US 2309075

I'll search for authoritative information on this patent number, including any CAFC 2026 docket activity.

US Patent 2,309,075 — Analyst Summary

⚠️ Two important threshold findings

1. No CAFC 2026 docket activity exists for this patent. I found no Federal Circuit docket, appeal, or 2026 litigation proceeding involving US 2,309,075. This is expected rather than surprising for structural reasons: the patent's own record shows "Anticipated expiration: 1960-01-19" with legal status "Expired – Lifetime." The Federal Circuit was not created until 1982, so a 2026 CAFC docket on a patent that expired in 1960 is not plausible. Any 2026 activity involving this number would have to be something other than an appeal of this patent's validity/infringement (e.g., a citation in a modern patent or a historical/defensive reference). I cannot confirm any such 2026 filing.

2. Numeric collision warning. Search results also surface EP 2309075 B1 — an entirely different patent, assigned/owned by Daifuku-type in-building storage interests, titled "In-building storage device and discharge method for the same" (EP app. 09758135, JP priority 2008-06-05, granted 2014-12-10). That is not US 2,309,075 and must not be conflated with it. I kept them separate throughout.

Bibliographic Data (US 2,309,075 A)

Uncertainty note on assignment: The OCR of the front page reads "…of one-hen to' Continental Oil Company," which is garbled. The conventional front-page phrasing this echoes is "assignor of one-half to…," which would suggest a partial assignment, but I cannot verify the assignment fraction from the available text and treat the assignee of record as Continental Oil Company.

Abstract

There is no formal abstract. This patent issued in 1943, well before abstracts were standard on US patents. What Google Patents labels "Definitions" is actually the opening of the specification ("My invention relates to a method of operating high pressure oil and'gas wells, and more particularly to a method of recovering liquid hydrocarbons from high pressure wells").

A faithful summary of the disclosed invention: high-pressure well fluids (wellhead pressures stated as 1,000–2,500 psi or higher) are produced at full wellhead pressure to an expander cylinder 9 (framed alternatively as a turbine, or multi-stage/tandem cylinders) instead of being throttled through a conventional choke. The expander performs three functions: (a) acts as the well choke, (b) expands and thereby cools the gas, and (c) converts pressure (potential) energy into work that drives a recompression compressor 14. The cooled, expanded stream (plus precipitated liquid droplets) is optionally chilled in a heat exchanger 16, then separated in a recovery tower 20; liquid hydrocarbons are withdrawn and the residual gas is recompressed and returned to the formation via the same well or a repressuring well. The specification also addresses (i) a high-pressure separator 4 upstream to drop water/foreign matter/entrained oil at formation pressure, (ii) a dehydrator 8 (e.g., silica gel) used only when expander temperatures would fall below the gas-hydrate formation point, (iii) withdrawal of a fuel-gas slipstream not exceeding state-law limits, and (iv) an optional auxiliary compressor 36 driven by a prime mover (internal combustion engine 37) for lean-gas cases. The motivation is explicitly economic: Texas conservation law capping gas removal at 2,000 cubic feet per barrel of oil requires returning excess gas, which is uneconomic if the return gas must be recompressed from scratch; the expander recovers that energy. The legal/technical backdrop relies on Andrews' 1870 critical-point work (Phil. Mag. (4) 39, 150) and the retrograde condensation range, with the critical pressure identified as roughly 500–700 psi.

Independent Claims — Plain-Language Overview

The patent has 7 claims total. Two are independent: claim 1 and claim 7. Claims 2–4 depend from claim 1; the text of claims 5 and 6 is not legible in the OCR (only the fragment "…hydrocarbons and a residual gas fraction" survives), so I flag those as indeterminate based on available text.

Claim 1 (independent — the core method). A method of operating an oil/gas well producing "desirable liquefiable hydrocarbons" from a formation whose pressure is above the critical pressure, with these steps in sequence:

1. Expand the produced hydrocarbons from formation pressure down to below the critical pressure and within the retrograde condensation range;
2. Convert pressure energy into kinetic energy during that expansion (i.e., do useful work rather than wasting the drop across a choke);
3. Separate the precipitated liquids from the uncondensed gases in a liquid recovery zone;
4. Withdraw the liquids as the desired liquid hydrocarbon product;
5. Withdraw the uncondensed gases from the recovery zone;
6. Recompress those gases using that same converted kinetic energy (self-powered recycle); and
7. Return the compressed gases to the formation they came from.

In plain terms: don't burn energy throttling the well — let the pressure drop run an expander, use the expander's work to push the gas back underground, and keep the liquids.

Claim 7 (independent — distillate-well variant). A method of recovering liquid hydrocarbons from fluid produced by a well of the distillate type at high wellhead pressure, comprising:

1. Expanding the fluid from a pressure on the order of wellhead pressure down into the retrograde condensation range while the fluid does work;
2. Separating the expanded fluid into a liquid-hydrocarbon fraction and a residual gas fraction;
3. Using at least a part of the work done by the expanding fluid to compress at least a part of that residual gas fraction; and
4. Cycling the recompressed gas to a subsurface formation.

In plain terms: same idea as claim 1, but framed more broadly — no express "above the critical pressure" limitation, no express requirement that the gas return to the same formation ("a subsurface formation"), and compression of only "at least a part" of the residual gas. Claim 7 is therefore the broader, more permissive independent claim.

Dependent claims (as legible):

• Claim 2 — adds a separation step at formation pressure before expansion (the separator 4).
• Claim 3 — adds a dehydration step before expansion (the dehydrator 8).
• Claim 4 — adds chilling by heat exchange with a cool external medium between expansion and recovery (heat exchanger 16).
• Claims 5–6 — text not recoverable from the provided OCR; given the fragment referencing "hydrocarbons and a residual gas fraction," they appear to add process refinements. Treat as unverified.

Caveats

• Claim sets 5 and 6 are OCR-corrupted in the source text; I have not reconstructed them and will not guess.
• The "Current Assignee: ConocoPhillips Co" entry carries Google's own disclaimer that it "may be inaccurate," and it is a modern successor entity rather than the 1943 assignee of record.
• No CAFC 2026 docket exists for US 2,309,075 that I could locate, and the patent's 1960 expiration makes such a docket logically implausible for this patent (as distinct from the unrelated EP 2309075).

Source: https://patents.google.com/patent/US2309075/en (fetched 2026-09-26).

I'll search for litigation records involving US 2,309,075.

Let me search additional litigation-specific sources.

Let me run a few more targeted searches to be thorough.

Litigation Search Results — U.S. Patent No. 2,309,075

Conclusion: No litigation involving U.S. Patent No. 2,309,075 was found.

I searched for litigation records keyed to the exact number 2,309,075 (and the unpunctuated form 2309075), including queries aimed at patent-litigation trackers (Unified Patents portal), CourtListener, Docket Alarm, Justia, and general case-law sources. I found no district court action, ITC investigation, PTAB proceeding (IPR/PGR/CBM), appeal to the CAFC, or reported decision in which U.S. 2,309,075 was the patent-in-suit or was otherwise at issue.

Because no case was identified, the requested fields (plaintiff, defendant, jurisdiction, case number, filing date, outcome) cannot be populated. I am stating this rather than filling in a table, since inventing case identifiers would be worse than reporting a null result.

Important caveats on scope of the null result

1. The number is not unique across jurisdictions. Searches for "2309075" return numerous unrelated records that must not be confused with the U.S. patent:

   • EP 2309075 B1 — "In-building storage device and discharge method for the same" (Japanese priority, Murata/ Daifuku-type storage apparatus).
   • DE 2309075 A1 — AGFA-Gevaert copier output/stacking device.
   • FR 2309075 A1 — Sony "Temperature effect compensation circuit."
   • RU 2309075 C1 — rail lubrication control method.
   • JP 2309075 / JPS5197955A, and a Brazilian STJ proceeding (AREsp 2309075) plus a Chilean and a Vietnamese administrative record.

   None of these is U.S. Patent 2,309,075. A keyword search on "2309075" alone will surface all of them; only the "2,309,075" / US-patent form with the correct bibliographic data is responsive.

2. Similarly numbered U.S. patents do appear in litigation and must be excluded. For example, U.S. Patent 7,232,075 (the "'075 Patent") was asserted by Rosen Technologies LLC against Lennox Industries Inc., N.D. Tex. Civil No. 3:22-cv-00732-K, and was resolved by a stipulated consent judgment (claims dismissed with prejudice). That is a different patent (7,232,075, a thermostat-related patent), not 2,309,075. I flag it explicitly so it is not mistaken for responsive.

3. Age and term. Per the Google Patents record, U.S. 2,309,075 issued 1943-01-19 on application US 260,724 filed 1939-03-09, inventor John L. Hill, original assignee Continental Oil Co.; legal status "Expired – Lifetime," with "Anticipated expiration" 1960-01-19. A patent that expired in 1960 is (a) beyond the modern litigation-tracking databases' coverage depth, and (b) incapable of supporting an infringement action for any post-1960 conduct. Any suit would have had to be brought in the 1943–1960 window. Pre-1960 federal patent litigation is not systematically indexed in the free sources I searched, so the null result should be read as "not found in the sources searched," not as an affirmative certification that no suit was ever filed.

4. What the record does show is citation activity, not litigation. The Google Patents "Cited By" list shows U.S. 2,309,075 cited as prior art by later patents (e.g., US 2,477,367, US 2,588,296, US 3,277,151, US 3,372,747, US 3,780,805, US 4,505,333, plus modern CNX Resources expander/wellhead-pressure patents such as US 2023/0193759 A1, US 11,123,433 B2, US 12,631,378 B2, and Leed Fabrication disk-turbine cases). Prior-art citation is not litigation and does not create case data. The "Similar Documents" list (e.g., US 2,582,148; US 2,203,954; US 2,423,156) is likewise a classification/text-similarity output, not an assertion record.

Recommended next steps if a definitive answer is needed

Free web search is a weak instrument for a 1943–1960 patent. To close this out with confidence you would need fee-based/archival tools:

• Docket Alarm and Unified Patents Portal — best for 2000s–present; will confirm a clean negative for modern proceedings but will not reach the 1940s–50s.
• LexisNexis / Westlaw full-text case databases — searchable for "2,309,075" in reported opinions back to the 1940s. A hit here would most likely be an infringement or declaratory-judgment action by or against Continental Oil Co. (or its successors, ultimately ConocoPhillips, per the current-assignee listing).
• FJC Integrated Database / NARA and the printed Federal Reporter / Decennial Digest volumes covering 1943–1960, which are the practical sources for pre-electronic district court patent dockets.
• USPTO Patent Assignment records for application 260,724, to identify any licensee or assignee that might have had standing to sue.

If you want, I can draft the specific search strings for those databases, or run additional targeted queries on any particular company (e.g., Continental Oil / ConocoPhillips) as a party in historical patent litigation.

I'll verify the structured data against live sources before writing.

Proceedings overview

Zero AIA trial proceedings exist for US Patent 2,309,075 — total count: 0. No IPRs, no PGRs, no CBM reviews: not active, not instituted, not denied, not settled, no claims canceled, no claims sustained by the Board, no FWDs, and no Federal Circuit appeals from any Board decision. The structured "PTAB proceedings on file" block (USPTO Open Data Portal) returns an empty set, and I could not surface any proceeding on the live web either.

The defensive posture this gives a defendant is not "hardened patent" and not "claims canceled" — it is "no case to defend." US 2,309,075 issued 1943-01-19, carried a 17-year term, and its record shows "Anticipated expiration: 1960-01-19" with legal status "Expired – Lifetime." An expired patent cannot be infringed (35 U.S.C. § 271 reaches only "a patent," i.e., an unexpired one), and § 286 caps back damages at six years. Any demand letter citing this patent is, on its face, asserting a right that lapsed 66 years ago. The absence of PTAB activity here is not a litigation signal about the patent's strength; it is a consequence of arithmetic — the earliest possible AIA filing date (2012-09-16) postdates the patent's expiration by more than half a century, and the PTAB itself did not exist until 2012 (its predecessors, including the Board of Patent Appeals and Interferences, never handled AIA trials, and the inter partes reexamination regime did not begin until 1999).

Source for bibliographic and status data: https://patents.google.com/patent/US2309075/en (fetched 2026-09-26).

No proceedings on file

• Type: n/a — no Inter Partes Review, Post-Grant Review, or Covered Business Method review identified.
• Filed: n/a.
• Status: The canonical ODP-derived block supplied in this prompt reports no AIA trial proceedings. My independent searches agree.
• Judge panel: n/a. There is no panel, so no APJ names exist to report.
• Petition grounds: n/a. I will not synthesize claims, art, or statutory bases for a petition that was never filed.
• Institution decision: n/a.
• Final Written Decision: n/a.
• Settlement / termination: n/a.
• Appeal: n/a. Consistent with the previously generated summary, no CAFC docket (2026 or otherwise) involves this patent; the Federal Circuit was created in 1982, 22 years after the patent expired.
• Defensive value: The patent is unenforceable because it is expired. There is also no § 315(e) or § 325(e) estoppel of any kind attaching to anyone, because estoppel is triggered only by a proceeding that "results in a final written decision" (35 U.S.C. § 315(e)) — there is none. Consequently, for any other patent you actually face, no prior-art ground is foreclosed by anything that happened (or failed to happen) here.

Categorical reasons no AIA proceeding was possible

• PGR (35 U.S.C. § 321) is available only for patents subject to first-inventor-to-file, i.e., effective filing date on/after 2013-03-16. This patent's effective filing date is 1939-03-09. Categorically unavailable.
• CBM (AIA § 18) required a patent claiming a financial product or service and a petitioner sued for or charged with infringement. Both elements fail, and the transitional program sunset on 2020-09-16.
• IPR (35 U.S.C. § 311) has no express expiration bar — the Board has instituted on expired patents where a live collateral dispute existed — but here there is no live dispute, no accused product, and no path to one, so no petitioner has ever had a reason to file. Note also the outcome the Board has reached in other cases when petitioners attack dead patents: the AIA's estoppel and "one bite at the apple" architecture presupposes an enforceable right to narrow.

⚠️ Do not import a "'075 patent" record from an unrelated matter

Multiple live PTAB and district court documents shorthand unrelated patents as "the '075 patent." Any of these, misread, would produce a fabricated proceedings list. Concrete examples surfaced in my search:

• IPR2020-01559, a US 7,286,075 B2 case (Confalonieri / Cai / Hiller grounds) whose institution decision reads: "inter partes review of the '075 patent is hereby instituted as to claims 1-25." That is a different patent with different claims, art, and panel (Morrison & Foerster for petitioner).
• IPR2023-00707 (Google LLC) involving a "'075 patent" about virtual reality viewer headsets and touchscreen electrical shields.
• A PTAB petition challenging claims 1–6 of a "'075 patent" directed to a catamenial/tampon applicator with curved face portions and "sawteeth."
• Apple/Samsung v. Smart Mobile in W.D. Tex., where the court noted "the PTAB denied IPR proceedings for the '075 patent" — again an unrelated patent, and a denial, i.e., the opposite of what a careless reader might assume.
• EP 2309075 B1 ("In-building storage device and discharge method for the same," JP priority 2008-06-05, granted 2014-12-10) — flagged in the earlier summary and repeated here as a cross-format collision.

None of the above is US 2,309,075. None may be cited as its PTAB history.

Strategic summary

Claim-status ledger for US 2,309,075. There is no Board outcome, so nothing is "CANCELED" by the PTAB and nothing is "SUSTAINED" by the PTAB. The correct characterization of all seven claims is UNTESTED — and jurisdictionally untestable in an AIA trial on any live dispute, because the patent expired 1960-01-19. If forced to map the claim set (per the earlier summary: independent claims 1 and 7; claims 2, 3, and 4 dependent on claim 1 and respectively adding formation-pressure separation, dehydration, and external-medium chilling; claims 5 and 6 not recoverable from the OCR), the practical answer for a defendant is that the distinction between canceled and sustained is moot: an expired claim set has no scope to litigate. One analytical wrinkle worth preserving: because the OCR truncates claims 5 and 6, any later document that purports to quote those claims verbatim should be verified against the USPTO PatentCenter / USPTO Patent Public Search copy or the printed original (US2309075.pdf) rather than the Google Patents HTML, which is corrupted at that point.

Estoppel landscape. There is no § 315(e)(2) estoppel, no § 315(e)(1) estoppel, no § 325(e) estoppel, and no In re Cygnus / Sotera-style stipulation on this patent — none was ever created. Practically, this means every prior-art ground remains available to every party as to any other patent in the same technology space. Separately, and because the patent's entire defensive value today is as a printed publication, note that US 2,309,075 published 1943-01-19, making it a clean pre-AIA § 102(b) reference for anything with a critical date after that. Its Google Patents "Cited By" list shows it has already been used this way by modern entrants — US 20230193759A1, US 12123433B2, and US 12631378B2 (all CNX Resources Corporation, expander/wellhead-pressure energy recovery) cite it, as do US 9,205,348 B2 and US 10,941,643 B2 (Randy Swan, vapor recovery). That is its live use: prior art, not an asserted patent.

Pattern signals. There is no petitioner to identify, no repeat-filer pattern, no Unified Patents–style defensive aggregator in the chain, and no patent-owner appellate aggression — all of those categories require a proceeding that does not exist. The only chain that exists is Continental Oil Company (assignee of record, Ponca City, Oklahoma; the 1943 front-page OCR is garbled as "of one-hen to' Continental Oil Company," which I cannot decode into a verified assignment fraction) → ConocoPhillips Co, flagged by Google with its own disclaimer that the current-assignee list "may be inaccurate." I found no recorded infringement action or validity challenge on US 2,309,075 in the 1943–1960 term; the Continental Oil litigation trail I could locate involves other patents (the '800 paravane patent in Continental Oil Co. v. Cole, 634 F.2d 188 (5th Cir. 1981); the coking patents in Great Lakes Carbon Corp. v. Continental Oil Co.). Treat the absence of a 1943–1960 suit as "not found," not as proof none occurred.

Recommended next steps

• If you are a defendant and the demand letter cites US 2,309,075: do not build an IPR budget. Build a one-page response. The record of the patent itself is the disposition: "Anticipated expiration 1960-01-19" / "Expired – Lifetime" (https://patents.google.com/patent/US2309075/en). The patent is unenforceable, there is no § 271 infringement to plead, and § 286 forecloses recovery. If the sender persists, the appropriate escalation is a declaratory-relief and fee posture, not a PTAB petition.
• Do not file an IPR/PGR on this patent. A PGR is categorically barred by the 1939 effective filing date (§ 321 / AIA § 3(n)(1)); a CBM is barred by subject matter and by the 2020-09-16 sunset; an IPR would burn statutory fees and expert hours to invalidate an already-dead right, with no estoppel or narrowing benefit worth capturing.
• Use it as prior art instead. If your real exposure is to a modern wellhead expander / vapor-recovery patent family (e.g., the CNX or Swan families in the Cited By list), pull the printed publication date of 1943-01-19 and treat US 2,309,075 as § 102(b) art. Retrieve the authoritative full text from USPTO PatentCenter or the original PDF (https://patentimages.storage.googleapis.com/15/28/a2/dda26d449d92fd/US2309075.pdf) so that claims 5 and 6 are legible rather than OCR-truncated.
• Verify the negatives independently before relying on them. Confirm the empty proceedings list at USPTO PTAB E2E / PTAB Decisions (https://developer.uspto.gov/ptab-api) and check CourtListener for both "2,309,075" and "2309075." When you do, expect heavy false positives from the unrelated '075 patents and from EP 2309075 B1 — screen every hit by patent number, filing date, and technology before citing it.
• If someone hands you a memo asserting PTAB outcomes for this patent, treat it as unreliable. There is no FWD to quote, no panel to name, and no disposition to link. Any such memo has almost certainly imported the record of a different "'075" patent.

I'll search for authoritative assignment-record information before drafting.

Let me try more targeted searches for the assignment record and for related family members.

Ownership-Chain Reconstruction — US 2,309,075 A

Method note / scope limitation, stated up front. I was unable to retrieve a single assignment record for this patent. The USPTO Assignment Center (https://assignmentcenter.uspto.gov/) became the sole access point for assignment data on 2025‑09‑27, when USPTO retired the legacy Patent Assignment Search and Assignments on the Web applications (USPTO bulletin, 2025‑08‑27). The new tool is a JavaScript application whose result sets are not exposed to web search, and I have no interactive query access to it. Every "no result" below therefore means "not retrievable by me," not "not on record at USPTO." I have not invented reel/frame numbers, correspondents, or conveyance types to fill the table. This is the single most important caveat in this report.

Inventors

Patterns observed — and not observed:

• Single-inventor patent. No co-inventor "departure cluster" is possible; the "all inventors left within 12 months" tell cannot be assessed on a one-inventor patent.
• Partial assignment on the face of the patent (flag). The front-page legend is OCR-garbled — the earlier section quoted "…of one-hen to' Continental Oil Company." The conventional pre‑1943 construction it echoes is "assignor of one-half to Continental Oil Company, Ponca City, Oklahoma, a corporation of Delaware." I flagged this in the prior summary as unverified and I maintain that flag: the assignment fraction (one-half vs. full) is not confirmed from the text I can read. If it is one-half, the practical implication for an ownership-chain analysis is material and worth stating plainly — a surviving half-interest in a then-living individual inventor would be a chain-of-title defect in any later derivative transfer, and half-interests in 1930s–40s oilfield patents not uncommonly surface decades later in probate or estate sales. This is a hypothesis about a risk, not a finding of a defect.
• Contradiction to flag against the earlier section: the prior summary states wellhead pressures of "1,000–2,500 psi or higher." That figure does not appear anywhere in the authoritative full text supplied to me. The description sentence that would have carried it is OCR-corrupted at exactly that point ("It is understood that the well is under ___ vided between the well and the after equipment"). Treat the pressure range as unverified, not as a document-sourced figure.

Original assignee

Entity as named on the issued patent: Continental Oil Company, Ponca City, Oklahoma, a corporation of Delaware (Google Patents lists "Continental Oil Co" as original assignee).

• Line of business: Fully integrated oil and gas — exploration, production (including the high-pressure distillate wells this patent addresses), refining, and marketing under the Conoco brand. This is a classic operating-company patent, filed by a corporate employer-inventor in its own business, covering a practice method used in its own field operations.
• Did they ship a product embodying the claims? The claims are method claims (a well-operating/processing method), so there is no "product" in the ordinary sense. There is no evidence in the record I can access that Continental Oil commercialized an expander-driven wellhead recycle facility under this patent, though the patent's own motivation — Texas gas-oil-ratio rules capping removal at 2,000 cu ft per barrel of oil — is a strong indicator the method was aimed at its own Texas/OK high-pressure distillate wells. I cannot confirm field practice; I will not guess.
• Current status of the original assignee: Acquired and successively absorbed. Verified succession:1981 → 1998/1999 → 2002:

Conclusion: Google Patents' "Current Assignee: ConocoPhillips Co" is almost certainly a corporate-succession inference, not a recorded assignment to ConocoPhillips. Its own disclaimer ("The listed assignees may be inaccurate") applies. No assignment from Conoco Inc. to ConocoPhillips need exist, because the 2002 merger transferred title by operation of law.

Assignment timeline

Plainly stated: I retrieved no reel/frame entries for US 2,309,075, and I therefore have no conveyance, assignor, assignee, correspondent, or execution/recording date to report.

Three independent structural reasons make this outcome ambiguous rather than conclusive:

1. Tooling. The Assignment Center's replacement of the legacy search was completed 2025‑09‑27; its results are not search-indexable. I could not execute a query by patent number.
2. Index coverage. The online assignment index is materially complete only for records entered in the electronic era. Reel/frame image numbering did not begin until the mid‑1950s, and this patent's only likely ownership event — the pre-issuance assignment reflected in the printed front-page legend, i.e. a 1939-era recordation — would have been made on a liber/page system that is generally not in the searchable online index. A negative result at the Assignment Center for a 1939 application is therefore expected even if a record exists.
3. Google Patents legal events corroborate only the negative, weakly. The legal-events list for US2309075 that I was given contains only filing (1939‑03‑09), priority (1939‑03‑09), grant (1943‑01‑19), publication (1943‑01‑19), and anticipated expiration (1960‑01‑19) — no assignment, security-agreement, merger, or change-of-name events. That is consistent with no post-issuance assignment in the digitized record, but Google Patents routinely omits pre‑1980 assignments, so it is not proof.

The single ownership fact I can ground: the printed front-page legend, which shows a pre-issuance assignment to Continental Oil Company (fraction unverified — see Inventors). Recording data for that transfer is unknown.

How to close this gap definitively (for a human analyst with access):

• Query https://assignmentcenter.uspto.gov/ by patent number 2309075 and, separately, by assignor Hill, John L. with execution-date window 1938–1943.
• For pre‑1955 liber/page records not in the electronic index, order a certified copy from the USPTO Assignment Recordation Branch (the manual/paper search is the only route).
• Verify with the original file wrapper (Record Group 241 at NARA, or the digitized file history), which will carry the original assignment document if one was recorded.

Timeline diagram

The chain below is documented corporate succession, not recorded assignments — that distinction is deliberate and load-bearing, because a merger passes title without a reel/frame entry. The only assignment-shaped event on the record is the 1939 pre-issuance transfer, whose date and reel are unverified.

timeline
    title Ownership of US 2309075
    1939 : Filed by Continental Oil Company
         : Hill legend shows pre-issuance assignment
    1943 : Patent issued
    1960 : Anticipated expiration
    1979 : Continental Oil renamed Conoco Inc
    1981 : Conoco becomes DuPont subsidiary
    1998 : DuPont begins selling Conoco stake
    2002 : Conoco merged into ConocoPhillips
    2012 : Phillips 66 spin-off takes Conoco brand

NPE / troll-pattern signals

Every signal below is answered against a record set I could not query. I have marked "unclear" wherever the honest answer depends on data I do not have, rather than converting absence of evidence into a negative finding.

Cross-check worth recording: none of the 20–24 forward-citing assignees surfaced (Texaco Development, Aluminum Co of America, National Tank Co, Constock Liquid Methane, Phillips Petroleum, Air Technologies, Leed Fabrication Services, Randy Swan, Flogistix, CNX Resources, Atlantic Richfield) appear in US 2,309,075's ownership chain. Those are prior-art citations in a 1939 patent's forward-citation graph — the modern turboexpander/wellhead-energy-recovery art — and several were surfaced under the unrelated EP 2309075 / CA 2309075 / DE 2309075 numbers. Flag for the file: any hit on the string "2309075" must be jurisdiction-checked; the numeric collisions are documented in the earlier section and were confirmed again in this pass.

Verdict

Insufficient data

Justification. The chain contains no recorded assignment entries that I could retrieve, so no NPE signal can be marked "present" without fabricating reel/frame citations — the one thing the task constraints forbid. What is documented is a benign corporate succession verified outside the assignment record: Continental Oil Company renamed Conoco Inc. on 1979‑07‑01, became a DuPont subsidiary in 1981, and merged into ConocoPhillips Company effective 2002‑12‑31 (per the Delaware certificate of merger reproduced in the Utah operator-change filing), with title passing by operation of law rather than by recorded assignment. Two factors make an NPE characterization affirmatively unlikely rather than merely unproven: the patent expired 1960‑01‑19, so its claims are unassertable by anyone, and the only front-page ownership event is a 1939 pre-issuance assignment to an integrated oil major whose assignee chain never touched a licensing entity.

To upgrade this verdict, the following single action is required: run claim 1's ownership question against the live Assignment Center at https://assignmentcenter.uspto.gov/ (patent number 2309075; assignor Hill, John L., execution window 1938–1943), capture any correspondent of record, and check the pre‑1955 liber/page records via the Assignment Recordation Branch. If that query returns a correspondent whose name recurs on other patents, signal 3 flips from "unclear" to a real finding and this report should be re-run.

I'll start by pulling the authoritative USPTO record for this exact number, then work through the citation layers.

The search results confirm an important structural fact about the citation layers. Let me verify the USPTO record and check the candidate pre-1939 art.

Prior-Art Analysis — US Patent 2,309,075 (John L. Hill / Continental Oil Co.)

Bottom line up front — three threshold findings

1. US 2,309,075 has no "References Cited" of record. There is no prior-art citation list to work through. The authoritative record for this patent contains zero backward patent citations. The Google Patents page for US2309075A shows only two citation-type fields — "Cited By" (forward citations by later patents) and "Similar Documents" (an algorithmic text/classification-similarity output). There is no "References Cited" / "(56) References Cited" field populated on this record. I verified this against the full text supplied and re-checked via search; nothing contradicted it.

This is not an OCR gap. It is consistent with the era: USPTO did not routinely print examiner-cited references on the face of the patent until well after this 1943 grant. (I am flagging my confidence here rather than asserting a precise adoption date — I do not know the exact year the face-citation practice became standard with high confidence.) Consequently the literal task ("look at each patent citation for 2309075") has an empty set on the backward side, and I will not manufacture references to fill it.

2. The "patent citations associated with 2309075" are all forward citations and therefore cannot be §102 prior art. Every entry in the "Cited By" list post-dates Hill's 1939-03-09 filing — the earliest, US 2,477,367, was filed 1943-05-15, more than four years after Hill. A reference that post-dates the filing cannot anticipate. I give the full list below (as requested) with a §102 verdict for each, but the verdict is uniform.

3. Legal-anachronism flag. This patent issued 1943, so its validity was governed by the Patent Act of 1870 as amended — 35 U.S.C. §102 did not exist until the 1952 Act. Applying "§ 102" here is therefore a retrospective, hypothetical exercise. I frame it as pre-AIA §102 (a)/(b)/(e)/(g), the closest modern analogue, and note that under the then-governing law the critical date was the date of invention, not the filing date.

Source-internal contradiction to flag (per cross-reference instruction)

The previously generated Bibliographic Data table records Application No. US 260,724. The OCR body text of the specification itself reads: "application Masch e, 1939, semi No. 260,124." 260,124 ≠ 260,724. The Google Patents structured data field says 260,724 in three places, so I treat 260,724 as the operative number, but the discrepancy matters for prior-art/family searching: a search on application 260,124 would retrieve a different case. My searches could not independently resolve which is correct.

Layer 1 — References actually mentioned inside the patent (non-patent)

There is exactly one technical reference cited in the body of US 2,309,075, and it is non-patent literature:

The specification's other "references" are not prior art: the Texas conservation-law gas-oil ratio limit of 2,000 cubic feet per barrel, and the statement that wellhead pressures are on the order of 1,000–2,500 psi. These are cited as legal/economic motivation, i.e., they define the problem, not the solution.

Implication: if a §102 rejection of this patent were ever at issue, it could not have been built on the patent's own reference list, because there wasn't one.

Layer 2 — The "Cited By" set (forward citations) — full citations and § 102 verdicts

These are the only patent citations Google associates with US2309075. None can anticipate any claim, for the single dispositive reason that each was filed after 1939-03-09. I include them because the task asked for each citation, dates, and description.

Analyst note on the CNX and Leed entries: these are the modern descendants of exactly Hill's concept — wellhead pressure driving an expander/turbine to recover energy and produce cold — which is why they cite US 2,309,075. That is a teaching-lineage signal, not a novelty problem. If anything, this citation pattern is evidence against the patent being anticipated by anything of its own era: later inventors kept re-patenting improvements on Hill's structure rather than pointing to a pre-1939 anticipating document.

Layer 3 — "Similar Documents" (NOT prior art of record)

I must be explicit: "Similar Documents" is a machine-similarity/classification output, not an examiner citation and not a §102 statement. Any §102 use of these requires independent verification of filing dates and disclosure, which my searches could not complete (the search budget was exhausted before I could pull filing dates). Treat this layer as a lead list only.

Within it, only these items issued before Hill's 1939-03-09 filing date and are therefore even candidates for pre-filing printed-publication art:

Items issued after 1939-03-09 but potentially §102(e)-eligible (US patent granted on an application filed before Hill's invention date) also need date verification: US 2,156,234 (1939-04-25, "Recovery of liquid hydrocarbons"), US 2,198,098 (1940-04-23, "High pressure gas process"), US 2,209,534 (1940-07-30, "Method for producing gas wells"), US 2,231,500 (1941-02-11), US 2,245,028 (1941-06-10).

Everything else in the Similar Documents list post-dates Hill and is irrelevant as art — e.g., US 2,357,703 (1944), US 2,364,660 (1944, "Method of and apparatus for recovering desirable petroleum hydrocarbon fractions from high pressure wells"), US 2,409,691 (1946), US 2,423,156 (1947, "...at substantially wellhead pressure"), US 2,582,148 (1952, "Method of recovering desirable liquefiable hydrocarbons"), US 2,601,599 (1952), US 3,397,138 (1968). The 1944–1952 titles are notable: they echo Hill's claim-1 language ("desirable liquefiable hydrocarbons") and describe the same object, strongly suggesting they are later members of the same technical lineage (Continental Oil / distillate-cycling patents). They are descendants, not antecedents.

One further document to flag, with an identifier conflict: a PDF at patentimages.storage.googleapis.com/…/US2617484.pdf recites a method of expanding a heated gaseous mixture "while said mixture does mechanical work," separating out a residual gas, and having "means driven by said means for expanding so that said work is employed for recompressing said residual gas… to reinject said residual gas into said reservoir." That is structurally Hill's invention. Its URL indicates US 2,617,484, but the OCR page header reads "3,617,484." Per the no-auto-correction rule I report both and correct neither. Either way it issues in the 1950s–60s and post-dates Hill, so it is a descendant/improvement, not prior art. Its relevance is evidentiary: it shows the art treated expander-driven recompression as Hill's contribution and built on it.

Claim-by-claim § 102 framework (what a reference would have to show)

Since no reference is of record, I give the anticipation test rather than a false mapping. Under pre-AIA §102, a single reference must disclose every element, arranged as claimed, or the claim is not anticipated (it could only be obvious under §103).

Claim 1 (independent). An anticipating reference would need to disclose, in one document: (a) a formation pressure above the critical pressure; (b) expanding produced hydrocarbons from formation pressure to below critical and within the retrograde condensation range; (c) converting pressure energy into kinetic energy during that expansion (i.e., an expander doing work, not a choke); (d) separating precipitated liquids; (e) withdrawing the liquids; (f) recompressing the uncondensed gases with that same kinetic energy; and (g) returning the gas to the same formation. The novelty/point of departure is the energy-integration of (c)+(f) — the expander doubling as the well choke and as the recompression prime mover. A reference showing expansion-to-do-work plus reinjection without using that work to drive the recompression would not anticipate claim 1.

Claim 7 (independent, broader). Drops the express "above the critical pressure" requirement, does not require the same formation ("a subsurface formation"), and requires compression of only "at least a part" of the residual gas. Its broader scope makes it the claim most vulnerable to a §102 challenge — but only to a reference showing expander-work-driven recompression of produced gas into a subsurface formation for a distillate-type well. I found no such pre-1939 reference.

Claims 2–4 (dependent). Anticipation of these requires the base claim plus the added step: claim 2 – separation at formation pressure before expansion; claim 3 – dehydration before expansion; claim 4 – chilling by external heat exchange between expansion and recovery. These are conventional oilfield steps, so if the base claim were anticipated they would fall readily; alone they add little.

Claims 5–6 — indeterminate. As noted in the previously generated summary, only the fragment "…hydrocarbons and a residual gas fraction" survives the OCR. I still cannot reconstruct them and will not guess. This is a genuine gap in the record, and it means any §102 conclusion about claims 5–6 is impossible on the available text.

Same-number/different-patent exclusions (per instruction)

A bare search on "2309075" returns many unrelated records. None is US 2,309,075, and none was used as prior art here:

• EP 2,309,075 B1 — "In-building storage device and discharge method for the same" (JP priority 2008-06-05).
• DE 2,309,075 A1 — AGFA-Gevaert copier output/stacking device.
• FR 2,309,075 A1 — Sony temperature-effect compensation circuit.
• RU 2,309,075 C1 — rail-lubrication feed control (this one re-appeared in my search results as a cited document in the search report of RU 2,425,768 C1, published 27.10.2007 — confirming it is an unrelated Russian railway patent, not the US oil-and-gas patent).
• Brazilian STJ proceeding AREsp 2309075; JP 2309075 / JPS5197955A; a USACE Columbia River engineering table containing the bare number.

What I could not do, stated plainly

• I could not retrieve a USPTO-hosted record (PatentCenter / Patent Public Search) directly. My searches surfaced the Google Patents rendering, which is the source of the authoritative full text; the USPTO-hosted pages did not return in results. The record's substantive content was not contradicted by anything I found.
• I could not verify filing dates for the candidate pre-1939 art (US 2,133,774; US 2,134,699; US 2,134,700) or for the potential §102(e) references. Without those dates, a defensible §102(e) determination is not possible.
• I will not assign a §102 rejection to any reference. The record contains no cited prior art, and the only citations that exist are forward citations that post-date the patent. Reporting that null result is the correct answer; inventing a "most relevant prior art" citation would be fabrication.

Source: https://patents.google.com/patent/US2309075/en (full text supplied; fetched 2026-09-26).

I'll verify filing dates for the references appearing in this page's prior-art/related-art sections before building the §103 combinations, since date qualification is dispositive here.

The US2133774 result is highly material. Let me pull filing/priority dates for the other candidate references in this page's related-art lists.

§103 Obviousness Analysis — U.S. Patent 2,309,075

Threshold problem with the "Prior Art" material on this page

Before analyzing, I have to flag a structural defect in the source record, because it determines what can legitimately be combined.

The fetched Google Patents record for US 2,309,075 contains no backward-citation table (no "Patent Citations / Cited By [examiner]" listing of the references the examiner actually applied). What the page labels near the prior-art section is only:

1. "Prior art keywords" — an index string: gas, pressure, well, pipe, high pressure. This is a search-field artifact, not a reference.
2. "Cited By (20/24)" — forward citations. Every entry postdates this patent (earliest priority: US 2,477,367, 1943-05-15). A reference that postdates a patent cannot be §102/§103 prior art against it. Treating, e.g., US 2,937,503 ("Turbo-expander-compressor units," Nat'l Tank Co., priority 1955-09-19) as prior art to a 1939 application is an anachronism and would be reversible error.
3. "Similar Documents" — an algorithmic text/classification-similarity list. Only a subset qualifies as prior art by date.
4. The specification's own admitted art — Andrews, Phil. Mag. (4) 39, 150 (1870), plus the patent's admissions that the well choke "expends energy which is not employed in any useful way" and that the expander may be "any suitable device in which a gas may be expanded to do work, such as a turbine."

So a true §103 analysis as to the references of record cannot be performed from this page — the file wrapper isn't here. What follows combines (a) the legally-qualifying "Similar Documents," and (b) the applicant's own admissions. Where I have not verified a reference's disclosure or filing date, I say so.

Governing law and the person of ordinary skill

The application was filed 1939-03-09 and granted 1943-01-19; the AIA does not apply. Modern invalidity analysis proceeds under pre-AIA 35 U.S.C. §103(a) with the Graham v. John Deere, 383 U.S. 1 (1966), factors, and the KSR Int'l v. Teleflex, 550 U.S. 398 (2007), motivation standard. (For historical completeness: at grant, the controlling standard was the "flash of genius" language of R.S. 4886 as construed in Cuno Engineering v. Automatic Devices, 314 U.S. 84 (1941) — a stricter patentability standard than §103(a). That cuts against, not for, the patent.)

PHOSITA (as of March 1939): a petroleum or mechanical engineer with ~3–5 years in natural-gasoline-plant design, high-pressure well production, and gas-repressuring/compression operations; familiar with retrograde-condensation behavior and with commercial refrigeration and air-separation practice (the Claude/Linde expansion-engine art).

Date screening of the page's "Similar Documents"

Critical gap: neither the qualifying 1938 references nor the cited "Cited By" list supplies a pre-1939 reference that expressly teaches replacing the well choke with a work-producing expander. The expander-plus-compressor art in this record (US 2,937,503, 1960; US 3,397,138, 1968; US 2013/0068314 and US 12,631,378, 21st century) is all later. That gap is the whole case.

What the qualifying references actually teach

US 2,133,774 (Google Patents: https://patents.google.com/patent/US2133774) — expansion of high-pressure well gas; "pressure relief valve 25 ... adjusted to reduce the pressure to that at which retrograde condensation will occur at the particular temperature"; multiple cooling stages (water cooling, heat exchange with expanded gas, and artificial refrigeration in cooler 16 using "expanded liquid propane, ammonia or other refrigerating fluid"); separator 17 for condensed constituents; and "the re-compressed gas recycled to well A." It states the process "is applicable particularly to the treatment of gases under a pressure above 700 pounds per square inch gage," and discusses hydrate formation in the cold separator. Note: expansion is through a valve — throttling, which by definition produces no recoverable shaft work.

US 2,198,098 (https://FreePatentsOnline.com/[2198098](/patent/2198098).html) — even closer to the claims: cooling, expansion into separator 16, recovery of condensate, and residue gas recompressed "in a multi-stage compressor C ... to a higher pressure sufficient to permit the gas to be recycled to the sub-surface reservoir from which it came and to which it is returned ... to the flow pipe 29 of a return well B"; expressly defining "recycling" as return to the source reservoir; offering the alternative of "repressuring lower pressure oil reservoirs"; and stating that "a dehydrating agent may be circulated in the system to remove water vapor." Again, expansion is by pressure reduction, not by work extraction.

The combination that would be asserted

Proposed rejection (claims 1, 2, 4, 7): US 2,133,774 or US 2,198,098 in view of a work-expansion/expansion-engine teaching.

Mapping (claim 1):

• above-critical formation pressure → 2,133,774 (">700 p.s.i.g.").
• expanding to below critical into retrograde condensation range → 2,133,774's valve 25; 2,198,098's separator expansion.
• separating liquids / withdrawing liquids and uncondensed gas → both, expressly.
• returning compressed gas to the formation from which it was withdrawn → 2,198,098 almost verbatim; 2,133,774 ("recycled to well A").
• "converting pressure energy into kinetic energy ... recompressing said uncondensed gases by means of said kinetic energy" → not taught by any qualifying reference in this record. It requires the expansion-engine/expansion-turbine teaching from the analogous refrigeration and air-separation arts.

Claim 2 (separator at formation pressure): 2,198,098's separator 16 and 2,133,774's separator 11; high-pressure gas-oil separators were ubiquitous and the specification itself calls separator 4 conventional. Obvious.

Claim 3 (dehydration before expansion): 2,198,098 expressly recites a dehydrating agent; 2,133,774 recognizes hydrate plugging. Where the cooling is intensified by work expansion, pre-dehydration is the predictable anti-hydrate measure. Obvious.

Claim 4 (chilling by heat exchange with a cool external medium): 2,133,774's cooler 16 teaches exactly this — artificial refrigeration with expanded liquid propane or ammonia. Obvious; in substance anticipated in the sense of being fully disclosed, though claim 4 depends from claim 1.

Claims 5–6: text unrecoverable from this record (consistent with the earlier section). No opinion.

Claim 7 (broader; "while said fluid does work"; "at least a part" of the residual gas; "a subsurface formation"): same crux, broader in scope. Its breadth (any subsurface formation; partial compression) makes the non-expander limitations even easier to meet from 2,198,098 alone, so claim 7 rises or falls entirely on the "does work" limitation.

Motivation to combine — the strongest part of the rejection

1. Express regulatory-economic pressure. The specification itself recites that Texas law capped gas removal at 2,000 cu ft per barrel, forcing return of excess gas and imposing "large capital expenditure ... for equipment enabling the return of excess gas ... under extremely high pressures." A recognized, quantified industry problem with a known regulatory deadline is powerful motivation.
2. Same field, same purpose. Both qualifying references are high-pressure well-gas liquid-recovery processes aiming at the same two goals (liquid yield; reservoir-pressure conservation by recycle). Combination of references directed to the same problem is the paradigm case.
3. The choke was already identified as pure waste. The patent's own words: pressure reduction "through the choke expends energy which is not employed in any useful way." A POSITA reading that sentence has the entire inventive concept handed to him.
4. Known technique, predictable result (KSR rationale 1). Work-producing gas expansion was conventional since Claude (1902) in air separation and in refrigeration cycles: substituting a reciprocating expansion engine or turbine for a throttling valve yields (a) a lower outlet temperature than isenthalpic throttling and (b) recoverable shaft work. Both results are predictable, and both are exactly what the claims recite.
5. Mechanical substitution + design incentive (KSR rationales 2 and 4). Swapping a valve for an expander is a substitution of one known flow-control/pressure-reduction element for another, driven by the acknowledged cost of purchased recompression power.
6. "Obvious to try" (KSR rationale 3). Facing "we must recompress a large gas volume and we are throwing away the pressure drop," the finite set of identified solutions includes "recover the pressure drop as work."

Countervailing considerations (patentee's side) — and why they are substantial here

1. No qualifying reference teaches work expansion. Every pre-1939 reference in this record that expands well gas does so through a valve or separator. Expansion while doing useful work is absent from the corpus. That is a genuine §103 gap, and it is the sole limitation carrying claim 1 and claim 7.
2. Prolonged failure of others / long-felt need. After 2,309,075, contemporaneous and later art continued to use valve expansion plus separately powered compression — US 2,477,367 (Texaco, priority 1943), US 2,364,660 (1944), US 2,423,156 (1947), US 2,582,148 (1952) — and the record's own first "Turbo-expander-compressor units" citation is US 2,937,503, with a 1955 priority. A ~16-year lag between the invention and the art's arrival at the expander-compressor combination is classic objective evidence of non-obviousness.
3. Potential teaching away. The retrograde-condensation art produces a cold, droplet-laden, hydrate-prone stream and deliberately routes it to a "relatively large" separator precisely because hydrates "will be unlikely to stop up such passages" (US 2,133,774). Feeding that same stream into a close-tolerance reciprocating expander cylinder — the exact structure the specification discloses — is arguably discouraged by the art. This is a weaker argument (hydrate mitigation, not hydraulic-slugging, is the stated concern), but it is colorable.
4. Secondary evidence of merit. The record's forward-citation profile (cited by US 2,937,503; US 3,397,138; Leed Fabrication's boundary-layer disk turbine cases US 2013/0071259 and US 2013/0068314; and CNX Resources' wellhead-pressure expander family, US 2023/0193759, US 12,123,433, US 12,631,378) shows the concept remained the reference point for wellhead-pressure energy recovery for eight decades. Post-1960 activity is not tied to the 1939 invention and carries limited weight, but it corroborates that the claimed combination was not a trivial design choice.

Bottom line

Overall: the patent does not read like a pioneering claim set. Claims 2–4 are plainly obvious over US 2,133,774 (1938) and US 2,198,098. The two independent claims survive only by a single limitation — recovering shaft work from the well stream's expansion and using it to drive the recycle compressor — and that limitation is exactly what the pre-1939 art is silent on in this record. Expect the outcome to turn entirely on whether expansion-engine practice is treated as analogous art and on the 16-year gap to US 2,937,503 as objective evidence.

Caveats

• The page contains no examiner's citation list, so this is not a reconstruction of the actual prosecution. A real §103 opinion requires the file wrapper and the references of record.
• Forward citations ("Cited By") are not prior art to this patent and I have not used them as such; they appear only as objective-evidence context.
• Furnace dates for US 2,156,234, US 2,198,098, US 2,203,954, US 2,231,500, and US 2,245,028 (§102(e) eligibility) and the disclosures of US 2,134,699 / US 2,134,700 are unverified; I did not retrieve them. My search budget was exhausted before I could.
• The patent expired 1960-01-19, so this analysis is historical/academic; no live invalidity challenge is possible.
• No contradiction with the earlier-generated sections: the "no CAFC docket / no litigation / claims 5–6 indeterminate / EP 2309075 is a different patent" findings stand and are consistent with what I found here.