US 10254261

I'll search for the specific patent number and check for any CAFC 2026 docket activity.

Let me search more specifically for litigation and PTAB/CAFC activity involving this patent.

One more targeted check for any PTAB/ITC/docket proceedings.

US 10,254,261 — Summary

Note on the searches first (important):

• The patent number as given, 10254261 (i.e., US 10,254,261 B2), resolves cleanly. No similarly-numbered patent was substituted; all results below are for US 10,254,261 B2 / application 15/213,100 / publication US 2018/0017536 A1.
• No CAFC 2026 docket activity, PTAB (IPR/PGR) proceeding, or ITC investigation involving US 10,254,261 was found. I ran repeated targeted searches on the patent number plus "CAFC," "IPR2025/IPR2026," "PTAB," "ITC," "District Court," and "litigation," and nothing tied to this patent surfaced. Caveat: I cannot query PACER/Docket Navigator or the CAFC docket system directly with the tools available, so I frame this as "no results located in public web/search sources," not as a certified absence of litigation. The Google Patents record shows no post-grant challenges; the only legal events are routine assignment and maintenance-fee entries.

Bibliographic Data

Abstract (verbatim)

"A microelectronic device capable of detecting multiple gas constituents in ambient air can be used to monitor air quality. The microelectronic air quality monitor includes a plurality of temperature-sensitive gas sensors tuned to detect different gas species. Each gas sensor is tuned by programming an adjacent heater. An insulating air pocket formed below the sensor helps to maintain the sensor at a desired temperature. A temperature sensor may also be integrated with each gas sensor to provide additional feedback control. The heater, temperature sensor, and gas sensors are in the form of patternable thin films integrated on a single microchip. The device can be incorporated into computer workstations, smart phones, clothing, or other wearable accessories to function as a personal air quality monitor that is smaller, more accurate, and less expensive than existing air quality sensors."

Plain-Language Overview of the Independent Claims

Claim 1 — Device (stacked cavity/heater/thermo-sensor/gas-film unit).
A chip with a substrate carrying several gas-species sensors arranged as elements of one integrated-circuit array. Each sensor element contains four things: (a) a cavity formed in the substrate; (b) a gas-sensitive material that detects a particular gas species; (c) a heating element next to that material; and (d) a temperature-sensing element located between the gas-sensitive material and the heater. Critically, the cavity, gas film, heater, and temperature sensor are aligned with one another — i.e., a vertically stacked, co-located structure rather than laterally separated components.

Claim 8 — Microelectronic air quality monitor (system-level claim).
A semiconductor substrate (claim 9 adds that it may be glass) supporting a micro-sensor array of multiple sensor elements. Each element has its own heater, its own gas micro-sensor configured to detect a particular gas species in ambient air, and its own temperature sensor positioned between the heater and the gas micro-sensor. A microprocessor is communicatively coupled to the array, and an electronic memory coupled to the microprocessor stores instructions for it to execute (claims 10–11 add that the gas micro-sensors are temperature-tunable and that the temperature sensor reports to the microprocessor).

Claim 12 — Method of fabricating a two-sensor (multi-sensor) chip.
Steps: form a first heater in a first region of an IC chip; form a first temperature sensor in that region on top of the first heater; form a first gas-sensitive material in that region on top of the first temperature sensor; form a through hole through the first heater and the first gas-sensitive material; then repeat the layered build in a second region of the same chip — second heater, second temperature sensor on the second heater, second gas-sensitive material on the second temperature sensor. (Claim 13 specifies low-TCR tantalum-aluminum heating elements; claim 14 lists ZnO₂, In₃O₃, SnO₂; claim 15 adds first/second air pockets with the sensitive material spaced from the air pocket by the heater; claim 16 requires the two gas-sensitive materials to be different materials.)

Claim 17 — Device (through-hole variant).
Substantively the same device as claim 1 (substrate + multiple gas-species sensors in a single IC array; each with a substrate cavity, a gas-sensitive material, an adjacent heating element, and a temperature-sensing element adjacent to the gas-sensitive material, all mutually aligned), plus the added requirement that each sensor has a through hole extending through the gas-sensitive material, the heating element, and the temperature-sensing element. Claim 18 narrows this to the temperature sensor being between the film and the heater; claim 19 adds a passivation layer with openings over the gas-sensitive material.

Notable Observations and Points of Uncertainty

1. The "temperature sensor between heater and gas film" is the recurring point of novelty. Three of the four independent claims (1, 8, 17) hinge on that vertical ordering or on a through-hole penetrating the stack. Claim 1 also requires mutual alignment of cavity/film/heater/thermal sensor.
2. Reactive-layer thickness limits appear only in dependent claims: <500 nm (claim 2) and <200 nm (claim 5). The specification's stated VOC film is ~30–100 nm SnO₂.
3. The specification mentions a CO₂ sensor and humidity/temperature sensors, but no independent claim recites CO₂, humidity, or a gas analyzer as such. The claim set is directed to the multi-element MOX sensor architecture. A sibling filing by the same inventors filed the same day (US 15/213,230 → US 10,429,330 B2, "Gas analyzer that detects gases, humidity, and temperature") covers the broader analyzer.
4. Literal identifier inconsistency in the source, preserved here: the description variously writes "indium oxide (In₂O₃)" and "In₃O₃," while claim 4 recites "In₂O₃"; SnO₂ is written elsewhere as "SnO 2." Per instructions I have not auto-corrected these.
5. Assignee chain is unsettled beyond the recorded events. Google Patents shows an original 2016 assignment to STMicroelectronics Pte Ltd and a 2024 assignment to STMicroelectronics International N.V. I did not independently verify the USPTO Assignment database entry (reel/frame 068433/0970 is referenced only for the 2024 transfer; the 2016 entry cites reel/frame 039186/0242).
6. No post-grant or appellate activity located. As stated above, I found no CAFC 2026 docket, IPR, PGR, reissue, reexam, or ITC matter for this patent. Treat this as "nothing found in the sources searched," not as a definitive negative.

Primary sources:

• https://patents.google.com/patent/US10254261/en (full text, claims, cited-by, legal events)
• https://patents.google.com/patent/US20180017536A1/en (pre-grant publication)
• Sibling: https://portal.unifiedpatents.com/patents/patent/US-[10429330](/patent/10429330)-B2 (US 10,429,330 B2, app. 15/213,230, same inventors)

I'll search litigation databases for US Patent 10,254,261.

Initial results are inconclusive. Let me query litigation-specific sources.

Let me try additional litigation-focused searches.

Summary Finding

No litigation involving U.S. Patent No. 10,254,261 ("Integrated air quality sensor that detects multiple gas species") was found. Based on searches of publicly indexed litigation and patent sources, there is no record of this specific patent having been asserted in any U.S. district court, the Court of Appeals for the Federal Circuit, before the PTAB, at the ITC, or in any foreign (e.g., Chinese or European) proceeding.

What I searched and what came back

I ran targeted queries against publicly indexed sources, including Google Patents, USPTO.report, Justia Patents, Docket Alarm/PACER-derived document indexes, CourtListener, and the Unified Patents litigation/PTAB caselist URLs. Results for the exact string "10254261" / "10,254,261" returned:

Per your instruction, I did not treat any of these as the patent in question.

Case-by-case table

Important caveats

1. The patent appears to be unasserted. U.S. 10,254,261 is a 2019 STMicroelectronics patent (application US 15/213,100, filed 2016-07-18, granted 2019-04-09, priority 2016-07-18). Its family includes Chinese counterparts CN107632113A (application 201710288521.5) and CN207300952U (application 201720459726.0U). No assertion activity against those counterparts surfaced either. STMicroelectronics is generally a licensing/product operating company rather than a serial litigant, which is consistent with an unasserted portfolio asset.
2. No PTAB or reexamination activity found. The patent is listed as Active with maintenance fees paid (4th-year fee paid 2022-09-21) and no IPR, PGR, or ex parte reexamination was identified.
3. Search limitations. Unified Patents' portal litigation caselist and PTAB caselist are JavaScript-driven and could not be fully enumerated via indexed search results; PACER and Docket Navigator require authenticated access and are not fully indexed by general web search. My conclusion rests on publicly indexed sources as of the search date.
4. Verification path. To confirm a negative with higher confidence, I would recommend checking directly: (a) Unified Patents Portal → Litigation caselist, filtered by patent number; (b) Docket Navigator or Lex Machina patent-number search for 10,254,261; (c) PACER Case Locator; (d) USPTO Patent Trial and Appeal Board ISI and the USPTO Patent Examination Data/reassignment records; and (e) the USPTO Patent Assignment Search for any security-interest or license records that might hint at enforcement activity.
5. Not a legal opinion. This is a factual survey of publicly available records, not a legal conclusion, and it should not be relied on as a substitute for a formal clearance or freedom-to-operate search.

If you'd like, I can run the same targeted check against the sibling patents in the same family — e.g., US 10,429,330 ("Gas analyzer that detects gases, humidity, and temperature"), US 10,768,133 ("Integrated SMO gas sensor module"), and US 11,009,477 ("Integrated multi-sensor module") — to see whether the assignee has asserted the broader sensor portfolio even if it has not asserted 10,254,261.

I'll verify the canonical "no proceedings" position against live sources before drafting.

Initial results are noise — let me run more targeted queries on the patent owner and any litigation/appeal chains.

One more check for district-court assertion or ex parte reexam activity, which would be the usual precursor to IPRs.

Proceedings overview

There are zero AIA trial proceedings on file against US 10,254,261. The structured "PTAB proceedings on file" block (USPTO Open Data Portal, most recent ingest) reports no IPR, PGR, or CBM — no institution decisions, no Final Written Decisions, no settlements, no adverse judgments. My independent web searches returned no PTAB petition, institution decision, or FWD naming this patent, so the canonical list stands unrebutted. Defensive posture: this patent has never been tested at the PTAB, so there is no invalidity finding to lean on — but equally, there is no § 315(e)(2) estoppel, no adverse claim construction, and no prior petitioner's roadmap. A defendant today starts from a clean slate and can file the first IPR on this patent without inherited baggage. The single most important correction to the framing in the task prompt: this is not a "hardened, survived-two-IPRs" patent, and it is also not a "troll's patent with claims already canceled." It is an untested operating-company patent owned by STMicroelectronics that has simply never attracted a challenge.

---

No proceedings to enumerate

Because the canonical list is empty, there are no proceeding entries to rank by impact. To be explicit about what I checked and did not find, and what I could not verify:

Diligence trap — do not conflate these similarly-numbered patents. Web searches for "the '261 patent" in a PTAB context surface several entirely unrelated patents, and a careless search will produce a false positive:

• US 9,215,261 (Verizon) — IPR petitions challenging claims 1–10 over Whitehead, Cristofalo, and Marcus '904. Different patent, different owner, different field (media programming).
• US 6,442,261 (Voxathon v. Hyundai/Toyota/Subaru/VW) — a telephone-set § 101 case.
• US 7,608,261 (Regeneron) — asserted as an ODP reference in a different patent's ex parte reexam.
• US 10,464,992's ex parte reexam request references a "’261 patent" that is the Regeneron patent, not this one.

None of these is US 10,254,261. I found no proceeding number attributable to this patent and have not invented one.

Also worth flagging for the record: the "Cited By" entries for US 10,254,261 are all Denso International America vehicle-cabin sensor patents (US 11,636,870 and siblings) — those are forward citations, i.e., later patents that cite this one as prior art. They are not PTAB challenges and carry no invalidity effect. Likewise, the "Similar Documents" list (US 10,429,330; US 10,768,133; US 11,009,477; US 11,743,422) is a family/portfolio-neighbor list, not a docket.

Checked and found no indication of: any district-court assertion of US 10,254,261 specifically; any ex parte reexamination; any Federal Circuit appeal. Note that STMicroelectronics is a frequent litigant — it is the petitioner in a long IPR campaign (e.g., IPR2022-00680/00681 v. Ocean Semiconductor, IPR2022-00252/00723/00724 v. Purdue, IPR2021-01230 v. Monterey Research), and it is a defendant in the Purdue and Ocean Semiconductor/Monterey NPE actions, plus a newer 2026 MEMS suit (Pacific Research Group Pte Ltd v. STMicroelectronics NV, 7:26-cv-00134, W.D. Tex., filed 2026-04-07). Those involve other STMicro patents and none of them names US 10,254,261. Confidence on the "no assertion" point is moderate, not absolute — I could not query a complete litigation docket for this patent number, so treat "never asserted" as a working assumption to be confirmed against Docket Navigator or Lex Machina.

---

Strategic summary

Claim status: everything is UNTESTED. All 19 claims stand as issued, never canceled, never confirmed in an AIA trial, never construed by the PTAB. The independent claims in play are claim 1 (device: substrate + array of gas species sensors, each with a cavity, gas sensitive material, heating element, and temperature sensing element between the gas sensitive material and the heating element, all aligned with each other), claim 8 (microelectronic air quality monitor: semiconductor substrate + micro-sensor array with heater, gas micro-sensor, and temperature sensor between them + microprocessor + memory), claim 12 (method of forming first/second heaters, temperature sensors, gas sensitive materials, plus a through hole), and claim 17 (device as in claim 1 plus a through hole extending through the gas sensitive material, heating element, and temperature sensing element). The narrowing pressure points a challenger would aim at are the relative-position requirement ("between," "aligned with each other") and the through-hole limitations of claims 6/17 — those are also, not coincidentally, where an infringement case would be hardest for a patent owner to prove against a competing die layout.

Estoppel landscape — there is none, and that cuts both ways. § 315(e)(2) estoppel attaches only to a petitioner that obtained an FWD. With no petitioner, there is no estoppel binding anyone, no § 315(b) one-year bar running against any party (no complaint on this patent has been served on anyone, per the sources I could check), and no IPR ground a defendant is foreclosed from raising. Practically, the entire prior-art space is open: the § 102/§ 103 art cited during prosecution, the SMO sensor art STMicro itself incorporated by reference (Shankar, U.S. Pub. 2016/0018356, from application Ser. No. 14/334,572), and the third-party metal-oxide gas sensor art (e.g., U.S. Pub. 2015/0323510 to Active-Semi; U.S. Pub. 2017/0168066 to IBM; U.S. Pub. 2005/0109081 to Zribi) are all live IPR candidates. The affirmative corollary: if you are the first to file, you take on § 315(e)(2) estoppel risk and, if you settle or lose, you may hand later defendants the same estoppel-free position you enjoy now.

Pattern signals. No repeat petitioner, because there is no petitioner at all. No PTAB appeal history, because there is no FWD to appeal. No defensive aggregator (Unified Patents, RPX, or similar) appears anywhere in the chain — the patent sits with STMicroelectronics Pte Ltd, now assigned to STMicroelectronics International N.V. (assignment recorded 2024-07-18, effective 2024-06-28), and is Active, with an adjusted expiration of 2036-12-18, maintained through at least the 4th-year fee paid 2022-09-21. That combination — a large operating company, a long remaining term, an unasserted-but-maintained portfolio asset — is the classic "sleeping patent" profile. It is not being monetized by an NPE today, which is precisely why it has never drawn an IPR.

Procedural windows that have already closed. The patent issued 2019-04-09, so the nine-month PGR window closed on approximately 2020-01-09 — PGR is no longer available. CBM was never realistically available (this is a technological invention in a technical field, not a financial-services business method) and the transitional CBM program has sunset for new petitions. IPR under § 311–§ 319 remains available for the life of the patent, and there is no § 315(b) bar against any party because no infringement complaint on this patent has been served (to the extent of my search).

---

Recommended next steps

1. If you are a defendant facing a demand or complaint citing US 10,254,261: there is no FWD to quote, no canceled claim to cite, and no PTAB record to hide behind. Your invalidity positions must be built from scratch. Start with the prosecution file history and with Shankar (U.S. Pub. 2016/0018356, application Ser. No. 14/334,572), which the specification expressly incorporates by reference and which describes overlapping integrated SMO gas sensor structure — that is the highest-value starting art for a § 103 petition against claims 1, 8, 12, and 17.

2. If you are considering filing an IPR: there is no § 315(b) clock, so you are not forced to file early — but note that filing is the trigger for your own § 315(e)(2) estoppel and that PTAB trial practice requires an FWD within 12 months of institution for most cases (35 U.S.C. § 316(a)(11)), with institution decided within 6 months of the petition's filing (§ 314(b)). Map the two independent device claims and the method claim separately; a petition that wins on claim 1 can still lose on claim 12.

3. Verify before relying on the absence. Because I could not query the complete PTAB E2E docket or a full litigation database directly for this patent number, confirm the null result yourself at the USPTO PTAB E2E portal (https://ptacts.uspto.gov) and the PTAB Decisions page (https://www.uspto.gov/patents/ptab/decisions), and check the patent's litigation history on CourtListener (https://www.courtlistener.com) or Docket Navigator. The authoritative source document for this report is the patent record at https://patents.google.com/patent/US10254261/en.

4. Watch for a first-mover. Well-maintained, long-lived, unasserted operating-company patents in a hot technology space (indoor air quality / MEMS gas sensors) are exactly the profile that attracts a first IPR once a large enough defendant is sued. If you are that defendant, you may be the first mover — and the first mover sets the estoppel trap for everyone behind it.

I'll verify the assignment records and cross-reference before drafting.

Assignment & Ownership Analysis — US 10,254,261 B2

Patent: "Integrated air quality sensor that detects multiple gas species"
App. No.: 15/213,100 · Filed: 2016-07-18 · Issued: 2019-04-09 · Adjusted expiration: 2036-12-18
Verification sources: Google Patents legal events · USPTO Assignment Center · Patent PDF

Scope caveat up front: only two post-filing ownership records exist for this patent, and both are visible in the indexed legal-events feed (reel 039186/0242 and reel 068433/0970). I could not retrieve the Assignment Center record images themselves through this session, so the correspondent-of-record field is not available for either entry — that is a real gap, not an omission, and I flag it rather than guess. Prosecution counsel of record on the sibling application (15/213,230) was Seed IP Law Group LLP / STMicroelectronics, which is the same outside-counsel program that would plausibly have handled the 2016 recording, but I have not confirmed that as the recorded correspondent.

---

Inventors

Notes and patterns:

• All four inventors are Singapore-resident and assigned to the same operating subsidiary. The 2016-07-19 cover-sheet entry names "LE NEEL, OLIVIER; LOH, TIEN CHOY; KAM, SHIAN YEU; AND OTHERS" with signing dates spread 2016-07-13 to 2016-07-17 — i.e., executed 1–5 days before the 2016-07-18 filing date. That is the signature of a routine, pre-filing employee-invention assignment package (ST's standard practice), not a departure or fire-sale precursor.
• No departure signal. Nothing in the record shows any inventor leaving ST within 12 months of filing; the same four inventors are co-named on the sibling patent US 10,429,330 ("Gas analyzer that detects gases, humidity, and temperature"), also filed 2016-07-18 and also ST-assigned, which indicates a continuing, funded internal sensor program rather than a dispersed inventor group.
• Small data-quality caution: the granted patent's OCR text renders inventor 3 as "Shian You Kam" while the assignment cover sheet and Google Patents use "Shian Yeu Kam." Same person; I am not auto-correcting the record, just noting the discrepancy.

---

Original assignee

STMicroelectronics Pte Ltd (Singapore — Ang Mo Kio). Listed as both applicant (71) and assignee (73) on the face of the patent.

• Primary line of business: operating semiconductor manufacturer — ST's Singapore entity is the group's Asian manufacturing, back-end/test and design hub, and ST is one of the world's largest IDMs (NYSE: STM). This is unambiguously an operating company, not a holding vehicle.
• Did they ship a product embodying the claims? Not established from the record. The specification describes an embodiment (conformal MOX thin film <0.2 µm, TaAl low-TCR heater, air-pocket thermal isolation) but I found no commercial ST part number and no ST datasheet identifying a shipping MOX multi-gas sensor built to these claims. Notably, ST's own air-quality monitoring application material points customers to partner (Sensirion) VOC/CO₂ sensors (SGP40/SGP41, SCD40/SCD41) alongside ST MCUs — which is at least consistent with ST not having driven this particular MOX architecture to a flagship commercial sensor. Treat this as unclear, not as a negative finding: absence of a datasheet in a web index is not proof of non-practice.
• Current status: operating. STMicroelectronics Pte Ltd remains an active, 100%-owned subsidiary listed in ST's group-entity disclosures. No bankruptcy, no dissolution, no assignment for the benefit of creditors.

---

Assignment timeline

Two recorded assignments. Both are intra-group; neither involves a third party.

• 2016-07-13 → 2016-07-17 (executed) / recorded 2016-07-19 — Reel 039186/0242

  • Conveyance: Assignment (Assignment of Assignors' Interest)
  • Assignor: Olivier Le Neel; Tien Choy Loh; Shian Yeu Kam; and others (Ravi Shankar) — individual inventors
  • Assignee: STMicroelectronics Pte Ltd (Singapore)
  • Correspondent: Not disclosed in the indexed record. The Assignment Center image would carry this field; I could not retrieve it. If it matches the sibling application's prosecution agent, it would read Seed IP Law Group LLP / STMicroelectronics — unconfirmed, and I am not asserting it as a finding.
  • Context: Routine pre-filing employee-invention assignment; original vesting of title in the operating employer. Filed the day after the last inventor signature and one day after the application was filed.

• 2024-06-28 (executed) / recorded 2024-07-18 — Reel 068433/0970

  • Conveyance: Assignment
  • Assignor: STMicroelectronics Pte Ltd (Singapore)
  • Assignee: STMicroelectronics International N.V. (Switzerland; legal seat Amsterdam, Netherlands per ST's group-entity table)
  • Correspondent: Not disclosed in the indexed record (would require the record image).
  • Context: Internal corporate reorganization — intra-group transfer of title from the Singapore operating subsidiary to ST's group IP-holding entity. This is the same intra-group pattern ST applied across its portfolio; it is a change of holder within one corporate family, not a sale.

Ancillary events (not assignments, but relevant to chain status):

• 2022-09-21 — maintenance fee, 4th year, large entity (fee paid; patent remains in force).
• 2017-04-27 — priority claims recorded to CN 201710288521.5 (CN 107632113 A) and CN 201720459726.0 (CN 207300952 U, a Chinese utility model). Family members only; the CN filings are downstream of the same ST parent, not ownership transfers.
• No litigation. I found no district-court or ITC infringement action naming US 10,254,261. The only "cited by" activity is third-party prosecution citation by Denso International America in its olfaction-sensor family (US 11,636,870 etc.), which is citation, not assertion.

---

Timeline diagram

timeline
    title Ownership of US 10254261
    2016 : Filed by STMicroelectronics Pte Ltd
         : Employee assignment reel 039186 frame 0242
    2019 : Patent issued
    2022 : 4th year maintenance fee paid
    2024 : Intra group transfer to ST International NV
         : Reel 068433 frame 0970

---

NPE / troll-pattern signals

---

Verdict

Insufficient data — in the specific sense that the record contains only the original employee assignment plus a single intra-group transfer, which is the functional equivalent of "original assignment only" for NPE purposes.

Justification: The entire chain is two entries — inventors → STMicroelectronics Pte Ltd (reel 039186/0242, executed 2016-07-13/17, recorded 2016-07-19) and STMicroelectronics Pte Ltd → STMicroelectronics International N.V. (reel 068433/0970, executed 2024-06-28, recorded 2024-07-18). Both assignees are inside one corporate family, the patent is unasserted, and no NPE marker (shell transferee, rostered asserter, cascading LLCs, pre-suit transfer, bankruptcy sale, privateering) is present. I am deliberately not selecting "Operating-company assertion," because that category requires the current assignee to be suing actual competitors — there is no suit naming this patent, and I could not confirm a commercial product embodying the claims. Nor is "Defensive / non-asserting" apt, since the chain does not terminate at a defensive aggregator. The honest answer is: clean intra-group ownership by a large operating semiconductor company, with a recorded-correspondent gap that prevents full clearance of signal #3.

Verify at: USPTO Assignment Center — search patent number 10254261. Pull the record images for reels 039186/0242 and 068433/0970 to capture the correspondent-of-record field, which is the one material datum missing from this analysis.

I'll verify the key cited references against live sources before building the § 102 analysis.

Verified Shankar (US 2016/0018356 A1 → US 9,810,653 B2) and Zribi (US 2005/0109081 A1 → US 7,104,113 B2). Now checking the other high-relevance cited references.

Prior-Art Analysis — US 10,254,261 B2

Patent analyzed: US 10,254,261 B2, "Integrated air quality sensor that detects multiple gas species"
App. No.: 15/213,100 · Filed: 2016-07-18 · Priority: 2016-07-18 · Granted: 2019-04-09
Assignee: STMicroelectronics Pte Ltd → STMicroelectronics International N.V.
Source of citation list: the "Patent Citations (89)" / "Cited By"/"Non-Patent Citations" tables at https://patents.google.com/patent/US10254261/en

Scope note (read first): the patent number as given resolves cleanly to US 10,254,261 B2. Every reference below is drawn from the citation table printed on the face of that patent or from the examiner-cited reference lists — nothing has been substituted from a similarly-numbered patent. I have independently live-verified only a subset (flagged "✔ verified live"); the remainder are reported as the record states them, which per your instruction is authoritative for identifying what was cited.

---

1. Claim-element map used for the § 102 screen

The single hardest claim element to anticipate is the vertical ordering "temperature sensing element between the gas sensitive material and the heating element" (claims 1, 8, 12, 18) and the through hole through all three layers (claims 6, 12, 17). No reference I reviewed discloses both in one document, and most do not disclose even the ordering.

---

2. § 102 framework and the effective filing date

• Filing/priority: 2016-07-18 → AIA § 102 governs (post-2013-03-16).
• § 102(a)(1) art: anything patented / described in a printed publication / in public use / on sale before 2016-07-18.
• § 102(a)(2) art: US patents and US patent-application publications effectively filed before 2016-07-18 (including § 102(d) foreign-priority dates).
• § 102(b)(1)(A) grace-period exception: a disclosure made by the inventor or a joint inventor, or by another who obtained the subject matter from them, ≤1 year before filing.
• § 102(b)(2)(C): a § 102(a)(2) disclosure commonly owned at the time of filing is excepted.
• Anticipation requires every element of the claim in a single reference, arranged as claimed. Where a reference discloses all but one element, the correct characterization is § 103, not § 102 — I flag that distinction explicitly below.

---

3. Tier 1 — References with realistic § 102 anticipation risk

These are the only cited references that come close to reading on an independent claim. Ranked by risk.

3.1 US 2016/0018356 A1 — "Integrated SMO Gas Sensor Module" ("Shankar") ⚠️ HIGHEST RISK BUT LIKELY EXCEPTED

3.2 US 2005/0109081 A1 — Zribi et al., "Miniaturized multi-gas and vapor sensor devices and associated methods of fabrication" ✔ verified live

3.3 US 2015/0323510 A1 — Huynh et al., "Olfactory Application Controller Integrated Circuit" ✔ verified live

3.4 US 2013/0010826 A1 — Le Neel et al., "Microsensor with integrated temperature control" ✔ verified live

3.5 US 9,105,479 B2 / EP 2,645,091 B1 — ams AG, "Integrated circuit including an environmental sensor"

---

4. Tier 2 — Strong § 103 references; conditional § 102 on dependent claims

---

5. Tier 3 — Complete enumeration of remaining citations (no realistic § 102 against the independent claims)

These are the examiner's remaining cited references. Their common thread: they were cited for micro-machining, cavity/suspended-membrane, housing/packaging, flow-sensing or electrode concepts — not for multi-species gas detection with an interposed temperature sensor. I list them so the record is complete and to flag which are formally unrelated art.

3A. Flow / metering references (field-mismatched — cited for cavity, channel and suspended-structure teachings only):

3B. MEMS/CMOS process, cavity, housing, packing and bonding references (structural background; § 103 fodder only):

3C. Sensing-layer, electrode and calibration references:

3D. Non-patent literature:

---

6. Cross-cutting findings

6.1 Nobody anticipates the independent claims. Across all 89 cited references plus 4 NPL items, I found no single reference that discloses, in one document:

• (a) a plurality of gas-species sensors arranged as elements of a single IC array;
• (b) a cavity in the substrate per element;
• (c) a gas-sensitive material configured for a particular gas species;
• (d) a heating element adjacent to it; and
• (e) a temperature-sensing element positioned between the gas-sensitive material and the heating element, with all four aligned.

The references split cleanly into two camps: multi-species MOX sensor arrays with heaters and thermal tuning (Shankar, Zribi, Active-Semi, BASF, ChipSensors) and stacked heater/temperature-sensor micro-hotplates with an interposed thermal sensor (Le Neel '826, ams '479, Stanford calorimeter). No cited reference bridges both camps. That bridge is the actual inventive contribution, and it is the reason the § 102 case is weak.

6.2 The two strongest § 102 references are the ones most likely disqualified. Shankar (US 2016/0018356 A1) is the best structural match on claim 1 and claim 8 — and it fails as § 102 art because of § 102(b)(1)(A) (same four joint inventors, published 2016-01-21, ~6 months before filing), § 102(b)(2)(C) (common ownership by STMicroelectronics Pte Ltd), and the fact that the '261 specification incorporates Shankar by reference so its content is part of the '261's own disclosure. Any invalidity theory must either use Shankar for § 103 purposes only, or build on Zribi/Le Neel '826/Active-Semi instead. Flagging this as the single most important practical point in the analysis.

6.3 The same-day siblings are not art. US 2018/0017513 A1 ("Miniature gas analyzer," filed 2016-07-18, STMicroelectronics Pte Ltd) and US 10,429,330 B2 (app. 15/213,230, "Gas analyzer that detects gases, humidity, and temperature," same four inventors, same filing date) are co-pending/co-owned same-day filings. They are not § 102 art (no earlier effective filing date; and § 102(b)(2)(C) would except them in any event).

6.4 The through-hole limitations (claims 6, 12, 17) are essentially unanticipated. Claim 12's "through hole through the first heater and the first gas-sensitive material" and claim 17's through hole "through the gas sensitive material, the heating element, and the temperature sensing element" have no counterpart in any cited reference. The '261 disclosure ties these openings to the polyimide-well release step (openings 256 leading to air pockets 260). This is a genuine point of novelty and the strongest non-infringement/validity differentiator after the "between" ordering.

6.5 Claim 9 (glass substrate) is the thinnest dependent claim. US 2013/0106813 A1 (Apple, chip-on-glass) plus the STMicroelectronics module references collectively disclose sensor-on-glass integration; a § 103 attack on claim 9 is more plausible than on any other dependent claim.

6.6 Chemistry and thickness claims (2–5, 14) are the softest. SnO₂/ZnO/In₂O₃ MOX films, sub-500-nm and sub-200-nm reactive layers, and ceramic (sintered, dense) films are extensively disclosed — Shankar, Zribi, Honk Kong PolyU '489, O'Brien '865, BASF '823, Hitachi '232. Claims 3, 4, 5 and 14 are vulnerable to § 103 combinations, and claims 3–4 arguably to § 102 on the Honk Kong PolyU reference alone.

6.7 Field-mismatch art dilutes the citation list. Roughly 25 of the 89 citations (flowmeters, flow metering, venturi nozzles, encoders, microphones, relays, CMUTs, photomasks, adhesives, packaging) have no gas-sensing disclosure at all. They were cited for isolated micro-machining/cavity/packaging teachings and have no § 102 significance. Any invalidity contention that leans on them, rather than on the six-to-eight genuinely on-point references, will look padded.

---

7. Verification status and caveats

Additional caveats:

1. "Potentially anticipates" is a screening judgment, not a legal conclusion. Anticipation requires every element in one reference, arranged as claimed. Where I flag a reference as "§ 103," it discloses all but one or two elements.
2. The Bing/Google citation table lists both a "Citations (85)" set and a "Patent Citations (89)" set — the 89-set adds WO 2005/087471 A1, US 2012/0168882 A1, US 2012/0171713 A1 and US 2012/0171774 A1 (the Cherian/Le Neel ST chemical-sensor trio). I have reported both.
3. The Google record gives an earliest-priority column alongside the publication date (e.g., US 8,487,387 B2 shown as 2010-08-23 / 2013-07-16, whereas the granted patent's own filing is 13/526,279 filed 2012-06-18). Where the two differ, I have used the record's figures as printed and noted the discrepancy rather than silently correcting it.
4. One identifier irregularity preserved as-is per instructions: the '261 specification writes "indium oxide (In₂O₃)" in the description of FIG. 3B but "In₃O₃" in the fabrication description, while claims 4 and 14 recite "In₂O₃" and "In₃O₃" respectively. This affects which dependent claims the O'Brien and Hong Kong PolyU references can reach, so it matters for the § 102 mapping — it is noted, not corrected.
5. Primary sources: https://patents.google.com/patent/US10254261/en (full text, claims, citation tables) · https://patentimages.storage.googleapis.com/2c/a8/60/d9e767332574b2/US10254261.pdf · https://patents.google.com/patent/US20160018356A1/en · https://patentimages.storage.googleapis.com/22/a0/73/7310c315cbf396/US9810653.pdf · https://patents.google.com/patent/US20050109081A1/en · https://patents.google.com/patent/US20150323510 · https://patentimages.storage.googleapis.com/b6/10/36/b8574cb362e5f2/US20130010826A1.pdf · https://patentimages.storage.googleapis.com/84/db/d7/e73569691add67/[US8387464](/patent/US8387464).pdf

---

Bottom line: for US 10,254,261, the most relevant cited prior art is Shankar (US 2016/0018356 A1 → US 9,810,653 B2), Zribi (US 2005/0109081 A1 → US 7,104,113 B2), Active-Semi (US 2015/0323510 A1 → US 9,664,661 B2), and Le Neel (US 2013/0010826 A1 → US 9,448,198 B2) — in that order. But the most usable § 102 art is Zribi and Le Neel '826, because Shankar and the same-day ST siblings are almost certainly excepted under § 102(b)(1)(A) / § 102(b)(2)(C) and, in Shankar's case, subsumed into the '261's own specification by express incorporation by reference. No cited reference anticipates claims 1, 8, 12 or 17 outright; the closest any comes is § 103 territory, and the temperature-sensor-between-heater-and-film ordering plus the through-hole limitations remain undisclosed across the entire cited corpus.

I'll verify a few of the highest-value references before building the obviousness combinations.

That first search was highly probative. Let me verify the Shankar reference the patent itself incorporates, and the through-hole/cavity art.

Two more targeted verifications — the Silicon Labs "integral heater" sibling, and through-hole/opening art for the claim 6/17 limitation.

Obviousness Analysis — U.S. Patent No. 10,254,261 B2

35 U.S.C. § 103

Patent: "Integrated air quality sensor that detects multiple gas species" · App. 15/213,100 · Filed/priority 2016‑07‑18 · Granted 2019‑04‑09
Claim set analyzed: all 19 claims; independent claims 1, 8, 12, 17
Primary source: https://patents.google.com/patent/[US10254261](/patent/US10254261)/en

Date discrepancy flagged (per operating rules): the task header states "Current Date: April 26, 2026," while the session date stamp reads 2026‑09‑29. Nothing in this § 103 analysis turns on which is correct — the critical date (2016‑07‑18) and the availability analysis are fixed — so I proceed without resolving it.

---

0. Frame, and the single fact that governs the whole analysis

Governing law. The application was filed 2016‑07‑18, after the AIA's 2013‑03‑16 cutover, and claims no pre‑AIA priority (the CN 2017 filings are downstream). AIA § 102/§ 103 therefore apply, with an effective filing date of 2016‑07‑18. Pre‑AIA case law remains persuasive but the § 102 structure is the AIA's.

The governing fact. The most structurally on‑point reference in the entire field — Shankar, US 2016/0018356 A1 / US 9,810,653 B2 (app. 14/334,572, filed 2014‑07‑17, published 2016‑01‑21), which the '261 specification expressly incorporates by reference — describes a stack of gas sensor + resistive heater + resistive temperature sensor over an insulating recess, on a common substrate with an ASIC, i.e., very nearly claim 1. But Shankar's named inventive entity is Ravi Shankar, Olivier Le Neel, Tien‑Choy Loh, Shian‑Yeu Kam — the identical four inventors as the '261 — and it published within one year of the critical date and is commonly owned. It is therefore, on its face, excepted from prior art by AIA § 102(b)(1)(A) (inventors' own disclosure within the grace period) and, as to its US patent document character, by § 102(b)(2)(C) (common ownership). The examiner cited it; it is in the file; but it may not be available as § 102 art at all.

The corollary that shapes every ground below. The § 102(b) exceptions are time‑limited. The same cannot be said for STMicroelectronics' older self‑published family — notably US 2013/0010826 A1 (filed 2011‑07‑05; published 2013‑01‑10; granted as US 9,448,198 B2; inventor Olivier Le Neel, assignee STMicroelectronics Pte Ltd). Because it published more than one year before 2016‑07‑18, the § 102(b)(1) grace period cannot reach it, and § 102(b)(2)(C) common ownership reaches only § 102(a)(2) art — not § 102(a)(1). The '826 is therefore unambiguously § 102(a)(1) prior art even though it shares an inventor and an assignee with the '261. That asymmetry — ST's pre‑2015 publications are fair game; ST's January‑2016 publication is not — is the hinge on which this challenge turns.

§ 102 status of the key references

Level of ordinary skill (POSA)

A POSA here is a person with a master's degree (or equivalent industrial experience) in electrical engineering, applied physics, or materials science, with 2–3 years of hands‑on experience in CMOS‑compatible MEMS micro‑hotplate/metal‑oxide gas sensors — including thin‑film deposition and patterning at ≤400 °C back‑end temperatures, micro‑heater design (TCR, serpentine geometry), and thermal‑isolation structures. This is corroborated by the '261 specification itself, which assumes familiarity with CVD/PVD/ALD, RIE, spin‑expose‑develop, and CMP as conventional (see Detailed Description; and the earlier-generated Prior Art section, which catalogues this skill level across the cited art).

Claim construction of the three dispositive limitations

Note also that claim 12's through hole runs through "the first heater and the first gas sensitive material" only — it does not require penetration of the temperature sensor. Claim 12 is therefore broader than claim 17 and is the softest independent claim in the set.

---

1. Which limitation actually carries the case

Everything else — materials (SnO₂/ZnO₂/In₂O₃), thicknesses (<500 nm, <200 nm), TaAl low‑TCR heaters, glass substrates, passivation windows, thermal tuning, air pockets, differing materials per element — is either expressly disclosed in the art or a predictable design choice. The patent's validity therefore rests almost entirely on the words "between" and "aligned with each other."

---

2. Ground 1 — Claim 1 obvious over US 2013/0010826 A1 (Le Neel/STMicroelectronics) in view of US 2005/0109081 A1 / US 7,104,113 B2 (Zribi et al./GE)

This is the strongest single ground, because every element is disclosed in one reference and the ordering limitation is taught explicitly, in the alternative, by the other.

The '826 reference — what it actually says

Verbatim from US 2013/0010826 A1 (verified against the published text and the granted US 9,448,198 B2):

"…the microsensors described by the present disclosure can be formed as a semiconductor device through a sequence of growing, deposition, patterning, and etching steps. In one embodiment, such a method involves providing a silicon substrate that includes a thermal insulation layer. A thermal energy source layer of a refractory material is formed over the thermal insulation layer. The resulting thermal energy source layer is patterned to form a thermal energy source. A thermal conducting layer is formed over the thermal energy source. After the thermal conducting layer is formed, a working electrode, a counter electrode, and a temperature sensor are formed over the thermal conducting layer. A passivation layer is then formed over the temperature sensor. The passivation layer is patterned to define a sensing region and expose at least a portion of the working electrode and counter electrode."

and, decisively:

"In some embodiments, the thermal energy source is provided below the transducer. In some embodiments, the thermal energy source is also provided below the temperature sensor. In yet other embodiments, both the thermal energy source and the temperature sensor are provided below the transducer."

with the stated purpose:

"The integrated temperature sensor detects the temperature of the sensing region and produces a signal indicative of that temperature. That signal can be used by a control unit to control the thermal energy produced by the thermal energy source. Through this feedback loop, localized heating of the sensing region can be adjusted and controlled… Localized temperature control allows analyte detection to be carried out at the same temperatures or substantially the same temperatures at which the sensor is calibrated."

The '826 is also assignee‑ and inventor‑common with the '261 (STMicroelectronics Pte Ltd; Olivier Le Neel).

Claim chart — claim 1

Result: Zribi supplies everything the '826 lacks (the substrate cavity + criterion of alignment + an actual gas/vapor sensing layer), and the '826 supplies the one thing Zribi lacks (a discrete temperature sensor stacked between the sensing layer and the heater). Neither requires physical modification of the other — the combination is a substitution of one known sensor architecture for another in the same stack, using only conventional BEOL/MEMS steps the '261 itself concedes are routine.

Motivation to combine (Ground 1)

1. Explicit teaching in the primary reference. The '826 does not merely permit the claimed ordering — it lists it as one of a small set of alternatives ("In yet other embodiments, both the thermal energy source and the temperature sensor are provided below the transducer"). Under KSR, a reference that "discloses a finite number of identified, predictable solutions" gives the POSA good reason to pursue that which is claimed. There is no hindsight reconstruction required; the reference itself articulates the claimed geometry.
2. Articulated problem, articulated solution. The '826 states the very problem the '261 claims to solve: sensing accuracy. Its stated mechanism — "carrying out the sensing near the calibration temperature [and] near their peak reaction rate" — is the identical rationale the '261 gives for integrating a temperature sensor with each gas sensor ("provides a feedback signal to ensure that the SnO₂ material is at the desired temperature"). The '261's own FIG. 5 table (100 °C → H₂; 200 °C → butane/propane; 300 °C → CO; 400 °C → methane) makes temperature fidelity the performance variable. A POSA seeking species selectivity therefore has a strong, stated motivation to put an accurate thermometer exactly where the reaction occurs — between heater and film.
3. The references are physically combinable and the result predictable. Both are CMOS/BEOL thin‑film, wafer‑scale, sequentially deposited and patterned devices. Substituting an MOX/vapor sensing layer for the '826's transducer is a materials swap within a known process window; moving a resistive thermometer from the Zribi "combined heater/thermometer" to a discrete interposed element is a de minimis layout change with no change in principle of operation. Predictable result: better temperature control → better selectivity and accuracy. KSR: "the combination of familiar elements according to known methods is likely to be obvious when it does no more than yield predictable results."
4. Same field, same problem, overlapping disclosures. The '826 is STMicroelectronics' own integrated-chemical-microsensor platform; Zribi is a GE MEMS multi‑gas platform; both are directed to miniaturized, thermally isolated, thin‑film sensing with dedicated heaters. KSR "familiar elements" and "same field of endeavor."
5. Design incentive: thermal isolation and low power. Zribi's stated objective — "faster response time… lower power consumption, and less crosstalk between unit sensor devices" — is the same objective the '261 recites ("ensure that the heat is confined to the local region… and is not transmitted to other VOC sensors"). A POSA building a multi‑element array is directly led to Zribi's cell/cavity per element.

Weakness of Ground 1. The '826's transducer is described in its working examples as an electrochemical transducer (working/counter electrodes). A patentee will argue that an electrochemical cell operates at ambient temperature and in a liquid/electrolyte medium, so combining its thermal architecture with a 200–500 °C MOX chemoresistor is improper. Rebuttal: the '826's own text disclaims any such limitation ("the reference to an electrochemical transducer and its electrodes is for illustrative purposes only and is not to be construed as limiting"; "the transducer may be an electrical transducer that measures surface or electrolyte conductivity"), and its stated mechanism is explicitly reaction-rate temperature control — i.e., it contemplates thermally driven surface chemistry. The argument is available but weak.

---

3. Ground 2 — Claim 1 (alternative) over Shankar US 2016/0018356 A1 — and the § 102(b) trap

For completeness and candor: if Shankar is available, claim 1 falls.

Shankar additionally supplies: glass substrate ("the substrate 148, which is made of glass in the embodiment described herein" — claim 9), polyimide pillars + cavity (claim 15), ≤400 °C process flow (claims 12), thermal tuning of a multi‑sensor array to monitor different gas species (claim 10), and TaAl heaters (claim 13).

The trap, stated plainly

Shankar is very likely not prior art. Its inventors are the same four people (Shankar, Le Neel, Loh, Kam) as the '261; it published 2016‑01‑21, i.e., within one year of 2016‑07‑18. Under § 102(b)(1)(A), a disclosure made ≤1 year before the effective filing date by the inventor or a joint inventor is not prior art. Under § 102(b)(2)(C), commonly owned subject matter is excepted from § 102(a)(2) art. And the '261 specification does not merely cite Shankar — it incorporates it by reference in its entirety and states the micro‑sensor array "may be implemented as described in [Shankar]." Incorporation by reference makes Shankar's disclosure part of the '261's own specification for § 112 purposes; it does not convert it into § 102 prior art.

Practical consequence. Shankar is best used not as the § 102 reference but as (a) evidence of the level of ordinary skill and of what was conventional, and (b) an admission in the patentee's own specification that a gas sensor + heater + temperature sensor stacked over an insulating cavity on an ASIC substrate was known technology. That admission matters: it undercuts any argument that the claimed architecture was inventive over the art. A challenger should lead with Ground 1 (which rests on unambiguously available art) and use Shankar only as corroboration.

---

4. Ground 3 — Claim 8 (and 9, 10, 11) obvious over US 9,164,052 B1 (Speer et al./Silicon Laboratories) in view of US 2013/0010826 A1

Claim 8 is the system claim (substrate + array + heater per element + gas micro‑sensor per element + temperature sensor between them + microprocessor + memory). Claim 8's weakest link is not the processor/memory — it is the interposed temperature sensor.

Motivation to combine. (i) The '052's own processor uses the temperature sensor's data to compute gas concentration — "the processor utiliz[es] the detected electrical characteristics of the gas sensitive material… and data from the temperature sensor to determine a gas concentration." A POSA is thus explicitly directed to the temperature‑sensing function as a gas‑concentration input, not a housekeeping parameter; improving its fidelity and its locality is the obvious next step. (ii) The '052's thermometer is a substrate bandgap device measuring die temperature, not the temperature of the heated MOX film — a POSA would recognize that a film‑adjacent thermometer is what the feedback loop needs. (iii) The '826 teaches precisely a deposited thin‑film thermometer placed in the stack between heater and sensing region, with the explicit rationale of matching calibration temperature. (iv) Both are integrated gas‑sensor platforms reading on the same problem ("improved gas sensing accuracy and consistency," per the '052's own Background). (v) Predictable result: tighter temperature control → better selectivity (the '261's FIG. 5).

Claim 9 (glass substrate): Shankar's express glass substrate; also generic in the art and conceded in the '261's own text ("a silicon substrate or a glass substrate").
Claim 10 (temperature‑tunable elements): Shankar's operating method ("thermally tuning the sensors to monitor different gas species"); IBM US 2017/0016866 A1 (MOX sensors at different temperatures for differential measurement); the '261's own FIG. 5 table.
Claim 11 (temperature sensor coupled to the microprocessor): the '826's "control unit" feedback loop; the '052's processor consuming temperature‑sensor data.

---

5. Ground 4 — Claim 12 (and 13–16) obvious over US 2013/0010826 A1 in view of Zribi and the multi‑gas‑array art

Claim 12 is the broadest independent claim in the patent (its through hole need not penetrate the temperature sensor), and it is method‑only. Its steps map almost one‑for‑one onto the '826's disclosed process flow:

Motivation (Ground 4). Two rationales dominate. First, the physical impossibility of species selectivity with one MOX film is common knowledge in the art and is expressly stated in the references — Shankar ("the SMO sensors may have difficulty distinguishing between two gases"); the '261's own FIG. 5 (one material, four different gases at four different temperatures). Reproducing a known unit sensor in a second region of the same die is the paradigmatic KSR "duplication of parts" / "arrangement of elements" case, particularly where, as here, the references expressly contemplate arrays. Second, process economy: the '261's own stated benefit ("a single process flow can be used to fabricate different types of VOC sensors") is delivered by the ordinary practice of masking one sensor region while patterning another — exactly the masking sequence the '261 describes and which Zribi's sacrificial‑layer method already contemplates.

Dependent claims:

• Claim 13 (TaAl low‑TCR heater): Shankar's 870 nm TaAl heater layer; the '261's own text ("TaAl… features a low thermal coefficient (TCR) that results in a stable resistance"); US 2014/0291677 A1 (STMicroelectronics, integrated multi‑sensor module) discloses a low‑TCR refractory‑metal heater common plate.
• Claim 14 (ZnO₂ / In₃O₃ / SnO₂): see § 8 below — and note the literal‑identifier defect: claim 14 recites "In₃O₃" while claim 4 recites "In₂O₃" and the specification writes both "indium oxide (In₂O₃)" and "In₃O₃." Because the challenge is § 103, not § 112, the practical effect is that a POSA reading claim 14 would understand "In₃O₃" to mean the indium oxide already disclosed for the third VOC sensor — so the usual § 103 reasoning applies.
• Claim 15 (first/second air pockets, film spaced from the pocket by the heater): Zribi's cell 18 with the heating element on the membrane above it; Shankar's polyimide‑filled recess (with the pillars supporting the stack) becoming an air volume upon polyimide removal; IMEC US 8,062,497 B2 (sealed cavity).
• Claim 16 (first and second gas sensitive materials are different materials): Zribi's multi‑gas/multi‑vapor sensor devices with different sensing chemistries; the '052's "multiple gas sensors with specific materials… selective to particular gases"; the '261's FIG. 5 (SnO₂/ZnO₂/InO₂ at different temperatures ⇒ different materials).

---

6. Ground 5 — Claim 17 (and 6, 18, 19) obvious over Ground 1 plus through‑hole art

Claim 17 = claim 1 + a through hole penetrating the film, the heater, and the temperature sensor. Claim 18 = claim 17 with the sensor between film and heater. Claim 19 = passivation with openings over the film.

Through‑hole. Etching an access aperture through a deposited thin‑film stack to reach and remove a sacrificial layer beneath it is a bedrock MEMS technique, taught in the same art:

• Zribi: "selectively removing a portion of the second thin film layer and selectively removing a portion of the silicon layer to form at least one cell"; the frame defines a cell "having at least one opening."
• Shankar: openings etched through the SiN layer to the polyimide (bulk polyimide 182) so that the sacrificial material can be removed / the cavity formed behind the pillars.
• US 2008/0308920 A1 (Wan): micro‑cavity fabrication by sacrificial removal through etch access.
• US 8,062,497 B2 (IMEC): access apertures for forming a hermetically sealed cavity.
• The '261 specification itself: "Openings 256 are formed by etching through the VOC sensor layers and ILD layers to expose the polyimide wells 226."

And the result is predictable and stated: thermal isolation. Zribi: "The resulting thermal isolation provides faster response time, greater mechanical strength, lower power consumption, and less crosstalk." Shankar: heat confinement lowered measured power "from 900 mW to less than about 5 mW." The '261's own air pocket rationale is identical. A POSA seeking the '261's stated goal would route the release etch through the stack as a matter of course.

Claim 19 / claim 7 (passivation with openings over the gas sensitive material). Directly met by the '826: "A passivation layer is then formed over the temperature sensor. The passivation layer is patterned to define a sensing region and expose at least a portion of the working electrode and counter electrode." Also met by the '052: "windows may be formed in the second RDL insulative layer 212 to provide the location for a gas sensitive layer 214." The '261's own spec does the same ("The passivation layer 254 is patterned to expose the VOC sensors"). This limitation is, for practical purposes, unavoidable in any integrated MOX sensor and adds nothing.

Claim 18 is simply claim 1's ordering, already addressed in Ground 1.

---

7. Motivation to combine — consolidated KSR rationales

---

8. Dependent claims — accelerated analysis

---

9. Where the patent pushes back — the credible counterarguments

For an honest assessment, the patentee's best defenses, and their weight:

1. "No reference teaches the combination." The strongest defense. Claim 1's four‑element stack with the thermal sensor interposed and aligned is not disclosed in any single available reference (Shankar is the only one, and it is likely § 102(b)‑excepted). The patentee will argue that the examiner considered all of this art and allowed the claims anyway — a weak argument (the examiner's allowance is not evidence of non‑obviousness), but it explains why the ground must be a two‑reference combination, which is always attackable as hindsight.
2. The '826's transducer is electrochemical, not MOX. An argument that combining the '826's thermal architecture with a 200–500 °C chemoresistor changes the principle of operation. Weight: low–moderate. The '826 disclaims the electrochemical limitation for illustrative purposes and expressly contemplates conductivity, optical, mass‑ and heat‑sensitive transducers; its stated rationale is reaction‑rate temperature control.
3. The '826 isolates thermally by a solid insulating layer, not a cavity. A "teaching away" argument from air‑cavity isolation. Weight: low. Zribi, Shankar, Wan, and IMEC all teach cavity/cell isolation, and the '826's solid layer is a design choice, not a denigration of cavities.
4. The '261 claims "aligned with each other" — a word absent from the '826. Weight: low–moderate, and construction‑dependent. Zribi uses "substantially aligned"; the '826's stack is inherently vertically aligned because layers are deposited one atop another. But if the patentee can persuade a tribunal that "aligned" requires strict coaxial registration (e.g., concentrically registered cavity/heater/thermometer/film), the combination weakens and the claim narrows sharply — a narrowing that likely costs infringement coverage.
5. § 102(b) safe harbor for Shankar. Disqualifying the best reference. Weight: high as a threshold matter — a challenger must build the case without Shankar as § 102 art. Ground 1 does that.
6. § 112 problems are not § 103 defenses. The In₃O₃/In₂O₃ and "ZnO₂" inconsistencies, and the claim/specification mismatch over CO₂ and humidity, are not invalidity arguments in an IPR (no indefiniteness ground). They do, however, undercut any argument that these claims represent a carefully scoped, non‑obvious advance, and they surface in a district‑court § 112 defense.

Net: Ground 1 is a strong but not overwhelming prima facie case on claim 1. The § 102(b) refuge for Shankar is the single largest vulnerability in the challenge, and the "aligned" limitation is the single largest claim‑construction risk.

---

10. Secondary considerations

There is essentially nothing to weigh on the patentee's side, on the present record.

• No commercial success evidence with nexus. The earlier-generated sections found no commercial ST part number and no datasheet for a MOX multi‑gas sensor built to these claims; ST's own air‑quality application material instead points to third‑party (Sensirion) VOC/CO₂ sensors. Without a covered product, there is no basis for a commercial‑success argument, and even a covered product would require a nexus to the claimed ordering/alignment, not to the '261's overall sensor functionality.
• No licensing, copying, or industry‑acclaim evidence — and no litigation to have surfaced any. The earlier litigation and PTAB sections found no assertion, no IPR/PGR/CBM, no CAFC activity, and no § 315(e)(2) estoppel on this patent.
• No unexpected results or long‑felt need supported by comparative data. The '261 contains no comparative testing against a non‑interposed‑thermometer device; its performance benefits (power confinement, accuracy) trace to the incorporated Shankar disclosure, not to the claimed "between"/"aligned" geometry.
• The strongest available objective evidence runs against the patent: the asserted benefit — a thin film "about 100 times thinner than the bulk sensor material" enabling integration — is squarely a known MOX miniaturization story already told by Zribi, Shankar, IBM, and Speer.

If the patentee later develops secondary‑consideration evidence tied to a shipping part, the strongest counter would be simultaneous invention / industry convergence — the very number of independent actors (GE, Silicon Labs, IBM, ams, InSyte, plus ST itself) arriving at integrated MOX sensors with stacked heaters and thermometers in 2011–2016.

---

11. Bottom line — ranked vulnerability

Confidence rating on the overall § 103 conclusion: Moderate‑to‑high on claims 12/17/6/18; moderate on claims 1/8; moderate on the dependent claims. The limiting factors are (i) the § 102(b) safe harbor that likely removes Shankar, the only single reference that maps the full stack, forcing a two‑reference combination; and (ii) the unresolved construction of "aligned with each other."

What would change my mind about claim 1: (a) a construction of "aligned with each other" requiring strict coaxial registration of all four elements (which the '826's sequentially deposited stack and Zribi's "substantially aligned" cell would not literally meet); (b) a persuasive showing that a POSA could not have substituted a 200–500 °C MOX chemoresistor into the '826's architecture without undue experimentation (unlikely — process temperatures ≤400 °C are the patent's own concession of the state of the art); or (c) objective evidence of unexpected results with nexus to the specific interposed‑thermometer geometry (none of record).

---

12. Caveats and verification gaps

1. Shankar's § 102 status is a legal conclusion I cannot certify. The exception analysis turns on the identity of the inventive entities and the ownership dates. I verified the inventor lists (identical four names on both) and common assignment to STMicroelectronics Pte Ltd, and the publication date (2016‑01‑21). Before relying on a Shankar‑based ground, confirm the § 102(b)(1)/(b)(2) analysis against the USPTO's assignment records and the AIA grace‑period guidance.
2. Full claim text not retrieved for US 9,105,479 B2 / EP 2 645 091 B1 (ams), US 2017/0336343 A1 (InSyte), US 2016/0290946 A1 (Lfoundry), CN 1684285 A, and several 2012–2014 STMicroelectronics applications. Their § 102/§ 103 weight is stated at a candidate level only.
3. The '826's application number/date. The granted counterpart I verified is US 9,448,198 B2, filed 2011‑07‑05, granted 2016‑09‑20; the publication US 2013/0010826 A1 (2013‑01‑10) is the operative § 102(a)(1) event. The grant itself postdates the critical date and is not the § 102(a)(1) event — the publication is.
4. Zribi filing‑date discrepancy. Google Patents/Justia show a 2003‑11‑21 filing; one aggregator record shows "18 November 2004" (likely a division/continuation). Either date predates 2016‑07‑18 by more than a decade, so the discrepancy is immaterial to availability.
5. No post‑grant record to leverage. Consistent with the earlier PTAB/litigation sections: no IPR/PGR/CBM, no FWD, no estoppel, no claim cancellation, no certified construction. This is a clean‑slate, never‑tested operating‑company patent — which cuts both ways: no adverse findings to cite, and no roadmap to inherit.
6. This is a § 103 analysis, not a legal opinion, and not an invalidity or FTO conclusion. It is intended to identify and rank grounds and the motivations that support them; a petition or invalidity contention would require full‑text review of every reference relied on, a claim‑construction position on "aligned with each other," and confirmation of each reference's § 102(b) status.

Authoritative source for the patent: https://patents.google.com/patent/US10254261/en
Key references verified in this session: https://patents.google.com/patent/US20130010826A1 · https://patentimages.storage.googleapis.com/b6/10/36/b8574cb362e5f2/US20130010826A1.pdf · https://patentimages.storage.googleapis.com/f8/9f/cd/5759aeb606141b/US9448198.pdf · https://patents.google.com/patent/US20050109081A1/en · https://patentimages.storage.googleapis.com/87/fe/3e/723383eeb70c1d/US7104113.pdf · https://patents.google.com/patent/US9164052 · https://patentimages.storage.googleapis.com/34/d9/68/8690bbff039c1f/US9164052.pdf · https://patentimages.storage.googleapis.com/22/a0/73/7310c315cbf396/US9810653.pdf · https://www.patents-review.com/a/20160018356-integrated-smo-gas-sensor-module.html · https://patentimages.storage.googleapis.com/cd/5e/58/bf06d6fa1f6ca8/US8852513.pdf