US 4923308

I'll run searches for this specific patent number across USPTO and CAFC-related sources.

Let me check for any CAFC or litigation activity involving this patent.

Summary of US Patent No. 4,923,308

Direct answer on the CAFC search: I ran targeted searches for "4923308" / "4,923,308" combined with CAFC, Federal Circuit 2026, PTAB, and reexamination. No hits. I found no authoritative indication of any Federal Circuit 2026 docket, PTAB proceeding, or district-court litigation involving this specific patent. Treat that as "no evidence found," not as a confirmed negative — my search tools do not give me direct docket access to the CAFC or PACER, so I cannot rule out a filing. What I can confirm from the record is that the patent lapsed for non-payment of maintenance fees in 2002, which makes active litigation unlikely.

---

Bibliographic data (from the authoritative patent text)

Abstract (verbatim)

"A process for determining a relative temperature distribution at a surface of a high temperature member to be used in a gas containing oxygen at high temperatures is disclosed. The process comprises the step of determining the relative surface temperature distribution of the member by measuring concentrations of oxygen at the surface of the member, which surface is composed of silicon carbide or silicon nitride. The process may also include the step of quantitatively determining a distribution of temperatures at the surface of the member by measuring concentrations of oxygen at the surface of the member, determining a relationship between temperatures and concentrations of oxygen on a surface of a standard member, and comparing the measured oxygen concentrations at the surface of the member with those of a standard member with reference to this relationship."

Plain-language overview of each independent claim

Claim 1 — relative (qualitative) temperature mapping. A method of finding the relative temperature distribution across the surface of a hot part that operates in an oxygen-containing high-temperature gas. The trick: the part's surface is either made of silicon carbide or coated with silicon nitride or silicon carbide. Because that surface grows a thin oxide film whose thickness tracks how hot it got, you map the surface by measuring how much oxygen is present across it. The result is a contour map showing where it's hotter and cooler — not absolute temperatures.

Claim 2 — quantitative temperature mapping. Same setup (oxygen measurement on a SiC-bodied or SiC/Si₃N₄-coated surface), but adds a calibration step: you separately establish the temperature-vs-oxygen-concentration relationship for a "standard member" (spec: same material, same treatment conditions), then compare the test part's measured oxygen concentrations against that relationship to convert the map into actual temperatures.

Claim 3 (dependent on claim 2): the relationship in claim 2 is determined using an "Arthenius" equation — this is spelled that way in the granted claim, and I interpret the identifier literally rather than correcting it to "Arrhenius."

How the measurement actually works (specification)

• Electron beams are directed at the surface; the intensity of the resulting characteristic X-rays of oxygen is measured (an EPMA — electron probe microanalyzer — is the preferred instrument). Electron penetration depth is ~1 µm, and X-ray intensity is proportional to oxide-film thickness up to 1 µm.
• Oxide film formation rate depends on temperature, oxygen partial pressure, and exposure time. Hold partial pressure and time constant, and film thickness becomes a proxy for temperature.
• Preferred materials: silicon carbide sintered bodies, silicon carbide coatings, silicon nitride coatings. The spec warns that silicon nitride sintered bodies may be unsuitable because their intergranular phase oxidizes fast enough to exceed the 1 µm measurement window.
• Working range claimed as practical: roughly 400–1,800 °C; no pressure limitation. Coatings formed by CVD or PVD.
• Ten examples cover SiC turbine wheels, SiC-coated metallic turbine wheels, Si₃N₄ wheels with SiC or Si₃N₄ coatings, and a SiC gas-turbine stator, with a standard test piece thermocouple-calibrated per the Arrhenius relation (e.g., 1,000 counts/sec ↔ 1,206 K / 933 °C).

Notable prior art and citations on the face of the patent

• Eight cited references (US 3,652,427; US 3,891,512; US 3,935,079; US 4,133,036; US 4,223,549; SU 640198; SU 892284; JP S62-81560) — these are largely molten-metal oxygen sensor art, which is a different field from thermal-history mapping.
• Non-patent citation: Russell, C.K. et al., "Probing For More Than Temperature," Journal of Metals, vol. 23, No. 11, pp. 44–47 (Nov. 1971).
• Later-cited-by art includes US 5,309,874 (Ford), US 6,062,811 (Siemens Westinghouse), US 7,654,734 / US 8,974,180 / US 2006/0256833 / US 2013/0129510 (General Electric), CN 116678514 A, and EP 4628861 A1 (GE Vernova, 2024).

Uncertainties / caveats

• No CAFC or PTAB activity located for this patent. I cannot independently verify absence through a primary docket source with the tools available; my searches returned nothing on point.
• The Google Patents rendering of the description has scrambled Example numbering (e.g., "Example 2" text appearing under Example 4's heading). I relied on the claims and the clean example run in the "Description" section, which is internally consistent.
• There is a minor internal tension worth flagging: the description states silicon nitride sintered bodies are often unsuitable, while claim 1 recites a surface merely "coated with silicon nitride" (a coating, not a sintered body). That is a claim-scope vs. enablement nuance, not a reading error on my part.
• The assignee "Current Assignee" on Google Patents is listed as NGK Insulators Ltd, matching the original assignee; no reassignment of record appears in the legal events beyond the original 1989 assignment.

I'll search for litigation involving US patent 4923308.

Let me search more specific litigation databases.

Search Results: Litigation Involving U.S. Patent No. 4,923,308

Bottom line: I found no known patent litigation involving U.S. Patent No. 4,923,308.

What I searched

I ran multiple targeted queries across open web sources (including Google Patents, CourtListener/ RECAP, Docket Alarm, Unified Patents-related materials, and general litigation coverage), searching for the exact number "4923308" and "4,923,308" in combination with litigation terms. I did not query PACER or the Unified Patents litigation caselist directly (see the limitations note below).

Why the apparent "hits" are not litigation

Every result that contained the string "4923308" or "4,923,308" was an unrelated, non-patent reference to the same digits:

None of these concern US Patent 4,923,308.

Patent status (context)

• Patent: US 4,923,308 A — "Process for determining the distribution of temperatures at surfaces of high temperature members"
• Assignee: NGK Insulators, Ltd. (Nagoya, Japan)
• Inventors: Keiichiro Watanabe, Tadaaki Matsuhisa, Hiroaki Sakai
• Priority date: 1988-04-13 ; Filed: 1989-04-06 ; Granted: 1990-05-08
• Legal status: Expired – Fee Related. Maintenance fees lapsed, and the patent expired for nonpayment of maintenance fees (effective 2002-05-08 per the legal-events record). An expired patent is generally not a vehicle for infringement suits, which is consistent with the absence of litigation.
• Foreign family: JP H0654264B2 / JPH01262429A (Japan) and DE3912044A1 / DE3912044C2 (Germany) — no litigation surfaced for these either.

Related NGK activity (not patent litigation on this patent)

The only litigation connected to the assignee that appeared in searches is the In re Automotive Parts Antitrust Litigation (E.D. Mich., Master Docket No. 2:12-cv-00601 and related, including ceramic-substrate, spark-plug, and oxygen-sensor class actions), in which NGK Insulators, Ltd. and affiliates were defendants and NGK Insulators pleaded guilty to a DOJ price-fixing charge (2015). This is antitrust litigation about ceramic substrates, not patent litigation asserting US 4,923,308.

Limitations / confidence

• I could not directly query PACER or the Unified Patents litigation caselist endpoints in this session, so I cannot rule out a docket I did not reach. However, the complete absence of any web-visible docket, complaint, or reporting tying the number to a case — combined with the patent's 2002 expiry — makes litigation highly unlikely.
• Confidence that no known litigation exists involving this patent: high, but not absolute, given those database-access limits.

Conclusion: Per the sources searched, there is no known litigation — filed, pending, or resolved — involving U.S. Patent No. 4,923,308. If you want a definitive negative, the next step would be a direct PACER docket search and a query of Docket Alarm / Lex Machina / Unified Patents caselist under the assignee "NGK Insulators."

Proceedings overview

Count on file: 0 AIA trial proceedings. The structured "PTAB proceedings on file" block (USPTO Open Data Portal) returns no IPR, no PGR, no CBM, and no derivation proceeding for US 4,923,308 — breakdown by status is therefore 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denied, and my independent web searches (PTAB E2E docket text, ptacts.uspto.gov, CourtListener/RECAP, and general web) surfaced no proceeding I could attribute to this patent. Bottom-line defensive posture: this is not a "hardened" patent and it is not a "canceled-claims" patent either — all three claims are UNTESTED at the Board, and the controlling fact for a defendant is that the patent lapsed for non-payment of maintenance fees effective 2002-05-08, a decade before the AIA trial regime existed (IPR/PGR/CBM petitions became available 2012-09-16). If a demand letter cites US 4,923,308, the answer is an expiration argument, not an IPR.

---

No proceedings to summarize — the per-proceeding sections are empty

There is no IPR20XX-XXXXX, PGR20XX-XXXXX, or CBM20XX-XXXXX number to report. I will not invent one. What I can report is the negative result and its evidentiary basis.

What I searched (all negative)

False positives you should not be misled by

Several search hits use the short form "the '308 patent" but are different patents:

• Gen-Probe v. Becton Dickinson — the "'308 Patent" there is a cap/stopper patent, not the NGK patent; the opinion text is at finnegan.com.
• Omnitracs v. Platform Science — the "'308 patent" there is a telematics communication-terminal patent (CASD 3:17-cv-01864, Doc. 621).
• IPR2020-01083 / IPR2020-01091 — those petitions, and the 2020-12-04 institution denial (Paper 14), concern US 8,273,308, a completely different patent.

None of these involve US 4,923,308 or NGK Insulators.

Why the absence is structural, not accidental

1. IPR (35 U.S.C. §§ 311–319): legally available for any patent, but no petitioner could have reached this one after 2012-09-16 for a practical reason — it had already lapsed in 2002. Additionally, the § 315(b) one-year clock only starts on service of a complaint alleging infringement, and per the litigation section no complaint asserting this patent exists.
2. PGR (35 U.S.C. §§ 321–329): categorically unavailable. PGR applies only to patents with an effective filing date on or after 2013-03-16. This patent's priority is 1988-04-13 and its filing date is 1989-04-06.
3. CBM: substantively unavailable. CBM review was limited to "covered business method" patents (a financial product/service claim plus a "technological invention" carve-out); these claims recite an electron-beam/characteristic-X-ray temperature-measurement process. The CBM window also closed to new petitions on 2020-09-16.
4. Pre-AIA inter partes reexamination: categorically unavailable — 35 U.S.C. § 311(c) (pre-AIA) restricted it to patents issuing from applications filed on or after 1999-11-29. This application was filed 1989-04-06.
5. Ex parte reexamination (35 U.S.C. § 302): theoretically requestable by anyone "at any time during the period of enforceability," and it is not an AIA trial. I found no control number or certificate for this patent, but this is the one channel where my confidence is lower, because pre-AIA reexamination files are less consistently surfaced by open-web search than PTAB files. Treat it as "no evidence found," not a verified negative.
6. Foreign family (JP H0654264B2 / JPH01262429A; DE3912044A1 / DE3912044C2): any JPO invalidation trial, opposition, or German nullity/BPatG proceeding is outside PTAB jurisdiction and outside the scope of this task. I have no data on foreign proceedings and will not speculate.

---

Strategic summary

Claim status of US 4,923,308. Claims 1, 2, and 3 are all UNTESTED. There is no IPR, PGR, or CBM Final Written Decision canceling any claim, and no FWD holding any claim patentable. Claim 1 (independent — relative temperature mapping on a SiC-bodied or SiC/Si₃N₄-coated surface) is alive on paper. Claim 2 (independent — same plus a "standard member" calibration step) is alive on paper. Claim 3 (dependent — relationship determined by an "Arthenius" equation, spelled that way in the granted claim) is alive on paper. The absence of any FWD means there is no claim-level disposition to quote, and I am not going to characterize the Board as having reached any claim it never considered.

Estoppel landscape — effectively empty. Sections 315(e)(2) and 325(e)(2) estoppel only attach to a petitioner that obtained a Final Written Decision in an IPR/PGR. Because nothing was ever instituted, no party is estopped from raising any prior-art ground against these claims. There is likewise no § 315(b) bar in the field, since no complaint was ever served. Conversely, there is no positive estoppel benefit for a defendant — you cannot borrow a PTAB invalidation that does not exist. The prior art available against claims 1–3 is, as far as the PTAB record goes, entirely open: the eight references cited on the face of the patent (US 3,652,427; US 3,891,512; US 3,935,079; US 4,133,036; US 4,223,549; SU 640198; SU 892284; JP S62-81560) are mostly molten-metal oxygen-sensor art, and the single non-patent citation is Russell, C.K. et al., "Probing For More Than Temperature," Journal of Metals, vol. 23, No. 11, pp. 44–47 (Nov. 1971) — none of it was ever tested by an APJ panel.

Pattern signals. There is no petitioner pattern to detect (no petitioner), no PTAB-appeal pattern (the patent owner, NGK Insulators, Ltd., never had a Board decision to appeal), and no defensive aggregator (no Unified Patents, RPX, or similar entity) appears anywhere in the chain. The only litigation touching this assignee that my searches surfaced is the In re Automotive Parts Antitrust Litigation (E.D. Mich., Master Docket 2:12-cv-00601 and related), an antitrust/price-fixing matter about ceramic substrates — not a validity challenge to this patent. The 1993–2025 "cited by" art (US 5,309,874 Ford; US 6,062,811 Siemens Westinghouse; US 7,654,734, US 8,974,180, US 2006/0256833, US 2013/0129510 GE; CN 116678514 A; EP 4628861 A1 GE Vernova) shows the technology remained live long after the patent died, but none of those citations is a validity challenge.

---

Recommended next steps

If you are a defendant or recipient of a demand citing US 4,923,308, do not build the defense around an IPR. The correct first move is a status defense:

1. Lead with expiration. Per the patent's legal-events record, the maintenance fee was not paid, a reminder issued 2001-11-27, lapse for failure to pay was recorded 2002-05-08, and the status was recorded as "PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362" on 2002-06-04. There is no reinstatement event in the record. An expired patent cannot support prospective injunctive relief and, absent narrow intervening-rights carve-outs, cannot ground a going-forward royalty demand. Confirm the current status directly at USPTO Patent Center for application 07/334,133 before you make the argument in writing.
2. Confirm the negative at the primary sources that I could not fully reach: a PTAB E2E / PTAB public information search on application 07/334,133 (not just the patent number), and — if you need to close the pre-AIA reexamination question — the patent's full file wrapper via Patent Center, which will show any reexamination control number (90/xxxxxx, 95/xxxxxx) that ever attached.
3. Do not file an IPR as a defensive gesture. Beyond the futility of challenging an expired patent, an IPR on a patent with no asserted claims buys you nothing: no § 315(e)(2) benefit to bargain for, and no parallel litigation to stay. If you nevertheless need an invalidity position preserved (e.g., for an indemnity dispute over past sales), the § 102/§ 103 art is wide open — the record shows no estoppel and no prior PTAB or court adjudication of claims 1–3 — but that argument belongs in a district-court or licensing posture, not at the Board.
4. If someone represents that a PTAB decision exists on this patent, demand the proceeding number. There is none on file, and any cited number should be checked against PTAB E2E; the likely explanation is confusion with one of the unrelated "'308" patents listed above.

Verification caveat. The existence and content of PTAB proceedings are confirmed through the USPTO ODP block supplied in this analysis. I could not query PACER, Docket Alarm, or Lex Machina directly in this session, and pre-AIA reexamination records are only intermittently web-visible. My confidence that there is no AIA trial proceeding on file for US 4,923,308 is high; my confidence that there has never been any Office validity proceeding of any kind (including ex parte reexamination) is moderate, and a file-wrapper check at Patent Center is the way to convert that to a definitive answer.

I'll verify the assignment record before writing anything.

Ownership-chain reconstruction — US Patent No. 4,923,308

Scope note / sourcing limitation (read first). I reconstructed this chain from the USPTO-derived assignment data reflected in the Google Patents legal-events record for US 4,923,308, plus bibliographic corroboration (Justia, Patent Leaderboard). I attempted live searches against the USPTO Assignment Center (assignmentcenter.uspto.gov) and its indexed mirror, but my search queries returned only unrelated hits (e.g., a Korean trademark number, a Florida EIN, a Hyundai/Kia assignment at reel 032882/0048 — none of which involve this patent). I therefore could not directly view the Assignment Center result page in this session, and the correspondent-of-record field for the single recording below is not exposed in the record I do have. Everything stated as a recorded fact comes from the reel/frame entry in the legal-events block; everything else is labeled as inference.

---

Inventors

Employer determination: All three are listed as conveying parties on the only recorded assignment (reel 005061/0370, "ASSIGNMENT OF ASSIGNORS INTEREST," recorded 1989-04-06), and the receiving party is NGK INSULATORS, LTD., 2-56, Suda-cho, Mizuho-ku, Nagoya City, Aichi Pref., Japan. The JP priority filing JP 63-90915 (1988-04-13) is likewise an NGK-family application. All three residences are in Aichi Prefecture, i.e., the assignee's home prefecture. This is a standard employee-inventor → corporate-assignee chain.

Unusual-pattern check — departures within 12 months of filing: NOT PRESENT. The classic "all inventors left right after filing" fire-sale precursor does not appear here. Tadaaki Matsuhisa alone is credited with 26 US patents assigned to NGK Insulators (Patent Leaderboard), which indicates continued long-tenure service rather than a post-filing exit. I have not verified comparable counts for Watanabe or Sakai — treat those two as unverified, not as a negative finding. (Note also that a change of employment by an inventor would not itself produce a recorded assignment; only a change of ownership would.)

---

Original assignee

NGK Insulators, Ltd. — 2-56, Suda-cho, Mizuho-ku, Nagoya 467-8530, Japan. Founded 1919. Listed on the Tokyo Stock Exchange (TYO: 5333; JP3695200000). Google Patents lists NGK Insulators Ltd as both original and current assignee, and no reassignment of record appears beyond the original 1989 recording.

Primary line of business: advanced ceramics. Environment Business (automotive exhaust ceramic substrates — HONEYCERAM — diesel particulate filters, NOx/oxygen sensors), Digital Society Business (semiconductor-processing-equipment ceramics, ceramic heaters, electrostatic chucks, beryllium copper), and Energy & Industry Business (power insulators, NAS® sodium-sulfur batteries, refractories, ceramic membranes). ~19,931 employees.

Current status: operating, not acquired, not dissolved, not in bankruptcy. FY ended 2025-03-31 net sales ¥619,513 million, operating income ¥81,241 million, net income attributable to owners of the parent ¥54,933 million. Material update to the prior sections: the company announced a name change to "NGK Corporation" effective April 1, 2026 (disclosed in its FY2025 interim business report and reflected in third-party company profiles). This is relevant to an ownership search — see the change-of-name note in the timeline.

Did they ship a product embodying the claims? Nuanced. Claims 1–3 are methods of measurement, not articles of manufacture, so there is no "product embodying the claims" in the classic sense — the method is an internal test/diagnostic technique. The substrate technology the method operates on (silicon carbide sintered bodies; SiC and Si₃N₄ coatings on turbine wheels and stators) is squarely within NGK's ceramic component business, and the specification's examples describe NGK SiC turbocharger turbine wheels and a SiC gas-turbine stator tested in an actual machine. I did not find evidence that this specific measuring process was ever commercialized as a product or licensed — no license, no security agreement, and no assignment is recorded.

---

Assignment timeline

One recorded assignment. No post-issuance transfers, no security interests, no licenses, no merger or change-of-name conveyances.

• 1989-04-01 (executed; effective date as recorded) / recorded 1989-04-06 — Reel 005061 / Frame 0370
  • Conveyance: Assignment — "ASSIGNMENT OF ASSIGNORS INTEREST"
  • Assignor: WATANABE, KEIICHIRO; MATSUHISA, TADAAKI; SAKAI, HIROAKI (joint inventors individually, not as a company)
  • Assignee: NGK INSULATORS, LTD., 2-56, Suda-cho, Mizuho-ku, Nagoya City, Aichi Pref., Japan
  • Correspondent: Not exposed in the available record. I will not guess a name into this field. The closest verifiable datum is Justia's "Law Firm" field for this patent — Arnold, White & Durkee — which is the prosecuting firm of record; the recording correspondent for a 1989 paper filing is frequently the same firm, but that is an inference, not a recorded fact. No recurrence analysis is possible: there is only one link in the chain for a repeat player to recur on.
  • Context: Original filing-date assignment of inventor rights to the corporate employer — the standard pre-issuance, first-link conveyance. Executed five days before the US filing date of 1989-04-06 and recorded the same day the application was filed.

Post-issuance legal events (not assignments; no ownership change): maintenance fees paid at 4 years (1993-10-22) and 8 years (1997-09-24); fee reminder 2001-11-27; lapse for failure to pay maintenance fees 2002-05-08, patent expired due to nonpayment under 37 CFR 1.362; anticipated expiration 2009-04-06. The patent died in NGK's hands. The JP family member JPH0654264B2 is recorded as not_active / expired-lifetime; DE3912044A1/C2 (filed 1989-04-12) shows as granted/active on the family table.

Change-of-name gap worth flagging: Because NGK Insulators, Ltd. → NGK Corporation (effective 2026-04-01) is a name change of the same legal entity and no Change of Name conveyance is recorded against this patent, Assignment Center searches will continue to return "NGK Insulators, Ltd." as assignee of record. Do not read that as a stale or transferred record.

---

Timeline diagram

timeline
    title Ownership of US 4923308
    1988 : JP priority filing JP 63-90915
    1989 : Inventors assign rights to NGK Insulators
         : US application 07 334 133 filed
    1990 : US 4923308 issued to NGK Insulators
    1997 : Maintenance fee paid at 8 years
    2002 : Lapsed for unpaid maintenance fee
    2009 : Anticipated expiration date
    2026 : NGK Insulators renamed NGK Corporation

---

NPE / troll-pattern signals

---

Verdict

Insufficient data — in the sense specified by the rubric: only the original assignment exists, so there is no chain to assess for NPE behavior. The single recorded conveyance is the joint inventors → NGK Insulators, Ltd. assignment, executed 1989-04-01 and recorded 1989-04-06 at reel 005061/frame 0370; from issuance (1990-05-08) the patent never left NGK, and it lapsed for non-payment of maintenance fees effective 2002-05-08.

Read affirmatively, this is a clean single-link chain with no NPE indicia whatsoever — not a "moderate-confidence NPE" and not a "stealth assertion" scenario. The correct operational conclusion for a diligence purpose is: title vested in NGK Insulators, Ltd. (now NGK Corporation, name change effective 2026-04-01, no conveyance recorded) and the patent is dead. One residual caveat: I could not reach the Assignment Center result page directly in this session, so a targeted records pull (see link below) is the appropriate confirmation step before relying on this as a definitive negative.

Verification link: https://assignment.uspto.gov/patent/index.html (mirror of https://assignmentcenter.uspto.gov/) — search by patent number 4923308, or by assignee "NGK Insulators, Ltd." Cross-check: https://patents.google.com/patent/US4923308/en (legal events / reassignment entries).

I'll research each cited reference for US 4,923,308 to build the prior-art analysis.

Let me research the remaining cited references.

Prior-Art Analysis — U.S. Patent No. 4,923,308

0. Scope, method, and confirmation of the subject patent

Target document confirmed. I searched for the exact number 4923308 (no fuzzy/similar-number matching) and located US 4,923,308 A — "Process for determining the distribution of temperatures at surfaces of high temperature members" — NGK Insulators, Ltd.; inventors Keiichiro Watanabe, Tadaaki Matsuhisa, Hiroaki Sakai; App. No. 07/334,133 filed 1989‑04‑06; priority JP 63‑90915 (1988‑04‑13); granted 1990‑05‑08. This matches the bibliographic data in the previously generated summary, so I build directly on that section rather than restating it.

The prior art "cited for 4923308" is the set of references cited on US 4,923,308 (examiner-cited, listed in the "Patent Citations" and "Non-Patent Citations" tables of the record):

Governing law / §102 posture. US 4,923,308 has a 1989 filing date, so pre‑AIA 35 U.S.C. §102/§103 applies. Every reference above published well before the 1988‑04‑13 priority date and more than one year before the 1989‑04‑06 filing date, so all eight patents and both NPL items are fully available as §102(a) and/or §102(b) prior art. The analysis below therefore reduces to the substantive question: does any of them disclose every element of a challenged claim?

---

1. The claims being measured against the art

Restating only the elements needed for the anticipation test (see the prior summary for full claim text):

• Claim 1 — (i) process for determining a relative temperature distribution at a surface; (ii) member used in a gas containing oxygen at high temperatures; (iii) the determination is made by measuring concentrations of oxygen at the surface; (iv) the member is composed of SiC or coated with silicon nitride or silicon carbide at least at a surface portion.
• Claim 2 — Claim 1's measuring step plus comparing measured surface oxygen concentrations to a standard member's temperature‑vs‑oxygen relationship to yield a quantitative temperature distribution.
• Claim 3 — Claim 2 where the relationship is derived using an "Arthenius" equation (spelled that way in the granted claim; I do not correct it).

The single most important observation: all eight cited patents are molten‑metal oxygen-metering / electrochemical‑probe art, not thermal‑history mapping of a solid ceramic surface. They measure dissolved oxygen in a liquid metal bath via an electrochemical (solid‑electrolyte) cell or a getter-wire concentration cell — a completely different physical quantity from the oxide-film thickness on a SiC/Si₃N₄ surface that the '308 patent exploits.

---

2. Reference-by-reference analysis

2.1 US 3,652,427 A — Arthur D. Little (pub. 1972‑03‑28; priority 1969‑12‑22)

• Description: Continuously monitors dissolved oxygen (and, indirectly, carbon) in a BOF molten‑iron bath. A getter-metal wire is fed into the melt to form a concentration cell; the cell EMF is the measure of oxygen content. FIG. 4 plots EMF vs. oxygen content for a series of temperatures; a thermocouple wire optionally gives melt temperature. It is a melt-analysis and process-control method, not a solid-surface temperature map.
• §102 vs. claims 1–3: No anticipation — all claims. It lacks (iii) measuring oxygen concentrations at a surface of a solid high-temperature member as a proxy for that surface's temperature, and lacks (iv) any SiC or Si₃N₄ surface. Its temperature sensing (if used) is a thermocouple, which is exactly the prior approach the '308 patent was designed to replace. The only arguable overlap — a temperature-dependent EMF-vs-oxygen relationship — concerns a liquid electrochemical cell, not oxide-film growth.

2.2 US 3,891,512 A — Ford Motor Co. (pub. 1975‑06‑24; priority 1972‑12‑06; inventor Jackson)

• Description: Rapid determination of dissolved oxygen in molten steel using a solid oxygen-ion electrolyte cell in which the internal oxygen reference material is a carbide (Cr₂C₂ preferred; also silicon carbide, niobium, tantalum, titanium carbides). Readout is a single EMF maximum. Notably, this is the one cited reference whose disclosure actually uses silicon carbide — but as a chemical oxygen reference standard inside an electrochemical cell, never as a structural surface whose oxidation is measured.
• §102 vs. claims 1–3: No anticipation — all claims. Silicon carbide appears only as an internal reference material, which is not the claimed "member composed of silicon carbide … at least at a surface portion." No step determines a surface temperature distribution by measuring surface oxygen concentration, and there is no standard-member comparison step. This is a §102 near-miss on one word ("silicon carbide") but a clear miss on every element of the claims. (This reference is also the source of the Russell et al. NPL citation discussed in §3.)

2.3 US 3,935,079 A — Fitterer Engineering Associates (pub. 1976‑01‑27; priority 1972‑11‑03)

• Description: A portable, self-contained display/readout apparatus for a spent solid-electrolyte oxygen probe used in liquid steel. It shows active-oxygen and sensor-temperature readings, with calibration charts (FIGS. 3–5) converting probe mV to ppm dissolved oxygen as a function of temperature (e.g., the Nernst-type relation E = (RT/4F) ln(P°₂/P*₂)).
• §102 vs. claims 1–3: No anticipation — all claims. This is instrumentation for a liquid-metal electrochemical probe. It does not detect oxide-film thickness on a solid surface, does not map a surface temperature distribution, and involves no SiC/Si₃N₄ structural surface. Its calibration-graph concept is the closest thing in the cited art to the "temperature vs. measured signal" relationship of claim 2, but the underlying measurand and the physical mechanism are different.

2.4 SU 640,198 A1 (pub. 1978‑12‑30; 1974‑05‑15)

• Description (per title/citation record): A sensor for measuring the oxidation degree and temperature of liquid metal — again a molten-metal (immersion) device.
• §102 vs. claims 1–3: No anticipation — all claims. Co-measurement of "oxidation degree and temperature" of a liquid metal is categorically different from measuring the oxidation state of a solid SiC/Si₃N₄ surface to infer that surface's temperature distribution. No disclosure of SiC/Si₃N₄ surfaces or of a standard-member calibration.
• Confidence caveat: I could not retrieve the full text of this Soviet document (SU documents are poorly OCR'd / not fully indexed). The analysis rests on the title and citation context; treat it as likely non-anticipatory with high (not absolute) confidence.

2.5 US 4,133,036 A — Republic Steel Corp. (pub. 1979‑01‑02; priority 1976‑02‑26; inventor Watson)

• Description: A signal-processing method and system that samples an analog sensor output, detects a "condition dwell" (a plateau), and computes an approximate dwell level; applied to molten-steel temperature and dissolved-oxygen measurement (correcting oxygen readings for steel temperature).
• §102 vs. claims 1–3: No anticipation — all claims. The reference is directed to data-processing of an unstable liquid-metal sensor signal. It contains no surface-oxygen measurement, no SiC/Si₃N₄ member, and no temperature-distribution mapping. Relevant only as generic background on temperature/oxygen correlation in a liquid medium.

2.6 US 4,223,549 A — Noranda Mines Ltd. (pub. 1980‑09‑23; priority 1979‑02‑16)

• Description: A monitoring circuit with built-in self-test for an oxygen probe immersed in molten metal: high-impedance follow-and-hold, antilog conversion to linearize the logarithmic probe output, high/low oxygen alarms, and simultaneous thermocouple temperature readout.
• §102 vs. claims 1–3: No anticipation — all claims. Purely an electronic signal-conditioning/self-diagnostic circuit for an electrochemical oxygen probe in liquid metal. Nothing about solid-surface oxide films, SiC/Si₃N₄ members, or surface temperature-distribution determination.

2.7 SU 892,284 A1 (pub. 1981‑12‑23; 1980‑03‑31)

• Description (per title/citation record): A device for measuring the content of components (e.g., carbon) in a liquid melt.
• §102 vs. claims 1–3: No anticipation — all claims. Liquid-melt composition measurement; no surface-temperature mapping, no SiC/Si₃N₄ surface element, no surface-oxygen concentration step.
• Confidence caveat: Same as SU 640,198 — full text not retrieved; analysis based on the citation record.

2.8 JP S62‑81560 A — Tokyo Yogyo Co. (pub. 1987‑04‑15; 1985‑10‑04)

• Description (per title/citation record): A hydrogen sensor for molten metal.
• §102 vs. claims 1‑3: No anticipation — all claims. It is a molten-metal gas sensor (hydrogen), not a surface-temperature-distribution process. Neither the oxygen-surface-measurement step nor the SiC/Si₃N₄ surface element is present.
• Confidence caveat: Full text not retrieved in this session; based on the citation record. Note the assignee, Tokyo Yogyo, is a ceramics company, so I flag this as the one Japanese citation worth a full‑text check if a rigorous §103 study is later needed.

---

3. Non-patent literature

Russell, C.K. et al., "Probing For More Than Temperature," Journal of Metals, vol. 23, no. 11, pp. 44–47 (Nov. 1971).

• Description: A 1971 metallurgical survey of immersion-probe technology for liquid steel, i.e., "probing for more than [just] temperature" — using electrochemical cells/probes to obtain oxygen and related data in molten metal. It is also cited as a reference within US 3,891,512 (see §2.2).
• §102 vs. claims 1–3: No anticipation. A molten-metal probe survey does not disclose determining a surface temperature distribution by measuring surface oxygen concentration on a SiC/Si₃N₄ member, nor the standard-member comparison of claims 2–3. It is background art showing that oxygen measurement at high temperature was a mature field by 1971 — relevant to a §103 obviousness argument, not to §102 anticipation.

---

4. Overall §102 conclusion

None of the eight cited patents, and not the Russell NPL item, anticipates any of claims 1, 2, or 3, whether taken alone or in the combinations in which the examiner cited them.

The reason is structural, not incidental: every cited reference is in the molten-metal electrochemical oxygen-sensor field and measures dissolved oxygen in a liquid bath via a cell EMF or getter-wire concentration cell. Claim 1 of the '308 patent instead claims a thermal-history method: using the oxide film on a solid SiC (or Si₃N₄/SiC-coated) surface as a temperature index and reading it by surface oxygen concentration (specifically characteristic oxygen X-rays via EPMA). No cited reference discloses:

1. measuring oxygen concentration at the surface of a solid member as the temperature indicator (claim 1, element (iii));
2. a member composed of SiC or Si₃N₄/SiC-coated as the temperature-sensing surface (claim 1, element (iv)) — US 3,891,512's "silicon carbide" is an internal electrochemical reference material, not a structural surface; and
3. the standard-member calibration / quantitative conversion steps of claims 2–3 applied to that surface-oxygen measurand.

Where these references actually bite is as §103 background: US 3,652,427, US 3,935,079 and US 4,223,549 establish that (a) high-temperature oxygen sensing was known and (b) temperature-vs-electrochemical-signal relationships (Arrhenius/Nernst form) were routinely plotted — potentially useful against claim 3's "Arthenius" equation as an obvious mathematical expedient, but only if combined with a teaching to use a solid SiC/Si₃N₄ oxide-film surface as the temperature index, which none of the cited art supplies.

---

5. Confidence and caveats (per my operating rules)

• High confidence on the identities, dates, and technical content of US 3,652,427; US 3,891,512; US 3,935,079; US 4,133,036; US 4,223,549 (full-text excerpts retrieved and reviewed).
• Lower confidence on the two Soviet references (SU 640,198; SU 892,284) and JP S62‑81560, whose complete texts I did not retrieve; those descriptions rest on the titles and the citation record on the '308 face. If a definitive negative is required, the next step is a full‑text pull of those three documents.
• I found no anticipation; I am not asserting any reference is invalidating, and I have not fabricated citations or claim mappings. Where a reference merely uses the word "silicon carbide," I have said so explicitly rather than over-reading it.
• This analysis is consistent with the earlier-generated litigation summary (no litigation located on US 4,923,308; patent expired for non‑payment in 2002) — a 2002 lapse is fully consistent with the absence of any later validity challenge built on this prior art.

I'll ground this §103 analysis in the actual prior-art references listed on the patent page. Let me pull what I can confirm about the key cited art.

Let me confirm the substance of the remaining cited oxygen-measurement references so I characterize them accurately.

Obviousness Analysis of U.S. Patent No. 4,923,308 under 35 U.S.C. § 103

Source of record: US4923308A on Google Patents (authoritative text used throughout; claims quoted verbatim).
Analysis date: 2026-04-26

Scope note / limitation up front. I do not have access to the file wrapper (prosecution history) for this patent, so I cannot tell you what the examiner actually rejected or what the applicant actually argued. Everything below is a defensive/hypothetical §103 construction built from (a) the granted claims, (b) the specification, and (c) the references listed in the Prior Art section of the patent page. I also could not retrieve the full text of SU 640198, SU 892284, US 3,652,427, US 3,891,512, or US 4,223,549 beyond the titles/assignees the page supplies; I characterize them at the level the page supports and flag where I'm inferring.

---

1. Legal framework applied

A §103 analysis follows Graham v. John Deere Co., 383 U.S. 1 (1966): (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. Under KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), the claims are obvious if the combination was "predictable" using any of the enumerated rationales (known-technique substitution, known-element substitution, design incentive/market force, "obvious to try"), even absent an explicit teaching to combine. A reference qualifies as analogous if it is from the same field of endeavor or reasonably pertinent to the problem (In re Bigio; In re Kahn).

---

2. Level of ordinary skill in the art (PHOSITA)

A PHOSITA here is a ceramics/materials engineer or an instrumentation engineer with ~3–5 years' experience in high-temperature structural ceramics (SiC/Si₃N₄) for turbomachinery, familiar with:

• passive-oxidation kinetics of SiC/Si₃N₄ and the role of temperature, oxygen partial pressure, and time;
• surface/electron-beam analytical techniques, including EPMA and characteristic X-ray (light-element) analysis; and
• the standard problem statement of the field — how to obtain a surface temperature distribution on a moving, unobservable, high-temperature part, which the patent's own Background admits was previously addressed only by thermocouple arrays and photothermometers.

Note that the "measure the oxygen" and "measure the temperature" skill sets come from different communities (pyrometallurgical process control vs. ceramic materials characterization). That gap is central to the non-obviousness argument in §7 below.

---

3. The claims as construed

Key observation: Claim 1 is very broad. It is a method-of-use/information claim whose real novelty resides almost entirely in element (c) — the choice of a SiC/Si₃N₄ surface as the temperature-integrating medium — plus the insight that the oxygen map is the relative temperature map (element (b)).

---

4. Prior-art inventory (as listed on the patent page)

The page presents prior art in four distinct buckets, and it matters which bucket a reference sits in:

(A) "Cited By / Patent Citations (8)" — the examiner's actual citations:

(B) Non-patent citation (1 substantive):

• Russell, C.K., Fruehan, R.J. & Rittiger, R.S., "Probing For More Than Temperature," Journal of Metals (JOM), vol. 23, no. 11, pp. 44–47 (Nov. 1971), DOI 10.1007/BF03355742. Abstract: "Without adding any delays, steelmakers can now measure oxygen and temperature of the bath with one immersion probe." — i.e., an immersion EMF cell for dissolved oxygen + bath temperature in liquid steel.

(C) "Similar Documents" — including the most dangerous reference on the page, which the examiner did not cite:

• US 3,839,898 — "Determination of thermal profile history of high temperature parts" (United Aircraft Corp., filed 1972-11-22, issued 1974-10-08) — US3839898A. Abstract: "A method for determining the thermal history encountered by gas turbine engine parts exposed to high temperatures. Using the diffusion response of a conventional coating applied to the parts at low temperature, the discernible, irreversible metallurgical changes are correlated to known standards to provide the desired thermal history of the parts."

(D) "Cited By (later art)" — not prior art (post-date the 1988/1989 priority): US 5,309,874 (Ford, 1993); US 6,062,811 (Siemens Westinghouse, 1998); US 7,654,734 / US 2006/0256833 and US 8,974,180 / US 2013/0129510 (GE); CN 116678514 A (2023); EP 4628861 A1 (GE Vernova, 2024). Usable only as evidence of the field's persistence, not as §103 art.

---

5. Element-by-element mapping

5.1 US 3,839,898 (the reference the examiner missed) — maps most of claims 1–2

So US 3,839,898 supplies the entire architecture — an integral, applied-coating, reaction-thickness-as-thermometer, standard-calibrated surface temperature-distribution map for exactly this class of part — and misses only two elements: the identity of the coating material and the oxygen readout.

5.2 The oxygen-measurement cluster (bucket A + Russell)

Every one of these teaches measuring oxygen dissolved in a liquid metal bath via a solid-electrolyte galvanic cell (ZrO₂–CaO, etc.), coupled in several cases with a thermocouple:

• US 3,935,079 explicitly builds conversion graphs (its FIG. 3) converting cell mV → ppm dissolved oxygen as a function of temperature, and its FIG. 4 relates thermocouple mV to sensor temperature. That is a temperature-indexed oxygen calibration curve — structurally the same kind of artifact as claim 2's "relationship," but inverted in use (temperature is the input used to read oxygen).
• US 4,133,036 likewise: "The dissolved oxygen content of molten steel varies according to the steel temperature so that oxygen content readings must be corrected to reflect the steel temperature at which the reading is obtained."
• Russell 1971 teaches simultaneous oxygen and temperature measurement in one high-temperature immersion probe.
• JP S62-81560 A (Tokyo Yogyo, 1987) is a solid-electrolyte ceramic sensor for a dissolved gas in molten metal that uses a thermocouple and applies the Nernst equation to compute content from the EMF — cited by the examiner presumably as routine proof that (i) ceramic electrolyte sensors operate at 400–1100 °C and (ii) expressing sensor-output-vs-temperature relations by an electrochemical (Arrhenius-form) equation is ancient.

What the cluster does not teach: surface oxidation of a structural ceramic, post-exposure spatial mapping, or inferring temperature from oxygen.

---

6. The combinations an Examiner would likely assert

Combination A — Primary: US 3,839,898 + Russell 1971 (or US 4,133,036 / US 3,935,079) + PHOSITA knowledge of SiC/Si₃N₄ passive oxidation + EPMA

Asserted motivation (rationales under KSR):

1. Same field, same problem (KSR rationale (C)/(G)). US 3,839,898 is directed to precisely the preamble of claim 1 — the distribution of temperatures along high-temperature gas-turbine parts — and expressly frames itself against the same inadequate alternatives (thermocouples, krypton tracers, temperature paints) the patent's own Background recites. A PHOSITA seeking a surface-temperature map would start here.
2. Known-technique substitution (KSR rationale (B)). US 3,839,898 already teaches that a thermally activated, time-and-temperature-dependent reaction layer thickness on/at a part surface is a usable thermometer, calibrated against a baseline. Substituting a second known reaction of the same type — the growth of a thin oxide film on SiC — is the substitution of one known indicator for another to obtain a predictable result.
3. Use the part's own material (KSR rationale (F), design incentive). Ceramic (SiC/Si₃N₄) turbine wheels and stators are themselves the parts of interest (the patent's own Examples 1, 5, 7, 9). Eliminating the need for an applied metallic coating by letting the structural ceramic oxidize is a straightforward design simplification.
4. The readout is routine. Characteristic-X-ray/EPMA light-element (oxygen) quantification of a sub-micron oxide film was standard surface analysis; the patent itself calls the instrument "preferably used" without describing its construction, i.e., it treats EPMA as conventional.
5. The calibration step of claim 2 is expressly in the primary reference ("correlated to known standards"), and Arrhenius-form rate relations for oxidation (claim 3) are textbook (KSR: "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").

Reasoning: Under this combination claims 1 and 2 fall out as a predictable substitution of a known indicator medium (oxide film on SiC/Si₃N₄) and a known readout (oxygen by electron-beam X-ray) into a known temperature-mapping architecture US 3,839,898, with the calibration step of claim 2 explicitly taught.

Combination B — US 3,839,898 + the temperature-indicating ceramic film art it cites (Yee, US 3,415,122, referenced inside US 3,839,898) + oxygen measurement

US 3,839,898's own background states: "In the patent to Yee, U.S. Pat. No. 3,415,122 … a temperature indicating ceramic film is used which records a temperature profile by means of comparative crystal growth as a unique function of exposure." Bolting a ceramic film indicator (Yee) onto the turbine-part temperature-mapping method (US 3,839,898) supplies an express motivation to move from a metallic coating to a ceramic film for this exact purpose. If the ceramic is SiC/Si₃N₄ and the readout is oxygen, claims 1–2 are met. This combination is stronger than the examiner's own citations because it stays within the temperature-mapping field.

Combination C — the examiner's actual list, taken alone: weak

Any combination drawn only from bucket (A) + Russell cannot, on its face, reach claim 1, because:

• all of them measure oxygen in the bath medium, not oxygen accumulated in the part surface;
• none teaches or suggests spatial mapping across a solid surface; and
• none involves SiC or Si₃N₄ as a sensing medium.

If the examiner's rejection rested on this cluster, it would be vulnerable to a non-analogous-art and missing-element attack (see §7.1). The real exposure is Combination A/B, which requires a reference the examiner apparently never cited.

---

7. Countervailing (non-obviousness) arguments a patentee would make

7.1 Non-analogous art and a physically different measurement principle

The eight citations are steelmaking process-metallurgy references (bath composition control in BOF/open-hearth/ladle operations). The invention is high-temperature structural-ceramics characterization. More importantly, the principles are categorically different:

• prior art = galvanic-cell EMF → dissolved oxygen activity in a liquid, measured in situ/real-time, with oxygen as the unknown and temperature as the known used to interpret it;
• the invention = post-exposure surface analysis of an oxide film whose thickness is the time-integral of the exposure history, with oxygen as the known-proxy and temperature as the unknown.

That is an inversion of the inference direction plus a change in physical observable (partial-pressure EMF vs. cumulative film thickness). In re Ratti / In re Harza-style arguments — that a combination changing the principle of operation is not "obvious" — carry some weight. This is the patentee's best argument against Combination C; it is much weaker against Combination A, which doesn't need the molten-metal art at all.

7.2 US 3,839,898 arguably teaches away from using oxidation as the indicator

US 3,839,898 warns that the coating must be adapted to a hot, oxidizing atmosphere and that "If the coating is so severely oxidized that it is lost or oxidation is so gross as to interfere with the endurance of the coating or reproducibility of the diffusion mechanism, the essential purpose of the mapping process will be lost." In other words, the closest art treats oxidation as a failure mode to be avoided, not as the sensing mechanism. The invention does the opposite. That is a legitimate teaching-away argument (though "oxidation is undesirable for that coating system" is not necessarily a teaching away from any oxide-thickness thermometry — expect the examiner to rebut that it addresses a different coating).

7.3 Real design work was required (reasonable-expectation-of-success friction)

• The method only works if the oxide film stays in the regime where electron-beam penetration (~1 µm) and X-ray intensity remain proportional to film thickness. The specification flags this as a genuine constraint (working range ~400–1,800 °C; above ~1,900 °C the film exceeds 1 µm and "X-ray measurement is made impossible").
• The inventors found that Si₃N₄ sintered bodies are unsuitable because their intergranular phase oxidizes too fast (>1 µm). Selecting the right material form (SiC sintered body, SiC coating, Si₃N₄ coating) was not automatic.
• Claim 2's "standard member" must be the same material and same treatment conditions — a non-trivial calibration protocol.

These cut against a bare "obvious to try" conclusion but are probably not enough to save the broad claim 1, because each is a routine optimization once the concept is in hand.

7.4 Secondary considerations (§103 Graham factor 4)

I found no evidence to support objective indicia for this patent:

• No commercial-success nexus evidence tying the claims to NGK's SiC turbocharger wheel products.
• No licensing, industry-praise, copying, or long-felt-need evidence located.
• The later art (bucket D) shows the problem persisted for decades — but it also shows later workers (GE, GE Vernova, CN 116678514) took different routes (metal articles, thermal coupons, neutron-irradiation SiC spectral shifts), which the patentee could spin as "others did not find this solution," while an examiner could spin as "the concept of ceramics-as-thermal-history-recorders remained a live field." Net: neutral. I'd not rely on secondary considerations here.

---

8. Flags and inconsistencies to carry forward

1. The most relevant prior art appears to be missing from the examiner's citation list. US 3,839,898 sits on the page under "Similar Documents," not under "Patent Citations." If it was genuinely never applied, that is the single most important vulnerability fact about this patent — the strongest §103 case does not come from the cited molten-metal art at all.
2. The cited art is largely non-analogous. The examiner's eight references are molten-metal oxygen sensors; the Russell NPL is an immersion-probe article. On their own they do not reach claim 1.
3. Claim 3's "Arthenius equation." Construed literally, the claim term is "Arthenius" (a typographical variant); the specification uses "Arrhenius' equation." For §103 purposes this is immaterial — Arrhenius-form expressions for oxidation/electrochemical rate constants (including Nernst-type cell relations, as in JP S62-81560 A) were notorious, so claim 3 adds nothing over claim 2. There is a separate §112 written-description/spelling question, but it does not create a §103 defense and, if anything, a court would read "Arthenius" to mean Arrhenius on the §112 ¶2 "correctable typo" line of authority.
4. Claim-scope vs. enablement tension (already flagged in the earlier section, and it is directly relevant here): claim 1 recites a surface "coated with silicon nitride," while the specification states Si₃N₄ sintered bodies may be unsuitable due to fast intergranular oxidation. The coating embodiment is supported (Example 7, CVD Si₃N₄ coating), so claim 1 is not per se invalid — but note the §103/§112 squeeze: the broader claim 1 is read, the more the prior art covers it and the more it invites §112 attacks; the narrower it is read (limited to passive-oxidation, <1 µm regimes), the harder it is to invalidate but the less it covers.
5. JPS6281560A vs. JPS6281560U. These are different documents sharing the same number: JP S62-81560 A is the Tokyo Yogyo molten-metal hydrogen sensor (the cited reference); JPS6281560U is an unrelated 1985 utility-model filing (a fishing-reel wrapping rod). Interpreted literally, only the "A" document is the prior art here — worth noting so the "U" document is not mistaken for the citation in a later search.
6. Directional-inversion caveat on the oxygen art. US 3,935,079 and US 4,133,036 teach using temperature to interpret oxygen, the reverse of the claimed inference. An examiner relying on them for the calibration step of claim 2 must argue the inverse use was obvious — a fair but contestable point.

---

9. Bottom line (confidence-rated)

Overall: As a validity matter, this patent is in significant jeopardy, but the jeopardy runs through US 3,839,898 — a reference that does not appear in the examiner's citation list on the page — rather than through the molten-metal oxygen sensors the examiner cited. A defense mounted solely against the eight cited references and Russell would likely succeed on non-analogous-art and missing-elements grounds. A defense against US 3,839,898 + oxygen-measurement art + common knowledge of SiC/Si₃N₄ passive oxidation and EPMA would face a real fight, and the strongest surviving arguments would be (i) the teaching-away in US 3,839,898's oxidation warning, and (ii) the material-selection work evidenced by the 400–1,800 °C / <1 µm / Si₃N₄-sintered-body-unsuitable caveats.

What would change this assessment: the actual file wrapper (to see whether US 3,839,898 or the SiC-oxidation literature was ever applied and how the applicant distinguished it), and evidence of secondary considerations (commercial success of NGK's SiC turbocharger wheels with an established nexus, or unexpected results beyond the routine oxidation-kinetics expectation).

---

References relied on

• U.S. Patent No. 4,923,308 — https://patents.google.com/patent/[US4923308](/patent/US4923308)/en
• U.S. Patent No. 3,839,898 — https://patents.google.com/patent/US3839898 (and PDF via patentimages.google)
• U.S. Patent No. 3,935,079 — https://patents.google.com/patent/US3935079
• U.S. Patent No. 4,133,036 — https://patents.google.com/patent/US4133036
• JP S62-81560 A (Tokyo Yogyo) — https://patents.google.com/patent/JPS6281560A/en
• Russell, Fruehan & Rittiger, "Probing For More Than Temperature," JOM 23(11), 44–47 (1971) — https://link.springer.com/article/10.1007/BF03355742
• U.S. Patent No. 3,619,381 (Fitterer) — cited within US 3,935,079 for the solid-electrolyte oxygen-probe background.