glass support, separated, and transferred to a flexible substrate — supplying the manufacturing-method steps (form on carrier, peel, attach flexible substrate) that the '165 method claim recites.

Motivation to combine (KSR articulation): Both references address the same field (flexible/plastic-substrate display backplanes) and the same problem (electrical instability of circuitry on a plastic substrate). The cited JP art makes the structural placement known; KR '487 makes the grounded shielding placement and its fabrication known. Combining them is the application of two known techniques to their known uses with predictable results — Vth stability improves, noise immunity improves, and process flow is unchanged. There is no new and unexpected cooperation between elements; each layer performs exactly its known function (barrier, planarize, shield, conduct).

---

5. Combination B: Semiconductor-Energy-style back-gate teaching + a flexible-substrate backplane

US 9,059,219 B2 (SEL, issued 2015-06-16) expressly teaches a conductive layer below the transistor as a "second gate electrode layer (what we call a back gate) which controls the electrical characteristics of the transistor," and states that "by setting the potential of the conductive layer to GND (or a fixed potential), the threshold voltage… is increased, so that the transistor can be normally off." US 2012/0002127 A1 (published 2012-01-05) likewise teaches a back gate electrode below the island-shaped oxide semiconductor film through an insulating film, with the potential "supplied with… a fixed potential such as a ground potential," the level controlling Vth — and adds that the back-gate layer also functions as a light-blocking film reducing negative-bias stress photodegradation, and gives a thickness range (10–400 nm). US 8,888,284 B2 and US 8,546,810 B2 supply the dual-gate/back-gate context and oxide-TFT interface engineering.

Motivation: These references are directly on point to the '165 patent's own rationale. The '165 patent itself concedes the shield "may serve as a 'second' gate electrode that controls the amount of flowing carriers from the back channel side." Once the art teaches (a) a conductor under the TFT with an insulator between, and (b) biasing that conductor to a fixed/ground potential to control Vth, the claimed structure is the predictable result of a finite, identified design choice — floating vs. grounded. KSR squarely forecloses patentability of that selection: "if a technique has been used to improve one device, and a person of ordinary skill in the art would recognize that it would improve similar devices in the same way, using the technique is obvious."

---

6. Combination C: US 5,859,677 + KR 10-2016-0001584 A

US 5,859,677 (issued 1999-01-12) discloses a TFT array with a conductive light-shielding film on the back channel side, formed over the substrate/insulating film, with a separate applied potential (Vsc) and specifically reporting Vth-shift/disclination suppression as a function of that potential. This is a very early, very explicit teaching of a biased conductive layer beneath a TFT.

KR 10-2016-0001584 A (Korean publication; numbering indicates a very early-2016 publication — if it published before 2016-02-25 it is § 102(a)(1) art; verify the date) claims a flexible OLED panel in which a lower protection metal ("하부보호금속") is formed in a buffer on the lower substrate, insulated from and overlapping the active of a transistor, with a planarization film and an OLED above. That is the claim-1 structure plus a display layer, in a flexible OLED context, with the island-like overlap feature.

---

7. Obviousness of the secondary/embodiment features

---

8. The patent's own disclosed advantage is a process-efficiency argument, not a technical one

The specification's strongest "unexpected" pitch for Embodiment 2 is against the patent's own Comparative Example 2 (a wiring pattern layer 47 penetrating the insulating film 13 through an opening H100). The stated benefits are: (i) fewer photolithography steps; (ii) no photosensitive material needed; (iii) ability to make the insulating film thicker; and (iv) better moisture/mobile-ion barrier. Points (i)–(iii) are process conveniences, and (iv) is the predictable consequence of not perforating a barrier film. Under KSR, "[a] known or obvious composition… does not become patentable simply because a previously unappreciated advantage is discovered." This is a strong § 103 point against the second-embodiment subject matter if it is claimed.

---

9. Teaching away, and why it fails

The only teaching-away candidate is the back-channel effect the patent itself describes: a grounded shield "causes an unintended decrease in a potential of the semiconductor layer… thereby increasing a threshold voltage." But:

• The patent does not avoid grounding — it keeps the shield grounded and simply increases the insulator thickness. That is routine spacing, not a departure from the art.
• The same reference art teaches that this effect is benign or beneficial: US 9,059,219 and US 2012/0002127 teach that a fixed/ground back-gate potential sets Vth, expressly to make the transistor normally off — a designer's objective, not a deterrent.
• KSR requires that a reference "criticize, discredit, or otherwise discourage" the claimed solution. A known side effect that is managed by ordinary design parameters is not a teaching away.

---

10. Objective indicia — thin, and the data does not help

• FIGS. 14A/14B compare the Example against Comparative Example 1 — the patent's own unshielded device, not against the prior art. Improvements over one's own control establish the invention works; they do not establish unexpected results over a § 103 combination. The asserted Vth-drift reduction is exactly the result the back-gate/shield references already teach (US 5,859,677's Vsc-dependent shift control; US 9,059,219's Vth-setting back gate).
• No commercial success: per the earlier assignment analysis, JOLED never reached mass production and entered civil rehabilitation in 2023 — a nexus argument for commercial success is essentially unavailable.
• No licensing-success, copying, or industry-praise record appears in the patent or the searchable record.
• Long-felt need cuts against the patent: the industry solved the same plastic-substrate Vth-instability problem by the same class of solution repeatedly and independently (barrier/charge-compensation layers; bottom shield metals).

---

11. Honest weaknesses in the invalidity case

1. I did not obtain the granted claims. If the issued independent claim contains a limitation I have not seen — e.g., requiring both the island-part overlap and a specific transparent-film composition and the sheet resistance, or a specific terminal-section topology (FIGS. 5A/5B) — the § 103 analysis tightens, though the individual features remain separately obvious.
2. JP-A 2013-235598 / JP-A 2014-103410 full texts unretrieved. My mapping of those two references rests on the applicant's own Background characterization. If they in fact place their electrode for antenna coupling only and nowhere suggest shielding a TFT from substrate charge, they are weaker as § 103 primary art — though they remain strong art for the structural element (conductor below TFT circuit).
3. The KR dates need verification (KR 10-2014-0081487 A and, especially, KR 10-2016-0001584 A). The latter is date-critical; if it published after 2016-02-25 it drops out as § 102(a)(1) art entirely.
4. US 10,446,634 B2, US 2024/0046861 A1, US 2016/0225842 A1 are NOT prior art to this patent and must not be cited as § 102/§ 103 references. They appear here only because they corroborate that the "shielding component between the substrate and the transistor, overlapping in projection, to stabilize Vth" concept became wholly conventional — which is KSR "design incentive" and "same problem/same field" support, not anticipation.
5. § 112 support: the functional naming of the "electric field shielding layer" invites a parallel § 112(b) indefiniteness attack and is itself evidence of claim breadth that aids § 103.

---

12. Bottom line

The strongest single-art case is KR 10-2014-0081487 A, which on its face appears to disclose substrate → shielding layer → rear insulating layer → cell array (with grounded connection to the shield) → light-emitting array. The strongest multi-reference case pairs the conductive back-gate/back-shield teachings (US 9,059,219; US 2012/0002127; US 5,859,677) with flexible-substrate OLED backplane art, with the KSR motivation supplied by the undisputed, long-known problem of plastic-substrate-driven Vth instability in oxide TFTs.

Recommended verification before relying on this in a brief: (1) pull the granted claim set from the US 15/207,047 file wrapper; (2) pull the English-language full texts of JP 2013-235598 and JP 2014-103410 (JPO/J-PlatPat) and the machine translation of KR 10-2014-0081487 A and KR 10-2016-0001584 A with their actual publication dates; (3) confirm the effective filing date given the 2015-08-31 / 2016-02-25 discrepancy, since it controls the marginal 2016 art.