US 9897487

I'll search for this specific patent number across USPTO and litigation sources.

Initial searches returned noise. Let me run more targeted queries.

Let me check specifically for any litigation, PTAB, or CAFC activity.

US Patent 9,897,487 — Verification Summary

Note on scope: I searched for this exact number (9897487 / "9,897,487") across patent databases and litigation sources. I found no evidence of any CAFC 2026 docket, IPR/PTAB proceeding, or district court litigation involving this patent. My CAFC/litigation queries returned unrelated results (Shockwave/CSI, Ethanol Boosting v. Ford, Realtime Data, etc.), none referencing 9,897,487. The only authoritative record located is the granted patent itself. I have flagged uncertainty where it applies below.

---

Bibliographic Data (confirmed)

Legal status: Google Patents labels it "Expired – Fee Related," with an adjusted expiration of 2036-09-11. However, the Legal Events record shows a maintenance-fee reminder (2025-10-13), a lapse for failure to pay maintenance fees on 2026-03-30, and an "FP – Lapsed due to failure to pay maintenance fee" with effective date 2026-02-20. So as of the current date, the patent appears lapsed/expired for non-payment of the 8th-year maintenance fee rather than merely "expired" on its original term. (Per your instruction, I report the numbers literally; I note that the "2036-09-11" adjusted-expiration figure and the 2026 lapse event are not obviously reconciled, and Google's status label is expressly an assumption, not a legal conclusion.)

---

Abstract (verbatim)

"A combined leg structure of a micro bridge unit of a focal plane array adopts a conductive polymer film or a doped conductive polymer film to serve as an extraction electrode in the micro bridge unit of the focal plane array, which contacts a vanadium oxide thermosensitive film or a doped vanadium oxide thermosensitive film of a bridge surface layer, so as to electrically connect the thermosensitive film of the micro bridge unit with a read-out circuit. The combined leg structure includes three layers: respectively an upper SiNx film layer, a lower SiNx film layer and a middle layer of the conductive polymer film or the doped conductive polymer film. The present invention adopts the conductive polymer film or the doped conductive polymer film having a low thermal conductivity to serve as an electrode material. A bridge leg absorption structure is arranged in the combined leg structure."

---

Independent Claim — Plain-Language Overview

Claim 1 is the sole independent claim. All of claims 2–20 depend, directly or indirectly, from it; there are no other independent claims.

Claim 1 (paraphrased): A "combined leg structure" for a micro bridge unit of a focal plane array, characterized in that the structure uses a conductive polymer film, or a doped conductive polymer, as the extraction electrode inside the micro bridge unit. That electrode contacts a vanadium-oxide thermosensitive film (or a doped vanadium-oxide thermosensitive film) located on the bridge's surface layer, and thereby electrically connects that thermosensitive film to the array's read-out circuit.

In plain terms: instead of the conventional NiCr or Ti metal electrode in a bolometer/microbolometer bridge leg, claim 1 claims using a conductive (or doped conductive) polymer as the conductive trace that carries the pixel's resistance signal from the VOx sensing film to the readout circuitry.

Notable literal detail (not corrected): Claim 1 recites "a conductive polymer film or a doped conductive polymer" (the word "film" is absent from the doped-polymer alternative), whereas claim 2, the dependent claims, the abstract, and the specification consistently use "doped conductive polymer film." This is an internal inconsistency in the claim language itself.

Key Dependent Claims (themes)

Because claim 1 is the only independent claim, the following are the substantive narrowing limitations worth noting:

• Claim 2 — the conductive polymer is polyacetylene, polythiophene, polypyrrole, or polyaniline.
• Claim 3 — the dopant is a Lewis acid or a protonic acid (specification: ferric trichloride; sulfamic acid).
• Claim 4 (and claims 5–6 via 2/3) — adds the "bridge leg absorption structure" architecture: an absorption structure near the bridge surface layer, with a first bridge leg connecting a first side to the electrode contact hole and a second bridge leg connecting a second side to the bridge column hole.
• Claims 7–9 — the first bridge leg, absorption structure, and second bridge leg are each three-layer stacks: upper SiNx / conductive-polymer electrode / lower SiNx.
• Claims 10–12 — thickness ranges: upper SiNx 50–100 nm, conductive polymer 50–150 nm, lower SiNx 100–200 nm.
• Claims 13–14 — bridge leg absorption structure length and width each 10–20% of the micro bridge unit's length and width.
• Claims 15–16 — first bridge leg width 0.5–1.0 µm, length 3.0–5.0 µm.
• Claims 17–18 — second bridge leg width 0.8–2.0 µm, length set by unit size and absorption-structure length.
• Claims 19–20 — upper SiNx deposited by RF sputtering or CVD at ≤200 °C.

Worked Embodiments (NETD results)

Four embodiments on a 128×128 array with 50×50 µm micro bridge units, tested at 25 °C with 6 µs / 150 µA bias pulses:

Prior Art Cited on the Face

• US 2007/0108383 A1 (Combes) — Thermal detector
• US 2015/0021479 A1 (Jian Cai) — Bolometer and preparation method
• US 9,677,946 B1 (Magnolia Optical Technologies) — IR detectors using CNT-silicon vanadium oxide / a-Si nanoparticle-CNT nanocomposites

---

Confidence & Caveats

• High confidence on title, assignee, inventors, dates, abstract, claim structure, and the claim-1 overview — these come from the full authoritative patent text.
• Uncertain / unverified: I could not locate any CAFC 2026 docket, PTAB proceeding, or litigation for 9,897,487. Absence of results in my searches is not proof of absence, but I found nothing linking this number to any 2026 appellate docket.
• The "expires 2036-09-11" adjusted expiration and the 2026-02-20/2026-03-30 lapse events appear inconsistent with each other in the source record; treat the precise expiry mechanics as unresolved.
• The specification contains several apparent typographical anomalies retained literally in the text (e.g., "polyacethlene," "polylyrrole," "the first bridge leg Two," "foal plane array," "facilities" for "facilitates"). I have not corrected these, per your instruction.

I'll search for litigation involving US patent 9897487.

Let me search additional litigation-focused sources.

Let me try a few more specifically litigation-oriented sources.

Litigation Report — U.S. Patent No. 9,897,487

Bottom line

No litigation involving U.S. Patent No. 9,897,487 ("the '487 patent") was found. I identified no district court complaints, no Court of Appeals for the Federal Circuit appeals, no PTAB proceedings (IPR/PGR/CBM), no ITC Section 337 investigations, and no declaratory judgment actions in which the '487 patent is the asserted or challenged patent.

I want to be explicit about the confidence level here rather than overstate it: my searches were web searches of patent/litigation aggregators (Google Patents, uspto.report, Unified Patents, Docket Alarm, CourtListener, and general queries), not a direct PACER or Docket Navigator query. Absent a full PACER/Docket Navigator run, I cannot rule out a low-profile filing that never surfaced in indexed sources. However, the absence of any hits is itself meaningful, and consistent with the circumstances described below.

Patent identification (verify against the source before relying on it)

• Patent: US 9,897,487 B2 — "Combined leg structure of micro bridge unit of focal plane array"
• Application: US 14/963,822; filed 2015-12-09
• Priority: CN 201510532320.6, filed 2015-08-26
• Granted: 2018-02-20
• Inventors: Tao Wang, Liangzhong Wei, Wanglong Zhu, Yan Liu
• Assignee: Wuxi Aleader Intelligent Technology Co., Ltd. (Chinese company; also appears as "Wuxi Allide/Ailide Intelligent Technology")
• Family: CN105060237A / CN105060237B (Granted; "active" per Google Patents)
• US legal status: Expired – Fee Related. Per the source text, maintenance fee reminder mailed 2025-10-13; lapsed 2026-03-30 for failure to pay maintenance fees, effective date of lapse 2026-02-20. Adjusted expiration listed as 2036-09-11.

Why no litigation is unsurprising

1. Owner profile. Wuxi Aleader is an operating sensor/MEMS company in Wuxi, Jiangsu, China, founded 2011, with a portfolio of roughly 79 patents. It does not exhibit the patent-acquisition-and-assertion pattern typical of NPEs that generate litigations.
2. Technology. The patent claims a microbolometer/focal-plane-array micro-bridge leg structure using a conductive-polymer extraction electrode (claims 1–20). This is a narrow MEMS fabrication-space patent.
3. No third-party challenges. The Google Patents record shows only three examiner-cited references (US20070108383A1, US20150021479A1, US9677946B1) and no post-grant challenges.
4. Fee lapse. The patent expired for nonpayment of the 8-year maintenance fee with an effective lapse date of 2026-02-20, indicating the owner did not view it as worth maintaining — a strong signal it was not being asserted.

CPC classification (for completeness)

G01J 5/0225, G01J 5/024, G01J 5/046 (radiation pyrometry / constructional details; materials selection for thermal detectors).

Recommended verification steps

Because a negative finding is only as good as the searches behind it, I recommend confirming with:

• PACER / Docket Navigator — full-text search on patent number "9,897,487" across all district courts
• Unified Patents Portal — https://portal.unifiedpatents.com/litigation/caselist
• CAFC docket search for any appeal referencing the '487 patent
• USPTO Patent Trials (PTAB) — confirm no IPR/PGR/CBM
• ITC EDIS — confirm no Section 337 investigation
• China — CN family member CN105060237B shows "active" status; Chinese enforcement actions would not appear in US dockets and were not searched

Note on search-result reliability

Several litigation-focused searches returned hits for unrelated patents (e.g., the Fall Line '748 patent, Ozmo '991, MCOM IP, Fractus '887) purely because of token overlap in the query. I have excluded all of these; none involve US 9,897,487. Flagging this because a casual keyword search on "9897487" can produce false positives of exactly that kind.

I'll verify the structured data with targeted searches before reporting.

Let me run a couple more targeted checks for litigation or PTAB activity that could have spawned trials.

Proceedings overview

Total AIA trial proceedings on file for US 9,897,487: zero. The USPTO Open Data Portal returns no IPR, PGR, or CBM for this patent as of the most recent ingest, and my independent web searches (Justia, uspto.report, PTAB petition/decision repositories, litigation reporters) surfaced no petition, institution decision, Final Written Decision, or appeal naming US 9,897,487 or application 14/963,822. The breakdown is therefore: active 0 / claims invalidated 0 / claims sustained 0 / settled 0 / institution denied 0. The defensive posture this gives a defendant is the opposite of "the patent is hardened" — it means the patent's 20 claims have never been tested by the Board, and the patent itself has now lapsed for failure to pay the 7.5-year maintenance fee (lapse effective 2026-02-20). There is no per-proceeding section to write, because there is no proceeding.

Two caveats on the "zero" finding, stated explicitly per the operating rules:

• I am reporting an absence, not a verified negative. Absence in ODP plus absence in surfaced web results is strong but not dispositive; a petition filed within the last few weeks may not be indexed yet.
• I found no district court litigation asserting 9,897,487 in my searches (including a search on the assignee, Wuxi Aleader Intelligent Technology Co., Ltd.). That absence is itself notable — AIA trials almost always follow an infringement assertion, so a patent with neither litigation nor IPR activity is a patent that has not been asserted in the US at scale.

---

No proceedings to itemize

Because the structured "PTAB proceedings on file" block reports no AIA trial proceedings, there is no {PROCEEDING_NUMBER}, no petitioner, no judge panel, no institution decision, no FWD, and no appeal to report. I will not invent proceeding numbers or attribute claim-level holdings to decisions that do not exist. Everything below is derived from the patent's face and its legal-status record rather than from PTAB decisions.

---

Strategic summary

Claim status of 9,897,487 — all 20 claims are UNTESTED. No claim has been canceled, confirmed, or construed by the PTAB. Importantly, the claim set is structurally brittle at the top: claim 1 is the only independent claim. Claims 2–20 are all dependent (2 and 3 directly on claim 1; 4–20 flow down through claims 4/5/6, 7/8/9, and 10–12). Every claim therefore incorporates claim 1's core limitation — a conductive-polymer or doped-conductive-polymer extraction electrode contacting a vanadium oxide (or doped vanadium oxide) thermosensitive film. A single successful § 103 challenge to claim 1 would take the entire patent with it; conversely, if claim 1 is valid, the dependent claims add only narrow structural and dimensional limitations (0.5–1.0 μm first bridge leg, 50–100/50–150/100–200 nm layer thicknesses, 10–20% absorption-structure sizing, ≤200 °C SiNx deposition). That is a thin secondary positional defense and it is entirely untested.

Estoppel landscape — the field is wide open. Because no IPR or PGR was ever instituted, § 315(e)(2) estoppel attaches to no one. There is no petitioner, no real party in interest, and no privy precluded from asserting § 102/§ 103 grounds in a district court or before the Board. A defendant today may file an IPR on any prior-art ground, and may raise any invalidity theory in litigation, without any estoppel shadow. The flip side: because nobody has run this art, the prior art itself is uncharacterized — no institution decision or FWD tells you which references the Board found persuasive, so a defendant bears the full cost of developing the invalidity case from scratch. The patent's own prosecution citation list is thin (three US references): US 2007/0108383 A1 (Combes, "Thermal detector"), US 2015/0021479 A1 (Jian Cai, "Bolometer and preparation method thereof"), and US 9,677,946 B1 (Magnolia Optical, CNT-silicon/vanadium-oxide nanocomposites). The Chinese family citations include CN 101357747, CN 101774530, CN 103940518 (a low–thermal-conductivity terahertz microbridge unit) and CN 204873819 U (the utility-model counterpart to this patent) — the CN 103940518 reference in particular looks like a plausible starting point for a low-thermal-conductivity-leg obviousness theory, though I have not reviewed its disclosure and do not represent that it anticipates.

Pattern signals — none in the US. No petitioner has filed once, let alone multiple times. The patent owner has never had occasion to appeal to the Federal Circuit (there is no FWD to appeal). No defensive aggregator (Unified Patents or similar) appears anywhere in the chain — consistent with the absence of assertions. The only family activity is the Chinese counterpart, CN 105060237 B (granted 2016-09-14, priority 2015-08-26 to CN 201510532320.6), which I located but whose validity has not been adjudicated in any source I found. The two third-party citations of this family are unrelated (a CEA anti-glare radiation sensor, FR 3064060 B1, and a Shandong University InGaAs FPA evaluation device, CN 108844637 B).

The single most important fact for a defendant today is not in the PTAB record at all: the patent expired 2026-02-20 for failure to pay the 7.5-year maintenance fee (fee reminder mailed 2025-10-13; notice of lapse and 37 C.F.R. § 1.362 discontinuation entered 2026-03-30, effective 2026-02-20). Google Patents lists the status as "Expired – Fee Related," with a nominal adjusted expiration of 2036-09-11 that the maintenance-fee lapse cuts short. An expired patent cannot be asserted against prospective conduct, so any demand letter threatening suit over ongoing manufacture, use, or sale is defective on its face. One caveat, and it is a real one: a lapsed patent can be revived under 37 C.F.R. § 1.378 upon a petition showing the delay was unintentional, and lapsed-for-fee patents are frequently revived. Before relying on expiration as a complete defense, confirm current status directly in USPTO PatentCenter — revival would restore enforceability, and intervening rights under 35 U.S.C. § 252 (as applied to lapsed/revived patents) would then become the fight.

---

Recommended next steps

• Verify the lapse before anything else. Pull the fee history and any revival petition in USPTO PatentCenter (Application 14/963,822) and the Patent Assignment/legal-events record. If no petition under § 1.378 has been filed, expiration is your lead argument. This is a status question, not a merits question, and it is dispositive.
• Do not expect an FWD to help you — there isn't one. There is no final written decision to cite, no claim canceled, and no Board reasoning to lean on. Your invalidity case is a clean slate.
• No active trials are pending, so there are no institution deadlines, oral hearing dates, or statutory § 316(a)(11) one-year FWD deadlines to track. Nothing will happen at the Board unless someone files.
• If a demand letter has actually issued (none surfaced in my search), the normal defensive playbook applies: (1) lead with expiration/revival status; (2) map the asserted claims — note that every one of claims 2–20 requires all of claim 1's limitations; (3) because no estoppel exists, preserve IPR as a live option, but weigh the expired status of the patent against the cost — a fee-lapsed patent frequently does not justify IPR spend.
• The absence of PTAB activity is itself the signal. Well-asserted patents eventually attract IPRs. Twenty claims, one independent claim, and zero AIA trials over a decade suggests this patent has never been a meaningful US assertion vehicle.

Sources: Google Patents – US9897487B2 (claims, specification, family, citations, legal events, maintenance-fee lapse); Justia – US20160097675A1 (claim text); uspto.report – 9,897,487 (grant and assignee data). For claim-level outcomes I would normally link USPTO PTAB E2E/CourtListener decisions — there are none to link, because no proceeding exists to cite.

If you have a specific IPR or PGR number in hand that postdates the ODP ingest, give it to me and I will retrieve and analyze that record directly.

I'll search for the recorded assignment history for this patent.

Let me check the USPTO assignment database directly and look for any recorded transfers.

Let me check a few more sources for any recorded ownership changes.

I was able to confirm the ownership picture but could not retrieve any assignment-center record for this patent. Here is the report.

---

Assignment & Ownership Report — U.S. Patent No. 9,897,487

Report scope note (read first): I could not retrieve a single recorded post-issuance assignment for US 9,897,487 from any indexed source (USPTO Assignment Center / Assignments on the Web / Google Patents legal events / uspto.report). Since the assignee named on the issued patent is the same entity that filed the application as applicant, the most probable state is that there is no recorded assignment document at all — the chain consists of the original filing only. The USPTO Assignment Center is a JavaScript-gated query interface that my search tooling cannot render directly; I am reporting the absence of indexed records, not a certified null result. Per your instruction, I am not inventing a chain to fill the gap.

---

Inventors

Employer basis: The application was filed with Wuxi Aleader as the sole applicant/assignee and claims priority to CN 201510532320.6, which was filed by the same Chinese company. The inventors are Wuxi-based and their names correspond to the company's technical staff.

Notable corroborating observation — inventor retention, not flight. The "all inventors gone within 12 months" precursor to a fire-sale is not present. Inventor Liangzhong Wei (韦良忠) still appears as the first-listed inventor on a 2025 Chinese application, CN 120086873 A ("基于区块链的智能数据加密传输系统及方法", filed 2025-02-13), assigned to the same company, 无锡艾立德智能科技有限公司. That is roughly a decade after the '487 filing and is strong evidence of continuity rather than a team breakup.

Separate, relevant naming fact: Inventor Yan Liu (刘燕) shares a name with 无锡艾立德智能科技有限公司's registered legal representative (法人代表), 刘燕, per Chinese corporate registry data (establishment date 2011-01-30; registration no. 320213000148937; unified social credit code 91320214569136017C). I flag this as a probable identity match on the Chinese-name characters; I have not independently verified that the inventor and the legal representative are the same natural person.

---

Original assignee

Wuxi Aleader Intelligent Technology Co., Ltd. (无锡艾立德智能科技有限公司)

• Named on the issued patent: Yes — "Wuxi Aleader Intelligent Technology Co LTD" appears as both original and current assignee in the Google Patents record, and uspto.report lists the grantee as Wuxi Aleader Intelligent Technology Co., LTD.
• Primary line of business: A fabless/IC-design and MEMS sensor company — sensors, integrated circuits, infrared detection systems and equipment, smart-control systems ("是一家集研发、制造、销售为一体的国家集成电路设计企业" — a national IC-design enterprise integrating R&D, manufacturing and sales). Registered capital RMB 40 million; ~3,500 m² of R&D/production space.
• Does it ship a product embodying the claims? Likely yes, on the balance of the record. The company's registered scope expressly covers the design, manufacture and sale of infrared detection systems and equipment, and the patent's own embodiments target an uncooled VOx infrared focal plane array at 128×128. I did not find a public product datasheet that maps specific product part numbers to claims 1–20, so treat "ships an embodying product" as strongly indicated but not documentarily nailed down.
• Current status: Operating, not dissolved. Still active: 18 employees reported (2025); recent filings as recent as 2026 (CN application 202610665617.8); named as a bid candidate on a 2026 China Unicom supply/ops framework procurement. It is also a regionally listed entity on the Jiangsu Equity Exchange Center (江苏股权交易中心), securities code 650472, short name 艾立德, listing date 2021-07-09.
• Financing behaviour relevant to patents (context only — do NOT conflate with the '487 chain): Chinese registry pledge records show the same company using patents as bank collateral — e.g. 专利权质押合同登记 for CN 103792011 A, pledge to 中国农业银行股份有限公司无锡科技支行 (Agricultural Bank of China, Wuxi Science & Technology Branch), registered effective 2018-08-14 (no. 2018990000683) and again effective 2022-08-25 (no. Y2022320000491). That pledge is on a different patent (a 2014-filed thermal-imager drift-compensation case), not on CN 105060237 B and not on US 9,897,487. It tells us the owner monetises its portfolio through secured lending, which is a benign operating-company financing pattern — not an NPE signal. I am flagging it precisely because it is the kind of record easily mistaken for an assignment in this chain.

---

Assignment timeline

No recorded assignments were located for US 9,897,487. Neither the Google Patents legal-events stream for US 9,897,487 nor the uspto.report grant record for the patent surfaces a reel/frame entry, an Assignments section, or a change of assignee. The only ownership entry anywhere in the file is the original grant to Wuxi Aleader Intelligent Technology Co., LTD.

• No execution date / no recording date — no reel/frame located
  • Conveyance: none recorded
  • Assignor: n/a
  • Assignee: n/a
  • Correspondent: not determinable — no recording correspondent to capture. (The Chinese-language prosecution agent of record on the assignee's other filings is 无锡佳信专利代理事务所 / Wuxi Jiaxin Patent Agency, agent 宋亚超, but that is a CN prosecution agency and not a US assignment correspondent; I will not present it as one.)
  • Context: the absence of any record is itself the finding — it indicates the original assignee has retained ownership continuously since filing.

The Chinese family member CN 105060237 B is shown as "active" in the Google Patents family table; that status likewise reflects continued ownership by the Chinese counterpart entity, with no recorded transfer.

Because there is only the original assignment, I am flagging — per your instruction — that the substantive chain reconstruction ends here.

Link for independent verification: https://assignmentcenter.uspto.gov/ (also https://assignment.uspto.gov/patent/index.html — search by patent number 9897487 or application 14/963,822).

---

Timeline diagram

timeline
    title Ownership of US 9897487
    2015 : Filed by Wuxi Aleader
         : CN priority 201510532320.6
    2018 : US patent issued to Wuxi Aleader
    2018 : Sibling CN patent pledged as bank collateral
    2021 : Owner listed on Jiangsu equity exchange
    2026 : US patent lapses for fee nonpayment

(The 2018 pledge line refers to a different Chinese patent, CN 103792011 A, and is shown only as owner-financing context.)

---

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT. No assignment record moves the patent out of the operating company. The patent remains with Wuxi Aleader Intelligent Technology Co., Ltd., a company with a physical address (江苏省无锡市新吴区景贤路52号乐东大厦五层), an 18-person payroll, and active manufacturing/sales scope — the opposite of a registered-agent service address or a single-member Delaware/Texas LLC. No reconveyance to any "IP / Holdings / Ventures / Licensing" suffix entity appears anywhere in the file.

2. Known asserter in the chain — NOT PRESENT. I compared the assignee against the named asserted-NPE list you supplied (Acacia, Marathon, Intellectual Ventures, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, Spangenberg entities) and found no overlap at any point. No third-party or Unified Patents / RPX high-frequency-plaintiff entity appears as assignee.

3. Repeat correspondent across the chain — NOT PRESENT / NOT DETERMINABLE. With zero recorded assignments, there is no correspondent to recur. I searched for a recording attorney and found none, because there is no recording.

4. Cascading transfers — NOT PRESENT. No consecutive assignments exist, so no chained-LLC pattern within any window, let alone <24 months.

5. Pre-litigation transfer — NOT PRESENT. There is no transfer, and per the prior litigation section of this analysis there is no infringement suit naming the '487 patent from which a 6-month clock could be measured.

6. Bankruptcy fire-sale — NOT PRESENT. The assignee is not in bankruptcy; it is actively bidding on procurement contracts in 2026 and filing new applications as recently as 2026. No Chapter 7/11 sale, no docket, no patent-sale record surfaced.

7. Privateering — NOT PRESENT. No transfer to any entity that asserts on the operating company's behalf; nothing in the record suggests the operating company is using an NPE proxy. No SEC-report basis exists either, since Wuxi Aleader is a Chinese regional-equity-market issuer, not an SEC registrant (no 10-K/8-K available for cross-reference).

8. Defensive aggregator — NOT PRESENT. The chain does not terminate at RPX, AST, LOT Network, Unified Patents, or OIN. No such assignment exists. The related inverse situation — the patent being neutralised — is not attested by an assignment either.

---

Verdict

Insufficient data (no records, or only the original assignment).

The assignment chain consists solely of the original grant to Wuxi Aleader Intelligent Technology Co., Ltd. — no reel/frame entry, no conveyance type, no correspondent, and no execution or recording date could be located for any transfer, so there is literally nothing to score against signals 1, 3, 4, 5 or 8. Framing it as "insufficient data" rather than "defensive / non-asserting" is deliberate: the absence of transfers here does not reflect a deliberate anti-NPE aggregation step; it reflects an operating company that simply kept its patent. It is also not "operating-company assertion," because that category requires the current assignee to be suing actual competitors, and the earlier litigation section found no litigation, PTAB or ITC activity involving US 9,897,487, and the patent lapsed 2026-03-30 (effective 2026-02-20) for failure to pay the 8th-year maintenance fee — a fee-lapse, not an assertion, profile.

Net characterisation: a Chinese MEMS/IC operating company retained continuous ownership of its own patent and let the US member lapse for non-payment. Every concrete, independently verifiable data point — the physical address, the 2025–2026 filing activity, the payroll, the 2018/2022 patent-collateral pledges on other patents — points away from an NPE pattern.

Verify the null result independently at https://assignmentcenter.uspto.gov/ and https://uspto.report/patent/grant/9,897,487.

I'll research each cited reference to build an accurate prior-art analysis. Let me search for the key references.

Good data on the core references. Let me research the remaining family-cited references.

Prior-Art Analysis — U.S. Patent No. 9,897,487 B2

"Combined leg structure of micro bridge unit of focal plane array"

Building on the previously generated summary and litigation sections (which I treat as authoritative and will not repeat), this section addresses only the prior-art question: what is cited, what qualifies as prior art, and what — if anything — each reference can anticipate under 35 U.S.C. § 102.

---

1. Analytical framework (critical for § 102)

The '487 patent claims priority to CN 201510532320.6, filed 2015-08-26, and was filed in the U.S. on 2015-12-09. Both dates post-date the March 16, 2013 AIA date of first application, so the AIA version of § 102 governs. For prior-art purposes the operative cutoff is the effective filing date of 2015-08-26.

That produces two distinct categories:

Two immediate consequences that are easy to get wrong:

• A foreign (CN/KR) document is not § 102(a)(2) art. It can only be prior art under § 102(a)(1), i.e., if it published before 2015-08-26.
• A U.S. reference published after 2015-08-26 is not § 102(a)(1) art, but may be § 102(a)(2) art if its effective filing date is earlier.

Overarching caveat (applies to every reference below): Claim 1 is the sole independent claim, and its characterizing limitation — the extraction electrode being "a conductive polymer film or a doped conductive polymer" — is narrow and specific. None of the references cited on the face of the '487 patent discloses that limitation in the claimed role. My honest conclusion, stated up front, is that no cited reference anticipates claim 1 under § 102; the citations are materially § 103 (obviousness) material. I map each one to the claims it comes closest to below, and flag where a § 102 theory could arguably be pressed.

---

2. The examiner-cited references (the three on the face of the patent)

Google Patents lists exactly three references with the "cited by examiner" flag: US20070108383A1, US20150021479A1, and US9677946B1.

2.1 — US 2007/0108383 A1 (Combes, "Thermal detector")

Full citation: US 2007/0108383 A1, "Thermal detector," inventor David J. Combes; U.S. application Ser. No. 10/563,056 (national stage).
Filing date: 2004-07-02; priority: 2003-07-03; publication date: 2007-05-17. Status: abandoned.
Source: https://patents.google.com/patent/US20070108383

Brief description: A thermal (infrared) detector that does not rely on resistance/capacitance of a temperature-sensitive layer. Instead it measures the shift in a resonant mode of a resonator element carried on a suspended portion (microbridge) spaced above the substrate. The suspension is on one or more legs (suspension elements) that provide both thermal isolation and electrical connection ("the legs may advantageously include conductive material to provide an electrical connection between the resonator element and the underlying substrate"). The supporting frame may be a multiple layer stack including an absorber layer — expressly a "metal absorber layer of matched impedance to free space, such as titanium with a sheet resistance of 377 Ohms/square" — in a dielectric–metal–dielectric stack, and "the absorber layer may perform both absorber and electrical connection roles."

§ 102 analysis:

• Qualifies as prior art: Yes, unequivocally, under § 102(a)(1) — published 2007-05-17, eight years before the effective filing date.
• Claim 1: Does not anticipate. Claim 1 requires the extraction electrode to be a conductive polymer film or a doped conductive polymer contacting the VOx thermosensitive film. Combes' conductive leg/absorber material is a metal (titanium), and its detection modality is resonant, not resistance-based (no VOx thermosensitive film at all). The conductive-polymer limitation is absent, so § 102 fails.
• Claims 4–6 (bridge-leg absorption architecture): Partially relevant, not anticipatory. Combes discloses the general concept of a suspended microbridge absorption/support region with legs carrying electrical signals, but not the specific "bridge leg absorption structure + first bridge leg to electrode contact hole + second bridge leg to bridge column hole" geometry.
• Best use: § 103 background art establishing that suspended microbridges with absorptive multilayer legs and electrically conductive legs were known — not a standalone § 102 reference against any claim.

2.2 — US 2015/0021479 A1 (Cai et al., "Bolometer and preparation method thereof")

Full citation: US 2015/0021479 A1, "Bolometer and preparation method thereof," inventors Cai Jian, Wang Qian, Liu Ziyu, Hu Yang; assignee Tsinghua University; U.S. application Ser. No. 14/353,020 (national stage of PCT/CN2012/079126).
PCT filing date: 2012-07-25; priority: CN 201110319208, 2011-10-19; publication date: 2015-01-22. (Granted as US 9,258,894 B2 on 2016-02-09.)
Source: https://patents.google.com/patent/US20150021479 ; https://patents.google.com/patent/[US9258894B2](/patent/US9258894B2)/en

Brief description: A vanadium-oxide microbridge bolometer in which the thermal radiation measurement element and the readout circuit are on separate substrates, interconnected by a silicon via (TSV) filled with conductive material. The microbridge comprises a reflecting layer, supporting layer, thermosensitive element layer (vanadium oxide), interconnecting layer, and protective layer, with an electrode hole at the edge of the element. Critically for this case, the via-fill conductive material is described as "one or more of copper, tungsten, polycrystalline silicon, conductive polymer and metal-polymer composite material," and the interconnect may be made with conductive polymer.

§ 102 analysis:

• Qualifies as prior art: Yes — published 2015-01-22, which is before the 2015-08-26 effective filing date, so it is § 102(a)(1) art. (Its grant, 2016-02-09, is later, but the pre-grant publication controls.)
• Claim 1: This is the closest § 102 candidate, but I assess it as not anticipatory. It is the only cited reference that expressly names "conductive polymer" as a conductive material used to make electrical connection in a VOx bolometer. That is why the examiner cited it. However, in Cai the conductive polymer is disclosed as a silicon-via fill / interconnect material, not as an "extraction electrode … for contacting [the] vanadium oxide thermosensitive film … on a bridge surface layer," which is the specific structural role claim 1 recites. The thermosensitive film is contacted by a separate interconnecting layer 6 coupled to the electrode hole 9. A petitioner asserting § 102 would have to argue that the conductive-polymer-filled via is the "extraction electrode," but the claim's "contacting the vanadium oxide … on a bridge surface layer" element is not squarely met by a via beneath the bridge. Expect this reference to be used in a § 103 combination, not as a clean § 102 anticipation.
• Claims 7–12 (three-layer SiNx / conductive-polymer / SiNx leg stack): Does not anticipate. Cai's bridge stack is support layer / VOx / interconnect / protective layer — an ordered sequence that does not read on "upper SiNx / conductive polymer / lower SiNx," and the middle conductive layer is not specified to be a conductive polymer in the leg.
• Best use: § 103 primary reference supplying the express "conductive polymer" teaching in a VOx bolometer conductive path; background for electrode-hole/interconnect architecture.

2.3 — US 9,677,946 B1 (Magnolia Optical Technologies, "Infrared radiation detectors using carbon nanotubes-silicon vanadium oxide and/or amorphous silicon nanoparticles-CNT nanocomposites…")

Full citation: US 9,677,946 B1, "Infrared radiation detectors using carbon nanotubes-silicon vanadium oxide and or amorphous silicon nanoparticles-CNT nanocomposites and methods of constructing the same," assignee Magnolia Optical Technologies, Inc.
Priority (provisional): 2014-12-22; grant date: 2017-06-13.
Source: https://patents.google.com/patent/US9677946B1/en ; https://patentimages.storage.googleapis.com/c6/0b/b5/40623823b5fc1e/US9677946.pdf

Brief description: Microbolometer structures with a top layer of amorphous silicon or vanadium oxide, onto which a CNT/nanoparticle nanocomposite is deposited. One formulation mixes carbon nanotubes with vanadium oxide or silicon in a "conjugated polymer or epoxy" matrix applied by spin-on or spray, to tune TCR and percolation. The detector comprises a substrate, a Si-CNT (or VOx-CNT) thin film suspended over a cavity between first and second conductive contacts, connected to CMOS readout circuitry; sensitivity targets below 1 mK NEDT.

§ 102 analysis:

• Qualifies as prior art: Yes, but only as § 102(a)(2) art — this requires care. The grant date (2017-06-13) is after the 2015-08-26 effective filing date, so it is not § 102(a)(1) art. It qualifies (if at all) under § 102(a)(2) because it is a U.S. patent naming another inventor that was effectively filed before 2015-08-26 — its provisional priority date of 2014-12-22 precedes the cutoff. This qualification is contingent on the patent being entitled to the 2014-12-22 provisional benefit (the record states the benefit claim). It is not commonly owned with the '487 patent, so the § 102(b)(2)(C) exception does not remove it.
• Claim 1: Does not anticipate. The "conjugated polymer or epoxy" in Magnolia is a matrix/host for a CNT + VOx or CNT + Si nanocomposite sensing film, not a conductive polymer extraction electrode contacting a VOx thermosensitive film and carrying its signal to readout. The conductive contacts are metal interconnects. The claim-1 limitation is absent, so § 102 fails.
• Claims 7–12 (three-layer SiNx/conductive-polymer/SiNx): Does not anticipate. Magnolia's stack (nitride dielectric + CNT-nanocomposite + graphene/monolayers) is different in kind and sequence.
• Best use: § 103 art showing (a) polymer-hosted VOx/CNT films and (b) suspended, cavity-backed pixels with two conductive legs to a ROIC. Note that the examiner's citation of a later-granted U.S. patent here is consistent with § 102(a)(2)/§ 103 treatment rather than § 102(a)(1).

---

3. Family-cited references (cited in the CN counterpart CN105060237 / "Family Cites Families")

These eight appeared in the Chinese family member's citation record. I verified what I could; where a search returned no confirming detail I say so rather than guess.

Notes on the most substantive ones:

• CN101357747A — "一种非致冷红外焦平面微桥结构的制备方法" (Method for preparing an uncooled infrared focal-plane-array microbridge structure). Discloses a one-step film-growth process forming the bridge pier, bridge legs, and bridge-surface support structure integrally, with hollow bridge piers and an upper electrode connected to a lower electrode via simple lithography, plus a thermosensitive film and passivation layer. Relevant to the microbridge-leg architecture (claim 4 family) and to background; it does not disclose a conductive-polymer electrode, so it cannot anticipate claim 1. (Notably, its inventor list includes 王涛 / Wang Tao, the same-named first inventor as the '487 patent — an earlier, separate work by an overlapping inventor.) Source: https://patents.google.com/patent/CN101357747A
• CN101774530B — "一种微测辐射热计及其制备方法" (Microbolometer and preparation method). A three-layer sandwich microbridge: bottom amorphous silicon-nitride support/insulation, middle CNT–amorphous-silicon composite thermoresistive + IR-absorbing layer, top amorphous silicon-nitride passivation/stress layer. This is the closest structural analogue to claims 7–9 (upper nitride / middle layer / lower nitride), but the middle layer is a CNT–a-Si composite, not a conductive polymer, and it is the sensing layer, not the extraction electrode. Relevant to claims 7–9 only as § 103 art. Source: https://patents.google.com/patent/CN101774530B/en
• CN102280455B — "非制冷式红外焦平面阵列探测器" (Uncooled IR FPA detector). Its bridge leg is expressly a stack of [thermal-resistance layer / conductive layer / passivation layer], and the bridge surface is [support / absorption / insulation / thermosensitive / protective], with the absorption layer and leg conductive layer sharing a material of titanium nitride or nichrome (NiCr); thermosensitive material is a-Si, a-SiGe, or VOx; anchor column is tungsten + silicon oxide. Most relevant for claims 7–9's "three-layer leg" concept — but the conductive layer is TiN/NiCr, not a conductive polymer, so § 102 fails against claim 1 for lack of the characterizing limitation. Source: https://patents.google.com/patent/CN102280455B/
• CN204873819U — "焦平面阵列微桥单元桥腿复合结构" (Focal plane array microbridge unit bridge-leg composite construction). This is the applicant's own Chinese utility model, filed 2015-08-26 (same day as the '487 priority) and published 2015-12-16. It is the family member, not prior art: it published after the effective filing date, so it is not § 102(a)(1) art, and as a Chinese (non-U.S.) filing it cannot be § 102(a)(2) art. Do not treat CN204873819U as prior art against the '487 patent. Source: https://patents.google.com/patent/CN204873819U/
• CN103940518B ("Low thermal conductivity terahertz detection unit microbridge structure") and KR101910573B1 ("Infrared detector including broadband light absorber") are plausibly § 102(a)(1) art as to their pre-2015-08-26 A-publications, but I was unable to retrieve and verify their disclosure text within this session's search budget. I am flagging these as unverified rather than asserting a limitation-by-limitation § 102 mapping. CN103575407A (Peking Univ. THz radiation detector) and CN103365504A (Delta capacitive touch device) are THz-detection and touch-panel art respectively and appear only tangentially related — CN103365504A in particular is a capacitive touch device and is more likely cited for a general conductive-polymer deposition/patterning teaching than for microbridge structure.

---

4. Bottom line — § 102 anticipation assessment

Key takeaways for any validity/invalidity analysis:

1. No cited reference discloses or anticipates the core claim-1 limitation — a conductive polymer (or doped conductive polymer) film serving as the extraction electrode that contacts the VOx thermosensitive film on the bridge surface layer. Every cited reference either uses a metal conductor (Ti, TiN, NiCr) or uses a polymer in a different role (via fill in Cai; CNT-matrix in Magnolia).
2. The § 102(a)(2) nuance matters for the Magnolia reference — it is prior art only because its 2014-12-22 provisional predates the cutoff; its 2017 grant date does not make it § 102(a)(1) art.
3. The strongest § 103 combination the citations support would be Cai (express "conductive polymer" in a VOx bolometer conductive path) + CN102280455B or Combes (three-layer/leg-with-conductive-layer microbridge architecture), optionally with Magnolia or CN101774530B (polymer/nitride composite stacks). That is obviousness territory, not anticipation.
4. Do not rely on CN204873819U as prior art — it is the applicant's own, later-published utility model and is excluded from both § 102(a)(1) and (a)(2).

Confidence: High on the qualification analysis and dates for the three U.S. citations and for CN101357747A, CN101774530B, CN102280455B, and CN204873819U (grounded in the fetched patent text and Google Patents records). Lower/unverified for CN103940518B, KR101910573B1, CN103575407A, and CN103365504A — I retrieved no confirming disclosure text within the search budget and have not asserted a specific § 102 mapping for them. No reference here was located through PACER/Docket Navigator or a PTAB record (consistent with the litigation section's negative finding), so nothing in this analysis reflects an adversarial prior-art challenge; it reflects only the citations of record on the patent and its family.

I'll ground this obviousness analysis in the actual prior-art references listed on the patent page. Let me pull details on the key references.

Key finding: Cai (US 2015/0021479) expressly lists "conductive polymer" as a conductive material for bolometer interconnects. Let me pull the Chinese family references and Magnolia.

Let me check Magnolia and look for art teaching conductive-polymer electrodes in bolometers, plus the "Electrode material" similar document.

Obviousness Analysis — U.S. Patent No. 9,897,487 B2

0. Scope, date, and framework

This analysis is under 35 U.S.C. § 103 as amended by the AIA (the application was filed 2015-12-09, after 2013-03-16). The effective filing date is 2015-08-26 (CN 201510532320.6 priority, perfected within the 12-month Convention window), so only art predating 2015-08-26 is available.

I build on the two earlier sections rather than repeating them. One cross-reference note: the earlier Patent Summary quotes claim 1 as reciting "a conductive polymer film or a doped conductive polymer," and flags that the word "film" is missing from the doped-polymer alternative. That literal reading is preserved below; it does not change the § 103 outcome but does create a § 112 boundary ambiguity worth noting.

Minor flag per operating rules: the system date (2026-09-29) and the task's stated date (April 26, 2026) differ. Either way, the maintenance-fee lapse events already recorded (reminder 2025-10-13; lapse 2026-03-30, effective 2026-02-20) have occurred, so the patent is in a lapsed state during this analysis. That is relevant only to the secondary-considerations discussion (§ 8).

---

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

A PHOSITA here is a MEMS/bolometer process engineer (or device engineer) with a graduate degree or 3–5 years' experience in uncooled IR/THz focal-plane-array fabrication, familiar with: VOx microbridge pixel stacks; microbolometer leg thermal-conductance design (G, thermal time constant, fill factor); electrodes for VOx pixels (NiCr, Ti); CMOS/ROIC-compatible back-end processing; and materials selection for thin-film MEMS. This is the art reflected by the patent's own CPC classes (G01J 5/02, 5/0225, 5/024, 5/046) and by the page's "Prior art keywords" (bridge, leg, micro, focal plane, film).

---

2. The prior art on this page, validated

Because the instruction is to use the Prior Art section of the page, here is that section, sorted by whether each item is actually available as prior art against 2015-08-26:

2.1 Examiner-cited references ("Patent Citations" — 3)

2.2 "Family Cites Families" (8) — upstream art read onto the family

2.3 "Families Citing This Family" (2) — not prior art

FR 3064060 B1 (CEA, priority 2017-03-15) and CN 108844637 B (Shandong Univ., priority 2018-07-25) are later documents that cite this family. They are downstream, not prior art, and are excluded.

---

3. Claim 1 mapped against the art

Claim 1 is the sole independent claim and carries no structural limitation beyond "in the micro bridge unit." It is broad and the art maps onto it tightly:

Result: every element of claim 1 is disclosed or explicitly suggested by Cai alone. Cai names "conductive polymer" as a conductive material in the exact electrical pathway from the VOx film to the ROIC. If one accepts that extension, claim 1 is obvious over Cai alone (KSR rationales: substitution of one known conductor for another for its known properties; and Cai's own list supplies the suggestion).

---

4. Grounds of rejection

Ground A — Claims 1, 2, 3, 4, 5, 6: Cai in view of Combes (and Magnolia as secondary)

• Claims 1, 2, 3. Cai discloses the full structure and expressly lists conductive polymer (and metal-polymer composite) as the conductive material. Doping conductive polymers to raise conductivity is the definition of the class: polyaniline doped with a protonic acid and polythiophene doped with a Lewis acid (e.g., FeCl₃) are textbook. Magnolia expressly contemplates a "conjugated polymer" host with VOx in a bolometer, and further confirms that the polymer/CNT route was within routine bolometer practice. Claim 3's "Lewis acid or protonic acid" and the spec's FeCl₃/sulfamic-acid species (dependent-claim level) add nothing beyond known dopant chemistry — MPEP 2144.03 (substitution of known materials with predictable results).
• Claims 4, 5, 6 (the "combined leg" architecture). Claim 4's diagram — absorption structure near the bridge surface layer, first leg to the electrode contact hole, second leg to the bridge column hole — is the classic suspended-pixel topology: Combes discloses the suspended IR-absorbing portion on legs with the legs providing both mechanical support and electrical connection, and Cai supplies the electrode hole and the column/via nomenclature. CN 103 940 518 B supplies the pier/leg/face triad plus an absorber layer in the face stack. The combination yields claim 4's arrangement as a predictable assembly of known microbridge parts (KSR: familiar elements according to known methods, yielding predictable results).

Ground B — Claims 7, 8, 9 (three-layer stack: upper SiNx / conductive polymer / lower SiNx): Combes and/or CN 101 774 530 B, in view of Cai

This is the single strongest ground, because the claimed tri-layer is expressly in the art:

• Combes: "the supporting frame could comprise a dielectric–metal–dielectric stack. Locating the absorber layer in the centre," with "DIEL1 and DIEL2 … may comprise silicon nitride." That is SiNx / conductor / SiNx with the conductor in the middle — the literal claim-7 architecture, differing only in calling the centre layer a "metal" rather than a conductive polymer, which Ground A's substitution rationale resolves.
• CN 101 774 530 B: a three-layer sandwich with a-SiNx on both faces of the middle layer, the top SiNx acting as passivation and stress-control and the middle film's stress being opposite that of the SiNx layers. That reference supplies an independent, explicit teaching that a symmetric SiNx/sensitive-layer/SiNx stack is used in a bolometer microbridge — and the '487's own background expressly identifies residual stress and unit warping as the problem ("the NiCr film has the large elasticity modulus, which leads to the large residual stress and accordingly the warping of the array unit"). CN 101 774 530 B's stress-compensation teaching therefore supplies a motivation to adopt the sandwich even apart from Combes.
• Cai already stacks SiNx (supporting layer 4) / VOx / interconnect 6 / SiNx (protective layer 7) — i.e., a conductor between two SiNx layers in a bolometer bridge.

Motivation to combine: all three address the same problem (protect/encapsulate the sensitive film, control stress, keep the conductive trace centered for symmetric stress) in the same device class.

Ground C — Claims 10, 11, 12 (50–100 / 50–150 / 100–200 nm): routine optimization

Thickness ranges are broad and overlap the working examples of the art. Combes' dielectric and absorber films are on the 0.005–0.1 µm (5–100 nm) scale; Cai's layers are CVD/PVD thin films whose thicknesses are design choices tied to sheet resistance and thermal mass. A PHOSITA optimizing (a) SiNx thickness for stress balance and (b) polymer thickness for sheet resistance does no more than routine optimization of a result-effective variable (MPEP 2144.04; In re Aller). No criticality or unexpected result is asserted for these bounds; the '487's own examples (h₁/h₂/h₃ = 80/80/100, 80/100/100, 60/100/100, 50/100/100 nm) sit comfortably inside the claims and vary with the polymer used, which is the opposite of a critical, narrowly-selected range.

Ground D — Claims 13, 14 (absorption structure at 10–20% of the unit's length and width): design choice

Combes expressly frames the trade-off between minimizing thermal conductance and preserving IR absorption/fill factor, and CN 103 940 518 B does the same in the plane (tilting legs to raise absorption area and duty cycle). Sizing a paddle inside the pixel to a stated fraction of the pixel is the kind of parametric design choice a PHOSITA makes, with predictable results (more area = more absorption, more length = more thermal isolation).

Ground E — Claims 15, 16, 17, 18 (leg dimensions): the ranges overlap the applicant's own admitted prior art

This is the cleanest § 103 point in the patent, because the specification admits the prior-art value:

"The width of the conventional bridge leg reaches 0.5 μm, and the manufacture cost … will greatly increase if the width … is further decreased."

Claim 15 recites a first-leg width of 0.5–1.0 µm. The lower endpoint is the admitted conventional width. Where a claimed range overlaps (or touches) the prior art and applicant shows no criticality or unexpected result, the claim is prima facie obvious (In re Wertheim; Titanium Metals v. Banner; MPEP 2144.05). The accompanying lengths (3.0–5.0 µm) and the second-leg width (0.8–2.0 µm) are likewise the routine output of the known "long, narrow and thin suspension" trade-off that Combes describes. CN 103 940 518 B is corroborating art that leg lengthening/shaping to cut G was an active, known design lever.

Ground F — Claims 19, 20 (upper SiNx by RF sputtering or CVD at ≤ 200 °C): inherent consequence of the polymer choice plus routine optimization

Once the middle layer is a conductive polymer, a low-temperature cap layer is forced: polymers degrade if the overlying SiNx is deposited at conventional a-SiNx temperatures. CN 102426060 B (the VOx–fullerene terahertz/IR microbolometer, from the same UESTC line of art) deposits SiNx at 300 °C — precisely because it has no polymer to protect. The ≤ 200 °C limitation is therefore the known way to make the known structure work (KSR: "a known technique used to improve a similar device in the same way"), and the specific 120–180 °C values in the '487's examples are routine optimization. RF sputtering and CVD are the two standard low-temperature SiNx tools named in the art.

---

5. Why a PHOSITA would have combined these references (the KSR rationales, itemized)

1. Same field of endeavor / reasonably pertinent to the problem. Every reference is a suspended-microbridge thermal IR/THz detector with legs. In re ICON Health; In re Bigio. Combes, although it detects via resonator frequency shift rather than resistance, is squarely in the microbridge-pixel structural art and was in fact cited by the examiner on the face of the '487, which undercuts any non-analogous-art argument.
2. Same recognized problem. Thermal conductance through the legs sets responsivity: Combes states "In order to maximise the temperature rise at the pixel it is necessary to minimise thermal conductance"; CN 103 940 518 B's stated object is a "low thermal conductivity" microbridge; the '487 background says the same. Substituting a lower-thermal-conductivity conductor in the leg attacks the same problem the art already identified.
3. The references themselves supply the suggestion. Cai expressly lists conductive polymer among bolometer conductive materials. That is not hindsight; it is the reference's own teaching. Any alternative reading would require ignoring the express list.
4. Predictable result / simple material substitution (MPEP 2144.03). Choosing a conductor with adequate conductivity for a low-current, high-impedance VOx pixel (sheet resistance ~100 kΩ/□ per the '487 examples) while lowering thermal conductivity is a predictable engineering trade — and the applicant's own specification states the expectation ("has a lower thermal conductivity than a conventional electrode material … by about one order of magnitude"). An applicant's own asserted property expectation is usable against it (MPEP 716.01(c)).
5. Design incentive / market forces (KSR). The '487 background itself enumerates the drivers: NiCr's large elastic modulus → residual stress/warping; NiCr's difficulty with dry etch; Ti's oxidation. That paragraph is effectively an admission of the motivation to find an alternative electrode material — and it is the applicant's own admission (MPEP 2129).
6. Obvious to try / finite number of predictable solutions. The class of bolometer conductors was small and known (NiCr, Ti, and per Cai, copper, tungsten, polysilicon, conductive polymer, metal-polymer composites). Selecting a member of that class for its known low thermal conductivity is the paradigm of an obvious-to-try selection.
7. Enablement + once-created modification. The references are all enabled, and the modification (swap the electrode material; retain the SiNx cladding; add a paddle in the leg) is within the mechanical/process skill described in the references themselves (spin-coating, CVD/PVD, lithography).

---

6. Rebuttal considerations and honest weaknesses in the prima facie case

I flag these so the analysis is not overstated:

• Claim 1's "extraction electrode" location. Cai's express "conductive polymer" is stated for the silicon via 8, which sits in the substrate/pier region, whereas claim 1 requires the electrode to contact the VOx film on a bridge surface layer. A patent owner would argue Cai discloses a polymer via, not a polymer leg/bridge electrode. The answer is that Cai's electrode hole 9 + interconnecting layer 6 + via 8 form one conductive path from the VOx to the ROIC, so the extension to the bridge-side interconnect is a design choice — but this is the point of attack, and it is why Ground A is stronger as a § 103 combination with Combes/Magnolia than as a clean single-reference case.
• Combes is a different detection modality. Its pixel detects a resonant-frequency shift, and it expressly says it "is not reliant on the measurement of the relative resistance … of a layer of temperature sensitive material." A patent owner may argue Combes is non-analogous to a resistance microbolometer. Counter: the structural teaching (dielectric–metal–dielectric stack, absorber doing double duty as electrical connection, legs as thermal isolators with conductors) is what is relied on, and KSR/In re Bigio ask only whether it is reasonably pertinent to the problem the inventor faced — here, microbridge leg/electrode structure.
• No reference expressly says "use a conductive polymer as the bolometer's leg electrode." The case is built on Cai's express list plus the '487's own admitted problem statement. A court or the PTAB might want a stronger articulation of the rationale; the KSR "known-technique-to-improve-a-similar-device" rationale is the right hook.
• Nothing in the cited art teaches the specific discrete "bridge leg absorption structure" paddle interposed in the leg as a distinct rectangular element with a first and second leg on either side (claim 4 et seq.). CN 103 940 518 B gets close in purpose (legs shaped to raise absorption area and fill factor) but not in form. Claims 4–6 therefore have the best non-obviousness argument of the dependent set, though the argument is about form, not about any demonstrated technical effect.
• CN 204873819U is not prior art (same applicant, published 2015-12-16), and FR 3064060 B1 / CN 108844637 B are not prior art (they cite this family). Any analytics tool that dumps "citations" without dates will wrongly sweep these in — the same failure mode the earlier Litigation section flagged for keyword hits.

---

7. Claim-by-claim vulnerability summary

---

8. Secondary considerations (why they do not rescue the claims)

• The NETD data are not comparative against the closest art in the same structure. Embodiments 1–4 report 46 / 37 / 42 / 51 mK at 128×128, 50×50 µm. Combes, on the face of this very page, reports "of the order of 60 mK NETD … with a pixel pitch of approximately 50 um and F1 optics" for a commercial VOx bolometer. Even accepting that as a baseline, the improvement is modest, and the '487 gives no NiCr control processed into the identical stack, so no causal link between the polymer electrode and NETD is established. Worse for the patent owner, the polymer choice does not correlate monotonically with the result (undoped polyacetylene = 46 mK beats FeCl₃-doped polythiophene = 42 mK and polypyrrole = 51 mK), which undercuts any "critical material/range" argument on which the ranges in claims 10–20 would depend.
• The test conditions are not normalized (25 °C, 6 µs/150 µA pulse, no stated F-number or frame rate), so a direct comparison to Combes' F1/30 Hz figures is imperfect in either direction. I flag this rather than letting the 60 mK figure do more work than it can.
• No unexpected-result showing for the thickness ranges, the 10–20% paddle sizing, or the 0.5–1.0 µm leg width.
• Commercial-significance signal is weak/negative. Per the page's Legal Events, the patent lapsed for nonpayment of maintenance fees (reminder 2025-10-13; lapse 2026-03-30; effective 2026-02-20), while the CN counterpart CN 105060237 B is listed Active. Non-payment is not itself a § 103 argument, but it is inconsistent with the kind of commercial success or long-felt-need narrative that would rebut a prima facie case — and it is consistent with the earlier Litigation section's finding of no enforcement activity for the '487.

---

9. Bottom line

Claim 1 is very likely obvious — US 2015/0021479 A1 (Cai) expressly names conductive polymer as a conductive material in a vanadium-oxide microbridge bolometer's signal path to the readout circuit, which is the entire inventive substance of claim 1. Claims 2 and 3 fall with standard conductive-polymer/dopant chemistry.

Claims 7–9 are the next most vulnerable, because Combes discloses a literal "dielectric–metal–dielectric stack" with the conductor in the centre and SiN dielectrics, and CN 101 774 530 B discloses a SiNx/sensitive-layer/SiNx sandwich whose stress-compensation function answers the exact warping problem the '487 admits — while Cai already stacks SiNx/VOx/interconnect/SiNx.

Claims 15–18 are vulnerable on the ranges themselves: the 0.5 µm lower bound of the first-leg width is the applicant's own admitted conventional dimension, and no criticality is shown.

The most defensible claims are 4–6 (and, weakly, 13–14), and only because no reference is found showing the discrete paddle-in-the-leg geometry with a leg on each side of it — a form argument, not a technical-effect argument, and one that CN 103 940 518 B's absorption-area-driven leg design partially undercuts.

Net assessment: a well-pleaded § 103 challenge would likely take out claims 1–3, 7–12, and 15–20 using Cai + Combes (with CN 101 774 530 B and CN 103 940 518 B), leaving claims 4–6 (and 13–14) as the only claims with a realistic survival argument.

Primary sources relied on: the Google Patents record for US 9,897,487 B2 (https://patents.google.com/patent/US9897487/en); US 2007/0108383 A1 (https://patents.google.com/patent/US20070108383); US 2015/0021479 A1 / US 9,258,894 B2 (https://patents.google.com/patent/US20150021479 and https://patents.google.com/patent/[US9258894B2](/patent/US9258894B2)/en); US 9,677,946 B1 (https://patents.google.com/patent/[US9677946B1](/patent/US9677946B1)/en); CN 101 774 530 B (https://patents.google.com/patent/CN101774530B/en); CN 103 940 518 B/A (https://patents.google.com/patent/CN103940518B; original text at http://www.xjishu.com/zhuanli/52/201410182730.html).