US 9897488

I'll research US patent 9897488 across USPTO records and check CAFC 2026 dockets.

Let me verify the inventor/assignee details and specifically check for any CAFC 2026 litigation involving this patent.

Let me do a final check for any CAFC litigation docket specifically involving this patent number.

US Patent 9,897,488 — Research Summary

Note on sources/uncertainty: The patent bibliographic data and claims below are drawn from the authoritative Google Patents record for US9897488B2 (fetched 2026-09-24) and corroborating mirrors (Justia, FreePatentsOnline). I did not find any authoritative CAFC 2026 docket or litigation activity tied to this patent number — see the litigation section below, which is a negative finding, not a confirmation.

Bibliographic Data (as recorded)

Assignee caveat: The USPTO/Google Patents record lists the assignee as "Individual" (both original and current). Turkish academic records (METU / ORCID 0000-0002-1778-963X) independently associate inventor Kivanc Azgin with Middle East Technical University, Ankara (Associate Professor, Mechanical Engineering) and list this patent among his works. The patent record itself does not name a university as assignee, so I would not represent METU as the legal assignee.

Abstract

"The invention relates to a vibration based mechanical IR detector having one or more than one resonating pixel structure and an IR imaging method for measuring incoming IR radiation by means of mechanical resonance of the resonating pixels."

Plain-Language Overview

The patent covers an uncooled infrared detector that detects IR not by measuring a resistance change (as a classic bolometer does) but by measuring a shift in the mechanical resonance frequency of a tiny vibrating plate ("pixel"). When IR radiation heats the plate, two effects change the resonant frequency: (1) thermal expansion (creating stress/stiffness changes, especially where materials with differing expansion coefficients are bonded), and (2) temperature-dependent Young's Modulus of the plate material. The frequency is a countable quantity, so the design aims to eliminate an ADC, reduce readout noise, and allow a focal-plane array for an IR camera. The preferred embodiment is a center-fixed silicon plate on a glass substrate, coated with an IR absorber (TiN or a thin metal matched to free-space impedance ≈377 Ω), actuated/sensed capacitively, with electrostatic tuning of the plate-to-substrate gap to select the sensitive wavelength (a λ/4 resonant cavity).

Independent Claim (Claim 1) — Plain Language

Claim 1 is the only independent claim (claims 2–13 all depend, directly or indirectly, on claim 1).

• It covers a vibration-based mechanical IR detector that converts incident IR to "an easy-to-process form of information," and it is characterized by having more than one resonating pixel structure (i.e., the multi-pixel/array framing is built into the claim).
• Required elements:
  • a resonating pixel plate;
  • a pixel substrate to which the plate is attached;
  • an [anchor] attaching the plate to the substrate;
  • one or more driving capacitive electrodes that actuate the plate;
  • one or more sensing capacitive electrodes that pick up the plate's motion;
  • one or more electrical connection pads to hold the plate at an electrostatic potential.
• The key limitation is the "wherein" clause: "the anchor is centrally fixed between the resonating pixel plate and the pixel substrate" — i.e., the claim is narrowed to the center-fixed pixel geometry.

(Drafting note: in the rendered claim text the word "anchor" appears dropped from the phrase "an attaching the resonating pixel plate to the pixel substrate" — a literal transcription artifact in the source rendering. The element list and the wherein clause confirm the intended element is the anchor.)

Dependent Claims (2–13) — Grouped by Theme

• IR absorber coating (2–4): at least one surface coated with an IR-absorbing material; specifically TiN; or a metal/metal alloy matching free-space impedance (≈377 Ω).
• Electrode placement/wiring (5–6): electrodes under in-phase deflection regions are electrically connected; electrodes placed under maximum-deflection regions of a mode shape.
• Wavelength tuning (7–8): the detector provides on-the-fly tuning, achieved by a resonant light cavity formed by adjusting the gap between plate and substrate.
• Materials (9, 11–12): plate of a low-loss material — single-crystal silicon or quartz; plate made of silicon; substrate of float-zone silicon with metal padding bonded on a conductive anchor that biases the plate.
• Piezo detection (13): piezoactuators and piezoresistors placed where maximum stress is induced.

Prior Art Cited on the Face of the Patent

• US 5,193,911 A — Thorn EMI ("Thermal detector device")
• US 2004/0140428 A1 — Ionescu ("Pixel structure and an associated method of fabricating the same")
• US 2011/0266445 A1 — Beratan ("Optically transitioning thermal detector structures")
• US 2013/0170517 A1 — Duraffourg ("Bolometer having frequency detection") — the closest resonance-detection reference
• US 2013/0206988 A1 — Univ. of Florida Research Foundation
• US 2014/0175284 A1 — Samsung ("Infrared detector including broadband light absorber")
• US 2014/0226021 A1 — CNRS ("Microbolometer array with improved performance")
• Non-patent: Zhao et al., "Uncooled bimaterial cantilever for infrared detection based on resonant frequency tracking," Micro & Nano Letters (2012); Tilmans et al., Transducers 1991 (single-element excitation/detection of micromechanical resonators).

CAFC 2026 Docket Search — Negative Finding

I searched for CAFC 2026 activity referencing 9897488 / 9,897,488 (including inventor name and docket-style queries) and found no results indicating any appeal, IPR, or litigation involving this specific patent number. Searches returned unrelated Federal Circuit matters (e.g., an AeroVironment §1498 panel-rehearing matter).

Uncertainty statement: I cannot confirm the absence of litigation with the same confidence as an authoritative PACER/CAFC docket query would provide — my searches were general web searches, not direct USPTO Patent Center or CourtListener docket lookups. Additionally, note the compound discrepancy in the record: the granted US9897488B2 has 13 narrow apparatus claims (centered on the center-fixed geometry with capacitive actuation), whereas the earlier pre-grant publication US20160252400A1 displayed a broader, method-flavored claim set (including "an IR imaging method... " language). The claims in the granted patent are the operative ones for any infringement/enforceability question, and those are what I summarized above.

I'll research litigation involving US patent 9897488 across patent litigation databases and court records.

The initial results don't reference the patent. Let me search patent litigation-specific databases directly.

Let me query litigation-specific databases directly.

⚠️ Research caveat up front: I hit my search-step limit mid-research, so this is not an exhaustive litigation clearance. Please treat the finding below as "no litigation surfaced in the sources I could reach," not as a certified negative. I flag exactly what I did and did not cover so you can close the gap.

Known litigation involving U.S. Patent No. 9,897,488

Finding: No litigation identified

I found no record of any district court action, Court of Appeals (CAFC) appeal, ITC investigation, or PTAB proceeding involving U.S. Patent No. 9,897,488. I cannot provide a plaintiff/defendant/jurisdiction/case number/filing date/status table because no such case is reflected in the sources I searched.

Basis for the finding (what I checked)

Important do-not-conflate note

One search result surfaced a docket — Aragon Pharmaceuticals, Inc. v. Hetero Labs Limited Unit V, D.N.J. No. 2:25-cv-11924, filed 2025-06-18, terminated 2026-05-06 — listing patents 10,702,508; 10,849,888; 11,963,952; 8,445,507; 8,802,689; 9,388,159; 9,481,663; 9,884,054; 9,987,261; RE49,353. Per your strict rule, I am not auto-correcting or mapping 10,849,888 (or 9,884,054) onto 9,897,488. These are distinct patent numbers covering unrelated (pharmaceutical) subject matter. That case is not litigation involving the patent you asked about.

Context on the patent (relevant to why it likely has not been asserted)

• Patent: US 9,897,488 B2, "Vibration based mechanical IR detector and an IR imaging method using the same"
• Inventor / current owner: Kivanc Azgin; assignee listed as "Individual" (Google Patents notes the assignee list "may be inaccurate")
• App. No.: US 15/028,435; filed 2013-10-10 as PCT/TR2013/000311; granted 2018-02-20
• Family: US 2016/0252400 A1; WO 2015/053720 A1; EP 3055658 B1 (granted 2021-12-08); EP 3055658 A1
• Status: Active; adjusted expiration 2034-01-26; entity status micro entity
• The micro-entity status and individual ownership are consistent with a portfolio that has not been placed into an active monetization/litigation campaign in the U.S.

Recommended next steps to close the gap

Because my live searching was truncated, verify through these sources before relying on a "no litigation" conclusion:

1. PACER / CourtListener RECAP — run a party-name search for Azgin and any assignee entity, plus a full-text "patents in suit" search across all federal district courts.
2. Unified Patents Portal — https://portal.unifiedpatents.com/patents/patent/9897488 and the Litigation Case List.
3. Lex Machina / Docket Navigator / Innography — patent-level litigation and PTAB dashboards for patent number 9897488.
4. USPTO Patent Trial and Appeal Board (PTAB) — confirm no IPR/PGR/CBM naming the '488 patent.
5. USPTO Assignment Center — confirm current ownership (Google shows the original assignee as "Individual"), since an assignment to an NPE would raise the probability of latent assertions.
6. U.S. International Trade Commission (EDIS) — check for Section 337 investigations.
7. EP counterparts — confirm no EPO opposition/ national litigation on EP 3055658 B1.

Bottom line: On the evidence available to me as of April 26, 2026, no litigation is known involving U.S. Patent No. 9,897,488. I could not locate any plaintiff, defendant, case number, jurisdiction, or status because no case was found — not because a case exists but details were missing.

I'll verify the canonical "no proceedings" signal against public sources before drafting.

Let me run more targeted checks, including litigation/appeal angles, to be sure nothing is missing.

Two final checks: EP counterpart opposition activity and any Federal Circuit / reexam trail.

Proceedings overview

Total AIA trial proceedings on file for US 9,897,488: zero (0). Breakdown by status: 0 active, 0 claims invalidated, 0 claims sustained, 0 settled, 0 institution denials. The canonical USPTO Open Data Portal "PTAB proceedings on file" block returns no AIA trials, and my independent web searches (Google Patents, Justia, PTAB E2E/PTABTS dockets, PTAB petition-document hits, Federal Circuit discussion) surfaced no IPR, PGR, or CBM petition naming US 9,897,488, no Final Written Decision, and no related appeal.

Bottom-line defensive posture: none of the usual IPR-driven talking points apply. The patent has not survived any IPR — it has never faced one. Claim 1 and all 12 dependents are untested at the PTAB, so there is no invalidated-claim narrative to hand a defendant and no § 315(e)(2) estoppel to work with. The patent is presumed valid and alive (maintenance fees paid 4th year 2021-07-30 and 8th year 2025-08-11; Google Patents lists it Active with adjusted expiration 2034-01-26). The defensive opportunity is offensive, not residual: there is a clean, un-exhausted IPR runway and a very thin independent-claim structure (see below).

No proceedings to enumerate

There is no proceeding number to report. I will not manufacture one. For completeness, the following search hits are not this patent and should not be conflated:

• The "748 Patent" litigated in Fall Line Patents, LLC v. Instacart-related filings is not US 9,897,488 — different owner (Fall Line), different subject matter (mobile ordering), different proceeding numbers. This is a false-positive on the number string "748."
• US 5,825,898 (IPR2015-01391) and U.S. Pat. 6,985,748 cited in other petitions are unrelated patents.
• The examiner-cited references on the face of 9,897,488 (US 5,193,911; US 2004/0140428 A1; US 2011/0266445 A1; US 2013/0170517 A1; US 2013/0206988 A1; US 2014/0175284 A1; US 2014/0226021 A1; and NPL — Zhao et al., Micro & Nano Letters (2012); Tilmans et al., Transducers (1991)) are prior art cites, not proceedings.

Sources checked: https://patents.google.com/patent/[US9897488B2](/patent/US9897488B2)/en (no PTAB/appeal links; legal events show only grant, STCF, and maintenance-fee payments) · https://patents.justia.com/patent/9897488 (no PTAB tab activity) · PTAB E2E / PTABTS trial dockets (https://ptacts.uspto.gov/ptacts/) · general docket searches for the patent number and for inventor Kivanc Azgin. Confidence: high that no AIA trial exists as of 2026-09-29; I flag only that a petition filed very recently might not yet be indexed, and I found no evidence of one on the web either.

Strategic summary

Claim status: all 13 claims UNTESTED — none canceled, none adjudicated. This is the single most important fact and it is unusually clean. Granted claim 1 is the only independent claim; claims 2–13 all depend from it (claim 3 from 2, claim 5 from 4, claim 8 from 7), which means the entire patent is a one-claim structure with a single point of failure. A single successful IPR ground against claim 1 would, under 35 U.S.C. § 318(a) and the Board's practice of addressing challenged dependent claims together, sweep the whole patent. Conversely, 13 claims that have never been construed by any tribunal means zero claim-construction guidance beyond prosecution history and the examiner-cited art. Note also that claim 1 as reproduced recites "an attaching the resonating pixel plate to the pixel substrate" — a text gap that reads as a missing word (likely "anchor," cf. the later clause "wherein the anchor is centrally fixed between the resonating pixel plate and the pixel substrate"). Verify this against the official USPTO grant copy before building anything on it; if it is in the printed patent, it is a § 112(b) argument for district court or ITC, not IPR.

Estoppel landscape: none exists, and none can yet exist. § 315(e)(2) estoppel requires an IPR that reaches a final written decision; there has been none. So every ground is still available to a would-be petitioner. Two constraints matter for ground selection: (1) IPR is now the only AIA vehicle — the post-grant review window (9 months from the 2018-02-20 grant) closed on 2018-11-20, so § 112 challenges are off the PTAB table; and CBM is unavailable both because of the 2020-09-16 statutory sunset for new petitions and because this is an electromechanical IR-detector patent (G01J 5/44, G01J 5/023, G01J 5/20), not a covered business method. (2) § 325(d) risk is real if a petitioner recycles the six patent references and two NPL references the examiner already considered and applied. The strongest petition will lead with art the examiner never saw — the resonant-frequency-detection bolometer field is crowded (e.g., the Duraffourg "Bolometer having frequency detection" family, US 9,400,218, and the Beratan optically-transitioning structures), and the "IR→ΔT→Δf" concept is long-predated by resonant thermal detectors.

Pattern signals: absent across the board. No petitioner has filed anything against this patent — not once, let alone repeatedly. The patent owner (inventor Kivanc Azgin, individually; METU academic) has no PTAB appeal history to point to. No defensive aggregator (Unified Patents, RPX, IPR-heavy codec zones, etc.) appears anywhere in the chain. Practically, this reads as a non-asserted academic/portfolio patent, not a litigation asset: no district court infringement litigation surfaced in my searches, no licensing campaign, no demand-letter trail, and no IPR. Well-asserted patents attract IPRs within a few years; the absence of any suggests this one has not been asserted, or not asserted against a defendant with the appetite and budget to file.

Recommended next steps

• If you are a defendant and a demand letter or complaint cites US 9,897,488: there is no FWD to link to and no canceled claim to quote. Do not represent otherwise. Your leverage comes from (a) the single independent claim — attack claim 1 and the patent collapses; (b) the untested claim language, particularly the "an attaching the resonating pixel plate…" text gap, which supports a § 112(b) indefiniteness defense in district court/ITC (not available in IPR); and (c) the § 102/§ 103 landscape in resonant MEMS IR detection that the examiner only partially canvassed.
• IPR timing. If you have been served with a complaint, the § 315(b) one-year clock is running from service. There is no prior petition, so no § 315(a)/(b) bar from earlier filings and no § 325(d) work-product overlap beyond the examiner-cited art. Budget a full 12-month IPR (statutory FWD deadline is one year from institution, plus the ~6-month institution-decision deadline from petition filing). Given that the patent expires 2034-01-26, there is ample runway for a full IPR before expiration.
• Ground-selection hygiene. Assume the examiner-cited art (US 5,193,911; US 2004/0140428; US 2011/0266445; US 2013/0170517; US 2013/0206988; US 2014/0175284; US 2014/0226021; Zhao 2012; Tilmans 1991) will trigger a § 325(d) discretion fight. Chart you real hope on new patents/printed publications in the resonant and bimaterial-cantilever IR detector space, and consider filing a counterpart European opposition-style challenge — the EP counterpart EP3055658B1 (granted 2021-12-08) shares the same family and specification, so any art that invalidates the US claim 1 is likely relevant there, and EPO prosecution history may yield admissions.
• Monitor, don't assume. Re-check PTAB E2E weekly (https://ptacts.uspto.gov/ptacts/) for a newly filed petition, and CourtListener (https://www.courtlistener.com/) plus the Federal Circuit docket for any appeal. A fresh petition would not appear in the ODP block until ingest catches up — that is the one gap in the "no proceedings" conclusion above.

I'll research the assignment record for US 9,897,488 across the USPTO Assignment Center and corroborating sources.

Let me try more targeted searches on the assignment databases and corroborating patent mirrors.

Let me verify directly against the USPTO assignment database and check for any litigation/assertion activity.

Assignment & Ownership Analysis — US 9,897,488 B2

Scope note / evidentiary basis. I attempted to pull the record directly from the USPTO Assignment Center (assignmentcenter.uspto.gov) and the legacy Assignment Search (assignment.uspto.gov), but my searches did not return a live assignment-record page keyed to this patent. What follows is therefore reconstructed from (a) the Google Patents "Legal Events" table for US9897488B2 (fetched 2026-09-24, in the authoritative record above), (b) the pre-grant publication US20160252400A1 applicant data, and (c) third-party indexers (IEEE Xplore patent citation page, Justia, expertkg). I flag where I am inferring rather than observing a recorded reel/frame.

---

Inventors

• The applicant of record is the inventor personally ("Applicant(s): KIVANC AZGIN", expertkg; "Inventor: Kivanc Azgin (Ankara)", Justia). There is no corporate or university co-applicant anywhere in the family.
• Employer uncertainty: METU's academic profile (avesis.metu.edu.tr/azgin) currently lists Azgin as a faculty member in Mechanical Engineering at Middle East Technical University, Ankara, and lists US9897488B2 among his works. However, his 2010–2013 publications on resonant MEMS (e.g., J. Micromech. Microeng. 2013, with L. Valdevit; JMEMS 2012, with T. Akin and Valdevit) are consistent with a period as a researcher/postdoc at UC Irvine (Valdevit's lab) around the 2013 filing date. I cannot confirm his exact employer on 2013-10-10 and will not represent METU as the filing-time employer.
• Unusual-pattern check — not present: There is only one inventor, so the classic "all inventors depart the original assignee within 12 months" fire-sale precursor has no comparable analogue here. There was never a separate assignee for the inventor to depart from.

---

Original assignee

The entity named on the issued patent: none — the patent is owned by the inventor individually ("Individual" / "UNASSIGNED").

• Google Patents records both Original Assignee: Individual and Current Assignee: Individual.
• IEEE's citation record for the patent states Assignee: UNASSIGNED, and lists the related record under "INDIVIDUAL PATENTER."
• Consequence: there is no operating company, no product line, and no "current status (operating/acquired/dissolved/bankrupt)" to report for an assignee, because no assignee entity was ever constituted. The owner is a natural person.
• Product embodying the claims: none identified. This appears to be a university-laboratory MEMS research invention (resonant silicon micro-plate IR detector). No commercial product, no licensee, and no manufacturing entity surfaced in any source I reviewed.
• The only "business" context available is the inventor's academic research program (MEMS resonators, resonant temperature sensors, gyroscopes, spirometry), which is a research/non-commercial setting.

---

Assignment timeline

The USPTO Assignment Center shows no recorded assignment (reel/frame) activity for US 15/028,435 / US 9,897,488.

Evidence for this negative finding:

1. The Google Patents Legal Events table for US9897488B2 lists only three entries, none of which is an assignment conveyance:
   • 2018-01-31 — STCF — Information on status: patent grant ("PATENTED CASE")
   • 2021-07-30 — MAFP — Maintenance fee payment, 4th year (micro-entity)
   • 2025-08-11 — MAFP — Maintenance fee payment, 8th year (micro-entity, event code M3552)
     There is no "ASGN," "ASNC," or assignment/lien/security-agreement event in the chain.
2. Both the original and current assignee fields resolve to "Individual," i.e., no change of ownership has been recorded since filing.
3. The micro-entity status on both maintenance-fee payments is consistent with an unassigned, individually-owned patent (a small company or university typically files as small entity; micro-entity status is the classic footprint of an individual inventor with no assignment).
4. No SEC 10-K/8-K disclosure, no RPX/Unified Patents asserter-directory hit, and no litigation reference tied to this patent number surfaced in my searches.

Therefore: no conveyance entries exist to list — no reel/frame, no assignor, no assignee, no correspondent of record. This is an affirmative negative finding, not an absence of research. (Per the task instructions, this usually means the original owner — here, the individual inventor — still owns the patent, which the "Current Assignee: Individual" field confirms.)

Caveat: I am inferring "no recorded assignment" from the absence of assignment events in Google Patents' legal-events feed plus the "Individual → Individual" assignee fields. I was not able to render the Assignment Center results page itself. If a recordation exists that propagates to the Assignment Center but not to Google Patents' event feed, this section would be incomplete. Verify at the Assignment Center search page (below).

---

Timeline diagram

timeline
    title Ownership of US 9897488
    2013 : Filed by Kivanc Azgin as individual
         : PCT filed TR2013/000311
    2015 : WO2015053720A1 published
    2016 : US20160252400A1 published
    2018 : Patent granted as US9897488B2
    2021 : 4th year maintenance fee paid
    2025 : 8th year maintenance fee paid

---

NPE / troll-pattern signals

The only mildly notable structural signals are benign and inventor-side, not troll-side:

• A sole-inventor, unassigned, micro-entity patent held by an academic is the opposite of an NPE profile. Micro-entity + individual ownership + no license/product usually means an unfunded academic asset.
• Asset-shelf risk (not an NPE signal): individually-owned academic MEMS patents are exactly the raw material that later gets marketed to brokers. If this patent is ever transferred, watch for (a) a first assignment to a licensing LLC with an address that is a registered-agent service, and (b) a repeat recording correspondent across sibling Azgin patents. Neither has occurred as of the latest data.

---

Verdict

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

Justification: the patent was filed 2013-10-10 by Kivanc Azgin as a sole individual applicant and the record shows no recorded assignment of any kind — Google Patents' legal-events table for US9897488B2 contains only the 2018-01-31 grant entry (STCF) and the 2021-07-30 / 2025-08-11 maintenance-fee payments (MAFP, both micro-entity), with Original and Current Assignee both listed as "Individual" and IEEE recording the assignee as "UNASSIGNED." There is consequently no reel/frame, no correspondent, and no chain to evaluate for NPE signals — all eight signals resolve to "not present" or "no data." This is a clean, unassigned individual-inventor title, not a suspected assertion vehicle.

Verify (Assignment Center search page):

• https://assignment.uspto.gov/patent/index.html#/patent/search (search patent number 9897488 / application 15/028,435)
• https://assignmentcenter.uspto.gov/
• Google Patents legal events (source used here): https://patents.google.com/patent/US9897488B2/en

---

Flagged discrepancy with the prior section (carried forward): the earlier patent summary noted a claim-set discrepancy between the pre-grant publication US20160252400A1 and granted US9897488B2. That discrepancy does not affect ownership — both documents name Kivanc Azgin as sole individual applicant — but it reinforces that there is no corporate applicant hidden in the pre-grant record.

I'll research the prior art references cited on the face of US 9,897,488, verifying each citation and enriching the descriptions.

I have strong detail on four references. Let me gather the remaining three patent citations and the non-patent literature.

Prior Art Analysis — US 9,897,488 B2 ("Vibration based mechanical IR detector and an IR imaging method using the same")

Scope note: This analysis is built on the authoritative Google Patents record for US9897488B2 (fetched 2026-09-24) already used in the summary section, supplemented by live retrieval of the cited references. I hit my search-step limit before fully retrieving two of the seven references (US2014/0175284A1 and US2014/0226021A1); I flag exactly where that leaves uncertainty. No result contradicted the patent record, so the previously-generated sections stand.

Governing law / timing: The '488 application (US 15/028,435) was filed 2013-10-10, after the AIA first-inventor-to-file date (2013-03-16). AIA 35 U.S.C. § 102 therefore applies. A reference is § 102(a)(1) prior art if "patented, described in a printed publication, or in public use … before the effective filing date" (2013-10-10); it is § 102(a)(2) art ("secret" prior art) if it is a U.S. patent/published application effectively filed before that date, even if published later.

Anticipation standard: § 102 anticipation requires a single reference disclosing every element of a claim, arranged as in the claim. This matters here because the granted claims are narrow — claim 1 is the sole independent claim, and its point of novelty is the "wherein" clause (the anchor is centrally fixed between the resonating pixel plate and the pixel substrate). Most cited references are § 103 (obviousness) art, not true § 102 art, for the granted claims.

Critical date table for the seven cited references

---

Reference-by-reference analysis

1. US 5,193,911 A — "Thermal detector device"

• Full citation: US 5,193,911 A; inventors Nix, Holcroft & Etienne; assignee Thorn EMI plc; filed 1991-04-24 (GB priority 9009117, 1990-04-24); granted 1993-03-16. (Source: https://patents.google.com/patent/[US5193911A](/patent/US5193911A)/en; http://www.everypatent.com/comp/pat5193911.html)
• Description: A pyroelectric thermal-imaging array. A continuous pyroelectric film carries an array of interconnected entrance electrodes on one face and discrete electrodes on the other, supported on gold pillars above a silicon substrate bearing readout circuitry. A network of thermally conductive mesh electrodes between detector elements reduces thermal crosstalk. Detection is pyroelectric (charge generated by temperature change), with signal-processing/ADC electronics.
• § 102 anticipation of the '488: None. This reference lacks every distinguishing element of claim 1: there is no resonating pixel plate, no driving capacitive electrode, no sensing capacitive electrode to pick up resonance motion, and no resonance-frequency-shift detection whatsoever. It also lacks the central-anchor geometry. It is relevant only as background to the "array of pixels over a substrate with readout" and thermal-isolation concepts (§ 103 context, not § 102).

2. US 2004/0140428 A1 — "Pixel structure and an associated method of fabricating the same"

• Full citation: US 2004/0140428 A1; inventors Ionescu & Howard; assignee DRS Sensors & Targeting Systems; filed 2003-01-17; published 2004-07-22; granted as US 6,891,161 B2; PCT counterpart WO 2004/068454 A3. (Sources: https://patents.justia.com/patent/20040140428; WO2004068454A3)
• Description: A bolometer-based focal-plane-array pixel. A VOx detector is spaced above an insulator having serpentine legs, supported over a substrate; a resonant layer between detector and insulator forms two resonant absorption cavities, the resonant layer being selected so it is impedance-matched with vacuum (sheet resistance ~150.3–225.7 Ω/square, most preferably ≈188 Ω/square) so emissivity approaches unity. Readout is by change in electrical resistance (ΔR), not resonance.
• § 102 anticipation of the '488: None for any claim. The detection physics (resistive bolometer) is fundamentally different from the claimed frequency-shift detection, and the combination of a resonating plate + capacitive drive/sense electrodes + central anchor is absent, so claim 1 is not met.
• § 103 relevance: Meaningful for claim 4 (metal/alloy impedance-matched to free space) — note the numerical discrepancy: this reference impedance-matches at ≈188 Ω/square, whereas claim 4 recites ≈377 Ω (the impedance of free space). Also relevant to claim 8 (resonant light cavity) and claim 2 (IR-absorbing/absorber structure). It is a classic "cavity + absorber" teaching, not an anticipatory reference.

3. US 2011/0266445 A1 — "Optically transitioning thermal detector structures"

• Full citation: US 2011/0266445 A1; inventor Howard Beratan; assignee L-3 Communications Corporation; priority 2010-04-28; published 2011-11-03; granted as US 8,513,605 B2. (Sources: https://patents.google.com/patent/WO2011139329A2/en; DE 11 2011 101 491 T5)
• Description: Microbolometer pixel-membrane thermal detectors incorporating an optically transitioning (thermochromic / phase-change) material — GST, VO/VO₂, tungsten-doped VO₂, Nb/Ta oxides, etc. The membrane is suspended over a silicon ROIC wafer with a metal reflector to form a tuned quarter-wave resonant cavity; absorption efficiency depends on membrane–reflector spacing and sheet resistance. It also expressly lists TiN among the IR-absorbing thin-metal absorber materials. Readout is resistive.
• § 102 anticipation of the '488: None. No mechanical resonance, no frequency-shift detection, no capacitive drive/sense electrodes, no central anchor.
• § 103 relevance: Directly pertinent to claim 2 (IR absorber coating) and claim 3 (TiN as the IR-absorbing material), and to claim 8 (resonant light cavity / tunable sensitive band). A strong secondary reference for a § 103 obviousness case on the absorber/cavity-dependent claims.

4. US 2013/0170517 A1 — "Bolometer having frequency detection" ← CLOSEST CITED REFERENCE

• Full citation: US 2013/0170517 A1; inventors Laurent Duraffourg & Philippe Andreucci; assignee Commissariat à l'énergie atomique et aux énergies alternatives (CEA); PCT/EP2011/066564 filed 2011-09-23 (FR priority 2010-09-23); published 2013-07-04; granted as US 9,400,218 B2 (2016-07-26). (Sources: https://patentimages.storage.googleapis.com/05/a7/c4/22d4d495773a6d/US9400218.pdf; https://www.patents-review.com/a/20130170517-bolometer-frequency-detection.html)
• Description: A frequency-detection bolometer using a MEMS/NEMS oscillator. A mobile mass (plate-shaped) is hung from suspension beams above a support, and the mass itself forms the optical-flux absorber. Electrostatic (capacitive) actuation electrodes set the mass vibrating at resonance; capacitive detection electrodes detect the variation in vibration frequency as temperature changes. It expressly discloses: an array of such bolometers; an anti-reflection layer; a layer of different coefficient of expansion (e.g., Ti/TiN, ~10⁻⁵ K⁻¹ vs. silicon's ~2.6×10⁻⁶ K⁻¹) to exploit the expansion-based frequency shift; an optical resonant cavity formed with a reflector; silicon/SiGe plates; and an electronic control circuit that self-oscillates the mass and measures the frequency shift. Suspension beams are described as "at least two beams arranged along vibration nodes, or at least four beams arranged along vibration antinodes," and the actuation/detection electrodes are arranged laterally relative to the mobile mass.
• § 102 anticipation of the '488 claim 1: Very close, but not a complete anticipation. Mapping to claim 1:
  • "more than one resonating pixel structure" — ✅ met (array of bolometers).
  • "resonating pixel plate" — ✅ met (plate-shaped mobile mass).
  • "pixel substrate" — ✅ met (support/substrate).
  • "an [anchor] attaching the plate to the substrate" — ✅ met (suspension beams + anchor pads).
  • "driving capacitive electrode" — ✅ met (electrostatic actuation electrodes).
  • "sensing capacitive electrode" — ✅ met (capacitive detection electrodes).
  • "electrical connection pads to keep the plate at an electrostatic potential" — ✅ substantially met (control circuit biases the mass).
  • "wherein the anchor is centrally fixed between the resonating pixel plate and the pixel substrate" — ❌ NOT disclosed. Duraffourg anchors the mass at discrete node/antinode points via multiple beams ("at least two beams … at nodes, or at least four beams … at antinodes"), not a single central anchor. Additionally, its drive/sense electrodes are lateral to the mass, whereas '488 claim 6 recites electrodes under the maximum-deflection regions.
• Conclusion: This is the most relevant piece of art and the strongest potential § 102 challenge if "centrally fixed" were read loosely, but as written the central-anchor and electrode-placement limitations likely distinguish it. Realistically it is a § 103 primary reference (frequency-shift IR detection via a MEMS plate/absorber), combinable with, e.g., a central-anchor microbolometer design. It also bears on claim 2 (absorber), claim 9 (silicon), and claim 8 (optical cavity).

5. US 2013/0206988 A1 — "Method and apparatus for infrared detection and display"

• Full citation: US 2013/0206988 A1; inventor Franky So; assignee University of Florida Research Foundation, Inc.; filed 2013-03-26 (continuation; priority 2006-09-29); published 2013-08-15; granted as US 9,276,048 B2. (Sources: https://patents.google.com/patent/US9276048; https://uspto.report/patent/app/20130206988)
• Description: An organic phototransistor (or a photoconductor with PbSe/PbS quantum dots) detects IR; the phototransistor is placed in series with an OLED so a detected IR image is shown to the user in visible light. The wavelength range is tuned by incorporating quantum dots. Detection is by photoexcited charge carriers — an optoelectronic, non-mechanical mechanism.
• § 102 anticipation of the '488: None. It shares only the generic field (IR detectors / IR imaging / display) and lacks any resonating plate, capacitive electrodes, frequency-shift detection, or central anchor. It is, at most, general background and is cited in the '488 specification itself as an example of prior-art thermal IR detectors.

6. US 2014/0175284 A1 — "Infrared detector including broadband light absorber"

• Full citation: US 2014/0175284 A1; assignee Samsung Electronics Co., Ltd.; priority 2012-12-20; published 2014-06-26.
• Description (from the citation record; full text not retrieved during this analysis): an IR detector incorporating a broadband light absorber.
• Prior-art status: Because it published after the '488 effective filing date (2013-10-10), it is available only under § 102(a)(2) (effectively filed 2012-12-20, before the '488 date). It is not § 102(a)(1) art.
• § 102 anticipation of the '488: None — it does not disclose the resonating-plate/frequency-shift architecture or the central-anchor limitation. Its likely relevance is § 103 for claim 2 (IR-absorbing material / broadband absorber).
• ⚠️ Uncertainty: I could not retrieve the full text within my search budget, so I cannot rule out that it also discloses a suspended-plate/tuned-cavity structure. Treat the description as provisional.

7. US 2014/0226021 A1 — "Microbolometer array with improved performance"

• Full citation: US 2014/0226021 A1; assignee Centre National De La Recherche Scientifique (CNRS); priority 2011-07-15; published 2014-08-14.
• Description (from the citation record; full text not retrieved): a microbolometer array with improved performance (typically improved thermal isolation/absorption).
• Prior-art status: Published after the '488 filing; available only under § 102(a)(2) (effectively filed 2011-07-15).
• § 102 anticipation of the '488: None for any claim — it is a resistive-microbolometer-array reference and does not disclose mechanical-resonance frequency-shift detection, capacitive drive/sense electrodes, or the central anchor.
• ⚠️ Uncertainty: full text not retrieved; its relevance is presumptively § 103 background for array/pixel architecture (claim 1 preamble) and thermal isolation.

Non-patent literature (cited on the face of the patent)

NPL-1. Zhao et al. (2012) — the most conceptually on-point NPL

• Full citation: Danqi Zhao et al., "Uncooled bimaterial cantilever for infrared detection based on resonant frequency tracking," Micro & Nano Letters, IET, vol. 7, no. 5, May 31, 2012, pp. 434–438.
• Description: An uncooled IR detector using a bimaterial cantilever whose resonant frequency is tracked as it shifts with temperature — i.e., IR → ΔT → resonant-frequency shift → measurement. This is the same core measurement principle recited in the '488 specification ("measure the shift in the resonance frequency of an adequately designed resonator in the presence of IR radiation").
• § 102 anticipation of the '488: None for the granted claims. It is a bimaterial cantilever (single-clamped; expansion-driven) rather than a center-fixed plate with capacitive drive/sense electrodes, and it is a single element, not "more than one resonating pixel structure" as claim 1 requires. So it does not meet claim 1 (missing array/multi-pixel, capacitive electrode pair, and central anchor).
• § 103 relevance: High. It discloses the expansion-based resonant-frequency-shift IR detection method and would be an obvious starting point combined with, e.g., Duraffourg (capacitive MEMS frequency-detection bolometer) and a center-anchor microbolometer geometry. It directly bears on claims 1, 2, and 8.

NPL-2. Tilmans et al. (1991)

• Full citation: H. A. C. Tilmans et al., "Single element excitation and detection of (micro-)mechanical resonators," TRANSDUCERS '91 (Int'l Conf. on Solid-State Sensors and Actuators), San Francisco, Jun. 24–27, 1991, pp. 533–537.
• Description: Excitation and detection of micro-mechanical resonators, including capacitive actuation and capacitive sensing and self-resonant/feedback operation.
• § 102 anticipation of the '488: None. It is not an IR detector; it has no IR absorber, no multi-pixel array, and no central-anchor-on-a-substrate geometry. It is relevant to the actuation/detection architecture (and thus to the '488's "capacitive actuation and detection" and "self-resonance with frequency counting" disclosures), i.e., § 103 background for claim 1's electrode elements and for claims 5–6 (in-phase electrode wiring; electrodes under maximum-deflection regions).

---

Summary: which references "potentially anticipate" which claims

Bottom line:

• No single cited reference anticipates the granted claims, because claim 1's controlling limitation — the anchor is centrally fixed between the resonating pixel plate and the pixel substrate — is absent from every reference (the closest, Duraffourg, uses multiple node/antinode suspension beams).
• The two references that come closest to the inventive concept — and the ones I would actually rely on for a validity challenge — are US 2013/0170517 A1 (Duraffourg/CEA) for the capacitive, frequency-detection MEMS bolometer architecture, and Zhao et al. 2012 for the resonant-frequency-tracking uncooled IR detection method. Combined (optionally with Ionescu/DRS for the tuned cavity and Beratan/L-3 for the TiN absorber), they would support an obviousness argument against claims 1–2 and 8, but not a § 102 anticipation without reading "centrally fixed" out of claim 1.

Caveats on confidence: (1) The full texts of US 2014/0175284 A1 and US 2014/0226021 A1 were not retrieved before my search budget was exhausted — my "no anticipation" conclusions for those two rest on their titles/abstracts and the '488 claim structure, and should be confirmed against the full documents. (2) All § 102 conclusions assume the AIA applies (PCT filed 2013-10-10); if an earlier effective date were ever established, the analysis of the two 2014 publications could change. (3) This is a technical prior-art assessment, not a legal opinion on validity.

I'll ground this §103 analysis by verifying what the cited prior-art references actually disclose.

Strong material. Let me verify the remaining cited references plus the center-anchor teaching.

Obviousness Analysis — U.S. Patent No. 9,897,488 B2 under 35 U.S.C. § 103

⚠️ Search-degradation notice: I exhausted my search-step budget mid-verification. I fully verified the disclosures of Duraffourg (US 2013/0170517 / US 9,400,218), Zhao et al. (2012), Tilmans et al. (1991), CNRS (US 2014/0226021 / US 9,417,134), and Thorn EMI (US 5,193,911). I did not independently verify the contents of US 2004/0140428 (Ionescu), US 2011/0266445 (Beratan), US 2013/0206988 (Univ. of Florida), or US 2014/0175284 (Samsung); for those four I rely on the patent's own characterizations plus the title/assignee metadata. That gap is flagged at each point of use.

---

1. Legal Framework and Governing Date

Because the application was filed after the AIA cutover, § 102(a)(1) (public use/on-sale/printed publication/public disclosure) and § 102(a)(2) (earlier-filed U.S. applications/published applications) both apply. This matters because one cited reference publishes after 2013-10-10 and must be qualified.

PHOSITA — level of ordinary skill

A person with an M.S. or Ph.D. in mechanical/electrical engineering, applied physics, or materials science, plus 2–5 years designing MEMS/NEMS resonant devices or uncooled IR detectors, and familiarity with: microfabrication (surface/bulk micromachining, SOI, anodic bonding), capacitive and piezoelectric transduction, resonant-sensor readout (PLL, self-oscillation, frequency counting), bolometer thermal design (thermal conductance G, NETD), and IR absorber stacks / quarter-wave cavities.

Prior-art qualification table (critical — do not overlook)

Analytical consequence: CNRS is only usable because of the WO 2013/010933 A1 publication (2013-01-24) and its 2012-07-13 PCT filing. Any invalidity contention must plead the § 102(a)(2)/WO-publication route, not the US 2014/0226021 publication date. This is exactly the kind of date that gets overlooked and then sinks an IPR petition.

---

2. What the Prior Art Actually Discloses

2.1 Duraffourg — US 2013/0170517 A1 / US 9,400,218 B2 (the primary reference)

Sources: Google Patents US20130170517, WO2012038524A1, US9400218 PDF

Verified disclosure (paragraph numbers from the A1 publication):

• Resonance-shift IR detection, expressly: "a bolometer comprising at least one electromechanical microsystem or nanosystem, said microsystem or nanosystem comprising a support and a mobile mass hung by suspending means above the support, said mobile mass forming an optical flux absorber, actuation means configured to set the mobile mass in vibration at its resonance frequency …, and detection means to detect the variation in vibration frequency of said mobile mass." ([0021])
• Plate geometry: "The mobile mass is preferably in plate shape … an element extending mainly along a plane and of reduced thickness." ([0023])
• Capacitive/electrostatic actuation and detection: "the actuation means are of electrostatic type and comprise at least one electrode acting on the support"; "the detection means are of capacitive type and comprise at least one detection electrode on the support." ([0024]–[0025])
• Array: "an assembly of bolometers according to the present invention arranged in an array." ([0034])
• Anchor/node placement teaching (important): "The suspending means may comprise at least two beams arranged along vibration nodes, or at least four beams arranged along vibration antinodes." ([0026])
• Thermal-isolation rationale: beams "very narrow thickness and width relative to the thickness of the mobile mass," "thickness … at least ten times smaller than that of the mass," "[i]t is therefore sought to reduce G"; narrow beams "minimize measurable NETD"; "highly advantageous thermal uncoupling is obtained between the absorber 4 and the support." ([0122]–[0130])
• IR absorber materials — including TiN: "of TiN, Ti/TiN in a thin layer typically a few nanometres, or a layer of dielectric material such as SiN, SiO2 … or an optical multi-layer." ([0116]–[0117]); "a layer of Ti/TiN allows optimization of optical flux transmission and has a coefficient of expansion … about 5 times greater than that of silicon." ([0121])
• Bimaterial / CTE-mismatch temperature response: a layer "in material having a different coefficient of expansion to that of the mobile mass … at least twice higher or lower." ([0029])
• Optical resonant cavity + λ/4: "an optical reflector … so as to form an optical resonant cavity between the optical reflector and the second face of the mobile mass" ([0031]); "The layer 22 has for example a thickness of 2.5 μm corresponding to the length λ/4 for an optical beam in the range 8 to 14 μm."
• Materials: "The mobile mass and/or suspending means are preferably in a semiconductor, most advantageously crystalline e.g. in silicon or SiGe." ([0033])
• Self-oscillation readout: "control the setting in self-oscillation of the mobile mass(es)"; "a set of phase shifters … zero input/output phase shift for each self-oscillation loop."

Gap: Duraffourg hangs the mass by beams from the support and never describes a single centrally-fixed anchor.

2.2 Zhao et al. 2012 (the closest "capacitive resonant IR array" reference)

Source: IET Micro & Nano Letters, DOI 10.1049/mnl.2012.0144; PKU record

Verified disclosure:

• "detecting infrared (IR) radiation by tracking resonant frequency of bimaterial resonant sensors."
• "A capacitive bimaterial resonant sensor (BRS) array consisting of 6 × 6 cells … fabricated by surface sacrificial layer process."
• "The resonant frequency responsivity of the BRS is better than 363 Hz/K by electrical measurement."
• Pixel anatomy: "Each presented pixel of BRS array is composed of a bimaterial cantilever, a back cavity, two driving electrodes, four anchors and five dimples."
• "CTE mismatch between the two different materials poly-Si and Au"; "The heavily P-doped Poly2 serve as the IR absorber"; "The bimaterial cantilever IR absorber is anchored to a silicon substrate with 2 μm spacing"; "back cavity in substrate is also designed for reducing thermal dissipation."
• Drive/readout: "AC electric force is applied to the cantilever … the amplitude-frequency characteristic is accessible by frequency sweep measurement. The resonant frequency can be retrieved from the amplitude-frequency curve."
• Uses vacuum operation and thermal-isolation nitride films.

Gap: Zhao's pixel is edge-anchored (four anchors), and the paper is silent on center-fixing.

2.3 Tilmans 1991 (the capacitive-transduction reference of record)

Source: U. Twente record, DOI 10.1109/SENSOR.1991.148931

Verified disclosure:

• "a single-element approach for the excitation and detection of the vibrational motion of (micro-)mechanical resonators."
• "An equivalent electrical one-port network is derived for an electrostatically and a piezoelectrically driven resonator."
• "The motion … is excited by applying an ac drive voltage u(t) and a dc polarization voltage Vp across the (air) gap."
• "detection is capacitive, based on … a variation of the distance between the plates results in capacitance fluctuations, inducing either an a.c. current in case of a d.c. voltage bias …"
• Requires "a Q-factor on the order of 10,000 … For resonators operating in a vacuum, this is achievable."
• Related Tilmans work (Sensors and Materials 9(8):521–540) expressly addresses "the relevance of the spatial placement of the transducer elements with respect to the position of the vibrating member."

2.4 CNRS/ONERA/CEA — US 2014/0226021 A1 / US 9,417,134 B2 (WO 2013/010933 A1)

Sources: US9417134 PDF, FPO US2014/0226021

Verified disclosure:

• "a microbolometer array … comprising a supporting substrate and an array of microbolometers … arranged in an array."
• "Each of said microbolometers comprises a membrane suspended above said supporting substrate" with a radiation absorber and a thermometric element.
• "arms 306 for thermally insulating the thermometric element, one of the ends of the arm being connected to the thermometric element and the other being connected to the anchoring point making contact with the supporting substrate."
• "electrical connection pads 315 of the read-out circuit"; the substrate "comprises circuits … for biasing and reading."
• "It is known to place the membrane … at a distance d = λ0/4 away from a mirror 102 … so as to form a quarter-wave cavity allowing high absorption, typically 90% over … band III ranging from 8 to 12 μm."
• MIM resonant absorber (metal/dielectric/metal) with "resonant absorption … at at least one wavelength."

2.5 Thorn EMI — US 5,193,911 A

Sources: Google Patents US5193911A, EveryPatent transcription

Verified disclosure — and this one is a direct hit on claim 4:

• "A thermal imaging device … an array of pyroelectric detector elements, 31, 32, 33 supported by an array of pillars 36, the pillars also being effective to enable the passage of electrical signals between the detector elements and an electrical signal processing means 38."
• "a substrate 1 of semiconductive material … formed with a plurality of electrical circuit components 2 … capable of receiving and processing electrical signals derived from a planar array of, for example 100 × 100 detector elements."
• The 377 Ω element: "The metallic material … does not need to be of very high electrical conductivity, and nichrome (resistivity ≈ 377 Ω per square) has been found to be suitable." — an IR-receiving electrode material selected at literally the free-space impedance value recited in claim 4.
• Thermal isolation by "suitable dimensioning of the thickness, width and length of the links."
• An IR-absorbent layer ("areas of infra-red absorbant material 32a"; "black gold" absorbent layer in the sibling EP 0 345 047).

2.6 The patent's own admissions about the remaining cited art

The '488 specification itself characterizes two cited references, which are therefore applicant admissions usable in a § 103 analysis:

• US 2004/0140428 A1 (Ionescu): "defines a pixel structure, forming one element of a focal plane array, including a bolometer having a detector and an insulator for measuring the incoming IR radiation…" (spec, Background). → supplies "array of pixel structures + substrate-anchored suspended bolometer pixel."
• US 2013/0206988 A1 (Univ. of Florida): "a detector having organic layers … a phototransistor for the detection of IR radiation … The wavelength range of the defined IR detector can be modified by incorporating materials sensitive to photons of different wavelengths." → supplies the wavelength-selectivity design objective.

I was unable to verify US 2011/0266445 (Beratan) and US 2014/0175284 (Samsung "broadband light absorber") beyond their titles; I treat both as confirmed-by-title background art only and do not build any limitation on them below.

---

3. Claim 1 — Element-by-Element Mapping and Motivation

Claim 1 is the only independent claim. Its body recites: (a) more than one resonating pixel structure; (b) a resonating pixel plate; (c) a pixel substrate; (d) an anchor attaching the plate to the substrate; (e) driving capacitive electrode(s) actuating the plate; (f) sensing capacitive electrode(s) picking up motion; (g) electrical connection pad(s) holding the plate at an electrostatic potential; and (h) "wherein the anchor is centrally fixed between the resonating pixel plate and the pixel substrate."

3.1 Motivating the combination (KSR rationales, expressly articulated)

(1) Duraffourg + Tilmans. Duraffourg already names electrostatic actuation and capacitive detection as its embodiments but does not elaborate the one-port electrical interface. Tilmans supplies precisely that: the canonical electrostatic-excitation/capacitive-detection scheme, including the mandatory DC polarization voltage and the bridged detection circuit. Rationale: use of a known technique (capacitive MEMS transduction with DC bias) to implement a known device (a capacitive MEMS resonator) in the same way, to yield a predictable result — the "electrical connection pad" limitation (g) is not an inventive addition but a physical prerequisite of the capacitive scheme Tilmans describes.

(2) Duraffourg + Zhao. Both are directed to the same problem — uncooled IR detection by tracking a mechanical resonance frequency — via the same mechanism (temperature-induced stiffness/frequency shift). Zhao supplies the multi-pixel capacitive array, IR absorber, thermal isolation, back cavity, driving electrodes, anchors, and — crucially — experimental proof that the concept works (a fabricated 6×6 array with measured Hz/K responsivity). Rationale: predictable result / reasonable expectation of success, and KSR's "finite number of identified, predictable solutions" (obvious to try). Zhao's measured data removes any argument that capacitive resonant IR imaging was a speculative, unproven avenue as of 2013.

(3) + CNRS (WO 2013/010933). Supplies (i) array-level readout circuitry integrated in the substrate, (ii) thermally insulating arms + anchoring elements + connection pads, and (iii) the expressly "known" quarter-wave (λ/4) cavity for high absorption. Rationale: market/design incentive — improving IR absorption (η) and thermal isolation (R_th) are the two recognized levers on detector sensitivity; Duraffourg's own text works the same equation (responsivity ∝ η·α_T·ω₀/G; "[i]t is therefore sought to reduce G"). Combining CNRS's cavity and arm geometry with Duraffourg's resonant readout is the predictable use of known techniques to improve a similar device.

(4) Motivating element (h) — the center-fixed anchor. Three independent motivations converge:

• Thermal. The entire field — Duraffourg expressly, CNRS expressly ("[i]ncreasing the thermal resistance R_th thus increases the sensitivity") — is dedicated to maximizing thermal resistance between absorber and substrate. A single, narrow, central anchor is the limiting case: it minimizes the number and cross-section of conduction paths and frees the entire plate perimeter. A PHOSITA optimizing R_th would predictably arrive at it.
• Mechanical/Q. Anchoring at a node of the mode shape minimizes anchor loss and maximizes Q, which directly improves the frequency stability (S_ω) that sets NETD. Duraffourg expressly teaches anchoring "along vibration nodes" and placing supports "along vibration antinodes" ([0026]). For the fundamental out-of-plane plate modes, the mode's node is at/near the plate center — so the node-anchoring teaching points directly at a central anchor. Tilmans independently supplies the reason Q matters ("a Q-factor on the order of 10,000 is required").
• Electrode real estate / fill factor. Claim 6 (electrodes under maximum-deflection regions) and CNRS's fill-factor discussion both favor an anchor that does not consume plate perimeter. Center-fixing makes plate edges and corners available as maximum-deflection, high-signal electrode regions — the very geometry the '488 spec relies on ("the corners on the same diagonal deflect together").

Supporting art: In my search I surfaced US 8,809,786 B2 (a microbolometer detector whose platform is "supported by the support structure from the central region thereof," with thermally insulating support arms "within the central region of the platform," and with a background acknowledgment of "conventional edge-supported microbolometer detectors"). This is direct evidence that center-underside support of a suspended bolometer platform was a known, recognized alternative to edge support. Caveat: I could not verify US 8,809,786's priority/filing date, so I cannot confirm it is § 102 prior art to 2013-10-10. It must be date-checked before use. Commentators should treat the center-anchor limitation as the one limitation whose invalidity case depends on finding a pre-2013-10-10 center-support teaching.

3.2 Why element (h) is nonetheless narrow

Claim 1's "centrally fixed" is a single geometric configuration of an otherwise wholly conventional suspended-plate resonator. Under KSR and MPEP 2144.04 (design choice / optimization of a result-effective variable), selecting among a small set of known suspension topologies (edge-anchored, corner-anchored, multi-arm, center-anchored) to optimize a known result (thermal conductance and anchor loss) is a textbook obviousness posture provided the center-anchor option was known or suggested. Duraffourg's node-anchoring sentence supplies the suggestion; US 8,809,786 (if dated correctly) supplies the concrete prior art.

---

4. Dependent Claims 2–13

Note on claim 13 and the "similar documents" list: the record's own "Similar Documents" section lists Hui et al., "High performance NEMS resonant infrared detector based on an aluminum nitride nano-plate resonator" (2013) and US 9,097,578 B2, "Infrared sensing using pyro/piezo-electric resonators." If either predates 2013-10-10 in a qualifying way, claim 13's invalidity position becomes materially stronger. I did not verify their dates.

---

5. The Patent Owner's Best § 103 Rebuttals (steelman)

A rigorous analysis must state where the invalidity case is beatable.

1. Teaching away / consistent art. Duraffourg, CNRS, Ionescu, and Thorn EMI all suspend the absorber element by multiple arms/beams from its periphery, and the '488's own evaluation of the art describes edge/arm suspension as the norm. Patent owner will argue the art taught away from a single central anchor, and that a single central anchor raises thermal-stress/mounting concerns (a symmetry/buckling argument). Counter: Duraffourg expressly teaches node-anchoring, which for the relevant plate modes is centrally located; thermal-isolation logic runs monotonically toward fewer/shorter conduction paths; and KSR forecloses "the prior art taught away" where the reference merely discloses a preferred embodiment among alternatives.

2. "The allowance turned on the anchor." The prosecution history likely shows that the "wherein the anchor is centrally fixed" clause was the limitation that overcame Duraffourg. If so, the patent owner will argue the Examiner already considered Duraffourg and found the center-fixed geometry absent. Counter: the examiner's search does not bind a later validity challenge, and the center-anchor feature is a design choice whose obviousness turns on art the Examiner may not have consulted.

3. Secondary considerations. None are supported on this record: no comparative data against the closest prior art; no unexpected-results evidence (the spec offers FEM mode-shape plots and statements of aim, not comparative measurements); no evidence of commercial success, licensing, copying, industry praise, or long-felt need — consistent with the previously-generated finding that the patent is individually owned, micro-entity, and unasserted. Any such evidence would also need nexus to the center-anchor limitation specifically.

Where the rebuttal may succeed: if no pre-2013-10-10 reference shows or suggests central anchoring of a resonant suspended plate, claim 1 (and claims 5–13, which depend from it) may survive. That is the hinge.

---

6. Practical Assessment and Recommended Next Steps

Strength ranking for a § 103 challenge (strongest → weakest):

Procedural note: PGR is time-barred (the 9-month post-grant window for this 2018-02-20 grant closed 2018-11-20). IPR remains available and is the realistic vehicle, based on the patents and printed publications identified above.

To close the one material gap, verify (in order):

1. A pre-2013-10-10 reference disclosing a center-anchored / center-supported suspended plate or membrane — start with US 8,809,786 B2 (check its § 102(e)/(a)(2) date), plus center-anchored disk/plate MEMS resonators and center-post microbolometer platforms.
2. The exact U.S. filing date of US 2014/0175284 (Samsung) to confirm its § 102(a)(2) availability.
3. The publication dates of Hui et al. (2013) and US 9,097,578 B2 for claim 13.
4. US 2004/0140428 (Ionescu) and US 2011/0266445 (Beratan) full texts (I could not verify these).
5. The prosecution history of US 15/028,435 — specifically which limitation was argued over Duraffourg — to confirm whether the center-anchor was the point of novelty.

Discrepancies / flags carried over

• As previously noted, the rendered claim 1 text is missing a word — "an attaching the resonating pixel plate to the pixel substrate" — evidently the anchor. This drafting defect is not merely cosmetic: the "wherein" clause refers to "the anchor," so the antecedent basis is defective, and the scope of the limitation that decides validity (element (h)) is arguably ambiguous. Claim construction should be confirmed against the USPTO-issued claim text, not the rendered HTML.
• Claim 1 requires "more than one resonating pixel structure" while reciting "a resonating pixel plate" in the singular — an internal singular/plural mismatch that bears on whether the multi-pixel feature is a structural limitation or a mere capability statement. If construed as a capability, element (a) adds nothing to the § 103 analysis.
• The grant's 13 claims are apparatus-only; the earlier pre-grant publication US 20160252400A1 displayed broader method-flavored claims. Only the granted claims matter for validity.

Bottom line: Claims 3, 4, 10, 11 face a strong § 103 exposure on the face of the patent's own cited references (Duraffourg for TiN and silicon; Thorn EMI for ≈377 Ω). Claims 6 and 13 are strong on ordinary-optimization and applicant-admission grounds. Claim 1 — and therefore the whole claim set — turns on the "centrally fixed" anchor, which is a known design option in view of Duraffourg's node-anchoring teaching but for which I have not yet confirmed a squarely on-point pre-2013-10-10 reference. Resolve that one point and the § 103 position on the patent as a whole is strong; leave it unresolved and claim 1 is defensible.