US 7309467

I'll ground this in both the authoritative patent text provided and live searches for USPTO/CAFC records.

Let me verify the claim set (the provided text is truncated before the claims) and check for any litigation activity.

I have good data on the specification and claim set. Let me confirm the independent claim text and the legal status/litigation picture.

US Patent 7,309,467 B2 — "Fluidic MEMS device"

Bibliographic data (from the authoritative patent text)

Abstract (verbatim)

"A fluidic micro electro-mechanical system (MEMS) device is described. In one aspect, at least one at least partially covered fluidic channel is formed between a polymer layer and a polymer substrate as the polymer layer is deposited on the substrate. The partially covered fluidic channel is fabricated as a unitary polymer layer structure. In one implementation, a strong exposure process is applied to the polymer layer to create a deep cross-linked polymer region. A weak exposure process is applied to the polymer layer to create a shallow cross-linked polymer region."

Plain-language overview of the independent claims

Important caveat on sourcing: the authoritative full text supplied to me ends mid-specification and does not include the printed claim set. My claim descriptions below come from the patent's own pre-grant publication (US 2004/0265182 A1, the publication of this same application 10/606/023), where claims 26–48 are visible. Those are the as-published claims; the issued claims may have been amended during prosecution, and I could not independently verify claim 1 (or claims 2–25) verbatim. Treat the claim summaries below as high-confidence in substance but not as verbatim issued-claim text.

• A pump-construction method (base claim appears at claim 26; claims 28–31 depend from "the method of claim 26") — a method including depositing a polymer material on a substrate to create a polymer layer. Dependents add: spinning the polymer layer to make it more planar; creating a resistor in the substrate for forming a bubble to create a pressure differential; creating first and second check valves by a strong-exposure cross-linking process; and fabricating some containment-portion parts with a strong exposure process and others with a weak exposure cross-linking process.

• Claim 32 — Pump apparatus: a fluid pump formed from a polymer layer and a substrate portion, where the polymer layer comprises (a) a first check valve including a deep cross-linked polymer region, (b) a second check valve including a deep cross-linked polymer region, and (c) a containment portion that together with the substrate encloses a fluidic channel — with the first check valve, second check valve, and containment portion all formed in the polymer layer as a unitary structure.

• Claim 33 — Integrated total chemical analysis system: a system fabricated on a substrate using a direct imaging process, comprising a portion of a deep cross-linked polymer region defining lateral fluid boundaries and a shallow cross-linked polymer region defining upper fluid boundaries, the deep and shallow regions forming a unitary structure. Dependent claim 34 requires at least two devices from the group of a filter, a pump, a waveguide, a PCR reactor, and a separator; claims 35–37 add that the deep region is made by a strong direct imaging exposure, the shallow region by a weak one, and that the system includes a fluidic MEMS device.

• Claim 38 — Fabrication method: fabricating, using a single process, a fluidic MEMS device on a polymer layer deposited on a substrate, by (i) defining lateral fluid boundaries with a strong direct imaging exposure process and (ii) defining upper fluid boundaries with a weak direct imaging exposure process. Dependent claims 39–43 add filtering, heating, separating, optically detecting, and pumping functions performed with the fluidic MEMS.

• Claim 44 — Method of making a reactor on a substrate: forming at least one heating element within or proximate to the substrate; depositing a polymer material on the substrate to create a polymer layer; and hardening portions of the polymer layer to create a containment portion formed as a unitary structure. (Note: dependent claims 46–47 as published refer back to "the first check valve and the second check valve" and to "the polymer layer" respectively — the claim-46 antecedent appears inconsistent with claim 44 in the published set, which is worth flagging as a drafting artifact rather than correcting.)

• Claim 48 — Reactor apparatus: a reactor formed from a polymer layer portion and a substrate portion, the polymer layer portion comprising a containment portion that together with the substrate encloses a fluidic channel, and a portion of at least one heating element applied to at least a portion of the fluidic channel, with the containment portion formed in the polymer layer as a unitary structure and with certain portions fabricated using a strong [exposure process — text truncated at the source I retrieved].

Technical gist in one sentence: Instead of aligning and anodically bonding two etched glass/silicon wafers to form fluidic conduits, the patent builds the fluidic channels, through-holes ("top-hat" structure), valves, filters, and reactor walls directly in a photosensitive polymer layer (SU-8, or ShinEtsu SINR-3010) on a single substrate, using a two-dose "direct imaging" scheme — a strong exposure that cross-links the polymer through its full depth to define lateral channel walls, and a weak exposure that cross-links only partway to define the channel's upper boundary — followed by development and cure.

Compliance check on your specific instructions

• USPTO records: The provided authoritative text is the Google Patents rendering of the USPTO record. It confirms the bibliographic data above. I did not independently query USPTO PatentCenter or PAIR in this session, so I cannot confirm current fee/status fields beyond what the fetched record shows.
• CAFC 2026 dockets: I searched specifically for "7309467" combined with CAFC/2026 docket and litigation terms and found no Court of Appeals for the Federal Circuit 2026 docket, and no district-court litigation, involving US 7,309,467. The only 2026 CAFC-adjacent hits returned were unrelated matters (e.g., ParkerVision v. Qualcomm, CAFC No. 26-1033; Tau Ceti Ventures v. HP, 1:26-cv-00716; Mobility Workx v. HP, 4:26-cv-00053), none of which assert this patent. I state explicitly that I cannot confirm the absence of litigation with high confidence — a negative search result is not proof of no suits, and I may lack authoritative coverage of every 2026 docket.
• No similar numbers substituted: Searches also surfaced a Brazilian corporate-registry entry "7309467" (Construtora Atlântida Ltda.) and a Lithuanian registry number "7309467" — these are not the patent and are flagged only to document that I noticed and excluded them.
• Literal interpretation: I have reproduced "Xia feng Yang" exactly as the patent record renders it rather than auto-correcting to "Xiaofeng Yang," and I have not normalized any identifiers.

I'll search for litigation records involving US patent 7309467.

Let me run additional targeted searches on litigation databases.

US Patent 7,309,467 — Litigation Search Results

Patent identification (as confirmed by the sources)

Sources: Google Patents US7309467B2, FreePatentsOnline 7309467, Justia patent listing, WO2011156847A1 family annex.

Finding: no litigation identified

I found no known litigation involving US Patent 7,309,467. My searches across patent-litigation-oriented sources did not surface any district court complaint, ITC Section 337 investigation, CAFC appeal, PTAB/IPR, or reexamination proceeding in which US 7,309,467 was asserted, challenged, or otherwise at issue.

Because the instruction is to report only matches for this exact patent, I want to be explicit about what did not count: my searches returned many hits for unrelated patents whose numbers merely end in "730" or are otherwise similar — e.g., Whirlpool's US 6,971,730 and US 7,309,980 (LG dispute, D. Del. 08-234 (GMS)); Avante's US 7,036,730 (Diebold/Election Systems); Stanford's '730 patent (Roche Molecular Systems, N.D. Cal.); and US 7,454,002 / 6,873,940 (Unified Patents IPRs). None of these is US 7,309,467, and none should be attributed to it.

Separately, the only HP-related enforcement activity that surfaced (ITC Inv. No. 337-TA-581, inkjet ink supplies, and the personal computer/monitor investigation at 70 Fed. Reg. — see USITC notice and Federal Register API record) asserted entirely different HP patents (e.g., 5,825,387; 6,793,329; 6,501,721; 6,803,865). US 7,309,467 was not among them.

Case table

Caveats and confidence level

• Confidence: moderate-to-high that no litigation exists, but not certain. Within the step budget I could not run a per-patent docket query directly against PACER, CourtListener/RECAP, Unified Patents' litigation portal, or the USPTO Patent Litigation Docket Report — my attempts to reach those specific per-patent queries (e.g., docketalarm "7309467", "7309467" Unified Patents portal) exceeded my search-step limit and returned no data. My conclusion rests on general web/patent-database searches that returned zero relevant hits.
• The patent's status is consistent with the absence of litigation: it is recorded as expired for failure to pay maintenance fees (adjusted expiration 2025-07-12), and it remained with its original assignee HP throughout — which reduces (but does not eliminate) the likelihood that it was ever asserted, since asserted patents are frequently transferred to assertion entities.
• Non-U.S. proceedings were not searched. The family includes EP 1635947 A2, JP 2007-526108, and CN 1809421; opposition or nullity actions abroad would not appear in U.S. litigation sources.
• Related continuation US 8,039,205 B2 could theoretically have its own enforcement history separate from the '467 patent; that is outside the scope of this query.

Recommended verification steps (to convert this to high confidence)

1. Query PACER Case Locator / CourtListener RECAP for the patent number "7309467" (patent-case search requires a PACER account for full coverage).
2. Run the USPTO Patent Litigation Docket Report (formerly "Patent Docket") at the USPTO Patent Center for patent 7,309,467, which compiles 35 U.S.C. § 271 complaints naming the patent.
3. Check the Unified Patents Portal (https://portal.unifiedpatents.com/litigation/caselist) and PTAB caselist for both US 7,309,467 and US 8,039,205.
4. Check HP's ITC/docket history for any post-2007 assertion (the two ITC investigations found predate or are unrelated to this patent).

I'll verify the structured-data finding with targeted web searches before writing up the analysis.

Let me run a few more targeted checks, including the related continuation and PTAB-specific databases.

Proceedings overview

Total AIA trial proceedings on US 7,309,467: zero (0). The structured PTAB-proceedings block — the canonical list from the USPTO Open Data Portal — returns no AIA trial proceedings for this patent, and my independent web searches returned no IPR, PGR, or CBM record either. The breakdown by status is therefore 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denied.

Bottom line defensive posture: neither of the two stock framings applies. This patent has not "survived IPRs and been hardened" (it has never been challenged at the Board), and no claims have been canceled — so a demand letter citing any of its claims is not vulnerable to a "those claims are already dead" rebuttal. All claims stand as issued at the USPTO. What a defendant actually gets is a clean slate: no § 315(e) estoppel has attached to anyone, and the full prior-art universe remains available in district court, in a new IPR, or in reexamination.

The absence is itself meaningful when read alongside the litigation analysis already generated for this patent: no district court, ITC, or CAFC litigation was identified either. A patent that was granted in 2007, held by HP, and never asserted in ~19 years is a patent that generates no infringement recoveries — and the whole point of an IPR is to kill a patent that is being used against you. The two findings (no litigation, no PTAB) are mutually consistent and reinforce each other.

(One structural note for the record: US 7,309,467 has a 2003 priority date and is pre-AIA. Its only realistic AIA trial vehicle was IPR; PGR is unavailable to pre-AIA patents, and CBM is limited to financial-services business-method patents, which this fluidic-MEMS patent is not.)

Proceedings detail

No proceeding entries are available to render. I am deliberately not producing the ### {PROCEEDING_NUMBER} — … blocks, because there is no proceeding number to insert, and inventing one would be fabrication. For completeness, the itemized fields that would normally follow — type, filing date, status, judge panel, grounds, institution decision, FWD, settlement, appeal, defensive value — are all N/A: no proceeding on file.

Strategic summary

Claim status — everything is untested and intact. No claim of US 7,309,467 has been canceled, confirmed, or even construed by the Board. There is no FWD to cite, no certificate of cancellation to point to, and no claim-level narrowing to work around. The claims as granted — including the independent claim(s) directed to the unitary polymer layer structure with deep/shallow cross-linked polymer regions, per the abstract and the FIG. 6a–6e / FIG. 7 disclosure — survive exactly as the examiner allowed them in 2007. CANNOT list surviving claims in the sense the template asks, because nothing was challenged; the correct statement is "all issued claims are live patentability questions that have never been adjudicated."

Estoppel landscape — nothing is barred. Because no petitioner ever reached an FWD, 35 U.S.C. § 315(e)(2) estoppel never attached to anyone: not to a petitioner, not to a real party in interest or privy, and not to any defendant. A defendant today can raise any § 102/§ 103 ground, in any forum, using any reference — there is no "raised or reasonably could have raised" bucket to police. Practically, this means an accused infringer has the full menu: file a fresh IPR on the best art, run parallel invalidity contentions in district court, and/or seek ex parte reexamination (available for this pre-AIA patent at any time).

Pattern signals — none of the usual ones are present. There is no repeat petitioner (there is no petitioner at all), no patent-owner appeal practice before the CAFC arising from a PTAB case, and no defensive aggregator (Unified Patents or similar) in the chain. I also found no indication that the related continuation US 8,039,205 B2 (from application 11/948,891) has been through an AIA trial — the searches surfaced only its Google Patents and Justia bibliographic pages, no Board papers. I could not, within my step budget, verify the '205 patent's PTAB status with the same confidence as the '467 patent's; treat that as a lower-confidence negative.

Caveats. (1) My conclusion rests on the ODP structured block plus general web/patent-database searches; I was not able to hit PTAB E2E, the PTAB Decisions database, or Docket Alarm directly for a per-patent query before exhausting my search steps, so I cannot rule out a very recently filed petition that ODP has not yet ingested — though for a fee-expired 2007 patent that scenario is remote. (2) I did not search ex parte reexamination or the discontinued inter partes reexamination records. Pre-AIA patents filed on or after 1999-11-29 (this one was filed 2003-06-24) were eligible for inter partes reexamination, and any patent is eligible for ex parte reexamination at any time. A reexam certificate would not appear in an AIA-trial query but would change the claim picture. If claims matter to your case, that check is worth running.

Recommended next steps

1. Treat this as a no-PTAB-activity patent, and say so in any invalidity workup. Do not assume the Board has given you a free claim construction or a helpful claim-cancellation certificate — nothing exists.
2. Verify the negative directly before relying on it: pull the PTAB E2E / PTAB Decisions record for 7,309,467 (and separately for 8,039,205), and run the USPTO Patent Center "Patent Litigation Docket Report" for the patent.
3. Check reexamination history (ex parte and the legacy inter partes reexam) via Patent Center's file wrapper for application 10/606,023. This is the one post-grant avenue I did not cover.
4. If you are a defendant facing a demand: the absence of any PTAB or litigation history, combined with the patent's expired-for-fee-nonpayment status (adjusted expiration 2025-07-12) and its retention by the original assignee HP, means the most likely reality is that you are dealing with a stale demand rather than a live campaign. Confirm current ownership/chain of title before spending on an IPR — but note that even for an expired patent, IPR remains technically available (expiry does not moot the trial), and a claim-cancellation certificate can still matter for back-damages exposure if a pre-expiry infringement theory is advanced.
5. If you are the patent owner and need to enforce: the clean record is a double-edged sword. You face no estoppel, but you also have zero Board validation to lean on, and the fee-expired status will invite a validity/standing fight.

I'll research the assignment record for US 7,309,467, starting with the USPTO Assignment Center and Google Patents legal events.

Let me try targeted assignments-database queries and the Google Patents legal-events record.

Let me make a few more attempts at the reel/frame record and the PEDS/legal-events trail.

Assignment & Ownership Analysis — US 7,309,467 B2 ("Fluidic MEMS device")

Data-access note (read first)

I attempted to reconstruct this chain from the USPTO Assignment Center and cross-referenced sources. I was able to confirm the existence, date, conveyance type, parties, and the absence of any further recordings, but I could not retrieve the reel/frame numbers, execution dates, or correspondents of record — the per-patent assignment query at assignmentcenter.uspto.gov / assignment.uspto.gov was not reachable within my tool budget, and no indexed third-party page republished the reel/frame for this patent. Rather than guess reel/frame digits, I am leaving those fields explicitly marked as not retrieved. Everything below that is stated as fact comes from the Google Patents legal-events record embedded in the authoritative patent text for US 7,309,467, or from the family/assignee records listed there.

Inventors

Pattern observations:

• Both inventors were HP employees at filing and assigned their rights to HP within roughly one month of the 2003-06-24 filing date — this is HP's standard, non-anomalous practice for employee inventions, not a departure signal.
• No "inventor exodus" data available. I could not determine whether either inventor left HP within 12 months of filing (that would require employment records or later patent-assignee analysis, which I did not retrieve). I note only that both names recur in later HP-assigned filings, which is inconsistent with an exit — but this is an impression, not a verified finding.
• No inventorship dispute, correction of inventorship, or inventorship-change record appears in the legal events for this patent (contrast with the unrelated Delta/Appoptronics US 9,024,241 inventorship litigation that surfaced incidentally in search and has nothing to do with this patent).

Original assignee

Hewlett-Packard Development Company, L.P. — named on the issued patent and still listed as current assignee.

• Identity / role: HP's patent-holding subsidiary, formed in the 2002 HP–Compaq reorganization (a Delaware LP headquartered in Houston, TX for registration purposes, with the IP administration function in Fort Collins, CO). It is the record owner of the large majority of HP's U.S. portfolio.
• Primary line of business: computing, printing/imaging, and related hardware and services. The technology here (SU-8 direct-imaging microfabrication, unitary polymer fluidic layers, on-chip pumps/filters/electrophoretic separators/PCR chambers) sits squarely in HP's inkjet-printhead and microfluidics R&D competence.
• Did they ship a product embodying the claims? Not confirmed. HP is an operating company with a large inkjet printhead business, and the disclosed fabrication approach (thick SU-8, two-level strong/weak exposure, unitary polymer channel layer over a substrate) is the type of process used in HP's MEMS/printhead manufacturing. However, I did not retrieve any product mapping, claim chart, or HP marketing/SEC statement tying a shipping product to these claims, so I record this as likely but unverified, not as a finding.
• Current status: Operating. HP Inc. and Hewlett Packard Enterprise separated in November 2015; HP Development Company, L.P. continues as an IP-holding entity. No bankruptcy, dissolution, or receivership event exists for HP or for this patent's chain.

Assignment timeline

One — and only one — assignment is recorded against US 7,309,467:

• Recorded 2003-07-21 (execution/signing date not retrieved) — Reel/Frame not retrieved
  • Conveyance: Assignment — recorded event text: "ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)."
  • Assignor: YANG, XIAOFENG; CHEN, CHIEN-HUA
  • Assignee: HEWLETT-PACKARD DEVELOPMENT COMPANY, L.P.
  • Correspondent: not retrieved (this is precisely the field I was unable to pull; see the data-access note)
  • Context: Initial assignment of the inventors' rights to their employer, recorded ~4 weeks after the 2003-06-24 filing. Ordinary course; not a fire-sale, reorg, securitization, or asserter transfer.

No other recorded events. The Google Patents legal-events record for this patent shows only three ownership-relevant entries across its entire life:

There is no change-of-name recording, merger recording, security agreement, license recordation, release, or correction, and — critically — no transfer out of HP at any point before expiration. The "Current Assignee" field still returns Hewlett-Packard Development Co LP.

Family note (continuity, not assignment): the priority chains list a continuation application US 11/948,891 (filed 2007-11-30) which issued as US 8,039,205 B2. Continuity links are not assignments, but that sibling case has its own independent assignment history and would need to be checked separately if the goal is coverage of the whole family.

Benchmark for what an HP chain normally looks like (different patent — do NOT attribute to '467): an Espacenet INPADOC legal-status page retrieved during this session for a different HP patent, US 6,934,943, shows HP's typical multi-link pattern — HP Company → HP Development Company L.P. (reel 014061/0492), later an HPE-side transfer (reel 037079/0001 for Hewlett Packard Enterprise Development LP), and subsequent security-interest / patent-escrow recordings (reels 055269/0001 and 056157/0492). None of those reels belongs to US 7,309,467, and none should be reported as such. They are cited only to show that when HP does record post-issuance transfers, they are visible — and here they are absent.

Timeline diagram

timeline
    title Ownership of US 7309467
    2003 : Application filed by HP
         : Inventors assign rights to HP
    2007 : Patent granted to HP
         : Continuation filed by HP
    2025 : Patent lapses for non payment

NPE / troll-pattern signals

Verdict

Insufficient data — in the rubric's sense of "only the original assignment."

Justification: the entire recorded chain is a single inventive-entity assignment, recorded 2003-07-21 (Yang, Xiaofeng and Chen, Chien-Hua → Hewlett-Packard Development Company, L.P.) off a 2003-06-24 filing, with the assignee unchanged through the patent's 2025-07-12 fee-related lapse. There is no second link to evaluate, so no shell transfer, asserter, cascading LLC chain, pre-suit transfer, or defensive-aggregator endpoint can exist; the substantive conclusion is that no NPE pattern is present — the patent never left its original operating-company owner. The verdict is "insufficient data" rather than "defensive/non-asserting" only because the term of art in this rubric for a defensive outcome requires a chain terminating at an aggregator, which is not what happened here.

Confidence caveat: my inability to pull reel/frame and correspondent from the USPTO Assignment Center is the main gap. It does not change the conclusion (the number of links and the parties are documented), but it means the reel/frame citations this task asked for could not be produced.

Verify at: USPTO Assignment Center (search patent number 7309467) · Google Patents US7309467B2 — Legal Events · USPTO Patent Center for the file history (assignment papers are in the IFW).

Suggested follow-ups: (1) pull reel/frame and the correspondent from the Assignment Center interface; (2) run the same reconstruction on US 8,039,205 B2 (from US 11/948,891); (3) confirm the 2003-07-21 recording's execution date from the assignment document itself, since a signing date materially before the filing date (i.e., pre-filing assignment) would be worth noting.

I'll research the prior-art citations of record for US 7,309,467.

I found the family's PCT search report citations. Let me verify what appears specifically on the US 7,309,467 front page.

The Justia record gives the actual front-page cited references. Let me confirm the complete list and gather details on each reference.

Prior-Art Analysis — US Patent 7,309,467 ("Fluidic MEMS device")

0. Scope, sourcing, and a hard caveat up front

What I was able to retrieve. The authoritative full text supplied to me for this patent ends mid-specification and contains no front page and no printed claims. So the prior-art citation record below does not come from the authoritative text — it comes from three independent third-party renderings of the USPTO record:

I did not directly query USPTO PatentCenter/PAIR or the USPTO full-text search in this session, so I cannot personally vouch that the citation lists are exhaustive. That is an important limitation, stated plainly.

Second hard caveat — citations ≠ §102 rejections. A reference appearing on a patent's face or in an IDS/PCT search report is not proof that the examiner considered it anticipatory. Most of these are almost certainly category "A" background art or §103 material. I have not retrieved the actual examiner's rejections, office actions, or the PCT search report's category letters (X/Y/A) for this case, so the §102 mappings below are my own element-by-element analysis, not a record of an official rejection. Where I say "potentially anticipates," I mean "contains elements touching that claim," and I flag where an element is missing.

Third caveat — claim numbering. The printed claims are unknown to me. The claim numbers I cite are the as-published claims of US 2004/0265182 A1, which I can see run to at least claim 54 and include independent claims 26, 32, 33, 38, 44, 48, 49, 53 and 54. The earlier analysis in this file already flagged that claims 1–25 and the issued claim text could not be verified verbatim. That flag stands and applies to everything below.

Numbers explicitly excluded (per the "no similar numbers" rule): US 7,091,730 (Seagate/Parshotam, cited in an unrelated WO); US 6,971,730 and US 7,309,980 (Whirlpool); US 7,036,730 (Avante); and "7309467" as a Lithuanian corporate registry code and a Brazilian registry number. None of these is US 7,309,467.

1. The patent under analysis (re-confirmed)

Because the application was filed 2003-06-24 and (so far as the record shows) not accorded an earlier priority, the §102 categories below are keyed to that date: a reference published on or before 2002-06-24 is §102(b) art; a reference published after that date is §102(a)/(e) art only.

2. The complete citation record, merged and source-attributed

The three renderings do not fully agree, and that disagreement is itself the finding. Merging them:

Discrepancy to flag explicitly: Justia's front-page block shows the foreign items and the Daniel NPL but its "U.S. Patent Documents" section is elided in the retrieved snippet; Google Patents' entry for the pre-grant publication shows a 4-item US citation list (US 5,375,979; US 2002/0117517; US 2002/0098097; US 2003/0007715); the PCT ISR shows 10 items and adds US 2003/0017467 and all five PCT/EP documents. I therefore cannot state with confidence whether US 2003/0017467 A1 appears on the US face. This is a contradiction between my sources, not something I will paper over.

3. Reference-by-reference §102 analysis

Reminder of the two load-bearing limitations that most of this art lacks: (a) the fluidic containment material, channel, and through-holes being formed in a single polymer layer as a unitary structure on one substrate, and (b) the two-dose direct-imaging scheme (a strong exposure cross-linking through the full polymer depth to define lateral boundaries, plus a weak exposure cross-linking only partway to define the upper boundary).

3.1 US 5,375,979 A — Robert Bosch GmbH

• Full citation: US 5,375,979 A, "Thermal micropump with valves formed from silicon plates," assignee Robert Bosch GmbH. Priority 1992-06-19; issued 1994-12-27. (Google Patents renders the title with a typo, "values"; I reproduce the correct "valves" but note the record's rendering.)
• §102 category: §102(b).
• Description: A micromachined thermal (bubble-generation) micropump in which the inlet/outlet check valves are themselves formed from silicon plates, with a heater creating a vapor bubble that drives flow through one-way valves.
• Potentially anticipatory of: claims directed to the bubble/resistor pumping mechanism and to a pump having two one-way check valves — i.e., published claims 29 ("creating a resistor in the substrate, for forming a bubble to create a pressure differential"), 26 (pump-making method), 32 (pump apparatus), and 43 ("pumping fluid with the fluidic MEMS"). This is the strongest §102 candidate against the bubble-generating resistor sub-limitation.
• Where it fails as anticipation: claim 26/element of claim 32 require the pump to be formed in a polymer layer as a unitary structure. Bosch's device is silicon-plate based. A single-reference anticipation of claims 26/32 therefore fails on that element; it is better characterized as §103 art against the bubble/check-valve concept.

3.2 US 2002/0098097 A1 — Angad Singh

• Full citation: US 2002/0098097 A1, "Magnetically-actuated micropump," inventor Angad Singh. Filed 2001-01-22; published 2002-07-25.
• §102 category: §102(a)/(e) (published 31 days after the one-year bar date).
• Description: A micropump driven by magnetic actuation of a diaphragm/plunger rather than thermal bubble or piezoelectric drive.
• Potentially anticipatory of: the generic pump apparatus concept of claim 32 and the pump-making method of claim 26 — but only at the level of "a micropump with valves and a pumping chamber."
• Where it fails: no polymer-layer unitary structure, no SU-8/photoimagable-polymer direct imaging, and a different actuation principle from the bubble resistor of claim 29 or the piezoelectric alternative described in the specification. Anticipation is unlikely; §103 at best.

3.3 US 2002/0117517 A1 — Fluidigm Corporation

• Full citation: US 2002/0117517 A1, "Microfluidic devices for introducing and dispensing fluids from microfluidic systems," Fluidigm Corporation. Filed 2000-11-16; published 2002-08-29.
• §102 category: §102(a)/(e).
• Description: Microfluidic devices with interface structures for introducing and withdrawing fluids from a microfluidic network — i.e., fluidic I/O ports and interfaces to the outside world.
• Potentially anticipatory of: claims directed to fluid I/O / through-hole interfacing — the FIG. 17–20 through-hole 126 and connector portion 2002 subject matter — and, at a generic level, claims 26 and 32 (devices formed of a polymer layer on a substrate with an enclosed channel).
• Where it fails: Fluidigm's systems are multilayer elastomer (soft-lithography) devices; there is no "deep cross-linked polymer region" vs. "shallow cross-linked polymer region" formed by strong/weak direct imaging.

3.4 US 2003/0007715 A1 — Hans-Peter Loock

• Full citation: US 2003/0007715 A1, "Optical loop ring-down," inventor Hans-Peter Loock. Filed 2001-05-29; published 2003-01-09.
• §102 category: §102(a)/(e).
• Description: Fiber-loop ring-down optical detection — an optical-cavity measurement technique for detecting analytes.
• Potentially anticipatory of: only the optical detection functionality, i.e., published claims 42 ("optically detecting material in a fluid using the fluidic MEMS") and the FIG. 16 optical waveguide 54 subject matter. Note claims 42 and 39–43 are dependent method claims; the reference cannot anticipate them alone because they incorporate all limitations of claim 38 (single-process fabrication in a polymer layer with strong and weak exposures).
• Where it fails: it is a measurement technique, not a fluidic MEMS fabrication. Weakest reference of the set as to structure.

3.5 US 2003/0017467 A1 — Aclara Biosciences, Inc.

• Full citation: US 2003/0017467 A1, "Multiple-site sample-handling apparatus and method," Aclara Biosciences, Inc. Filed 2000-02-18; published 2003-01-23.
• §102 category: §102(a)/(e).
• Description: Apparatus and methods for handling samples at multiple sites — an integrated multi-function microfluidic platform.
• Potentially anticipatory of: the integration concept of published claim 33 and, more specifically, claim 34 ("at least two devices from the group of a filter, a pump, a waveguide, a PCR reactor, and a separator") — i.e., the "lab-on-a-chip" plurality limitation.
• Where it fails: claim 33 still requires the deep cross-linked polymer region defining lateral fluid boundaries and the shallow cross-linked polymer region defining upper fluid boundaries forming a unitary structure. Aclara's art is directed at multi-site handling, not at dual-dose polymer cross-linking depth. §103 material against the integration idea.

3.6 WO 98/45693 A1 — Aclara Biosciences

• Full citation: WO 98/45693 A1, "Methods for fabricating enclosed microchannel structures," Aclara Biosciences. Priority/filed 1997-04-04; published 1998-10-15.
• §102 category: §102(b).
• Description: Methods for making enclosed microchannel structures — i.e., covering and sealing microfabricated channels to yield closed fluidic conduits.
• Potentially anticipatory of: published claims 38 (fabricating a fluidic MEMS device on a polymer layer deposited on a substrate) and the unitary enclosed structure limitation in claim 33. This is one of the two most relevant references in the set, because the patent's stated advance over the prior art is precisely the elimination of the align-and-bond two-substrate step in favor of one-piece enclosure.
• Where it fails as §102: claim 38 requires both the strong-exposure lateral-boundary step and the weak-exposure upper-boundary step. WO 98/45693 does not disclose depth-differentiated cross-linking in a photosensitive polymer. So no literal anticipation of claim 38; strong §103 candidate combined with, e.g., 3M WO 01/96958.

3.7 WO 01/38844 A2 — Motorola, Inc.

• Full citation: WO 01/38844 A2, "Capillary electrophoresis devices incorporating optical waveguides," Motorola, Inc. Priority 1999-11-12; published 2001-05-31. US counterpart US 6,592,733 B1 (2003-07-15).
• §102 category: §102(b).
• Description: Capillary electrophoresis devices with integrated optical waveguides — combining electrophoretic separation with on-chip optical interrogation.
• Potentially anticipatory of: published claims 41 ("separating fluid with the fluidic MEMS"), 42 (optical detection), 33 (integrated analysis system), and the FIG. 15 separator 52 / FIG. 16 waveguide 54 combination. This is the single most thematically relevant reference — it discloses the very functional combination (electrophoresis + waveguide on one substrate) that the '467 specification presents as its "integrated total chemical analysis system."
• Where it fails as §102: it does not disclose the polymer-layer unitary structure or the two-dose exposure scheme. Anticipation of claim 33 specifically fails on the "deep cross-linked polymer region"/"shallow cross-linked polymer region" elements. §103 candidate, and a strong one for the separator + waveguide dependent claims.

3.8 WO 01/96958 A2 — 3M Innovative Properties Company

• Full citation: WO 01/96958 A2, "Process for producing microfluidic articles," 3M Innovative Properties Company. Priority 2000-06-15; published 2001-12-20. EP 1295180 B1 granted 2013-05-22.
• §102 category: §102(b).
• Description: A process for producing microfluidic articles — i.e., fabrication of microfluidic structures in/on polymer layers (lamination/patterned-layer approaches).
• Potentially anticipatory of: the process claims — published claim 26 (depositing a polymer material on a substrate to create a polymer layer), claim 38 (single-process fabrication on a polymer layer), claim 44 (depositing polymer and hardening portions to create a containment portion), and claim 54 (filter-making method). Because it is a generic "make a microfluidic article in a polymer layer" teaching, it is the highest-risk reference against any of the unverified independent claims 1–25 that may be drafted broadly as "polymer layer on a substrate enclosing a channel."
• Where it fails: it does not appear to disclose a strong vs. weak exposure producing different cross-link depths in the same polymer layer, which is the core novelty hook of the '467 disclosure. I could not verify the internal content of WO 01/96958 in this session, so if it in fact discloses depth-differentiated exposure, my "fails" statement would need revision — I flag that as an open verification item rather than asserting it.

3.9 WO 02/43615 A2 — California Institute of Technology

• Full citation: WO 02/43615 A2, "Microfabricated elastomeric valve and pump systems," California Institute of Technology. Priority 2000-11-28; published 2002-06-06. US counterpart US 6,899,137 B2 (2005-05-31).
• §102 category: §102(b) (published 18 days before the bar date).
• Description: The multilayer elastomeric (soft-lithography) microvalve and micropump architecture — valves and pumps built as stacked elastomer layers bonded to a substrate.
• Potentially anticipatory of: published claim 32 (pump apparatus: first check valve, second check valve, containment portion, unitary structure) and claim 30 (check valves created by a strong-exposure cross-linking process) — the closest prior art to the valve-in-polymer-layer concept. It is the most structurally analogous reference for the pump claims.
• Where it fails as §102: the Caltech architecture is multilayer elastomer bonded to a substrate, which is exactly the "connected composite member" approach the '467 specification contrasts with its own unitary polymer layer; and it uses no deep-vs-shallow cross-linked regions formed by differential-dose direct imaging. Anticipation fails; §103 is a serious risk.

3.10 EP 1 283 563 A2 — Polymatech Co., Ltd.

• Full citation: EP 1 283 563 A2, "Connector and method for manufacturing the same," Polymatech Co., Ltd. JP priority 2001-08-09; published 2003-02-12. Family: JP 4767450 B2; US counterpart published as US 2005/0112036 A1 (Jun Funazaki et al., "Micro flow passage device, connection device, and method of using the devices," 2005-05-26).
• §102 category: §102(a)/(e)-type art (published 2003-02-12, after the 2002-06-24 bar date; its JP priority of 2001-08-09 predates the invention).
• Description: A connector for a micro flow-passage device and its manufacture — i.e., the mechanical interface for coupling external fluid lines to an on-chip microchannel.
• Potentially anticipatory of: the through-hole/connector subject matter of FIGS. 17–20 and the claim set directed to a fluid connector — specifically the concept of a connector portion mounted in/over a through-hole and secured (the '467 glue 2010 / anchor 1902 features). Also relevant to the generic fabrication-method claims.
• Where it fails: the '467 anchoring feature (glue biting into a cross-linked polymer "top-hat" anchor 1902) is a specific structural refinement; EP 1 283 563 discloses a connector but not this anchor geometry. §102 unlikely for the anchor claim; §103 plausible.

3.11 Daniel et al. — non-patent literature

• Full citation: J.H. Daniel, D.F. Moore, et al., "Silicon microchambers for DNA amplification," Sensors and Actuators A: Physical, Vol. 71, Nos. 1–2, 1 November 1998, pp. 81–88, ISSN 0924-4247 (also cited as XP004140078 in the PCT ISR).
• §102 category: §102(b).
• Description: Silicon micromachined reaction chambers for DNA amplification (PCR) — i.e., an on-chip thermal cycling chamber.
• Potentially anticipatory of: published claims 44 (method of making a reactor: forming a heating element in or proximate the substrate, depositing polymer, hardening portions to create a containment portion), 45 (spin-coating the polymer for planarity), and 48 (reactor apparatus with a heating element applied to the fluidic channel) — and the FIG. 11–12 serpentine PCR reactor 50 subject matter generally.
• Where it fails as §102: it discloses silicon microchambers with heating, not a polymer-layer unitary containment structure. It is a solid §103 reference for the "on-chip PCR reactor" idea and for the "heating element formed in the substrate before polymer deposition" element of claim 44.

4. Ranking: most relevant prior art

Bottom line on §102: On the record I can see, no single one of these eleven references contains every element of any independent claim, chiefly because none of them discloses the depth-differentiated, two-dose cross-linking of one photosensitive polymer layer (strong exposure → deep cross-linked region defining lateral boundaries; weak exposure → shallow cross-linked region defining the upper boundary) that the specification and abstract treat as the point of novelty. The realistic invalidity theory is a §103 combination — e.g., 3M WO 01/96958 (polymer-layer microfluidic fabrication) or Aclara WO 98/45693 (enclosed microchannels) in view of Bosch US 5,375,979 (thermal bubble pump) and/or Caltech WO 02/43615 (valves in polymer), optionally with Daniel et al. for the PCR chamber — not clean anticipation.

5. Things I could not confirm (stated rather than invented)

1. The complete, verified US front-page citation list. I have three partially inconsistent renderings and no direct PAIR/PatentCenter query. The US-only vs. PCT-ISR-only overlap is unresolved (specifically whether US 2003/0017467 A1 is on the US face).
2. The examiner's actual rejections. I did not retrieve office actions, the ISR's X/Y/A category letters, or any appeal record. I therefore cannot say which (if any) reference was applied as a §102 rejection versus cited as background.
3. Issued claim numbers 1–25. My claim mapping uses the as-published US 2004/0265182 A1 numbering; the issued set may differ.
4. The internal content of the references. For all eleven I am relying on titles, dates, and family data, not full-text review. The WO 01/96958 "fails on strong/weak exposure" conclusion in particular is a hypothesis I have not verified and should be checked before it is relied upon.
5. Forward citations are not prior art. Several databases surface US 7,309,467 in the "Referenced Cited" tables of later documents (e.g., US 2008/0108932 A1, "MEMS filter module with multi-level filter traps"; US 2006/0019420 A1, "MEMS device polymer film deposition process"; and design/assay patents such as the Justia "Test card" and US 10,519,493 entries). These are citations to the '467 patent and have been deliberately excluded from the prior-art analysis.

I'll start by looking for the actual prior-art references associated with this patent, since the authoritative text I was given cuts off mid-specification and does not include the "References Cited" front-page section the task tells me to use.

I now have the actual front-page citations. Let me retrieve the issued claim set and the key references' disclosures.

Let me confirm the full issued claim set and pull the two HP "direct imaging" references that appear on the patent's face.

Obviousness Analysis — US 7,309,467 B2 under 35 U.S.C. § 103

Note on the "Prior Art section" instruction: the authoritative full text supplied for this patent terminates mid-specification and contains no bibliography / "References Cited" section. I therefore reconstructed the prior art of record from the family's front pages (the WO publication of the same application, WO 2005/005046 A3, and the family listing on the sibling publication US 2008/0128390 A1), which reproduce the examiner-cited references for US 10/606,023. I flag title-only relevance where I could not retrieve full text.

⚠️ Contradiction flag vs. the earlier generated sections. The previously generated "Patent summary" section described the claim set as running to claims 26–54 and flagged uncertainty, because it relied on the as-published claims of US 2004/0265182 A1. The issued patent is now verifiable: it carries claims 1–13 only, and claim 1 is an apparatus claim, not a method claim at "claim 26." The earlier caveat ("may not be the verbatim issued claims") is now resolved — the issued set is materially different from the published set. The earlier text's inference that the base claim "appears at claim 26" should be treated as superseded.

I. The issued claim set being analyzed (as displayed on Justia / FreePatentsOnline for 7309467)

I could not retrieve a verbatim independent claim with an appended list of claims 14+, but both Justia and the Google Patents claim display terminate at 13, so I treat 1–13 as the operative set (moderate-to-high confidence).

Practical context: the patent is recorded Expired – Fee Related (adjusted expiration 2025-07-12), so this is a retrospective/hypothetical validity analysis rather than one driven by live enforcement.

II. Legal standard applied

Under Graham v. John Deere, 383 U.S. 1 (1966), and KSR Int'l v. Teleflex, 550 U.S. 398 (2007), a claim is obvious if the differences between the claim and the prior art are such that the subject matter as a whole would have been obvious to a person of ordinary skill at the time of the invention. KSR controls the combination question: where a known technique is used to improve similar devices in the same way, and the results are predictable, the combination is obvious; a "finite number of identified, predictable solutions" is enough; and a motivation may be found in the express teaching, suggestion, or the knowledge of a skilled artisan, including the reference's own statement of the problem it solves.

Because US 10/606,023 was filed June 24, 2003, pre-AIA §§ 102/103 govern. Importantly, both of the primary references below issued more than one year before that filing date, so they are § 102(b) bars usable in § 103 regardless of inventorship overlap — the "by others" requirement of pre-AIA § 102(a)/(e) does not apply to art that is one-year-bar art.

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

A bachelor's degree in electrical/mechanical engineering, materials science, or chemistry, plus 2–4 years of experience in micromachining/BioMEMS and photolithography, or a master's degree plus ~1 year. Such a person would know: negative-tone photoresists (SU-8 and its commercial equivalents), dose-dependent cross-linking depth (the Beer–Lambert attenuation that makes "underexposure = shallow cure" predictable), spin-coating and dry-film lamination, sacrificial-layer ("lost wax") release, anodic/frit bonding of etched wafers, and the thermal-inkjet printhead art that HP had been publishing for a decade. This is exactly the profile of the named inventors and of the examiners of this family (primary examiner Brian Sines).

IV. The prior art of record (reconstructed front page)

V. Ground 1 — Claim 1 is obvious over US 6,136,212 in view of US 6,162,589

Element-by-element mapping

Every element of claim 1 is thus met by the two references combined. Claim 1 is not novel over the combination, and the combination is not a mere aggregation: the '589 dual-dose scheme is applied to the '212 channel-forming structure to give exactly the claimed construction.

The "unitary structure" limitation is the crux — and it is squarely met

The '467 patent's own specification calls its structure a "top-hat structure" and emphasizes that it "ensures an improved sealing effectiveness compared to connected composite members" and avoids "align[ing] … a plurality of substrates." The prior-art '589 family uses the identical phrase — it claims a "top-hat re-entrant structure" produced in one photoresist layer by one low-dose exposure plus one high-dose exposure, and touts that this "simplifies manufacturing, which lowers the cost of production, enables high volume run rates … increases the quality, reliability and consistency of the printheads." The problem the '467 purports to solve (alignment/bonding cost and leakage between bonded wafers) is the same problem the '589 reference expressly says its process solves. That is a textbook KSR motivation.

VI. Grounds for the dependent claims

VII. Why a PHOSITA would have combined these references

1. Same field and same problem. Both '212 and '589 concern forming fluid-handling microstructures by photopatterning a polymer on a semiconductor substrate. The '467 specification itself concedes the prior art alike — it cites the wafer-bonding approach (its FIGS. 1–2) as the baseline and states the goal is "reduc[ing] the expense, time, and effort associated with micromachining MEMS devices … and then bonding the two substrates together." '589 states the same goal.
2. Express teaching to use high/low dose on one photosensitive layer. '589 not only permits but claims the two-dose method and the single-layer underexposure/overexposure variant. A skilled artisan needing lateral walls (deep) plus a channel ceiling (shallow) in one polymer layer would predictably select this known dose-vs-depth relationship.
3. Predictable result. Cross-link depth as a function of dose/attenuation is a well-characterized, deterministic property of negative resists (and '589 even quantifies the doses: ~300 mJ full-depth vs. ~60.3 mJ partial). KSR: a finite, identified, predictable solution set.
4. Common assignee / common origin. Both '162,589 and '303,274 are HP patents from the same Corvallis, OR printhead program as the '467 application (the '589 inventor list includes a "Chen, Chien-Hau, Corvallis, OR"; the '467 inventor is "Chen, Chien-Hua, Corvallis, OR" — I reproduce both spellings literally rather than normalizing). Whether or not the same person, the references are within the same corporate technical community, removing any "non-analogous art" objection; the '589 and the '467 are directly analogous art.
5. No change in principle of operation. Adding the '589 dual-dose cure to the '212 polymer-channel structure does not alter how either works; it merely improves the channel-forming step — the classic KSR "improvement of a prior-art device … by the mere substitution of one known element for another" with predictable results.
6. Reasonable expectation of success. '589 reports working devices (printheads) and even empirical geometry data (orifice/fluid-well height ratios vs. refill time), so a skilled artisan would expect the dual-dose cure to transfer to channel fabrication without undue experimentation.

VIII. Anticipated counterarguments and secondary considerations

IX. Bottom line

• Claim 1 would likely be held obvious under § 103 over US 6,136,212 (polymer microfluidic channels on a substrate) in view of US 6,162,589 (high-dose/deep + low-dose/shallow direct imaging of a photosensitive polymer layer on a substrate, forming a unitary "top-hat" structure). The motivation is explicit and shared: eliminate wafer alignment/bonding to cut cost, improve sealing, and improve yield/consistency.
• Claims 2–13 are each obvious over that base combination further in view of the specific secondary references tabulated in § VI, including the face-cited Daniel et al. 1998 PCR microchamber NPL (claim 7), WO 01/38844 (claims 8–9), US 5,375,979 and WO 02/43615 (claim 6), and the sacrificial polymer of '212 (claim 12).
• Confidence: high on the claim 1 mapping against the fully-verified texts of '212 and '589; moderate on the dependent claims that rest on title-level-only references (I have not personally read the specifications of US 5,375,979, WO 98/45693, WO 01/38844, WO 01/96958, WO 02/43615, US 2003/0017467, US 2002/0117517, US 2002/0098097, US 2003/0007715, or EP 1 283 563, and I state that limitation explicitly rather than overclaiming).
• Caveat on the record: I could not retrieve a verbatim, complete issued-claim listing beyond claim 13, and the earlier generated sections were built on the published claim set (claims 26–54). If any issued claim differs materially from the Justia/Google display, the mapping above should be re-run against the certified copy from USPTO PatentCenter.

Recommended verification steps: (1) pull the certified issued claims from USPTO PatentCenter for 10/606,023; (2) retrieve full texts (not titles) of US 5,375,979, WO 98/45693, WO 01/38844, and WO 02/43615 to convert the title-level rows in § VI to verified rows; (3) confirm the '589/'627/'259 single-layer embodiment figure numbers against the version of record.