US 6666545

I'll search for this patent's records and check for any litigation activity.

Let me run two more targeted checks for litigation/assignment activity and confirm the claim set.

US Patent 6,666,545 — Search Results

No CAFC 2026 docket activity found. My searches of public sources returned no litigation, appeal, IPR, or PTAB proceeding involving patent 6,666,545. I could not query the CAFC docket system directly, so I cannot rule out activity exhaustively — treat this as "nothing found," not as a verified negative.

Bibliographic Data

Assignment chain: ITRI → Transpacific IP I Ltd. (2009-10-27) → Chinchiko Ko Group Ltd., LLC (2015-11-20) → Intellectual Ventures Assets 104 LLC (2019-01-15) → 92104 LLC (2019-08-21).

Abstract (paraphrased)

The patent describes distributing several body contacts throughout the source of a large-area MOSFET active region in an inkjet print-head driver. Because the body contact sits close to the channel, the equivalent body resistance R_B drops, which the patent says prevents secondary breakdown. Placing the body contact inside the active region — rather than defining a separate body region in the field oxide outside the active region — is stated to save roughly 20% of the driver transistor area and lower per-chip manufacturing cost.

Claim Analysis

Only one independent claim exists (claim 1). All others depend from it. Note that despite the title's reference to "the method for making the same," the granted claim set contains no method claims — all seven claims are structural.

Claim 1 (independent) — plain language:
A driver transistor structure for an inkjet print-head chip. The structure has an active region containing multiple MOSFET elements wired in parallel, which control current to an ink actuator (e.g., a heater) electrically connected to the driver transistor. The characterizing feature: at least one body contact is located inside the active region itself (rather than in the field oxide outside it), and that body contact is electrically connected to the MOSFET source so both are held at the same voltage level.

Dependent claims 2–7 (plain language):

• Claim 2 — the body contact is located in the source region of the MOSFET in the active region.
• Claim 3 — the body contact is located close to (adjacent to) the source region in the active region.
• Claim 4 — the body contact extends to the boundary of the field oxide region adjacent to the active region.
• Claim 5 — the distance between the body-contact dopant region and the oppositely-doped source region is no more than 5 micrometers.
• Claim 6 — the actuator is a heater that generates thermal bubbles to eject ink.
• Claim 7 — the resistance R_B from the MOSFET channel in the active region to the body contact is reduced, thereby avoiding secondary breakdown and improving reliability.

Notable Prosecution / Prior Art Details

• Only one prior-art reference was cited: US 6,102,528 (Xerox Corporation, "Drive transistor for an ink jet printhead," issued 2000-08-15).
• The patent is classified in B41J2/1642, B41J2/14072, B41J2/1626, B41J2202/13, and related inkjet/manufacturing subclasses.
• A closely related ITRI family member exists: US 6,841,830 B2 ("Metal oxide semiconductor field effect transistors (MOSFETS) used in ink-jet head chips and method for making the same"), which cites 6,666,545 among related documents.
• The specification describes processing details relevant to construction: p-type Si substrate; LOCOS field oxide 8000–18000 Å; gate polysilicon and a "dopant barrier layer" 28′ etched to define the body-contact region; n+ source/drain implant blocked from that region; p+ (boron) body implant; TEOS/PSG/BPSG interlayer dielectric; contact holes 26b/26c/26d; heater and conductor layers.

Uncertainty Flags

1. Assignee/ownership: The "92104 LLC" designation comes from Google Patents' Google-maintained assignment data, which explicitly disclaims being a legal determination of ownership records. Recent ownership after the 2019 Intellectual Ventures transaction chain should be confirmed against USPTO assignment records if it matters.
2. Expiration: Google lists an anticipated expiration of 2022-01-08 (20 years from the US filing date). This is an assumption, not a legal conclusion — any terminal disclaimer or PTA in the file history could alter it. I did not retrieve the full file wrapper.
3. CAFC 2026 dockets: Absence of results is not proof of absence; I could not run an authoritative docket query.
4. Publication date nuance: Issue date is 2003-12-23, while the pre-grant publication (2003/0081070) appeared 2003-05-01 — consistent, not a discrepancy.

I'll search for litigation involving US patent 6666545.

Let me search more targeted litigation databases.

Let me try more specific searches for this patent's assertion history.

Litigation Search Results for U.S. Patent No. 6,666,545

No litigation involving U.S. Patent No. 6,666,545 was identified in my searches.

Summary of findings

I searched litigation-oriented sources (including Unified Patents' litigation portal, PTAB/PTACTS records, CourtListener, and general web queries keyed to the exact number "6666545" / "6,666,545") and found no district court complaints, CAFC appeals, PTAB proceedings, or other adversarial proceedings asserting or challenging this specific patent.

Careful to avoid false matches: search results surfaced references to different patents with similar-looking numbers (e.g., 6,616,397; 6,658,464; 6,557,054; 9,338,449; etc.), and to the patent's own related sibling case (US 6,841,830, also assigned to ITRI). None of those are U.S. 6,666,545, and I have not attributed any of that activity to this patent.

Case details table

There is therefore no plaintiff, defendant, jurisdiction, case number, filing date, or outcome to report.

Relevant context (not litigation)

The patent has a documented chain of ownership that is characteristic of patent-monetization transfers, though I found no record that it was actually asserted:

• Original assignee: Industrial Technology Research Institute (ITRI), Taiwan (assignment recorded 2002-01-08).
• 2009-10-27: assigned to Transpacific IP I Ltd. (effective 2009-09-25).
• 2015-11-20: assigned to Chinchiko Ko Group Ltd., LLC (effective 2015-09-22).
• 2019-01-15: assigned to Intellectual Ventures Assets 104 LLC (nunc pro tunc, effective 2018-12-03).
• 2019-08-21: assigned to 92104 LLC (effective 2018-12-21).
• Status: Expired – Lifetime (anticipated expiration 2022-01-08).

These successive transfers to well-known IP-holding/monetization entities are the type of ownership pattern often associated with litigation, but I found no evidence that U.S. 6,666,545 was ever asserted in suit.

Caveats / confidence

• Confidence: moderate-to-high that no litigation exists, but I was not able to run a complete, exhaustive check. Unified Patents' portal, PACER, and Docket Navigator are not fully indexable via open web search, and my tool access was limited. A definitive negative would require a paywalled search (e.g., Docket Navigator, Lex Machina, RPX) and a PACER query for the patent number across all districts.
• I am not aware of any CAFC appeal or PTAB proceeding for this patent, but absence of a search hit is not proof of absence.
• If you need a certified litigation negative, I recommend a focused Docket Navigator / Lex Machina search restricted to Patent No. 6,666,545.

Bottom line: Based on the searches performed, no litigation is known involving U.S. Patent No. 6,666,545.

I'll verify the structured "no proceedings" signal against public sources before finalizing.

Let me check for any assertion/litigation trail and confirm no AIA trials exist under the current owner.

PTAB Proceedings — U.S. Patent No. 6,666,545

"Driver transistor structure of inkjet print head chip and the method for making the same"
Assignee of record: 92104 LLC (apparent IV/Chinchiko Ko acquisition chain) · Priority: 2001-10-26 · Filed: 2002-01-08 · Granted: 2003-12-23 · Expired — Lifetime (anticipated expiration 2022-01-08)

Proceedings overview

Total AIA trial proceedings on file: zero. Per the structured USPTO ODP block supplied in this prompt — the canonical source for this task — the Office's Open Data Portal returns no IPR, PGR, or CBM proceedings for US 6,666,545, so the breakdown is 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution-denied. My independent web checks (Google Patents, PTAB/PTABTS document mirrors, general docket searches for the current assignee 92104 LLC and predecessor Chinchiko Ko Group) surfaced no AIA trial petition and no Federal Circuit appeal of any PTAB decision involving this patent; the only "6666545" hits were unrelated OSTI and prosecution materials. The bottom line for a defendant: the patent has never been tested at the PTAB — all seven claims are UNTESTED, and no § 315(e)(2) estoppel attaches to anyone. That cuts both ways: there is no invalidity judgment you can borrow, but there is also no hardened, PTAB-blessed claim set you have to overcome — and the patent is now expired, which changes the defensive calculus more than any IPR would.

Confidence / caveat: I could not confirm the absence from a first-party PTAB E2E or Docket Alarm query within this session; the structured ODP block is the authoritative signal I was given and it says "no AIA trial proceedings … as of the most recent ingest." ODP ingest can lag, and pre-AIA-era papers are sometimes thin in the API. Counsel should re-run a PTAB E2E search by patent number and a Docket Alarm / CourtListener PTAB query before relying on the null result in a filing. I flag this explicitly rather than assert a verified negative.

Proceedings

None to report. There are no proceeding numbers to list, and I will not invent any. Because the task template calls for claim-level outcomes, panels, institution decisions, FWDs, settlements, and appeals, I state plainly that none of those exist for this patent — there is no FWD to link, no disposition to quote, and no panel to name. Any output purporting to describe an IPR on the '545 patent would be fabricated.

For completeness, here is what the record does show, which is the absence-of-activity story:

• No petition ever filed — not by a competitor during the 2002–2015 ITRI/Transpacific ownership window, not during the 2015–2019 Chinchiko Ko window, and not after 92104 LLC took title in 2019.
• Prosecution art of record is thin: a single U.S. reference, US 6,102,528 (Xerox, "Drive transistor for an ink jet printhead," 2000-08-15). That reference was cited by the examiner and is not IPR-estoppel-encumbered as to any party, because no IPR exists.
• Forward citations show a live technology neighborhood (e.g., US 6,841,830 and US 7,527,360 to ITRI/Liu; US 7,018,012 to Lexmark/Edelen), confirming that overlapping art and skilled artisans exist — but that is prior-art ammunition, not a proceeding.

Strategic summary

Claim status: everything is untested. US 6,666,545 has 7 claims total: independent claim 1 (driver transistor structure of an inkjet print head chip in which "at least one body contact is installed in the active region and in electrical connection with the source of the MOSFET element for keeping them at an equal voltage level"), and dependent claims 2–7 (body contact in the source region; body contact close to the source region; body contact extending to the boundary of the adjacent field oxide region; dopant-region-to-source spacing "not over 5 micrometers"; actuator is a thermal-bubble heater; and the functional result of reduced R_B avoiding secondary breakdown). Not one of claims 1–7 has been canceled, confirmed, or construed by the PTAB. There are no surviving-claims-to-cite and no dead-claims-to-exploit from the trial side.

Estoppel landscape: a blank slate, and the one real defense is the calendar. Because no IPR, PGR, or CBM was ever instituted, § 315(e)(2) estoppel bars no one — no petitioner, no privy, no real party in interest. Any defendant being asserted against today can raise any § 102/§ 103 ground on any prior art, in the district court or in a new PTAB petition, without a "raised or reasonably could have raised" limitation. Two structural points matter more than estoppel here: (1) PGR is unavailable by statute — the '545 patent is pre-AIA with a 2001-10-26 priority date, well before the 2013-03-16 AIA cutoff; and (2) CBM is foreclosed — the transitional program sunset on 2020-09-16, and in any event this is a semiconductor print-head structure patent, not a "financial services" patent under the PTO's covered-business-method definition. So the only AIA vehicle that was ever open — IPR — is the only one that remains, and nobody used it.

Pattern signals: none of the classic aggregator/troll choreography. No serial petitioner, no repeat filings, no Federal Circuit appeal of a PTAB decision involving this patent, and no PTAB participation by a defensive aggregator such as Unified Patents. The ownership chain (ITRI → Transpacific IP → Chinchiko Ko Group → Intellectual Ventures Assets 104 → 92104 LLC, 2019-08-21) is a classic patent-monetization conveyor, and 92104 LLC is an IV-descended entity — yet even that chain produced no litigation or PTAB record I could locate for this patent. In practice the absence of any IPR on a patent that lived for ~20 years and passed through three monetization owners is itself a signal: the asset was either never asserted against a party with the incentive and budget to petition, or was asserted only in campaigns that settled quietly.

The decisive fact is expiration, not validity. The patent's 20-year term ran from the 2002-01-08 filing date and it expired on or about 2022-01-08 ("Anticipated expiration," Google Patents legal status; status "Expired - Lifetime"). All three maintenance fees were paid (year-4 in 2007, year-8 in 2011, year-12 in 2015), so this is a natural end-of-term expiration, not a lapse. Practically, that extinguishes prospective injunctive exposure and all post-2022 damages. Recovery, if any, is limited to past infringement within the § 286 six-year lookback measured from the date any suit is filed — which for a complaint filed now (2026) reaches back only to roughly 2020-01-08, and any earlier claim is time-barred. An expired patent also remains a valid IPR target (the Board can institute, applying Phillips-style construction to expired claims), but with no forward damages, a defendant's incentive to petition is low unless past damages are in play — in which case a § 102/§ 103 IPR filed now, with no estoppel and no prior PTAB record to fight, is unusually clean.

Recommended next steps

1. Treat the "no PTAB activity" finding as a provisional negative and verify it first-hand. Run the patent number in PTAB E2E (https://ptab.uspto.gov/#/login → Patent Trial search) and in USPTO PatentCenter (https://patentcenter.uspto.gov/) for any AIA review filing or certificate, and cross-check Docket Alarm / CourtListener PTAB dockets. The ODP block is authoritative for this exercise, but a null result is the one result you should independently confirm before asserting it to a court or in an invalidity contention.

2. If you are a defendant and the demand letter cites claims 1–7: there is no FWD to lean on, but there is an expiration defense and a clean IPR runway. No claim is canceled, so you cannot say "claim 1 is dead." What you can say, and should lead with, is that the patent expired on 2022-01-08, so the asserted claims cannot support prospective relief, and past damages are capped by § 286's six-year lookback from the filing date. Build your first line of defense on the calendar, not on § 315(e)(2).

3. If past damages are material, consider an IPR — it is unimpaired. With zero prior petitions there is no § 315(e)(2) estoppel, no FWD record to distinguish, and no PTAB panel history to overcome. The natural ground is § 103 over the neighborhood art — starting with US 6,102,528 (Xerox) plus the forward-citation cluster (ITRI/Liu US 6,841,830; Lexmark/Edelen US 7,018,012) — aimed at limitations your products do not practice, e.g., the "not over 5 micrometers" spacing of claim 5 or the field-oxide-boundary feature of claim 4, if your mapping avoids them. Note that no § 315(b) one-year clock is running until you are served with a complaint alleging infringement.

4. Do not plan around PGR or CBM. PGR is statutorily unavailable (pre-2013 effective filing date) and CBM has sunset (2020-09-16) and never fit this art. Do not let an outside counsel memo mislabel a hypothetical trial as a "CBM" — that would be a factual error about this patent.

5. Watch the related family for PTAB spillover. The '545 family is cited by/related to ITRI's US 6,841,830 and US 7,527,360 and to Lexmark's US 7,018,012; if any of those has PTAB history, its record may supply invalidity reasoning with no estoppel consequence for you on the '545 patent. Verify each separately — I did not confirm proceedings on those here.

Sources consulted: structured "PTAB proceedings on file" ODP block (canonical, no proceedings); Google Patents US6666545B2 — https://patents.google.com/patent/US6666545/en (claims, legal events, expiration, citations); PTAB E2E — https://ptab.uspto.gov/; USPTO PatentCenter — https://patentcenter.uspto.gov/. Web searches returned no AIA trial proceeding, PTAB decision, or CAFC appeal for US 6,666,545.

I'll research the assignment records for this patent, focusing on reel/frame numbers, correspondents, and NPE patterns.

Let me search for more specific assignment record details and correspondent information.

Let me research the entities in the chain and the correspondents.

Assignment Record Reconstruction — US 6,666,545 B2

Source note: Reel/frame values below come from the patent's recorded assignment data as reflected in Google Patents legal events (which mirror the USPTO assignment index). I was not able to retrieve the "correspondent of record" fields for these recordings within my available tool calls — the Assignment Center's correspondent data is not exposed in open search results for this patent. Where correspondent data is unavailable I say so rather than infer it. This is the single biggest gap in the analysis below and is flagged again in the signals section.

Inventors

All five appear to be ITRI employees/inventors under obligation to assign, consistent with the institute's standard practice. Assignment instruments were signed 2001-12-10 to 2001-12-12 — i.e., before the 2002-01-08 US filing date, which is the normal ITRI in-house practice and not an anomaly.

Departure pattern: No public evidence was found (or is expected) that the inventors left ITRI within 12 months, because ITRI is a research institute where inventors are career research staff rather than founders of a spun-out venture. Not a finding.

Original assignee

Industrial Technology Research Institute (ITRI) — Hsinchu, Taiwan. ITRI is Taiwan's largest government-backed applied research institute (a nonprofit R&D organization, not a commercial manufacturer). It is a serial patent seller: a Transpacific IP presentation dated to 2014 states that "912 US patents [were] assigned (sold) by ITRI," out of 1,133 located US assignments from ITRI (source: Transpacific IP 2014 deck, sris.com.tw mirror).

• Product embodying the claims: ITRI itself did not commercialize an inkjet print-head chip embodying the claims; it developed the IDH driver-transistor technology and licensed/assigned its patent estate. It is an operating R&D organization, not a shell.
• Current status: Operating (government-funded institute, active).

Assignment timeline

All reel/frame numbers below are as recorded in the USPTO assignment index and mirrored in the patent's legal-events data. Execution dates are the "effective date" fields.

• 2001-12-10 / 2001-12-12 (executed) / recorded 2002-01-08 — Reel 012461/0472

  • Conveyance: Assignment (employment/obligation-to-assign, original)
  • Assignor: Chien-Hung Liu; Jian-Chiun Liou; Charles C. Chang; Je-Ping Hu; Chun-Jung Chen
  • Assignee: Industrial Technology Research Institute (ITRI)
  • Correspondent: not retrievable from available sources.
  • Context: original institute assignment at filing — internal, not a monetization event.

• 2009-09-25 (executed) / recorded 2009-10-27 — Reel 023419/0985

  • Conveyance: Assignment
  • Assignor: Industrial Technology Research Institute (ITRI)
  • Assignee: Transpacific IP I Ltd. (Taiwan/Singapore)
  • Correspondent: not retrievable from available sources.
  • Context: transfer-to-asserter — ITRI divested this patent to a firm that self-describes as an NPE (see signals §2).

• 2015-09-22 (executed) / recorded 2015-11-20 — Reel 037107/0555

  • Conveyance: Assignment
  • Assignor: Transpacific IP I Ltd.
  • Assignee: Chinchiko Ko Group Ltd., LLC (Delaware)
  • Correspondent: not retrievable from available sources.
  • Context: transfer-to-asserter / portfolio aggregation — "Chinchiko Ko Group" is a single-purpose Delaware LLC name of the style used by Intellectual Ventures' holding-entity family; the next recorded link runs directly into an IV Assets entity, which is corroborating.

• 2018-12-03 (executed) / recorded 2019-01-15 — Reel 048014/0009

  • Conveyance: Nunc pro tunc assignment
  • Assignor: Chinchiko Ko Group Ltd., LLC
  • Assignee: Intellectual Ventures Assets 104 LLC (Delaware)
  • Correspondent: not retrievable from available sources.
  • Context: transfer-to-asserter / internal aggregation — movement into the named Intellectual Ventures asset-holding family.

• 2018-12-21 (executed) / recorded 2019-08-21 — Reel 050114/0899

  • Conveyance: Assignment
  • Assignor: Intellectual Ventures Assets 104 LLC
  • Assignee: 92104 LLC (Texas)
  • Correspondent: not retrievable from available sources.
  • Context: transfer-to-asserter / securitization-style SPV transfer — final hop to a numbered single-purpose LLC, the terminal holder of record.

No other recorded assignments were found. The chain is continuous and terminates at 92104 LLC.

Timeline diagram

timeline
    title Ownership of US 6666545
    2001 : Inventors assign to ITRI
         : Instruments signed in Dec
    2002 : US application filed Jan 8
    2003 : Patent issued Dec 23
    2009 : ITRI sells patent to Transpacific IP I Ltd
    2015 : Transpacific transfers to Chinchiko Ko Group LLC
    2019 : Nunc pro tunc transfer to IV Assets 104 LLC
         : IV Assets 104 transfers to 92104 LLC

NPE / troll-pattern signals

1. Shell-entity transfer — PRESENT. The patent leaves an operating R&D organization (ITRI, reel 012461/0472) for a licensing-only intermediary (Transpacific IP I Ltd., reel 023419/0985), then continues through two single-purpose Delaware/Texas LLCs (Chinchiko Ko Group Ltd., LLC, reel 037107/0555; 92104 LLC, reel 050114/0899). The LLC names and the terminal numbered entity are characteristic of holding SPVs.

2. Known asserter in the chain — PRESENT. Transpacific IP is self-described and externally recognized as an NPE. Its founder/CEO states: "we buy patent portfolios, then keep and monetise them ourselves… our primary focus now is finding high-quality portfolios that we can monetise by filing patent infringement cases — primarily in the United States — against big targets" (IAM Special Report Q3 2023, transpacificip.com). A Taiwanese academic report describes Transpacific IP as "亞洲最大的 NPE" ("Asia's largest NPE"). Intellectual Ventures — the next link (IV Assets 104 LLC, reel 048014/0009) — is one of the most-cited NPE/aggregator entities on public lists. Two named NPE-linked entities in the chain, each with recorded reel/frame.

3. Repeat correspondent across the chain — UNCLEAR (data gap). I could not retrieve the correspondent-of-record field for any of the four post-issuance recordings, so I cannot confirm or deny a recurring recording attorney/firm. This is the signal most likely to be decisive and it is exactly the one not available to me here. To close it, pull each reel/frame directly in the USPTO Assignment Center and read the "correspondent" column.

4. Cascading transfers — PRESENT. Reels 037107/0555 (2015-11-20) → 048014/0009 (2019-01-15) → 050114/0899 (2019-08-21): three successive transfers through chained LLCs, with the final two recorded only ~7 months apart, both executed in December 2018. Consecutive single-purpose entities with back-to-back execution dates is the classic cascading pattern.

5. Pre-litigation transfer — NOT PRESENT (no litigation found). No infringement suit naming this patent was identified; there is therefore no transfer dated within 6 months of a complaint. Caveat: absence of a hit is not proof of absence (see prior litigation section).

6. Bankruptcy fire-sale — NOT PRESENT. The original assignee, ITRI, is a government-funded nonprofit research institute and is not in bankruptcy. Its divestiture is a portfolio-monetization program, not a distress sale.

7. Privateering — UNCLEAR. ITRI is not a product competitor suing rivals through a proxy; it is a research institute. Transpacific IP's model (buying third-party portfolios and asserting "against big targets") is consistent with independent NPE behavior rather than a single operating company's proxy war, but I found no evidence of a specific operating-company beneficiary.

8. Defensive aggregator — NOT PRESENT. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. It terminates at a Texas LLC (92104 LLC) that is an assertion-capable holder of record.

Cross-reference check: These findings do not contradict the previously generated summary and litigation sections. Those sections correctly noted no litigation found and the same ownership chain; the reel/frame values here are consistent with the legal-events data those sections relied on.

Verdict

NPE — high confidence.

Two independent strong signals drive the call: (a) an operating research institute (ITRI) transferred the patent to a self-described NPE, Transpacific IP I Ltd., reel 023419/0985 (recorded 2009-10-27); and (b) the chain then passed through Chinchiko Ko Group Ltd., LLC, reel 037107/0555, to Intellectual Ventures Assets 104 LLC, reel 048014/0009, and finally to 92104 LLC, reel 050114/0899 — cascading single-purpose LLC transfers with back-to-back December 2018 execution dates. No litigation was found, so this is an ownership-pattern verdict, not an assertion verdict; the missing correspondent fields (Signal §3) are the one caveat that could either reinforce or blunt the repeat-player finding.

Verify at: USPTO Patent Assignment Search — assignment.uspto.gov/patent/index.html (search patent 6,666,545) or assignmentcenter.uspto.gov, then read reel/frame 012461/0472, 023419/0985, 037107/0555, 048014/0009, and 050114/0899 — specifically the correspondent column that I could not retrieve.

I'll research the cited prior art for this patent, focusing first on the sole examiner citation, then the related/similar documents.

Let me get the full claim set of the cited reference and check the related family member.

Let me verify one more reference that appears in the "similar documents" list and is highly relevant to the body-contact concept.

Prior-Art Analysis — U.S. Patent No. 6,666,545 B2

1. Confirming the patent (literal number 6,666,545)

Search results for the exact number 6,666,545 return the correct document — U.S. Patent 6,666,545 B2, "Driver transistor structure of inkjet print head chip and the method for making the same," App. No. 10/038,909, filed 2002-01-08, granted 2003-12-23, ITRI (TW priority 090126507, 2001-10-26). A USPTO-mirror record is available at:

• https://uspto.report/patent/grant/6,666,545
• https://patents.google.com/patent/[US6666545](/patent/US6666545)/en

I did not retrieve the USPTO PatentCenter file wrapper directly, so the citation list below is taken from the granted patent's front page as mirrored by Google Patents (which reproduces the USPTO "References Cited" data). No similar-but-different number (e.g., 6,666,455; 6,666,545-family lookalikes) is attributed here.

Important scope note on the citation list. The face of 6,666,545 lists exactly one patent citation:

Everything else (US 5,075,250, US 5,969,392, JP H026138, etc.) appears in Google's "Similar Documents" / "Cited By" tables, not in the USPTO citation list. I separate those into §3 so the distinction is explicit. Under AIA-transition rules, this application (filed 2002-01-08) is governed by pre-AIA 35 U.S.C. §§ 102/103.

2. The sole examiner citation — analyzed in full

Reference A — US 6,102,528

Brief description. Xerox's '528 patent is a high-resolution thermal-inkjet power MOSFET drive-transistor layout. Its stated problem is the opposite of 6,666,545's problem: as nozzle spacing shrinks, source-to-drain spacing shrinks, causing drain-depletion "punch-through" and loss of breakdown voltage. The solution is a p-type "pocket implant" that extends beneath the gate oxide and channel region (and optionally ~1 µm into the n⁻ drift region), letting source/drain/gate regions be shortened without losing breakdown voltage. Claim 1 (as reproduced in the search results) recites: an elongated drain region connected to a heating element; source regions on opposing sides of the drain, parallel to it, "connected to ground"; elongated gate regions; a gate oxide layer; an n⁻ drift extension with an n⁺ drain contact implant; and an n⁺ source region together with a p-type pocket implant extending beneath the gate oxide and surface channel region. The patent also describes two side-by-side parallel-connected transistors with three parallel source regions, all sources grounded, and addressing circuitry monolithically integrated on the same p-type silicon substrate. ("[P]arallel source regions 26 are formed on opposite sides of each drain and drift region, so that there are three sources for each pair of drains… The source regions are connected to ground.")

§ 102 element mapping against 6,666,545's claims:

Conclusion on US 6,102,528. It is genuine § 102(b) prior art and is the closest cited reference — it is a same-field, same-problem-space Xerox drive transistor for a thermal inkjet printhead with parallel MOSFETs, grounded sources, and a p-type implant near the source — but it does not disclose the single characterizing limitation of claim 1 (a body contact installed in the active region and strapped to the source). Accordingly, US 6,102,528 does not anticipate any of claims 1–7; it would be, at most, a § 103 starting point. Its presence on the face more likely reflects the examiner's field-of-search context (drive-transistor geometry in inkjet head chips) than a § 102 rejection position.

3. Additional prior art of substantive relevance (not on the 6,666,545 face)

These were surfaced from the "Similar Documents" table for 6,666,545 and from the background of US 6,102,528. They matter because the actual anticipating art for the "body contact in the active region" concept appears to be in the Hawkins/Xerox lineage, not in the one cited reference. Confidence is moderate — I worked from Google Patents, EveryPatent and EP-EPO document excerpts rather than full certified copies, and I have flagged where a limitation's presence is inferred rather than quoted.

Reference B — US 5,075,250 (the strongest anticipation candidate)

Brief description. This Xerox patent fabricates monolithic TIJ chips (MOS logic + drivers + polysilicon heater resistors) with fewer masks. Directly on point, it discloses that "some of the driver source contacts, preferably one in every six, are masked to shield them from the phosphorous implant. These contacts are dispersed about the wafer surface to enable p⁺ substrate contacts for grounding through the top surface of the wafer, thereby eliminating the parasitic bipolar effect generally associated with grounding through the bottom surface of the wafer." It further discloses opening vias in reflow glass (PSG/BPSG), then a self-aligned boron (p⁺) implant through those vias to form the substrate contacts, with vias to be made into substrate contacts sitting in n⁻ regions while sources/drains/gates are n⁺ doped. The EP 0 494 076 B1 counterpart states the same in the first person: "The present invention further eliminates a second mask level by self-aligning the boron substrate contact to the etched vias in the reflow glass layer… Areas which are desired for substrate contact are shielded from the phosphorous n⁺ implant."

§ 102 mapping against 6,666,545:

Why this matters. US 5,075,250 (and its EP 0 494 076 family member) is the reference I would expect a challenger to lead with against claim 1 and claims 1+2/3/5: it appears to disclose p⁺ substrate contacts formed inside the driver active region at the source, grounded to the top surface, with multiple such contacts "dispersed" across the wafer (i.e., distributed body contacts). That is the core of the 6,666,545 "BES (Body-contact Embedded in Source)" idea. The two surviving differentiators for 6,666,545 would be (i) the specific dopant-barrier-layer (polysilicon 28′) masking method and (ii) the explicit ≤ 5 µm spacing — and note that 6,666,545's granted claim set contains no method claims, so the barrier-layer process is not itself claimed.

Reference C — Background/context references cited within US 6,102,528

Reference D — Other documents in the 6,666,545 "Similar Documents" table worth a file-wrapper check

• US 5,969,392 — "Thermal ink jet printheads with power MOS driver devices having enhanced transconductance" (1999-10-19). Same problem space (power MOS driver transconductance/geometry in TIJ heads). I did not retrieve its full text; do not treat this as a § 102 finding.
• JP H026138 A / JPH026138A — "Silicon integrated circuit chip of bubble-ink jet printing mechanism" (1990-01-10). Pre-dates the priority date; a full-text (machine-translated) review would be needed to assess whether it shows a source-embedded body contact.
• US 2003/0136999 A1 (Hodges, "Semiconductor device with deposited oxide," priority 2002-01-18) — post-dates the 2001-10-26 priority; not § 102 art.

Reference E — Items that are NOT prior art (flagged to prevent mis-attribution)

• US 6,841,830 B2 (ITRI) and US 2004/0124449 A1 — the related sibling MOSFET/injet-head-chip application. Its listed priority is 2002-12-31, i.e., later than 6,666,545's 2001-10-26 priority. It appears in the "Cited By" table as a later document citing 6,666,545, not as prior art against it.
• TW 502379 B (family member of the same TW 090126507 priority) and CN 1421316 A ("Drive transistor structure and manufacturing method of inkjet printing head chip," 2003-06-04) — same-family counterparts of 6,666,545, not prior art.
• US 6,419,335 (Xerox/Gooray, electrostatic drive systems) — appeared only as search noise; different technology, not relevant.

Reference F — Admitted prior art inside 6,666,545 itself

The specification's FIG. 1 and accompanying "Related Art" text are an admission that the conventional driver transistor places body contacts 23 in a body contact region 20′ outside the active region 20 (in the thick field oxide, 9,000–17,500 Å), with each driver transistor surrendering ~⅙–⅓ of its area to it. This admitted prior art supplies every element of claim 1 except the characterizing clause ("body contact installed in the active region"); it therefore anticipates nothing but is the correct § 103 baseline and is directly relevant to the "~20% area saving" and "R_B reduction" motivations recited in claims 1 and 7.

4. Bottom-line § 102 assessment

Key takeaway: The one reference actually cited on 6,666,545 — US 6,102,528 — is a same-field Xerox drive-transistor patent (filed 1997-10-17, issued 2000-08-15; § 102(b)) that discloses parallel grounded-source MOSFETs driving TIJ heaters and a p-type pocket implant near the source, but not a body contact located in the active region and electrically strapped to the source, which is the sole characterizing limitation of independent claim 1. The more dangerous art for claim 1 is the earlier Hawkins/Xerox US 5,075,250 / EP 0 494 076 family, which describes masking selected driver source contacts from the n⁺ implant and converting them by self-aligned boron implant into top-side-grounded p⁺ substrate contacts dispersed across the wafer — essentially the "body-contact-embedded-in-source" concept.

Confidence and caveats

• High confidence that US 6,102,528 is the only patent citation on the 6,666,545 face and that it does not anticipate claim 1.
• Moderate confidence on the US 5,075,250 mapping: the quoted masking/self-aligned-boron passages are from Google Patents/EveryPatent/EPO excerpts, but I did not verify every figure, and the reference does not state a numerical ≤ 5 µm spacing for claim 5.
• I could not reach USPTO PatentCenter or PACER directly; the citation list was reproduced from Google Patents' mirror of the USPTO "References Cited" data and corroborated by uspto.report. A certified § 102 study should pull the certified copies of US 5,075,250 and EP 0 494 076 B1, plus the full machine translation of JP H026138 A, against pending/remaining claims.
• No contradiction was found with the previously generated summary and litigation sections; the single-citation finding is consistent with the prior "Notable Prosecution / Prior Art Details" entry.

I'll research the cited prior art to ground the obviousness analysis.

Good — I found a highly relevant reference. Let me verify its dates and US equivalent.

Let me pin down the EP0783966 family dates and US equivalent, plus check the other prior-art candidates.

Obviousness Analysis — U.S. Patent No. 6,666,545 under 35 U.S.C. § 103

Building on the claim analysis, litigation, and PTAB sections above: the claim set is seven structural claims (one independent), never tested at the PTAB, expired 2022-01-08. This section addresses only validity under § 103.

1. Framework

Assessed under the Graham v. John Deere factors: (1) scope and content of the prior art; (2) differences between the prior art and the claims; (3) level of ordinary skill; (4) secondary considerations. Because the '545 patent issued in 2003, KSR Int'l v. Teleflex (2007) governs the motivation analysis as applied today: the rationale may come from "common sense," from "known techniques used to improve similar devices in the same way," or from a "finite number of identified, predictable solutions." MPEP § 2143 lists the acceptable rationales I apply below (A: predictable combination of known elements; C: known technique, similar device; D: obvious to try; F: design incentive/market forces).

Level of ordinary skill (PHOSITA): a semiconductor process/device engineer with a B.S. in EE, applied physics, or materials science plus ~3–5 years designing power MOSFETs on silicon for thermal inkjet printheads, or equivalent — familiar with LOCOS isolation, polysilicon-gate MOS processing, n+/p+ masking and implants, and the reliability failure mode of parasitic-bipolar "secondary breakdown"/snapback in large-W/L power MOSFETs.

2. Prior art available

The asserted prior-art basis supplied on the patent page is thin — exactly one reference was cited on the face of the patent. That is itself a signal: it means the examiner did not reach the most on-point art in the field.

Caveat / flag: EP 0 783 966 and US 6,309,053 were not cited during prosecution of the '545 patent. They surfaced through my searches, not the page's face-cited art, and I flag them as such. The EP 0 783 966 dates come from the EPO Global Patent Index and Google Patents; the US family member's exact issue date should be confirmed before it is relied on in a filing.

3. Key claim constructions that drive the analysis

• "active region" — the area of the substrate not covered by the thick field oxide, i.e., where the LOCOS field oxide is absent. (The '545 specification defines the active region exactly this way.)
• "body contact … in the active region … in electrical connection with the source … keeping them at an equal voltage level" — a body/substrate tie inside the active area, strapped to source potential. This is the whole point of the claim.
• "body contact" is not limited to a p+ region in a field-oxide opening; a substrate-potential-fixing doped region within the active area reads on it.

4. Ground 1 — EP 0 783 966 (Canon), alone or as the primary reference

This single reference maps onto nearly every limitation. If it is anticipatory under § 102(b), it is a fortiori obvious — a reference that discloses the whole structure cannot be combined away.

Claims 2 and 3 — the contact unit is literally within/embedded in the source region ("intercepts the respective source region"), the strongest possible disclosure of claim 2; claim 3 ("close to the source region") is met by the same structure.

Claim 6 — thermal-bubble heater: expressly disclosed ("film boiling … air bubbles").

Claim 7 — reduced R_B / avoiding secondary breakdown: EP 0 783 966 is directed at the unintended parasitic transistor (Fig. 5) formed in the large-W/L MOS driver, and fixes the back-gate potential by bringing the contact unit into the source. The functional result is inherent. Independently, claim 7 is a statement of intended result and adds no structural weight under In re Schreiber / In re Umberger.

Remaining limits — claim 4 and claim 5. Claim 4's "body contact extends to the boundary of a field oxide region adjacent to the active region" is not cleanly disclosed by EP 0 783 966's embedded contact unit. Claim 5's "not over 5 micrometers" spacing is not expressed numerically, but is inherently satisfied because the contact unit and source are adjacent and separated only by the minimum design-rule spacing (well under 5 µm at the relevant lithography), and because US 6,102,528 independently discloses a p-type pocket implant extending "about 1 µm" into the drift region and to the adjacent source — under 5 µm.

5. Ground 2 — US 6,102,528 (Xerox) in view of EP 0 783 966 (Canon) / US 6,267,470

This is the defensive, primary-reference-plus-secondary-teaching ground for a defendant, and it also shores up claims 4 and 5.

• What '528 supplies: the complete inkjet printhead environment (integrated address circuitry, heaters, two parallel drive transistors per heater, grounded parallel source regions on opposite sides of an elongated drain, elongated gate, channel under the gate) and the express, printed motivation — reduce the silicon area per drive transistor, raise resolution (to 1200 spi), avoid punch-through, without losing breakdown voltage. '528 also places a p-type implant at the source/channel boundary ("pocket implant … extending from the channel region into the substrate to an adjacent source region"), which is the physical mechanism the '545 claim 5 spacing is aimed at.
• What Canon supplies: the teaching that in a large-W/L printhead power MOSFET, a substrate-potential-fixing contact unit is placed inside the source regions in the active area, and that the source and the contact unit are tied to the same potential, to defeat the unintended parasitic device.

Nothing in either reference must be bodily incorporated into the other: the combination is effected at the mask-layout level — the p+ body/contact window that the '545 patent's own admitted prior art already formed "by the barrier layer 24 of a polysilicon doped layer" is simply patterned inside the source (Canon) instead of in the field-oxide opening ('545 Fig. 1), and strapped to source metal.

6. Ground 3 — adding area-reduction art (Torgerson, and the printhead-driver background)

Where the asserted advantage is the ~20% area saving / lower cost per chip, that motivation is explicit in the art:

• US 6,309,053 B1 (Torgerson/HP) — grounds conductor over the drive-transistor active regions expressly to shrink die size; supplies the same "reduce area per driver" motive that the '545 patent claims as its benefit.
• US 6,102,528 — "the number of printheads which can be made from each silicon wafer decreases, thus driving up the manufacturing cost"; the express design incentive.
• US 5,969,392 / US 5,075,250 / JP H02-006138 — establish that integrating a large number of power-MOS drivers on the printhead die, and the associated area/cost pressure, was the settled problem in the field. (These are listed among the '545 page's "Similar Documents.")

7. Why a PHOSITA would have combined these references (articulated rationales)

1. Same field, same problem, same solution. All references concern MOSFET drive transistors monolithically integrated with heater resistors on a thermal-inkjet printhead substrate. KSR / MPEP 2143(A).
2. Known technique applied to a similar device in the same way. Tying the body/substrate to the source (a "body tie" / back-gate potential fixing) to suppress parasitic-bipolar snapback was a standard, documented power-MOSFET technique (Wolf, Silicon Processing for the VLSI Era — cited of record in '528). Applying it to the printhead power MOSFET is MPEP 2143(C): improvement of a similar device in the same way. Expected improvement, not new principle.
3. Finite, predictable locations. A body contact on an isolated active region can be placed in the field oxide (admitted prior art) or in the active region. Two choices; the second predictably shortens the R_B path. KSR "obvious to try" / MPEP 2143(D).
4. Explicit design incentive. Higher nozzle counts → more drivers per die → smaller drivers per die without losing breakdown voltage or increasing resistance. '528 states this; Torgerson acts on it. MPEP 2143(F).
5. The patent's own background concedes the pieces. The '545 specification admits that defining body contacts via "the barrier layer 24 of a polysilicon doped layer" was known, that the comb-gate large-W/L MOSFET was known, and that the only problem was distance from channel to body contact (>400 µm). Moving an already-known contact to a shorter distance is the definition of an obvious mechanical/jurisdictional relocation with a predictable result. No teaching away is present — placing contacts in the field oxide was layout convenience, not a stated recognition that in-active-region contacts would fail.
6. Reasonable expectation of success: high. The modification is a mask redefinition in a mature LOCOS/polysilicon-gate flow; both structures use p-type substrate, n+ source/drain, p+ contacts, and the same interlayer dielectric. No new materials, no new process modules, no unpredictable physics.

8. Secondary considerations / rebuttal

There is no evidence of record (and none found in my searches) of commercial success, copying, long-felt unsatisfied need, or unexpected results. The asserted ~20% area reduction is an arithmetic consequence of eliminating a separately defined field-oxide body region — precisely the kind of predictable result that does not rebut obviousness. Nexus is absent. The '545 patent's "less than 5 µm" spacing is not shown to be a critical limit; the specification attributes no unexpected result to it, so claim 5 is routine optimization of a result-effective variable with a broad claimed range.

9. Weaknesses in the invalidity case (be candid if you rely on this)

1. Claim 4 is the weakest. "Body contact extends to the boundary of a field oxide region adjacent to the active region" is not clearly disclosed in EP 0 783 966. The best support is that claim 4 is a routine layout choice — the admitted-prior-art body contact is in the field oxide immediately abutting the active region (see '545 Fig. 1), so extending the in-source contact to that same boundary is an obvious spatial allocation. Consider adding a reference showing a p+ guard/body region abutting field oxide.
2. Claim 5's numeric limit needs a "routine optimization / result-effective variable" argument, or the inherency argument (adjacent = minimum design-rule spacing ≪ 5 µm).
3. "In the active region" construction could be contested for a contact that merely "intercepts" the source; expect a claim-construction fight rather than a clean § 102 kill.
4. Date/identity diligence. Confirm (a) EP 0 783 966 A2 publication date 1997-07-16 and B1 grant, and (b) the exact issue date and family relationship of US 6,267,470 B1, before citing either as § 102 art in a filing. I verified the EP priority dates (JP 271396, 1996-01-11; JP 34952996, 1996-12-27) via the EPO Global Patent Index but did not pull a certified copy.
5. No method claims exist in the granted set, so the "dopant barrier layer 28′" process details do not narrow the claim scope — the inquiry is purely structural, which favors the invalidity case.

10. Bottom line (tying back to the earlier sections)

• The '545 claim set is broad and structurally thin; the sole independent claim requires only "a body contact in the active region tied to the source," which EP 0 783 966 / US 6,267,470 (Canon) appears to disclose outright — art the examiner never cited against a patent that cites only US 6,102,528 (Xerox) on its face.
• US 6,102,528 (Xerox) + EP 0 783 966 (Canon) is the most defensible § 103 combination, with US 6,309,053 (Torgerson) added for the area-reduction motivation and US 5,969,392 / US 5,075,250 for background.
• Because no IPR/PGR/CBM ever issued (per the PTAB section above) and the patent expired 2022-01-08, this invalidity analysis is fully unencumbered by § 315(e)(2) estoppel, and an IPR remains statutorily available for past-damages defense — with claims 1, 2, 3, and 6 the most vulnerable and claims 4 and 5 the residual fight.

Confidence: high that claims 1–3 and 6 are invalid over EP 0 783 966 (Canon) alone or in combination with US 6,102,528; moderate-to-high for claims 5 and 7; moderate for claim 4, which needs a supplementary reference or a routine-optimization showing. All prior-art dates other than those I directly quoted should be independently confirmed against certified copies before use.

Sources: Google Patents — US6666545B2, US6102528, EP0783966A2, US6267470B1; EPO Global Patent Index (EP 0 783 966 priority); EveryPatent (US 6,102,528 claims/description); freepatentsonline (US 7,784,914 reference list, showing US 6,309,053).