US 5274434

I'll verify the patent record and check for any litigation or CAFC docket activity.

Let me retrieve the claim set and confirm there's no litigation involving this patent.

Let me pin down the exact independent claim text.

US Patent 5,274,434 — Verification Summary

Caveat on method: I ran targeted web searches against public patent databases (Google Patents, patentimages PDF, EveryPatent, USPTO.report, Justia) and general litigation-docket search. I did not have direct API access to USPTO PatentCenter or the CAFC/PACER docket system in this session, so docket-level statements below are based on absence of search hits rather than a direct docket query. I flag each uncertainty inline.

1. Bibliographic Record

Small discrepancy flagged: the patent's cover page is cited as "Morioka et al." (confirmed by the reference listing in US 6,894,773). One EPO search report citing this document renders it as "OHSHIMA YOSHIMASA ET AL." I treat the printed patent ("Morioka et al.", Morioka first-named) as authoritative; the EPO rendering appears to be a database formatting artifact.

Family (partial, from EPO family annex): JP 3285339 A (1991-12-16); JP 2843424 B; JP 4056245 A; JP 4152545 A; KR 9513433 B; US 5,274,434; US 5,463,459; US 5,233,191.

2. Abstract (as printed)

The inspection apparatus is built as a small-sized unit placed at the inlet/outlet of production-line processing apparatuses or in the transfer system between them. It includes at least one monitor for real-time sampling of foreign particles on a wafer being carried by the transfer system, enabling simplification of the production line and reduced manufacturing cost. The apparatus may comprise a refractive index changeable type lens array, a spatial filter, and a pattern data elimination circuit, permitting inspection of repetitively patterned portions of wafers during transfer. With the spatial filter eliminating repetitive-pattern data, the compact apparatus can inspect wafers for foreign particles at high speed in real time.

3. Plain-Language Overview of the Independent Claims

Uncertainty note: the full claim set was not reproduced verbatim in the retrieved sources. I have verbatim text for claim 19 (independent apparatus claim) and verbatim text for claim 20 (dependent). Claim 1 and the intervening dependent claims were only partially recovered (e.g., a claim reciting that it depends from claim 1 and adds mass-production start-up sampling of a wafer through a gas supplier and production apparatuses). I therefore describe claim 1's probable scope and clearly mark it as inferred, not quoted.

Independent Claim 19 — Apparatus (verbatim, recovered)

An apparatus for inspecting foreign particles on a product substrate in a mass-production line, comprising:

1. Transferring means at at least one predetermined position in the line, moving the product substrate along a path;
2. Means for detecting the rotation direction of repetitive patterns of the circuit pattern formed on the transferred substrate;
3. Illumination of substantially slit-shaped light onto the substrate surface as it is transferred;
4. Detecting means that detect surface information through a detecting lens, including scattered light from foreign particles on the illuminated substrate;
5. Means enabling relative alignment between the detected direction of the repetitive patterns and the direction of the detecting means; and
6. Means for determining a state of foreign particles on the transferred substrate based on the scattered-light detection.

In plain terms: bolt a compact optical head onto the wafer transport path, sense which way the wafer's repeating circuit patterns are oriented, shine a line of light on the moving wafer, catch the scatter off particles, rotate/align the detector to match the pattern direction, and report particle status — all without taking the wafer out of the line.

Dependent Claim 20 (verbatim)

Adds that the detecting means includes a spatial filter on a Fourier transform plane; the alignment means relatively aligns the pattern direction with the spatial filter direction; the spatial filter shields diffracted light from the repetitive circuit patterns imaged onto that plane through the detecting lens; and the determining means outputs a foreign-particle distribution in accordance with the detection output and the alignment.

Independent Claim 1 — Method (scope inferred, not verbatim)

Based on the specification's own summary of the invention and on the dependent claim that refers back to it, claim 1 appears to be the method counterpart of the claim 19 apparatus: inspecting foreign particles on a product substrate in a mass-production line of a semiconductor fabrication process, using a stage/transfer path, an illumination system illuminating a measuring position in linear (slit) form by a substantially single-wavelength plane wave, a lens imaging that position onto a detector, a spatial filter in the imaging optical path blocking diffraction light from repetition patterns, a detector, a binarizing circuit, a circuit extracting only isolated binarized signals within the wafer, and means for erasing signals recurring repeatedly between chips. One dependent claim adds the mass-production start-up step of feeding sampling wafers through gas supplier(s) and production apparatuses to determine generation of a foreign substance and contaminant.

In plain terms: scan a product wafer as it moves down the line with a thin sheet of coherent light, image it through a filter that optically deletes the repeating memory-array pattern, and use signal processing (binarize → isolate isolated hits → erase chip-to-chip repeats) to flag only genuine particles.

Claim count: the EPO search report maps this document to "claims 1–10" for citation purposes, which does not establish the total count. Given claim 19 and claim 20 exist, the patent contains at least 20 claims, with claim 1 (method) and claim 19 (apparatus) as the identified independents. I cannot rule out an additional independent claim between 1 and 19 from the sources retrieved.

4. Key Technical Content (context for claim scope)

• Spatial-filter-based dark-field inspection: oblique coherent illumination (semiconductor laser + anamorphic prism) produces a sharp Fourier transform of the wafer pattern; a spatial filter blocks the regular diffracted orders from repetitive patterns.
• Filter implementations disclosed: dry-plate (photographic) filters, liquid-crystal (switchable-format) filters, and an AND filter combining several production steps to reduce filter count.
• Lens options: ordinary lens, lens array, refractive-index-changeable lens array (selfoc-type, the subject of the "compact" emphasis), and microlens groups with zigzag-arranged CCD line sensors / 2D CCD / TV camera.
• Alignment/rotation handling: orientation-flat positioning, or rotating the spatial filter / whole detecting optical system (up to 45°) to follow wafer rotation; wafer rotating-direction detector also derives transfer speed to synchronize sensor scan rate.
• Coordinates management: two reference frames — a wafer/orientation-flat (or circuit-pattern) reference and an apparatus reference independent of wafer rotation, used to recognize a regular particle distribution indicating in-apparatus dust generation.
• Yield rationale: the specification asserts (based on the inventors' experiments) that particle counts rise suddenly, not gradually, so once-a-day or per-lot sampling misses excursions; only near-100% real-time sampling prevents mass defects.
• Alarm/control: FFT + repetition-portion elimination generates a particle map; exceeding an allowable count triggers an alarm and can stop the production line; an unusually low "inspection area ratio" also alarms (possible apparatus or process fault).

5. Litigation / CAFC / PTAB Check — Important Disambiguation

No litigation, CAFC appeal, or PTAB proceeding involving US 5,274,434 was found in the sources searched. Given the patent expired (recorded expiration 2011-04-02), this is consistent with expectations.

⚠️ Do not conflate. Search results for "the '434 patent" in litigation are almost entirely about a different patent — US 5,247,434 (Baxter International, hemodialysis machine with touchscreen interface), the subject of Fresenius USA, Inc. v. Baxter International, Inc. (Fed. Cir. 2009, 582 F.3d 1288; Fed. Cir. 2013, 12-1334) and In re Baxter Int'l, 678 F.3d 1357 (Fed. Cir. 2012), where the PTO cancelled the asserted claims in ex parte reexamination on 2013-04-30. That case, the IPR2020-00407 Hitachi/Maxell matter re US 6,748,317, and an ANDA suit referencing a different '434 patent are unrelated to Hitachi's US 5,274,434.

Uncertainty statement: because I could not directly query PACER/CAFC RSS or USPTO PatentCenter in this session, I cannot state with absolute certainty that no minor or unreported proceeding exists. What I can state is that no search result identified US 5,274,434 as a patent-in-suit or as an appealed patent.

6. Continuation Family (for prosecution/prior-art context)

US 5,274,434 sits in a long Hitachi continuation chain over wafer particle inspection, all descending from 07/679,317 / 07/778,363. Downstream members include US 5,463,459, US 6,650,409, US 6,411,377, US 6,894,773, US 7,037,735, US 7,177,020, US 7,098,055, US 7,692,779, US 7,443,496, US 7,639,350, and US 7,940,383 (several later ones showing Renesas Technology Corp. as assignee). US 6,894,773 and related patents expressly describe themselves as "an improvement over U.S. Ser. No. 07/778,363" — i.e., over this patent — and cite US 5,274,434 as prior art against themselves.

I'll research litigation involving US Patent 5,274,434. Let me search multiple sources.

The initial searches returned results for a different patent (5,247,434, a Baxter hemodialysis patent) — I must not conflate those. Let me search specifically for the correct number.

The searches keep surfacing the Baxter patent 5,247,434 (not our patent) and unrelated matters. Let me target litigation databases directly.

Result: No litigation found involving U.S. Patent No. 5,274,434

Based on my searches of patent litigation resources (Google Patents, USPTO records, Unified Patents‑style portals, CourtListener/RECAP decisions, and general web coverage), I found no litigation — no district court complaints, no ITC Section 337 actions, no Federal Circuit or PTAB appeals — that asserts or involves U.S. Patent No. 5,274,434.

What U.S. 5,274,434 actually is

• Title: "Method and apparatus for inspecting foreign particles on real time basis in semiconductor mass production line"
• Assignee: Hitachi, Ltd. (original and current)
• Inventors: Hiroshi Morioka, Minori Noguchi, Yoshimasa Ohshima, Yukio Kembo, Yuzo Taniguchi
• Application: 07/778,363, filed 1991‑10‑17; granted/publication date: 1993‑12‑28
• Priority: 1990‑04‑02; a continuation‑in‑part of U.S. Ser. No. 07/679,317 (now U.S. 5,233,191)
• Status: Expired – Lifetime (anticipated expiration 2011‑04‑02)

This is an optical wafer/foreign‑particle inspection patent, and it has only surfaced in the record as prior art / cited reference in later patents and prosecution proceedings (e.g., it is cited as a reference in several Hitachi and third‑party optical‑inspection patents). No case has been identified in which it is the patent‑in‑suit.

Important caution about the number (do not conflate)

Several high‑profile cases involve a different patent number — U.S. 5,247,434 — and must not be confused with 5,274,434:

• Fresenius USA, Inc. v. Baxter International, Inc. (N.D. Cal. No. 03‑cv‑1431; Fed. Cir. Nos. 12‑1334/‑1335) concerned U.S. 5,247,434 (Baxter hemodialysis touchscreen patent). That number (5,247,434) is not the patent at issue here (5,274,434).
• Other "'434" references encountered (e.g., a Bosch wiper‑blade '434 patent; a Perrigo/pharmaceutical '434 patent) likewise involve different patent numbers.

Because the digits differ, none of those matters qualify as litigation involving 5,274,434. I have interpreted the identifier literally and have not substituted the similar‑numbered patents.

Bottom line

• Plaintiff(s): None identified
• Defendant(s): None identified
• Jurisdiction: None identified
• Case number: None identified
• Filing date: None identified
• Outcome/status: N/A — no litigation known

Caveat on confidence: I could not complete a full sweep of every paid litigation database (e.g., Docket Navigator, Lex Machina) within the available searches, so I cannot state with absolute certainty that no suit was ever filed. However, based on the public sources searched and the patent's age (expired 2011), the absence of any hit identifying 5,274,434 as an asserted patent, I have no evidence of any litigation involving U.S. Patent No. 5,274,434. If you need a definitive negative, a paid docket service (Docket Navigator/Lex Machina) or a PACER/Unified Patents case search keyed exactly to "5,274,434" would be the confirming step.

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

Let me run two more targeted checks before concluding.

PTAB / AIA Trial Report — U.S. Patent No. 5,274,434

Patent: Method and apparatus for inspecting foreign particles on real time basis in semiconductor mass production line
Patent Owner: Hitachi, Ltd. (original assignee) · Inventors: Morioka, Noguchi, Ohshima, Kembo, Taniguchi
App. No. 07/778,363 · Filed 1991-10-17 · Granted 1993-12-28 · Earliest priority 1990-04-02 (JP 84805/90)
Legal status per source: Expired – Lifetime · Anticipated expiration: 2011-04-02
Report date: 2026-09-30

Proceedings overview

Total proceedings on file: 0. Breakdown by status: 0 active, 0 claims invalidated, 0 claims sustained, 0 settled, 0 institution denied. The USPTO Open Data Portal structured block returns no AIA trial proceedings for US 5,274,434, and four independent web searches (PTAB/petition-level queries, patent-number queries, and litigation queries) surfaced no IPR, PGR, or CBM referencing this patent. Bottom line for a defendant: there is no PTAB record to work from — and, critically, there is no § 315(e)(2) estoppel, no FWD, and no cancelled claim. This patent is not "hardened by surviving IPRs"; it is simply old and expired, and the entire prior-art universe remains un-adjudicated and available to you.

Why zero is the expected answer, not a data gap. The AIA trial regime opened for petitions on 2012-09-16. This patent's term ran from the 1990-04-02 earliest priority / 1991-10-17 filing and Google Patents records an anticipated expiration of 2011-04-02 — roughly seventeen months before the first IPR could be filed, and this patent is a pre-AIA, pre-2013 patent, so it was never PGR-eligible and its subject matter (semiconductor wafer inspection hardware) is not a "covered business method" under § 18 of the AIA. PTAB can in principle institute on an expired patent, but there was no commercial window here: no live injunction exposure, no ongoing royalty base, and therefore no economic reason for a competitor, a defensive aggregator, or an LP to fund a petition. No AIA trial was ever filed against 5,274,434.

⚠️ Identifier-confusion warning — flag this to your team. A search for "4-3-4" style patent numbers quickly surfaces U.S. Patent No. 5,247,434 — Baxter International's hemodialysis touchscreen patent — which was the subject of Fresenius USA, Inc. v. Baxter Int'l, Inc., 582 F.3d 1288 (Fed. Cir. 2009) (Fresenius I), 733 F.3d 1369 (Fed. Cir. 2013) (Fresenius II), and the parallel reexamination appeal In re Baxter. That is a different patent (note the transposed digits: 5,247,434 vs. 5,274,434) in an unrelated field, and it has no bearing on US 5,274,434. Do not import Fresenius' validity or mootness holdings into a 5,274,434 analysis. See Fresenius II, 733 F.3d 1369.

Per-proceeding detail

No proceeding sections are generated because there are no proceedings. Per the task constraint against fabrication, I am not inventing proceeding numbers, petitioner names, panels, or dispositions. There is no FWD to link, no institution decision to quote, and no Federal Circuit appeal docketed from a PTAB trial on this patent.

What was actually checked (all negative for 5,274,434):

Strategic summary

Claim status: everything is UNTESTED; nothing is CANCELED or SUSTAINED. Because no AIA trial ever reached a Final Written Decision, no claim of US 5,274,434 carries a PTAB validity adjudication in either direction. I did not verify the complete claim set from an authoritative full text (the Google Patents rendering supplied here reproduces the specification and abstract but not the claims), so I will not state a total claim count. The apparatus claims I could confirm from secondary sources include at least claim 19 (apparatus comprising transfer means, rotation-direction detection means, slit-shaped illumination, detecting means through a detecting lens, alignment means, and foreign-particle-state determining means) and claim 20 (claim 19 with a spatial filter on a Fourier transform plane and a shielding function). Anyone building a claim chart should pull the granted claims directly from USPTO Patent Center or the Google Patents full text rather than relying on this report.

Estoppel landscape: there is none, which is entirely to a defendant's advantage. Section 315(e)(2) estoppel attaches only to a petitioner that obtained an institution decision (§ 315(e)(1)) and is asserted in a later civil action or ITC proceeding. With zero petitions filed, no party is estopped, no ground has been "raised or reasonably could have been raised," and no § 325(e) bar exists. Practically, that means the full prior-art field described in the specification is open: JP-A-62-89336 (pattern-comparison laser scatter detection), JP-A-63-135848 (laser scatter plus photoluminescence/XMR analysis), the admitted prior-art stand-alone inspection tools, and the applicant's own admitted knowledge that repetition-pattern elimination and spatial-filtering of the Fourier plane were "known conventionally." Note that the patent itself repeatedly admits the spatial-filter/repetition-elimination technique into the prior art — an express-admission argument that is not foreclosed by any PTAB record.

Pattern signals: the family, not this patent, is where the PTAB risk lives. US 5,274,434 is the CIP-filial node of a long-lived continuation chain owned by Hitachi / Hitachi High-Technologies: 5,233,191 (parent, filed 1991-04-02) → 5,274,434 → 5,463,459 → 6,411,377 → 6,650,409 → 6,897,733 → 7,037,735 → 7,177,020 → 7,098,055 → 7,692,779 → 7,443,496 → 7,639,350 → 7,940,383, per the priority chain on Google Patents and the benefit clauses reproduced in US 7,037,735. Those later members were in force during the AIA era and are the realistic IPR targets in this family — a defendant facing a family-wide assertion should run the PTAB E2E check on each of those numbers, not just on 5,274,434. I found no evidence that the same petitioner filed multiple IPRs on 5,274,434 (impossible — zero filings), no aggressive PTAB-appeal posture by Hitachi on this patent, and no defensive aggregator (Unified Patents, RPX, or an LP) in the chain.

Recommended next steps

1. If you are being accused of infringing US 5,274,434, lead with expiration, not validity. The patent expired 2011-04-02. Any demand letter asserting it today can only reach pre-expiration conduct within the § 286 six-year damages look-back, which from 2026-09-30 reaches back only to 2020-09-30 — a date five years after expiration. There is no recoverable damages window. That is a complete practical defense and should be your first response, before spending a dollar on invalidity contentions.
2. Verify status yourself and preserve the record. Confirm the expiration and fee/term status at USPTO Patent Center and the expiration date at Google Patents, and confirm the absence of AIA trials directly at PTAB E2E. Absence-of-proceeding evidence is worth documenting contemporaneously.
3. Do not expect to use an IPR as your vehicle. A petition against an expired patent offers you no stay leverage (no co-pending district court case can be stayed by a patent that expired fifteen years ago) and no meaningful estoppel benefit. Any invalidity attack belongs in the district court or as an affirmative defense, not at the Board.
4. Distinguish the transposed number in every internal memo. If your search results, docket research, or an adversary's brief cites Fresenius or "the '434 patent" in the hemodialysis context, that is 5,247,434 (Baxter), not 5,274,434 (Hitachi). Correct it on the record early; a confused record on the patent's identity is easy to exploit.
5. If the assertion is family-wide, pivot. Screen each live sibling (especially 6,411,377, 7,037,735, 7,443,496, 7,639,350, 7,692,779, 7,940,383) on PTAB E2E for actual IPRs, institution decisions, and FWDs, and check their expiration dates individually. Those, not this patent, are where IPR-based defenses will be won or lost.

Known limitations

• No AIA trial proceedings exist for this patent, so there are no FWD, institution-decision, panel, or CAFC-appeal citations to provide. I have not manufactured any.
• I could not verify the complete granted claim set from an authoritative source; claim 19 and claim 20 are the only claims I could independently confirm, and only from a secondary reproduction.
• I found no litigation, ITC action, or reexamination involving US 5,274,434. If a defendant holds a complaint or demand letter referencing this patent, that document is the best available evidence of current assertion activity and is not reflected in any public source I could reach.

I'll research the assignment record for US 5274434 across the sources you specified.

Inventors

Five named inventors, all listed on the face of US 5,274,434:

Basis for the employer inference: the sole recorded assignment on the Google Patents legal-events record for this patent is a filing-date "ASSIGNMENT OF ASSIGNORS INTEREST" listing all five inventors as assignors and HITACHI, LTD. A CORP. OF JAPAN as assignee. An inventor→company assignment executed at filing is the standard indicator of an employment/obligation-to-assign relationship. I could not determine the specific Hitachi division or works (Hitachi's wafer-inspection work of this era sat across Hitachi, Ltd. Central Research Laboratory, Hitachi Electronics Engineering, and later Hitachi High-Technologies).

Unusual-pattern check: Not determinable from the sources retrieved. I found no evidence of inventor departures from Hitachi within 12 months of filing, and no evidence of a portfolio fire-sale. I am explicitly flagging this as unverified rather than negative — the Assignment Center index does not carry inventor-employment timing data, and I could not retrieve the underlying assignment instrument.

Original assignee

Hitachi, Ltd. (Tokyo, Japan) — named as assignee on the issued patent and as the assignee in the only recorded assignment.

• Primary line of business: diversified electrical/electronics manufacturer. The relevant division for this patent is semiconductor manufacturing and inspection equipment — the patent's own FIG. 2 / FIG. 3 describe mounting a foreign-particle monitor directly into the loader of a sheet-type CVD apparatus. Hitachi's inspection-tool business was subsequently consolidated into Hitachi Electronics Engineering Co., Ltd. and then into Hitachi High-Technologies Corporation, which was spun off from Hitachi, Ltd. in 2001 and re-absorbed as a wholly-owned Hitachi subsidiary in 2020 (now Hitachi High-Tech Corporation).
• Did they ship a product embodying the claims? Likely, but not confirmed at claim-mapping level. Hitachi and its Hitachi High-Tech line commercialized wafer-surface / particle inspection systems (the LS-/IS-/RS-series inspection tools). I could not, within this research pass, confirm a specific model's claim chart to US 5,274,434. Treat "shipped a product" as probable, unverified rather than established.
• Current status: Operating. Hitachi, Ltd. is a live, publicly listed operating company (TSE: 6501). There is no bankruptcy, dissolution, or assignment-for-the-benefit-of-creditors event in the record for this patent.

Assignment timeline

⚠️ Important sourcing limitation, stated plainly: I was not able to retrieve USPTO Patent Assignment Center reel/frame records for this patent. The Assignment Center search index did not return a hit for patent number 5,274,434 in the searches I ran, and Google Patents does not expose reel/frame numbers in its legal-events listing. I therefore cannot supply reel/frame identifiers, execution dates, recording dates, or correspondent-of-record names below without fabricating them, which I will not do. What follows is limited to what the authoritative patent text and Google Patents legal events actually establish.

• 1991-10-17 (date of filing; execution date not exposed) / recorded on or about 1991-10-17 — Reel/Frame not retrievable
  • Conveyance: Assignment ("ASSIGNMENT OF ASSIGNORS INTEREST")
  • Assignor: Kembo, Yukio; Morioka, Hiroshi; Noguchi, Minori; Oshima, Yoshimasa; Taniguchi, Yuzo
  • Assignee: Hitachi, Ltd. (Hitachi, Ltd. a corp. of Japan)
  • Correspondent: not retrievable from the sources retrieved — cannot be flagged for recurrence
  • Context: routine employment/inventor-to-company assignment executed contemporaneously with the filing of the CIP; not an acquisition, fire-sale, or securitization.

No post-issuance assignment was found. Google Patents' legal-events record for US 5,274,434 lists, after the 1993-12-28 grant, only continuation-application priority links — US 08/046,720 (→ US 5,463,459), US 08/617,270 (→ US 6,650,409), US 09/362,135 (→ US 6,411,377), US 09/805,188 (→ US 6,894,773), US 10/170,378 (→ US 7,037,735), US 11/128,352 (→ US 7,177,020), US 11/244,078 (→ US 7,098,055), US 11/244,080 (→ US 7,692,779), US 11/681,996 (→ US 7,443,496), US 11/681,981 (→ US 7,639,350), US 12/630,307 (→ US 7,940,383). Those are filings, not conveyances — they are new applications by the same original assignee, all still recorded to Hitachi. Do not mistake them for a transfer chain.

The patent's status is Expired – Lifetime, with anticipated expiration 2011-04-02 (20 years from the earliest priority date, JP 1990-04-02).

Timeline diagram

timeline
    title Ownership of US 5274434
    1990 : Priority applications filed in Japan
    1991 : US application filed as a CIP
         : Inventors assign to Hitachi Ltd
    1993 : Patent US 5274434 issues
    2011 : Patent term expires

Note: the 1991 assignment year is date-of-filing based; the underlying execution date was not retrievable.

NPE / troll-pattern signals

1. Shell-entity transfer — Not present. No post-issuance assignment exists in the accessible record at all, so the patent never moved to an "IP/Holdings/Licensing/Ventures" entity. The only assignee ever named is Hitachi, Ltd., a publicly listed operating company. No registered-agent-service address, no single-member LLC.

2. Known asserter in the chain — Not present. Neither the original assignee nor any successor matches Acacia, Marathon Patent Group, Intellectual Ventures, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, or any Spangenberg-linked entity. My searches surfaced no RPX or Unified Patents listing of this patent number as an asserted asset.

3. Repeat correspondent across the chain — Unclear / not assessable. There is only one conveyance in the record, so the recurrence test cannot be satisfied by construction. The correspondent of record on that single 1991 recording could not be retrieved from the sources available to me, so I make no finding either way. On the merits, the single link in this chain is a filing-date inventor assignment handled by Hitachi's own prosecution channel — not the signature pattern of an NPE filing agent.

4. Cascading transfers — Not present. Zero transfers in the 18 years between issue (1993-12-28) and expiration (2011-04-02).

5. Pre-litigation transfer — Not present. No infringement suit naming this patent was located, and the patent term expired 2011-04-02, removing assertion value entirely.

6. Bankruptcy fire-sale — Not present. Hitachi, Ltd. never filed Chapter 7/11; it remains an operating listed company. No Kodak/Nortel/Polaroid-style sale proceeding touches this asset.

7. Privateering — Not present. No operating-company→NPE transfer is recorded, so there is no vehicle through which Hitachi could have asserted against competitors off-balance-sheet.

8. Defensive aggregator (anti-NPE) — Not present. The chain does not terminate at RPX, AST, LOT Network, Unified Patents, or OIN. Absence of a defensive-aggregator assignment is expected here given the patent simply stayed with Hitachi until expiry.

Related-family cross-check (negative finding): none of the twelve US continuation/divisional family members listed above shows an assignee change away from Hitachi in the legal-events data I retrieved. The whole family appears to have remained Hitachi-owned through expiry.

Verdict

Insufficient data — but with an affirmative negative finding on NPE status.

The record contains only the original inventor-to-assignee conveyance to Hitachi, Ltd. executed at filing on 1991-10-17 (reel/frame not retrievable), followed by issuance on 1993-12-28 and expiration on 2011-04-02. Because there are no post-issuance assignments, there are no reel/frame entries to cite for any of the eight NPE signals — signals 1, 2, 4, 5, 6, 7, and 8 are affirmatively not present, and only signal 3 is merely unclear because a single-link chain cannot demonstrate correspondent recurrence. This is the "no records" case the rubric anticipates, and it means the original assignee — an operating semiconductor-equipment manufacturer — retained ownership of an expired patent. No NPE involvement is supported by any evidence retrieved.

Verification links:

• USPTO Patent Assignment Search: https://assignmentcenter.uspto.gov/ (alternative portal: https://assignment.uspto.gov/patent/index.html) — search patent number 5274434
• Google Patents legal events: https://patents.google.com/patent/US5274434/en
• Predecessor application (CIP parent, US 5,233,191 / Ser. No. 07/679,317): https://patents.google.com/patent/[US5233191A](/patent/US5233191A)/en

Two caveats you should carry forward before relying on this: (a) my inability to pull a reel/frame record may reflect an index-coverage gap for pre-1998 grants rather than a true absence of post-issuance activity — a direct Assignment Center query and, if needed, a manual assignment-abstract pull would settle this definitively; and (b) the "shipped a product embodying the claims" determination for Hitachi is probable but claim-unmapped.

I'll research the prior-art citations for U.S. Patent No. 5,274,434 specifically, being careful not to conflate it with the similar-numbered 5,247,434.

Let me locate the specific citation lists (both the references cited during prosecution and the examiner's prior-art citations) for 5,274,434.

The citation list didn't surface in that pass. Let me target the USPTO/full-text record for the "References Cited" section of 5,274,434 specifically.

Prior Art Analysis — U.S. Patent No. 5,274,434

Important scope and evidence caveat (read first)

U.S. 5,274,434 (Hitachi, Ltd.; Morioka, Noguchi, Oshima, Kembo, Taniguchi; App. 07/778,363 filed 1991‑10‑17; priority 1990‑04‑02; granted 1993‑12‑28) is confirmed as the optical wafer foreign‑particle/particle monitor patent — not the similarly numbered Baxter 5,247,434 hemodialysis patent. Consistent with the litigation section already generated, I have kept those identifiers strictly separate.

What I was able to retrieve vs. what I could not: The authoritative full text you supplied (Google Patents US5274434A) contains the description, abstract, and drawing list, but the fetch omitted both the machine‑readable claim set and the front‑page "References Cited / (56)" table. My subsequent searches surfaced the "Cited By" lists (later patents that cite 5,274,434 as prior art) but not a reliable, verifiable OCR of the examiner's own cited‑references table for 5,274,434**. Rather than fabricate citation numbers or dates, I have:

1. Grounded the analysis in the prior art the specification itself expressly cites, plus family/related documentation that is verifiable from the sources returned; and
2. Explicitly flagged which entries I could not independently verify.

Where an exact publication date of a Japanese Kokai could not be confirmed from a retrieved source, I have so stated rather than guessing.

A. References expressly identified in the patent's own text

These are the only prior‑art references that the authoritative document itself names, so they carry the highest evidentiary weight for a §102 analysis.

A.1 — JP‑A‑62‑89336 (Japanese Patent Application Laid‑Open 62‑89336; Showa 62 = 1987)

• Full citation: Japanese Kokai (unexamined published application) No. 62‑89336; publication year 1987 (Showa 62). Exact month/day not verifiable from the sources retrieved — do not treat as confirmed.
• Filing/priority date: Japanese filing is necessarily earlier than the 1987 publication; the patent does not state it.
• What the patent says it discloses (verbatim, from the spec): "irradiates a laser beam on the wafer, detects scattering light from the foreign particles generated when they adhere to the wafer, compares this detection result with the inspection result of the same kind of the wafer that is immediately before inspected so as to eliminate lie detection [i.e., false detection] resulting from patterns and makes it possible to carry out high sensitivity and high reliability foreign particle inspection."
• Brief description: A laser‑scatter wafer‑surface particle inspector using die‑to‑die / same‑kind‑wafer comparison to cancel pattern‑induced false counts. This is the classic "comparison inspection" architecture.
• Claims potentially implicated under 35 U.S.C. § 102: The wafer‑comparison inspection subject matter. In 5,274,434 this maps to the FIG. 21 embodiment (first wafer image stored in memory 62; first and second memory images 1461/1462 compared by comparison circuit 1463). A §102 rejection would only reach apparatus/method claims reciting wafer‑to‑wafer comparison as the particle‑discrimination mechanism; it would not anticipate claims requiring the spatial‑filter‑on‑the‑Fourier‑transform‑plane element, the plane‑wave (parallel‑flux) elongated illumination, or the FFT‑based inter‑chip repetition‑elimination circuit, none of which JP‑A‑62‑89336 is described as containing. Claim numbers should be verified against the actual claim set before any mapping is relied upon.

A.2 — JP‑A‑63‑135848 (Japanese Patent Application Laid‑Open 63‑135848; Showa 63 = 1988)

• Full citation: Japanese Kokai No. 63‑135848; publication year 1988 (Showa 63). Exact month/day not confirmed from retrieved sources.
• Filing/priority date: Earlier than its 1988 publication; not stated in the patent.
• What the patent says it discloses (verbatim): "irradiates a laser beam on the wafer, detects scattering light from the foreign particles when they adhere to the wafer, and analyzes the detected foreign particle by analytical technique such as laser photoluminescence or XMR [X‑ray microanalysis/XRF]."
• Brief description: Laser‑scatter particle detection coupled to elemental/compositional analysis (laser photoluminescence; X‑ray microanalysis) of the detected particle.
• Claims potentially implicated under §102: This reference is an analysis/identification teaching, not a real‑time in‑line monitor. It could only anticipate claims expressly reciting analysis of the composition of a detected particle (a concept 5,274,434 assigns to the off‑line mass‑production‑start‑up system 81 / analysis unit 60 — SEM 61, SIMS 62, STM/STS 63, IR spectrometer 64), and even then only if a claim recited laser‑photoluminescence or X‑ray microanalysis specifically. It does not anticipate the on‑line real‑time-in-transfer‑system claims.

A.3 — U.S. Ser. No. 07/679,317 → U.S. Pat. No. 5,233,191 (Noguchi et al., Hitachi)

• Full citation: U.S. Patent 5,233,191, "Method and apparatus of inspecting foreign matters during mass production start‑up and mass production line in semiconductor production process," application Ser. No. 07/679,317, filed 1991‑04‑02; granted 1993‑08‑03; assignee Hitachi, Ltd. (inventors Noguchi et al.).
• Why it is important: 5,274,434 is a continuation‑in‑part of this application, and its disclosure is expressly incorporated by reference. The priority chain (US07/679,317 → US07/778,363) is confirmed in the Google Patents family data.
• §102 status — flag clearly: Because it is the CIP parent and is incorporated by reference, it is not ordinary prior art against claims of 5,274,434 that are fully supported by the parent. It could only function as §102(e) art for claims (or claim elements) not entitled to the 1991‑04‑02 parent filing date — i.e., new matter first added in the CIP. Note the different inventive entity (Noguchi et al. vs. Morioka et al.), though common assignee Hitachi, so the pre‑AIA §103(c) common‑ownership disqualification may bear on obviousness but not on straight §102 anticipation. This is a genuine gray area and should be resolved against the actual claim set and the parent's disclosure. I cannot state with confidence which specific claims this affects without the claim text and the parent's specification.

B. Family members (NOT prior art — cross‑checked for the record)

The Google Patents family data confirms these JP applications are priority/foreign‑priority members of the same invention, i.e., not available as prior art against 5,274,434:

C. "Cited By" material — NOT prior art to 5,274,434 (clarification)

Searches repeatedly returned 5,274,434 appearing inside the citations of later patents. These are forward citations (the later patents cite 5,274,434), so they are not §102 art against it. Listed to prevent confusion:

• US 5,623,340 (Hitachi) — lists 5,274,434 among its citations.
• US 5,463,459 (Morioka et al., Hitachi) — continuation‑line relative; itself cited as [XY] art in EP 1 075 652.
• US 5,658,390 (Inspex) — "detecting particles on a surface of a semiconductor wafer having repetitive patterns."
• US 5,854,674 (Optical Specialties) — spatial‑filter high‑speed repetitive/random pattern inspection.
• US 5,233,191 — the CIP parent (see A.3).
• The 5,604,134 (TSMC) listing and EP 1 058 111 / EP 1 075 652 search reports show 5,274,434 used as [X]/[Y] art against other parties' later applications, confirming its status as a landmark reference in this field — again, forward‑looking, not citation art against it.

D. Bottom‑line answers to the request

• Most relevant prior art for 5,274,434, on the record actually in hand:

  1. JP‑A‑62‑89336 (1987) — laser scatter + same‑kind‑wafer comparison for pattern‑false‑count cancellation. Highest §102 relevance, directed at the FIG. 21 comparison‑inspection claims.
  2. JP‑A‑63‑135848 (1988) — laser scatter + photoluminescence/X‑ray analysis of particles. Narrow §102 relevance (only to claims reciting particle composition analysis).
  3. U.S. 5,233,191 / Ser. No. 07/679,317 — CIP parent, incorporated by reference; not ordinary §102 art except possibly as §102(e) art against CIP‑only claims.

• Claims at risk: Without the verified claim text I will not assign specific claim numbers. The vulnerable subject matter is (i) claims whose novelty rests solely on wafer‑to‑wafer/same‑kind comparison (JP‑A‑62‑89336), and (ii) any claim element supported only by matter added in the CIP (as against U.S. 5,233,191 under §102(e)). The spatial filter on the Fourier transform plane, the plane‑wave/parallel‑flux elongated linear illumination, the FFT inter‑chip repetition‑elimination circuit, the orientation‑flat detecting/rotating‑direction detector, and the real‑time in‑transfer‑system placement are the features most likely to survive these particular references — they are the core of the granted claims.

• What I could not confirm (stated explicitly per operating rules):

  • The complete examiner "References Cited (56)" table for 5,274,434, including any U.S. patents cited during prosecution.
  • Exact publication dates for JP‑A‑62‑89336 and JP‑A‑63‑135848 (years inferred from the Showa‑era Kokai numbering; not independently verified).
  • The actual granted claim set of 5,274,434 (the authoritative fetch stopped within the description).
  • Any §102(e)/§103(c) disqualification analysis of U.S. 5,233,191, which requires the parent's specification and the CIP's added matter.

Recommended confirming step: pull the USPTO PatentCenter "References Cited" and full claim set for 5,274,434 (and the file wrapper for Ser. No. 07/778,363) — these are the authoritative sources for the examiner's citation list and the exact claim language needed to finalize any §102 mapping. I have not substituted the similarly numbered U.S. 5,247,434 or any other "'434" patent anywhere in this analysis.

I'll research the prior-art references cited on the patent's page and the actual claim language before building the § 103 analysis.

Let me retrieve the actual claim set and the patent's cited-references list.

I have partial claim text. Let me pin down the full claim set and the front-page references of US 5,274,434 itself.

Obviousness Analysis — U.S. Patent No. 5,274,434 under 35 U.S.C. § 103(a)

Patent: US 5,274,434 A — "Method and apparatus for inspecting foreign particles on real time basis in semiconductor mass production line" (Hitachi, Ltd.; Morioka, Noguchi, Ohshima, Kembo, Taniguchi). Filed 1991‑10‑17; priority 1990‑04‑02 (JP 84805/90); CIP of USSN 07/679,317 (now US 5,233,191); granted 1993‑12‑28; expired 2011‑04‑02.

Source of record: https://patents.google.com/patent/[US5274434A](/patent/US5274434A)/en (authoritative text supplied in the prompt). Claims retrieved from https://patents.justia.com/patent/5274434 and claim 19 from http://www.everypatent.com/comp/pat5274434.html.

Two housekeeping flags before the analysis:

1. Date conflict in the record. The task header states "Current Date: April 26, 2026," while the tool environment reports 2026‑09‑30. This has no legal significance for a § 103 analysis of a patent that expired in 2011, but the discrepancy is noted rather than silently reconciled.
2. The "Prior Art" section of the fetched page is incomplete in my copy. The Google Patents fetch truncated before the "References Cited" table and the claim set. The only prior-art references expressly identified on the page are the two Japanese publications discussed in the Background of the Invention — JP‑A‑62‑89336 and JP‑A‑63‑135848 — plus the applicant's own parent application, USSN 07/679,317. My analysis is anchored on those, supplemented by the specification's own admissions (which are prior art under MPEP 2129) and by references I could verify from the surrounding family/search record. Where I cannot verify a reference's disclosure, I say so rather than assume it.

I. Legal framework

Governing statute: Pre‑AIA § 103(a), because the application was filed 1991‑10‑17, well before the AIA's first‑inventor-to-file provisions took effect (March 16, 2013). Under pre‑AIA § 103(c), subject matter that qualifies as prior art only under § 102(e), (f), or (g) and that was commonly owned or subject to an obligation of assignment to the same person may not be used in a § 103 combination. This matters enormously here (see § IV.C below), because most of the U.S. patents in this crowded field were Hitachi's.

Graham factors: (1) scope and content of the prior art; (2) differences between the prior art and the claims; (3) level of ordinary skill; (4) objective indicia.

KSR v. Teleflex (2007): Obviousness is not limited to a formal "teaching, suggestion, or motivation" test. It is satisfied where (a) a known technique is used to improve a similar device in the same way, (b) a known technique is applied to a known device ready for improvement to yield predictable results, (c) a combination of familiar elements according to known methods yields predictable results, or (d) the combination is "obvious to try" over a finite number of identified, predictable solutions. Note KSR post‑dates this patent by 14 years, so in an ex parte reexamination today the Office would apply KSR to a 1991 priority date — a point of substantial significance for this patent, whose inventive core is architectural integration of known optical-inspection techniques.

POSITA for purposes of this analysis: a person with a B.S./M.S. in optics, applied physics, or electrical engineering and 2–5 years of experience in wafer-surface inspection or semiconductor process metrology, familiar with (i) dark-field/laser light-scatter particle detection, (ii) Fourier-optics spatial filtering and imaging of repetitive patterns, (iii) line-scan (linear CCD) imaging, and (iv) fab automation/wafer-handling equipment.

II. The claims to be analyzed

The claims break into method claims 1–18 and apparatus claim 19 et seq. Claim 1 recites:

transferring the product substrate along a path of transferring means provided at at least one predetermined position in the mass production line …; detecting a rotation direction of repetitive patterns of a circuit pattern being formed on the transferred product substrate; illuminating light of a substantially slit shape on the surface of the product substrate transferred along the path …; detecting with a detecting means information of a surface of the product substrate including detecting scattered light from foreign particles … through a detecting lens; aligning relatively the detected direction of the repetitive patterns of the product substrate and a direction of the detecting means; and determining a state of foreign particles on the product substrate transferred along the path … in accordance with the detection of the scattered light.

Claim 2 adds the spatial filter on a Fourier transform plane and alignment of the pattern direction to the filter direction. Claim 3 adds a photodetector array; claim 4, orientation-flat rotation-direction detection; claim 5, rotation control of the wafer; claim 6, zigzag slit illumination with overlap; claim 7, lens-array length ≈ wafer diameter; claim 8, monitor display; claim 9, calibration by measured refractive index; claim 10, focal depth 0.1–0.5 mm; claim 11, opposite-side illumination; claim 12, white-light oblique illumination; claim 13, deducing production-apparatus characteristics; claim 14, oblique illumination; claim 15, 2‑D coordinates from the detected rotation direction; claim 16, coherent light; claim 17, changeable spatial-filter pattern; claim 18, mass-production start-up sampling wafers fed to gas/water suppliers. Claim 19 recites the corresponding apparatus (transfer path, slit illumination, detecting lens, "means for enabling relative alignment of the detected direction of the repetitive patterns … and a direction of the detecting means").

⚠️ Provisional caveat: I retrieved claims 1–18 verbatim and claim 19 in part. I did not obtain the full official claim set (e.g., the tail of claim 19 and any claims 20+). Conclusions beyond claim 19 are therefore conditional.

III. The prior art of record

A. JP‑A‑62‑89336 (published 1987 — § 102(a)/(b) art)

The specification's own characterization (authoritative as an admission):

"irradiates a laser beam on the wafer, detects scattering light from the foreign particles … compares this detection result with the inspection result of the same kind of the wafer that is immediately before inspected so as to eliminate lie detection resulting from patterns and makes it possible to carry out high sensitivity and high reliability foreign particle inspection."

Disclosed: laser/light-scatter detection of on-wafer particles; elimination of pattern-induced false positives by reference/compare processing; high-sensitivity inspection of patterned product wafers.
Not shown by this characterization: in-transfer-path mounting; slit illumination; spatial filtering on a Fourier plane; explicit rotation-direction detection (although wafer-to-wafer image comparison presupposes angular registration of the two images).

B. JP‑A‑63‑135848 (published 1988 — § 102(a)/(b) art)

The specification's characterization:

"irradiates a laser beam on the wafer, detects scattering light from the foreign particles when they adhere to the wafer, and analyzes the detected foreign particle by analytical technique such as laser photoluminescence or XMR."

Disclosed: laser scatter detection + compositional/elemental analysis of the detected particle. Per the operating rules I read "XMR" literally; I note it is most likely an OCR/typographical rendering of a microanalysis acronym (e.g., X‑ray microanalysis) but I do not auto‑correct it.
Significance: this reference supplies the motivation for rapid, high-frequency detection — the value of detecting a particle rises sharply if you can identify what it is and therefore which tool/chemical/gas is the source.

C. USSN 07/679,317 / US 5,233,191 (parent; incorporated by reference)

Discloses the overall start-up-vs-production architecture and monitors 22 (in the transfer system) and 24 (inside the processing apparatus). Prior-art status is problematic for a § 103 attack on this CIP's claims:

• As the parent application of a CIP sharing the same inventive entity/assignee, it is not "by another" and therefore is not § 102(a)/(e) art against claims whose subject matter it supports — it is the priority document, not a reference. See also the EPO family-annex listing, which shows US 5,233,191 and US 5,274,434 as members of the same family (https://patentimages.storage.googleapis.com/85/13/e9/2ae8162ca658a0/EP0898300A3.pdf).
• US 5,233,191 issued 1993‑08‑03, after the 1991‑10‑17 filing, so it cannot be § 102(a)/(b) art either.
• To the extent it could be reached under § 102(e), pre‑AIA § 103(c) would bar its use in a § 103 combination (common ownership by Hitachi).
  Practical consequence: the parent is usable only to show what the art knew (and for claim 18's subject matter, to test written-description support), not as a § 103 reference.

D. Applicant-admitted prior art (AAPA) — highly damaging admissions embedded in the specification

These are prior art even though they appear in this patent (MPEP 2129; In re Fout):

E. Additional references verified in the surrounding record (used cautiously)

• US 5,410,400 (Shishido & Noguchi, Hitachi; filed 1992‑06‑23) — https://patentimages.storage.googleapis.com/3e/37/90/17cab687aad1e4/US5410400.pdf — discloses slant illumination, an imaging optical system with NA > 0.4, "shielding diffracted light from the circuit pattern by a spatial filter mounted on the Fourier transform plane," and linear detector imaging. Filed after the patent at issue, so not § 102 art to claim 1 — but decisive evidence that the spatial-filter-on-Fourier-plane architecture was the team's routine state of the art. Because it is commonly owned by Hitachi, it is also shielded under § 103(c) — i.e., usable as background, not as a combination reference.
• EP 0 898 300 A3 (Motorola) European search report — https://patentimages.storage.googleapis.com/85/13/e9/2ae8162ca658a0/EP0898300A3.pdf — cites US 5,274,434 itself as an "X" (single-reference) document against claims 1–10 of an in-situ robotic-track particle-monitoring application, and cites US 4,895,446 (Allemand, publ. 1990‑01‑23) as X against claims 1, 2, 9. This is objective evidence that examiners in this art treated line/spot illumination of a wafer with sensor-array detection as general background knowledge by the late 1990s.
• WO 00/68673 ISR — https://patentimages.storage.googleapis.com/ae/a0/f8/074472d1735142/WO2000068673A1.pdf — likewise cites US 5,274,434 as X for claims directed to "illuminating a line spot … collecting scattered light … at least one sensor array." Again: line-spot illumination plus linear sensor array was regarded as conventional. (The WO publication itself post-dates the patent; I cite it only as evidence of the art's baseline, not as prior art.)
• US 5,463,459 (Hitachi, 1995‑10‑31; priority to the same 1993‑04‑16 US filing per the Google Patents family listing) — https://www.freepatentsonline.com/[5463459](/patent/5463459).html — a continuation whose claims 17–30 mirror this patent's claim 1 subject matter, useful to confirm claim scope but not prior art.
• Sibling Hitachi reference list (US 5,410,400 front page) shows the field was dense by 1990: US 4,342,515 (Akiba), 4,669,875 (Shiba), 4,681,442 (Wagner), 4,816,686 (Hara), 4,922,308 (Noguchi), 5,046,847 (Nakata), 5,235,400 (Terasawa), plus ~14 Japanese publications including JP 63‑315936, JP 1‑117024, JP 1‑153943. ⚠️ I have not verified the individual disclosures of these documents, so I treat them as evidence of a crowded field rather than as named § 103 references.

§ 103(c) trap on the U.S. members of that list: US 4,922,308 (issued 1990‑05‑01, i.e., after the 1990‑04‑02 priority date) can at most be § 102(e) art; and because it is commonly owned by Hitachi, pre‑AIA § 103(c) disqualifies it from an obviousness combination. The same is true of US 4,816,686, US 4,669,875 (if Hitachi), and the other Hitachi U.S. members. Any credible § 103 attack must therefore rest on § 102(a)/(b) printed publications — which points squarely at the two Japanese publications the applicant itself cited.

IV. The obviousness combinations

Combination I — JP‑A‑62‑89336 in view of JP‑A‑63‑135848 (two-reference combination)

Motivation to combine (KSR/MPEP 2143 rationales):

• (C) Predictable result: Both references address the same problem (finding particles on product wafers) using the same physics (laser scatterometry); combining a detector with an analyzer is the archetype of "familiar elements according to known methods."
• (B) Known technique improving a similar device: JP‑A‑63‑135848 supplies the reason to detect particles more often and faster — a particle you can identify tells you which chamber/gas/chemical to fix. That is a direct motivation to raise sampling frequency, which JP‑A‑62‑89336's compare-to-previous-wafer scheme permits without a stored golden reference.
• (F) Design incentive / market pressure: Yield economics. The specification itself concedes that "if a long time passes away from the generation of the foreign particles to the detection … a fabrication yield drops."
  Assessment: These two alone do not reach claim 1 (no slit illumination, no in-path transfer, no express rotation-direction/alignment step), but they occupy most of the problem space and supply the motivation layer for every wider combination below.

Combination II — JP‑A‑62‑89336 + the admitted, well-known Fourier-plane spatial filtering of repetitive patterns (as evidenced by the patent's own AAPA), optionally + JP‑A‑63‑135848

This is the strongest § 103 theory for claim 2 (and the theory most likely to have succeeded in a reexamination).

Motivation (rationale (B)/(C)/(E)/(G)):

• Obviating the reference wafer. JP‑A‑62‑89336's compare-to-immediately-preceding-wafer scheme requires retention of a full reference image and a same-kind preceding wafer; a Fourier-plane filter achieves the same pattern-rejection without a reference and without the wafer-to-wafer latency. Substituting a cheaper, faster known technique for an equivalent known technique is textbook obviousness.
• Throughput/memory. The motivation to eliminate the full-frame image memory (and the compute cost of a compare) is a recognized engineering incentive that, if followed, leads predictably to optical filtering.
• Enablement of single-pass inspection of memory devices. The specification concedes that repetition density is "at least 80% in DRAMs" — for which a repetition-eliminating filter is the natural choice.

On the "rotation-direction detection/alignment" element: the applicant will argue this is the point of novelty. Rebuttals:

1. The specification itself equates "detection of the rotation direction of the repetitive patterns" with orientation-flat and notch detection (claim 4; and FIG. 4's illumination-point method). Orientation-flat/notch prealignment is ubiquitous in wafer handling (steppers, tracks, spin-coaters) and was in 1990 old and well-known. Nothing more than routine prealignment is claimed.
2. Where a spatial filter must be angularly registered to a grating, the necessity of aligning the filter to the pattern's azimuth is a physical inherency of using a spatial filter at all — not an inventive step.
3. Reasonable expectation of success is high: the alignment is a mechanical/control task with no unknown physics.

Combination III — JP‑A‑62‑89336 (+ JP‑A‑63‑135848) in view of the admitted conventional practice of on-line/transfer-system monitoring (AAPA + parent disclosure), to reach claim 1's "transferring along a path of transferring means … in the mass production line"

Motivation:

• (A)/(B) Same technique, known improvement, predictable result: Re-locating an existing measurement from an off-line station to the wafer's existing path removes a transport step. Any engineer optimizing sampling rate would first eliminate the round-trip. The result is entirely predictable — the optics do not change.
• (D) Design incentive / market pressure: The specification admits the industry problem in the applicant's own words — "real time sampling is preferably made for a mass production line," the stand‑alone apparatuses "are great in scale," and a large number of "large-scale inspection apparatuses" would be needed for a real-time monitor, which "is practically unfeasible." KSR treats precisely this posture — a recognized, admitted problem with a known engineering solution (miniaturize and relocate) — as evidence of obviousness, not patentability. The disclosure of the problem and the assertion that the solution is "compact and high-speed" is the invention's entire stated contribution.
• Additional admitted motivation: eliminating wafer withdrawal removes the new-particle contamination caused by removing and re-loading wafers (specification: "new foreign particles adhere onto the inspected wafer").

Combination IV — Reaching claim 1's "substantially slit shape" illumination and claim 3's detector array

⚠️ Honest limitation: JP‑A‑62‑89336 and JP‑A‑63‑135848, as characterized on the page, do not show slit-shaped illumination. A complete § 103 case against claim 1 requires one more reference (or the general-knowledge route). Candidates from the record, in descending order of usefulness:

1. Line-scan / linear-array dark-field inspection as general knowledge, corroborated by the fact that two separate examiners (EPO 2001, EPO 2000) cited this very patent as an X reference for claims directed to no more than "illuminating a line spot … at least one sensor array" (EP 0 898 300 A3; WO 00/68673 ISR). If the claimed combination of line illumination + linear sensor was, by 1999, being treated as disclosed by a single 1993 document's FIG. 7, it is a fortiori a combination of background elements.
2. US 4,895,446 (Allemand, publ. 1990‑01‑23) — cited as X against in-situ particle-detection claims in EP 0 898 300 A3. ⚠️ I have not verified its disclosure; it appears, from its citation context, to concern wafer surface inspection. Do not rely on it without pulling the document.
3. JP‑A‑63‑315936 / JP 1‑117024 / JP 1‑153943 and the other pre‑1990 Japanese publications in the Hitachi sibling reference list. ⚠️ Unverified.

Motivation for line/slit illumination with a linear array: it is the only way to image a full wafer diameter in a single relative scan without a two-dimensional stage — a simple throughput argument (the specification itself makes exactly this argument at FIG. 9: "the entire surface … can be inspected in a lump by one scan"). Where the prior art discloses scanning and the claim recites a light shape chosen for scan efficiency, the choice is a design choice with predictable results — KSR; In re Rose; MPEP 2144.02.

Combination V — Dependent claims (day-one obviousness)

Combination VI — Priority-date attack (a predicate to Combinations I–V)

Before any of the above can be relied on, the effective date of each claim must be fixed. The Google Patents "Critical" priority date is 1990‑04‑02 (JP 84805/90), with further foreign priorities on 1990‑06‑26 (JP 2165743) and 1990‑10‑17 (JP 2276253), and a U.S. CIP filing on 1991‑10‑17.

• Any claim limitation not supported under § 112 ¶1 by the 1990‑04‑02 (or later) Japanese priority documents — e.g., the fully integrated FIG. 23–28 line-scan/multi-detector embodiment, the refractive-index-changeable lens-array spatial filter (FIG. 27), and the start-up sampling architecture of claim 18 — is entitled only to a later date.
• A later effective date moves several documents from "not prior art" to "prior art," and (equally important) moves documents already on the list from § 102(e)-only status into reach.
• This is the single highest-leverage issue in the validity of this patent and should be resolved first.

V. Objective indicia (the applicant's counterarguments, fairly stated)

1. Long-felt need / failure of others. The specification frames a genuine, admitted industry problem (sampling "one wafer … everyday," inadequate detection latency). If the record shows that competitors tried and failed to build in-line monitors, or that the industry long accepted the off-line sampling regime, that is probative nonobviousness (WBIP; Graham factor 4). I have no verified evidence of such a failure-of-others record and do not assert one.
2. Unexpected results. The applicant asserts an unexpected empirical finding — that particle counts increase "quite sudden[ly]" rather than gradually. If the prior art taught gradual accumulation and the applicants discovered abrupt events, that discovery is (a) a recognition of a problem rather than a solution, and (b) arguably not commensurate with the claims, which are apparatus/method claims to an inspection architecture rather than to any detection-of-abruptness technique. KSR and In re Montgomery counsel that an unrecognized problem's discovery does not by itself confer nonobviousness on a known solution.
3. Teaching away. None of the identified references teaches away from in-line monitoring; JP‑A‑62‑89336 and JP‑A‑63‑135848 are agnostic as to location. The absence of a teaching-away is fatal to any "but the art discouraged combination" argument.
4. Commercial success / industry adoption. In-line/in-tool wafer particle monitoring became a dominant architecture across the industry in the 1990s (this patent's own family was cited by KLA‑Tencor's later filings, e.g., https://patents.google.com/patent/JP4527205B2/en). If a nexus to the claimed features and a causal link to a Hitachi commercial monitor can be shown, this weighs for patentability. ⚠️ I have no verified sales/market-share data and do not rely on this.

VI. Conclusion

Claims 1–18 (method) and claim 19 (apparatus): On the prior art appearing on the page, the most credible § 103 attack is:

JP‑A‑62‑89336 (laser scatter detection of particles on product wafers with pattern-induced-false-detection elimination by comparison) in view of JP‑A‑63‑135848 (scatter detection plus particle analysis), further in view of the admitted, conventional repetitiveness-based pattern-discrimination art (spatial filtering of repetitive patterns, including programmable/liquid-crystal filters), further in view of the admitted conventional practice of on-line/transfer-system monitoring at a process tool's input/output port.

Motivations are supplied by: (i) the predictable-result rationale for joining a particle detector to a particle analyzer; (ii) the throughput/memory rationale for replacing wafer-to-wafer comparison with optical filtering; (iii) the admitted industry need for real-time sampling and the admitted infeasibility of achieving it with stand-alone, large-scale tools; and (iv) routine engineering optimization for the dependent claims (line illumination shape, detector array size, focal depth from the stated equation, white light vs. laser, programmable filter).

Honest gaps. Two claim-1 limitations are not shown by the two Japanese references as characterized on the page: (a) slit-shaped illumination and (b) detecting a rotation direction of repetitive patterns and aligning it with the detecting means. My Combination IV and Combination II address these only by general-knowledge/design-choice argument, which is weaker than a documentary showing. A confident, defensible invalidity position requires:

• the actual texts of JP‑A‑62‑89336 and JP‑A‑63‑135848 (Japanese; I could not retrieve or verify their full disclosures — my characterization is taken from the patent's own Background, which is authoritative as an admission but not as a complete disclosure);
• at least one pre‑April‑1990, non-commonly-owned printed publication showing line/slit illumination with a linear photodetector array (candidate: US 4,895,446, unverified);
• and a resolution of the effective filing date issue for the CIP-supported subject matter.

Confidence. High that the patent's own specification supplies powerful obviousness evidence (the AAPA admissions and the admitted-suboptimal-practice framing). Moderate that claims 1–3 and 19 would be held obvious on the art identified on this page alone. Low-to-moderate on the dependent claims requiring the unverified references. High on the § 103(c) common-ownership shield barring use of Hitachi's own U.S. patents in any § 103 combination — a point frequently overlooked in informal invalidity analyses of this family.

Not legal advice; not an opinion of counsel. Adverse facts may exist that I have not seen — in particular, the unretrieved portions of the Google Patents page (the full "References Cited" table, the "Cited By" table, and the official claim set), the prosecution histories of USSN 07/778,363 and its Japanese priority applications, and any IDS content. Also note that this patent expired on 2011‑04‑02, which limits — but does not entirely eliminate — the practical utility of an invalidity determination.