US 11402323

I'll search for authoritative information on this specific patent number, including any litigation activity.

Let me check specifically for any CAFC appeals or PTAB proceedings involving this patent, and verify assignee details.

Search scope and results

I searched for the literal number 11402323 / US 11,402,323 across available patent and court sources. Results:

• USPTO (patent record): Confirmed. Full text retrieved from the Google Patents mirror of the USPTO record.
• CAFC 2026 dockets: No appeal or docket was found involving US 11,402,323. I found no Federal Circuit proceeding (2026 or otherwise) naming this patent. The only 2026 Exxon-related patent case surfaced was Nearby Systems LLC v. Exxon Mobile Corp., No. 2:26-cv-00260 (E.D. Tex.), which is a district-court case, not a CAFC docket, and it asserts different patents (U.S. 9,532,164; 10,469,980; 11,937,145; 12,185,177) — not 11,402,323. That case was dismissed with prejudice on 06/05/2026. I am not asserting it is unrelated to this patent family in any business sense, only that 11,402,323 is not one of the asserted patents.
• PTAB: No IPR/PGR petition referencing 11,402,323 was found.

⚠️ Uncertainty: The searches returned no evidence of litigation or PTAB challenge specifically about this patent, but absence of results is not proof of absence. I could not access PACER or the USPTO Patent Public Search / PTAB API directly, so litigation status should be verified against those primary sources.

Bibliographic summary

Assignee caveat: The Google Patents record lists "ExxonMobil Technology and Engineering Co" as both the original and current assignee. However, at the 2019 filing date the applicant entity was more likely ExxonMobil Research and Engineering Company (which appears as assignee on the sibling application US 2020/0103390 A1, "Systems and methods for implicit chemical resolution of vacuum gas oils and fit quality determination"). Google Patents frequently normalizes assignee names post-hoc, so I flag the original-assignee identification as not fully verified. A USPTO Assignment Abstract of Title search would resolve it.

Citations of record (8): US 6,662,116; US 7,904,251; US 8,322,200; US 8,682,597; US 2017/0023538; US 2017/0190991; US 2018/0134978; US 2020/0103390.

Abstract (as issued)

Systems and methods for determining performance properties of an unknown composition are disclosed. The performance properties can include a Research Octane Number (RON), a Motor Octane Number (MON), and/or a cetane number. The systems and methods include utilizing an optical spectrum of an unknown composition to determine a model of composition, where the model of composition includes a molecular identity and a relative abundance for components therein. The model of composition is then utilized to calculate one or more performance properties. Additionally, the fit quality for the model of composition is determined by performing a partial least squares analysis on specific spectral regions of interest in the optical spectra of the unknown composition.

Plain-language overview of the independent claim

The granted patent contains 7 claims, with only one independent claim (claim 1); claims 2–7 all depend on claim 1. The diesel- and jet-fuel-covering subject matter recited in the specification's "Additional Embodiments" (Embodiments 8–14) and in the pre-grant publication did not survive into the granted claim set — the issued protection is limited to naphtha boiling range compositions. This is a significant prosecution-history point worth noting.

Claim 1 — the sole independent claim

What it covers, in plain language: A method of estimating octane performance (RON and/or MON) for an unknown naphtha boiling range sample, implemented on a distributed computer system (a "performance property modeler" plus a known-composition spectra database, a compositional-class database, and a user computing device connected via a network), where the modeler itself has three sub-components (composition determiner, fit-quality determiner, property determiner).

The steps:

1. Obtain an optical spectrum of the unknown naphtha boiling range composition.
2. Represent that spectrum as a first weighted combination of library spectra of known compositions (i.e., fit the unknown spectrum as a blend of reference spectra, per the U.S. 6,662,116-style least-squares/non-negative least-squares approach).
3. Generate a model of composition for the unknown sample from that weighted combination — using the composition determiner of the modeler. The "model of composition" gives the molecular identities and relative abundances of the sample's components.
4. Determine at least two "partial representations" — separately, for two different compositional classes drawn from the compositional-class data source — by partial least squares (PLS) analysis. Each partial representation is its own weighted combination of library spectra, corresponds to spectral regions associated with that class, and yields both a leverage value and a residual value. In practice this means: pick, e.g., n-paraffins and olefins, restrict the spectrum to their characteristic IR bands, and run PLS on each restricted region.
5. Calculate a combined leverage value from the (at least) two leverage values (e.g., Eq. 6, using a weights vector and the 90% confidence interval).
6. Calculate a combined residual value from the (at least) two residual values (e.g., Eq. 3).
7. Identify the weighted combination as having "good fit quality" if the combined leverage is ≤ 1, the combined residual is ≤ 1, or both. This is the core diagnostic: it tells the operator whether the reference library actually represents the spectral regions (i.e., compound classes) that matter for the unknown sample.
8. A concurrency limitation: at least part of the model-of-composition generation and at least part of the good-fit-quality identification must occur concurrently, with the fit-quality work done by the fit quality determiner using data from the compositional-class data source.
9. Estimate RON, MON, or both from the model of composition, using the property determiner of the modeler.

Practical gist: It's a two-stage analytical workflow — first invert the IR spectrum into a molecular composition via library blending, then compute the octane number from that composition, while running a class-specific PLS "goodness-of-fit" check on the spectrum to validate that the reference library is adequate before the operator relies on the result. The system-architecture and concurrency recitations are notable claim-narrowing additions relative to the pre-grant publication.

Notes on the literal claim text

• The claim-1 preamble contains a duplicated transition: "…comprising, comprising:". I have not corrected it; it is reproduced literally from the record as retrieved.
• Dependent claims narrow as follows: claim 2 — the compositional classes are n-paraffins, iso-paraffins, olefins, naphthenes, or total aromatics; claim 3 — specific FT-IR wavenumber ranges for those regions (e.g., 1350–1550 cm⁻¹ and 2750–3000 cm⁻¹ for n-paraffins; 900–1200 cm⁻¹ and 1300–1600 cm⁻¹ for iso-paraffins; etc.); claim 4 — naphtha boiling range definition (T5 ≥ ~50 °F/10 °C, T95 of 350 °F/177 °C or more); claim 5 — FT-IR spectrum; claim 6 — library includes spectra from at least two separate refineries; claim 7 — library includes at least 50 individual known naphtha boiling range compositions.

Confidence assessment

• High confidence: All bibliographic data, the abstract, claim text, and claim scope, taken directly from the authoritative patent record provided.
• Low confidence / unverified: (a) the identity of the original assignee at filing; (b) whether any litigation, PTAB, or CAFC activity exists for this specific patent — no such proceeding was found, but I could not reach PACER, PTAB E2E, or the CAFC docket system directly to confirm a negative.

I'll search for litigation involving US patent 11,402,323.

Let me search patent litigation databases specifically.

Let me try more targeted litigation database searches.

Findings: Litigation Involving U.S. Patent No. 11,402,323

No litigation involving U.S. Patent No. 11,402,323 has been identified in the sources searched.

I searched patent litigation aggregators and databases (including general web searches targeting the specific patent number, litigation dockets, and Patent Trial and Appeal Board / IPR materials). The only substantive matches for the number "11402323" concerned the patent's bibliographic and technical record — not litigation:

• Patent record (Google Patents): U.S. 11,402,323 B2, "Systems and processes for performance property determination using optical spectral data," Application No. 16/565,638, filed 2019-09-10, granted 2022-08-02, priority 2018-10-02 (provisional 62/740,032). Inventors Bryan T. Bowie, John E. Lee, and Bryan E. Hagee. Assignee: ExxonMobil Technology and Engineering Co. Legal status: Active (anticipated expiration 2039-09-10).
• Inventor profile pages (patentleaderboard.com) listing the patent — again, no litigation content.
• Unrelated results citing the number string "11402323" as a product code (a pipe-fitting catalog) and references to different patents (e.g., "the '023 patent" in unrelated cases, and U.S. 5,885,023) — these are not U.S. 11,402,323 and were excluded per the instruction not to substitute similar numbers.

Why no litigation is expected to appear

• The patent is owned by ExxonMobil Technology and Engineering Co. and is a process/analytical-method patent covering refinery octane/cetane estimation technology. Such patents are typically practiced internally or licensed rather than asserted in public infringement suits, and no NPE activity appears associated with it.
• The record shows a family spanning US, EP (EP3861321B1), and WO (WO2020072177A1), all with ExxonMobil as assignee — no evidence of re-assignment to a litigation-oriented entity.

Important caveat

My search results are based on publicly indexed sources (general web search, PTAB/IPR materials, and publicly accessible docket aggregator excerpts). I did not obtain a direct, verified query result from the Unified Patents litigation caselist, PACER, or Docket Navigator specifically for this patent number (the one attempt to reach a docket-navigator result was cut off). Therefore:

• I can state that no litigation was found in the sources accessible to me.
• I cannot conclusively certify that no litigation exists, because litigation databases paywall or index with lag, and a district-court complaint that was filed and quickly resolved (or filed under seal) might not surface in open web search.

Recommended verification: a direct, docketed search of PACER (all district courts), the Unified Patents litigation portal (portal.unifiedpatents.com/litigation), RPX Insight / Docket Navigator, and the CAFC docket for "11,402,323" or "16/565,638" as the specific identifiers. If those confirm no hits, the correct answer is that no known litigation exists against U.S. Patent No. 11,402,323.

No auto-correction of identifiers was performed; all numbers are reported exactly as found.

Proceedings overview

Total AIA trial proceedings on US 11,402,323: zero (0). The USPTO Open Data Portal (ODP) structured "PTAB proceedings on file" block returns no IPR, PGR, or CBM for this patent, and targeted web searching surfaced no petitions, institution decisions, final written decisions, or Federal Circuit appeals naming the '323 patent. There are therefore no claims canceled, no claims sustained, no settlements, and no institution denials — the breakdown is simply 0 active / 0 invalidated / 0 sustained / 0 settled / 0 denied. Bottom line for a defendant: the patent is entirely untested at the PTAB — it has neither been hardened by surviving an IPR nor weakened by one. Everything in the claims as granted (claims 1–7, all live) remains available to the patent owner, and no estoppel or adverse PTAB record exists that you can leverage.

⚠️ Confidence caveat: The above rests on the ODP block supplied in this prompt plus public search coverage. I could not run a live PTAB E2E docket query in this session, so treat this as "no proceedings found," not "affirmatively confirmed none exist." If you are relying on this for a real dispute, re-run the query directly at the PTAB E2E / PTABIS portal and cross-check on CourtListener's PTAB docket search before filing anything.

Proceedings

None to report. No proceeding numbers exist to list, and I will not invent any.

Strategic summary

Claim status. Because no AIA trial has ever been instituted against US 11,402,323, all claims remain in their as-granted form: claims 1–7 are UNTESTED (not canceled, not amended). There is no Certificate of Correction, no adverse judgment, and no statutory disclaimer on the record that I located. Claim 1 is the sole independent claim; claims 2–7 are dependents directed to (2) compositional classes, (3) FT-IR wavelength ranges, (4) naphtha boiling-range endpoints, (5) FT-IR as the optical spectrum, (6) a spectra library drawn from ≥2 refineries, and (7) a library of ≥50 known naphtha compositions. Claim 1 is a method claim with a substantial system-architecture element (a "performance property modeler" with a composition determiner, fit-quality determiner, and property determiner, plus a known-composition spectra data source, compositional class data source, computing device, and network) and a concurrency limitation ("at least a portion of the generating … and at least a portion of the identifying … occur concurrently"). Those are the kinds of limitations that invite both § 103 challenges (the architecture may be conventional) and § 112 written-description/enablement attacks on the concurrency recitation — but none of that has been tested.

Estoppel landscape. There is no § 315(e)(2) estoppel of any kind, because estoppel attaches only to petitioners who obtain a final written decision (or whose challenges are otherwise terminated after institution). No petitioner exists, so no one is barred. Practically, this means a defendant today has the entire universe of prior art available — §§ 102, 103, and 112 grounds alike, before the PTAB and in district court — with no risk of being precluded by, or collaterally estopped through, a prior IPR. It also means there is no petitioner-side 35 U.S.C. § 315(b) one-year clock running anywhere that could be inherited.

Pattern signals. None. There is no repeat-petitioner pattern, no defensive aggregator (e.g., Unified Patents, RPX) in the chain, and no indication the patent owner has litigated it at the Board or the Federal Circuit. Equally, I found no district-court assertion of the '323 patent to date — the patent reading (an FT-IR/chemometrics method for determining RON, MON, and cetane of naphtha, jet, and diesel fractions) suggests a refinery-process/analytical-use asset for an operating company rather than the typical troll litigation pattern. That profile is consistent with the observed absence of IPRs, but it also means the assertion history is thin and hard to characterize from public records. One flag worth noting: the citation list shows a sibling ExxonMobil application, US 2020/0103390 A1 ("Systems and methods for implicit chemical resolution of vacuum gas oils and fit quality determination," priority 2018‑10‑02) — same inventors' family, same fit-quality concepts. I have not verified whether that sibling or any related family member has PTAB activity; if your case involves related patents, check the family separately.

Recommended next steps

• Treat the patent as fully live and un-narrowed. Do not assume any claim has been canceled. Any invalidity position must be built from scratch on your own art, not borrowed from a PTAB record.
• If you are a defendant: the absence of IPRs cuts both ways. It means (a) no helpful FWD to cite and no estoppel to exploit, but also (b) no time bar under § 315(b) has been triggered by you or anyone else, so you retain a full menu of PTAB options. If you intend to file an IPR, do it now — there is no possibility that a prior petitioner's filing has pre-seeded a General Plastic / Fintiv discretionary-denial problem for you, but those doctrines will spring up if parallel litigation exists.
• Highest-value attack vectors to evaluate first (subject to your own prior-art search): § 103 over the ExxonMobil family itself — U.S. Pat. Nos. 6,662,116; 7,904,251; 8,322,200; and 8,682,597, all cited on the face of the '323 patent — combined with routine optimization of FT-IR + PLS fit-quality metrics; and § 112 on the concurrency limitation in claim 1, which is functional and arguably lacks structural/algorithmic support.
• Verify the null result yourself: run the patent number through PTAB E2E and CourtListener and, if budget permits, pull the patent's full prosecution file for any later-filed continuation or terminal-disclaimer events. PTAB FWDs are public at the USPTO's PTAB Decisions page.
• Timeline note: since no trial is instituted, there are no institution deadlines, oral hearing dates, or statutory one-year FWD due dates to track for this patent.

Sources consulted: Google Patents full text for US 11,402,323 (claims, priority 2018‑10‑02, filing 2019‑09‑10, granted 2022‑08‑02, assignee ExxonMobil Technology and Engineering Co.); the USPTO ODP "PTAB proceedings on file" block provided in this prompt (empty); and public web searches that returned no IPR/PGR/CBM or Federal Circuit appeal tied to this patent. No live PTAB E2E query was executed in this session — see caveat above.

I'll research the assignment chain for US 11,402,323, starting with the USPTO Assignment Center and cross-referencing ownership signals.

Let me search the assignment databases directly and cross-check assignee history.

Key finding: "ExxonMobil Technology and Engineering Company f/k/a ExxonMobil Research and Engineering Company." Let me pin down the recorded assignment details.

⚠️ Tool-budget note: I exhausted my search-step budget while attempting to pull the actual reel/frame records. I retrieved the corporate-identity evidence below but not a verified USPTO Assignment Center reel/frame listing. I flag every place this limits the analysis rather than filling the gap with invented data.

Inventors

• The provisional was filed 2018-10-02 (62/740,032) and the non-provisional 2019-09-10 (16/565,638), so the working employer at filing was the ExxonMobil research organization headquartered at 1545 Route 22 East, Annandale, NJ.
• No unusual pattern found. I found no evidence of inventors departing the assignee within 12 months of filing, and no evidence of individual (non-corporate) ownership that would typically precede a portfolio sale. All three appear as long-tenure ExxonMobil inventors with multiple patents assigned to the same corporate entity. Absence of a departure signal is a weak-negative — I did not run per-inventor employment-history searches.

Original assignee

• Entity on the issued face of the patent (as recorded in the source text provided): ExxonMobil Technology and Engineering Co.
• Entity at the 2019 filing date: ExxonMobil Research and Engineering Company — this is now resolved, not merely inferred. A publicly filed contract states verbatim: "ExxonMobil Technology and Engineering Company f/k/a ExxonMobil Research and Engineering Company." (Source: Law Insider — Amendment No. 2 to Joint Development Agreement, FuelCell Energy / ExxonMobil Technology and Engineering Company). The FuelCell JDA effective date is 2019-10-31, i.e., contemporaneous with this patent's filing, and the amendment provides that EMTEC "shall replace all references to ExxonMobil Research and Engineering Company."
• This resolves the caveat flagged in the previously generated Bibliographic summary. The earlier section marked the original-assignee identity as "not fully verified." It is now verified as a corporate name change / corporate reorganization, not a transfer of title.
• Primary line of business: ExxonMobil's centralized upstream/downstream/chemical technology and engineering research arm. Announced 2022-01-31, effective 2022-04-01, ExxonMobil consolidated its technology operations into a single organization, "ExxonMobil Technology and Engineering," led by Linda DuCharme (Source: ExxonMobil restructuring release via Rohstoff-Welt, Gulf Oil & Gas). This reorganization is the documented cause of the assignee-name normalization.
• Product embodying the claims: ExxonMobil is an operating integrated energy company. Whether a discrete shipped product practices this analytical method is not determinable from the record — the claims cover a refinery analytical/spectroscopic workflow (octane and cetane estimation), which is typically practiced internally as a process rather than sold as a product. This internal-practice characteristic is the relevant fact for the NPE analysis, not product sales.
• Current status: Operating subsidiary of Exxon Mobil Corporation (NYSE: XOM). No bankruptcy, no dissolution, no Chapter 7/11 proceeding found.

Assignment timeline

Plainly stated: I was unable to retrieve a verified reel/frame record from the USPTO Assignment Center for patent 11,402,323 / application 16/565,638 within my search budget. I therefore cannot confirm whether post-issuance assignment records exist, and I will not invent reel/frame numbers.

What the available evidence supports:

• 2019-09-10 — Application 16/565,638 filed; applicant/assignee ExxonMobil Research and Engineering Company (Annandale, NJ). Standard employer-assignment filing; if an inventor-to-company assignment was recorded, it would be a pre-issuance Assignment of Assignors' Interest.
• c. 2022 — Change of Name / corporate reorganization. ExxonMobil Research and Engineering Company → ExxonMobil Technology and Engineering Company. This is corroborated by the executed contract language ("f/k/a"), which is stronger evidence than a naming convention. Whether this was recorded at USPTO as a formal Change of Name conveyance, and its reel/frame, is unverified.
  • Conveyance: Change of Name (company reorganization, not a sale)
  • Assignor: ExxonMobil Research and Engineering Company
  • Assignee: ExxonMobil Technology and Engineering Company (same corporate person, continued identity)
  • Correspondent: not retrieved
  • Context: internal corporate reorganization into the consolidated technology organization announced 2022-01-31.
• No third-party assignment, security interest, license-and-record, or release was found. Google Patents' legal-events timeline for this patent shows only filing → publication → grant, with no assignment events surfaced — consistent with the assignee never having parted with title.

Interpretation: This is the pattern of a patent that remains with its original operating-company owner. The name on the certificate changed; the owner did not.

Verification still required

Direct lookups I could not complete and that you should run:

• USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search patent 11402323 / app 16/565,638) — retrieve reel/frame, correspondent of record, and execution/recording dates.
• USPTO Patent Center / PEDS for the issued patent's assignment tab.
• EP counterpart EP3861321B1 (granted 2023-04-26) and PCT WO2020072177A1 (status: ceased) for parallel ownership records.

Timeline diagram

timeline
    title Ownership of US 11402323
    2018 : Provisional 62 740 032 filed
    2019 : Nonprovisional filed by ExxonMobil Research and Engineering
    2020 : Application published as US 2020 0103340 A1
    2022 : Patent granted 2022-08-02
         : Assignee renamed ExxonMobil Technology and Engineering
    2039 : Anticipated expiration

(Kept deliberately sparse: no verified reel/frame events exist to plot as transfers.)

NPE / troll-pattern signals

Verdict

Operating-company assertion — not applicable / non-asserting in practice; on ownership criteria, this is an ordinary operating-company-held patent.

Strictly against the menu: this is best characterized as Insufficient data to find any NPE pattern combined with a clear positive finding that the owner is an operating company. There is no asserter in the chain to justify any NPE verdict, and there is no defensive-aggregator termination to justify "Defensive / non-asserting." The record shows a single corporate identity, ExxonMobil Research and Engineering Company → ExxonMobil Technology and Engineering Company, established as a name change (f/k/a language in an executed 2019-context contract), with the current assignee an operating subsidiary of Exxon Mobil Corporation.

Justification in 2–3 sentences: The only ownership event evidenced is the ExxonMobil Research and Engineering → ExxonMobil Technology and Engineering corporate rename, corroborated by third-party contract text "f/k/a ExxonMobil Research and Engineering Company," which is a change of name, not a transfer of title [Law Insider, FuelCell Energy / EMTEC Amendment No. 2]. No reel/frame transfer to an NPE-typic entity was found, and the patent remains with an operating energy company whose business is refining technology — the technology is practiced internally rather than asserted. Caveat: I could not retrieve reel/frame records or the correspondent of record from the USPTO Assignment Center within my tool budget, so "no assignments recorded" is a weak-negative, not a verified negative.

Verification link: USPTO Assignment Center — https://assignmentcenter.uspto.gov/patent/index.html (search by patent number 11402323; the older mirror is https://assignment.uspto.gov/patent/index.html).

Cross-reference flags vs. previously generated sections

1. Resolved: The Bibliographic summary's uncertainty about the original assignee is now closed — the original assignee was ExxonMobil Research and Engineering Company, later renamed to ExxonMobil Technology and Engineering Company. No contradiction; the earlier caveat was correct to raise and is now resolved in favor of a name change.
2. Still open: The prior Litigation summary's conclusion (no litigation found) is consistent with my findings — a patent held by an operating company with no transfer to an asserter has no ownership-chain predicate for an NPE suit.
3. Unresolved due to tool limit: reel/frame numbers, correspondent of record, and confirmation of whether any Change of Name conveyance was formally recorded at USPTO. Do not treat the "no assignments" implication above as verified.

I'll research the cited prior-art references for US 11,402,323 to provide accurate citations, dates, and descriptions. Let me search for the key references.

Let me gather details on the remaining cited references.

Prior-Art Analysis — U.S. Patent No. 11,402,323

Note on the number search: I searched the specific number 11402323. The record returned is U.S. Patent No. 11,402,323 B2 (App. No. 16/565,638), "Systems and processes for performance property determination using optical spectral data." I did not substitute any similar number, and I excluded matches for other numbers (e.g., the "11402323" pipe-fitting product code, unrelated "023 patent" references).

A. The "patent citations" of record for US 11,402,323

The granted patent lists 8 references on its face (examiner/applicant citations). Four are also acknowledged directly in the patent's own Background/Definitions section. All 8 are reproduced below with full citation, dates, description, and the §102 assessment.

Because the effective filing date of US 11,402,323 is the priority date 2018-10-02 (provisional 62/740,032), the AIA (§102) applies. Each reference's availability as prior art is keyed to that date.

B. Reference-by-reference analysis

1. US 6,662,116 B2 — Method for analyzing an unknown material as a blend of known materials calculated so as to match certain analytical data and predicting properties of the unknown based on the calculated blend

• Dates: filed 2001-11-30; granted 2003-12-09.
• Description: The foundational ExxonMobil "virtual assay" patent. It determines IR (or other multivariate) data of an unknown material, fits it to a linear combination of known spectra in a database, and derives an assay property of the unknown from the coefficients and the reference materials' known properties. Its dependent claims expressly recite octane number (claim 37) and cetane number (claim 38) as the predicted "performance property," and saturates/paraffins/naphthenes/aromatics content and aromatic-ring distribution as molecular composition. The 11,402,323 patent adopts this method expressly (equations (1)–(2), non-negative least squares, same orthogonalizations) and incorporates it by reference.
• §102 assessment: This is the single most relevant reference and the conceptual heart of claim 1. It anticipates a hypothetical broad claim to "represent a spectrum as a weighted combination of a library → generate a model of composition → estimate octane/cetane." However, it does NOT anticipate issued claim 1, because it discloses none of: (i) the PLS partial representations per compositional class; (ii) the leverage and residual values; (iii) the combined leverage / combined residual ≤ 1 "good fit quality" test; (iv) the system architecture (performance property modeler with composition determiner / fit quality determiner / property determiner; compositional-class data source; network); or (v) the concurrency limitation. → No clean §102 anticipation of any claim; strongest §103 art.

2. US 7,904,251 B2 — Method for modification of a synthetically generated assay using measured whole crude properties

• Dates: provisional 60/644,908 filed 2005-01-19; granted 2011-03-08. Inventors Martin & Brown.
• Description: Modifies a synthetically generated ("Virtual") assay of a whole crude/condensate/resid by using measured whole-crude property values to adjust the predicted whole-crude and boiling-point-distributed properties so the assay is more accurate and in material balance.
• §102 assessment: Prior art in date, but directed to correcting a virtual assay with bulk property measurements — not to spectral library blending, a molecular model of composition, or PLS fit-quality diagnostics. Does not anticipate claim 1 or any dependent claim. Marginal §103 relevance at best.

3. US 8,322,200 B2 — Method for certifying composition and property variables of manufactured petroleum products

• Dates: provisional 2007-12-21; granted 2012-12-04.
• Description: Analyzes whether a manufactured product meets regulatory/contractual requirements by re-introducing a representative sample back into the on-line process analyzers.
• §102 assessment: Prior art in date, but concerns product certification via on-line analyzers — unrelated to forming a model of composition from an optical spectrum or to PLS fit-quality. Does not anticipate any claim.

4. US 8,682,597 B2 — Estimating detailed compositional information from limited analytical data

• Dates: provisional 2007-10-16; granted 2014-03-25.
• Description: Determines the composition of a material by fitting multivariate analytical data of an unknown whole crude to a database of multivariate data of known whole crudes to obtain an initial composition estimate, then refining the estimate by matching a set of additional analytical data (synthesis to bulk/distributed targets). Notably, its "model of composition" output tabulates n-paraffins, iso-paraffins, saturates, and 1-, 2-, 3-, 4- and higher-ring aromatics — the same compositional classes recited in dependent claim 2 of 11,402,323.
• §102 assessment: Prior art in date. Relevant to the "model of composition from spectral data" concept and to the compositional-class taxonomy of claim 2. But it explicitly uses an iterative bulk/distributed-property synthesis (not PLS leverage/residual partial representations), and it discloses no combined leverage/residual ≤1 fit-quality threshold, no PLS statistics, and none of the system-module/concurrency limitations. Does not anticipate claim 1; good §103 combination material with reference 1.

5. US 2017/0023538 A1 — Methods and Apparatuses for Spectral Qualification of Fuel Properties (Tesoro; inv. D. C. Mertens)

• Dates: filed 2015-07-23; published 2017-01-26; granted as US 10,295,521 B2 (parent disclosures US 9,709,545 B2).
• Description: Determines property values of fuels from spectral data, including construction of a representative spectrum by weighting the spectrum of one fuel with spectra of fuel components, then comparing to calibration data to derive property values. Its claim 2 expressly lists RON, MON, aromatics, olefins, cetane, etc. It uses multivariate regression (e.g., ASTM E1655) calibration models and FT-NIR/MIR/FTIR or Raman spectra.
• §102 assessment: Prior art in date, and the most relevant non-Exxon reference because it ties spectral weighting/combination to RON/MON/cetane output. However, it operates by spectral construction + calibration regression, not by generating a model of composition with molecular identity/relative abundance via library blending, and it discloses no compositional-class PLS leverage/residual analysis or ≤1 fit-quality threshold. Does not anticipate claim 1; §103-relevant for the "weighted spectral combination → fuel property" concept.

6. US 2017/0190991 A1 — Process for producing high octane gasoline component from renewable raw material (Neste Oyj; Sundberg et al.)

• Dates: FI priority 20156043 filed 2015-12-31; published 2017-07-06; granted as US 10,023,815 B2 (WO 2017/114715 A1).
• Description: A refining/process-chemistry invention: co-processing tall-oil pitch (TOP) and vacuum gas oil (VGO) in a catalytic cracking unit to make a high-octane gasoline component; the inventors measure the RON/MON increase of the light distillate.
• §102 assessment: Prior art in date, but non-analogous to the claimed subject matter. It mentions RON/MON only as measured fuel properties — it discloses no optical spectroscopy, no library blending, no model of composition, and no fit-quality analysis. Does not anticipate any claim.

7. US 2018/0134978 A1 — Fuel compositions for controlling combustion in engines (ExxonMobil)

• Dates: filed 2017-10-20; published 2018-05-17; granted as US 10,584,292 B2 (related US 10,550,344 B2).
• Description: Naphtha boiling range fuel compositions with improved combustion/ignition behavior; controls the combined wt% of n-paraffins and isoparaffins containing a straight-chain propyl group; characterized against RON and MON (and sensitivity RON−MON); optionally T5 ≥ ~10 °C and T95 ≤ ~233 °C.
• §102 assessment: Prior art in date (§102(a)(1) printed publication as of 2018-05-17 — before the 2018-10-02 priority date). Relevant to the naphtha boiling range definition of dependent claim 4 and the RON/MON and n-/iso-paraffin subject matter. But it is a fuel-composition disclosure, not an analytical method — it discloses no spectral library blending, no model of composition, and no PLS fit-quality. Does not anticipate any claim.

8. US 2020/0103390 A1 — Systems and methods for implicit chemical resolution of vacuum gas oils and fit quality determination (ExxonMobil)

• Dates: effective filing 2018-10-02 (identical to 11,402,323); published 2020-04-02. Commonly owned companion/sibling application.
• Description: Highly related subject matter — implicit chemical resolution of VGOs with fit-quality determination (the same applicant's parallel disclosure).
• §102 assessment (important caveat): Although cited on the face of the patent and closely related in subject matter, this publication cannot serve as §102 prior art against claim 1. It shares the same effective filing date (2018-10-02) as the patent, so under §102(a)(2)/§102(e) it is not "before the effective filing date" of the claimed invention; and even if dated earlier, it would be removed by the §102(b)(2)(C) common-ownership exception (same ExxonMobil applicant). No anticipation. Its citation reflects family/companion status, not prior art.

C. Overall §102 conclusion

No cited reference — singly or on its face — anticipates any claim of US 11,402,323 under 35 U.S.C. § 102.

The reason is structural: issued independent claim 1 is a narrow, specific combination requiring (a) a defined system architecture (performance property modeler with composition determiner, fit quality determiner, property determiner; known-composition spectra data source; compositional-class data source; computing device via network); (b) at least two PLS partial representations, each tied to a compositional class and each yielding both a leverage value and a residual value; (c) a combined leverage and combined residual, with a ≤1 "good fit quality" identification; and (d) a concurrency limitation. No single cited reference discloses all of these.

Best §102/§103 candidates, ranked:

1. US 6,662,116 B2 — discloses the core "spectrum → weighted combination of library spectra → model of composition → performance property (incl. octane/cetane)" pipeline; the strongest §103 base reference against claim 1's general concept, but silent on the PLS fit-quality elements.
2. US 8,682,597 B2 — supplies the "model of composition including n-/iso-paraffin and 1–4-ring aromatic classes" concept (claim 2) and refining via additional data; combinable with #1.
3. US 2017/0023538 A1 (Tesoro) — supplies weighted spectral combination to derive RON/MON/cetane, but via calibration rather than a molecular model of composition.

A §103 obviousness rejection could plausibly be built from US 6,662,116 + US 8,682,597 (and/or US 2017/0023538) for the broad method, but the PLS compositional-class leverage/residual fit-quality ≤1 test and the concurrency + system-module recitations in claim 1 remain the distinctive, non-anticipated features. References 2, 3, 6 and 7 appear to be context/background citations with little or no bearing on patentability of claim 1.

D. Additional notes and caveats

• Forward citations (not prior art): The record also lists three later patents citing US 11,402,323 (CN 111899802 B; CN 114741953 B; CN 118464877 B). These are cited-by references published 2024–2025 and are therefore not §102 prior art; I excluded them accordingly.
• Common-ownership family issue: Reference #8 (US 2020/0103390 A1) and the patent share the same 2018-10-02 priority date and the same applicant. This is a recurring feature of the ExxonMobil portfolio (references #1–#4, #7, #8 are all ExxonMobil), meaning several "citations" are self-citations of a common research program rather than third-party art.
• "Cited by examiner" vs. "cited by third party": In the record, references #1–#4 are the applicant-acknowledged/background art, and #5–#8 are marked as third-party or companion citations. I treated all 8 as the patent's "patent citations" per the task, since the record does not, in the retrieved text, definitively segregate examiner vs. applicant origin.
• Verification limits: Reference numbers and dates above are taken literally from the authoritative patent record (USPTO/Google Patents mirror) and corroborated by search results; the descriptions of references #5 and #7 were verified against their published full-text PDFs. I did not auto-correct any identifier. If a formal prosecution-history determination is needed, the USPTO PatentCenter IFW for App. 16/565,638 and the USPTO Assignment Abstract of Title should be consulted to confirm which references were applied (versus merely listed) and by whom.

No patent or technical identifier was auto-corrected; all numbers are reported exactly as retrieved.

I'll analyze obviousness under § 103 using the prior art of record, and verify the substance of the key references before reasoning.

Key discovery: the record's own citation US 2020/0103390 is the same-family sibling, and I've found a directly-on-point ExxonMobil reference on Fit Quality Ratio (FQR). Let me verify a few more threads.

§ 103 Obviousness Analysis — U.S. 11,402,323 B2

Scope note. As established in the earlier sections, the granted patent has one independent claim (claim 1) and six dependents, all limited to naphtha boiling range compositions; the diesel/jet subject matter from the pre-grant publication was cancelled. Everything below is directed at claim 1 and its dependents as issued.

Important addition to the earlier work: my prior-art sweep of the record turned up a reference that is not in the § 102 citation list but is squarely on point for the hardest limitation in claim 1 — US 2006/0047444 A1 (Brown et al., ExxonMobil Research & Engineering; published 2006‑03‑02; also WO 2006/023800 A2/A8) — which discloses the "Fit Quality Ratio" (FQR) and, critically, uses FQR = 1 as the pass/fail cutoff and computes it against a confidence interval. I flag this because the earlier litigation/prior-art sections stated that the FQR/cutoff-1 concept was not located in the record; it is located, just not in the examiner's citation list. It is § 102(a)(1) art (published 12+ years before the 2018‑10‑02 effective filing date) and nearly overlapping in disclosure with the inventor entity's own earlier work.

1. Governing framework and person of ordinary skill

PHOSITA (proposed): a person with a B.S./M.S. in chemical engineering, analytical chemistry, or chemometrics, plus 2–5 years' experience applying multivariate spectroscopy (FT‑IR/NIR) and multivariate calibration statistics to petroleum streams, familiar with: (a) least‑squares/NNLS spectral blending against reference libraries; (b) PLS regression per ASTM E1655; (c) leverage/residual outlier statistics; (d) refinery-stream molecular composition (HDHA/PIONA-type class breakdown); and (e) composition-based octane correlations.

Legal posture: KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007) (predictable combination of known elements; "obvious to try" where finite number of identified, predictable solutions); Graham v. John Deere, 383 U.S. 1 (1966) (scope/content, differences, PHOSITA level, secondary considerations). Motivation may come from the problem to be solved, the teachings of the references, common assignee/field, or design incentives — a teaching-away or bodily-incorporation requirement is not required. Where a range is claimed, In re Aller / In re Woodruff allow routine optimization absent criticality.

Critical structural point for § 103: the patent's own specification admits away the final step. It states that RON/MON are calculated "using various conventional techniques" and that "any convenient method" can be used, citing Ghosh et al., Ind. Eng. Chem. Res. 2006, 45, 337‑345 (and, for cetane, 45, 346‑351). Those Ghosh papers are therefore applicant-admitted prior art, and the composition → RON/MON conversion cannot supply patentability.

2. The prior-art references and what each teaches

(a) References of record (from the § 102 citation list)

(b) Supplementary art located (not in the record's list, but § 102(a)(1) printed publications)

• US 2006/0047444 A1 (Brown et al., ExxonMobil) — the improvement to US 6,662,116 that the '323 inventors' employer itself filed. It discloses "Fit Quality (FQ)" and the "Fit Quality Ratio (FQR) = the ratio of the Fit Quality to a Fit Quality Cutoff (FQC)," and in Example 3 states that "the vertical lines represent the point at which FQR equals 1" — i.e., the pass/fail threshold of one. It further discloses that FQR can incorporate confidence intervals ("the FQC is set such that the Confidence Interval for the prediction … is approximately 3 volume percent"), and that FQR values "calculated using different sets of known (reference) materials and/or different inspection inputs can be compared." This reference speaks directly to claim 1's fit-quality gate, its threshold of one, its weighting/pooling of fit statistics, and its cross-reference-set motivation.
• ASTM E1655 (Standard Practices for Infrared Multivariate Quantitative Analysis, ‑00/‑05/‑17) — a printed publication evidence of the knowledge of the art. It: names PLS as a standard technique (§ 4.2); requires outlier statistics including "high leverage samples (samples whose spectra contribute a statistically significant fraction of one or more of the spectral variables)" and samples whose reference values are inconsistent (§ 4.3); uses outlier statistics "to detect when the analysis involves extrapolation of the calibration" (§ 6.1.6); and notes that scanning instruments allow "additional options for selections of spectral areas less sensitive to band shifts and extraneous noise" (§ 7.5).
• The Ghosh octane/cetane composition-based models — applicant-admitted prior art (cited in the '323 specification itself).

3. Element-by-element mapping of claim 1

Net result: every element is disclosed or is a predictable, routine variant of a disclosed element, and the only arguably "new" aspect is the combination of (i) blend-based MoC, (ii) class-partitioned PLS sub-region statistics, (iii) a pooled unit-threshold fit gate, and (iv) concurrency. Under KSR, a combination of familiar elements according to known methods, yielding no more than predictable results, is obvious.

4. Enumerated § 103 combinations and the motivation to combine

Combination A — the primary 3‑reference case

US 6,662,116 + US 8,682,597 + US 2017/0023538 (Mertens/Tesoro), with ASTM E1655 as evidence of the level of skill

• Motivation: All three are in the same field (spectroscopic determination of petroleum/fuel properties) and address the same problem (avoiding slow wet-chemistry/engine assays). US 8,682,597 is an express improvement on US 6,662,116 by the same assignee, and expressly acknowledges that "the quality of the eventually estimated composition crucially depends upon the reference" — a direct, articulated reason to add a fit-quality check on the reference blend. US 2017/0023538 supplies the PLS/ASTM‑E1655 multivariate-regression practice and the IR band ranges, and confirms RON/MON as the target property.
• Covers: limitations 1 (file-sharing/analyzer context), 2–4, 5–6 (PLS on bands, leverage/residual per ASTM E1655), 11 (RON/MON).
• Gap: the class-specific partitioning of the PLS models and the pooled threshold-of-one gate are not expressly stated.

Combination B — Combination A + US 2006/0047444 A1 (the strongest case)

• Adds the Fit Quality Ratio with a unit (1.0) cutoff and confidence-interval-based normalization — the very numeric anchor of claim 1's "one or less."
• Motivation: same assignee, same analytical program, same problem statement; the reference's own stated purpose is "comparing the quality of property predictions made using different sets of known (reference) materials and different inspection inputs," which is exactly the '323 patent's stated use case ("valuable when utilizing these processes at multiple refineries or after slate changes"). A PHOSITA starting from 6,662,116/8,682,597 would consult the assignee's own published improvement to 6,662,116 as a matter of ordinary course.
• This is the combination I would expect to carry an obviousness challenge.

Combination C — Combination B + US 8,322,200

• Adds the networked on-line analyzer architecture (data sources + computing device + network) and the process-control motivation for concurrent/fast computation.
• Motivation: refinery on-line analysis has always demanded results within the decision window; running the composition fit and the fit-quality check in parallel rather than serially is a routine engineering choice with a predictable benefit (latency), satisfying KSR's "predictable results" prong and In re Zurko-type deference to routine engineering decisions.

Combination D — Combinations A–C + US 2018/0134978 (and the applicant-admitted Ghosh 2006 papers)

• Supplies the naphtha-specific composition↔octane relationship at the n-paraffin/isoparaffin class level, providing the motivation to build class-partitioned PLS models precisely for the classes that drive RON/MON — the last gap in Combinations A–C for the "first/second compositional class" limitations.
• Motivation: a PHOSITA seeking to validate a RON prediction would naturally validate the fit in the spectral regions corresponding to the compositional classes that control RON (olefins, n-paraffins, iso-paraffins — which is exactly what the '323 specification itself says an operator "may be most interested in"). This is the reference the patent's own disclosure recites as the operator's priority, i.e., a design choice dictated by the art.

Overall motivation statement: the references are analogous art in a single field, several are commonly owned, several expressly cross-reference each other (8,682,597 → 6,662,116; 2006/0047444 → 6,662,116; 2018/0134978 → same assignee and same naphtha/octane subject matter), the problem to be solved is expressly identified in the prior art ("systematic reference selection techniques do not exist"), and every added element (class windows, PLS, leverage/residual, unit-threshold FQR, parallel computation) is a known technique with a predictable result. This is a textbook KSR combination.

5. The dependent claims

6. Anticipated counter-arguments and how they fare

1. "No single reference discloses the whole combination." Correct, and irrelevant — § 103 permits combination. The strongest available rebuttal to the cited combination would have to show the references teach away or that the combination was unpredictable; nothing in the record suggests either. To the contrary, 8,682,597 invites the improvement.
2. "The class-partitioned PLS sub-region statistic is the point of novelty." This is the only genuinely strong non-obviousness argument. Neither 6,662,116, 8,682,597, nor 2006/0047444 expressly partitions its fit statistics by spectral sub-region keyed to compound class. A patentee could argue that a spectrum-wide FQR (2006/0047444) is a different diagnostic from a pooled, class-weighted, sub-region statistic, because a whole-spectrum fit can be excellent while a diagnostically critical narrow region (e.g., olefins) is poorly represented — which is precisely the '323 specification's stated insight. Rebuttal: KSR and In re Aller make partitioning/optimizing a known statistic into known sub-ranges an obvious design choice; ASTM E1655 § 7.5 expressly directs the analyst to select spectral sub-areas; and 2017/0023538 performs PLS on defined IR bands. The patent supplies no comparative data showing that the pooled sub-region statistic produces an unexpected or qualitatively superior result.
3. "Concurrency of the two computations is a process-control advance." Weak. The claim says "at least a portion … occur concurrently," with no specified degree of overlap, so any pipeline/parallel implementation reads on it. Absent a demonstrated criticality (e.g., a required throughput or control-loop period), this is routine engineering.
4. "The system-architecture recitations are structural." Either they are generic functional language performed by a general-purpose computer (little patentable weight, In re Katz/WMS Gaming), or they are conventional and disclosed by US 8,322,200. They add nothing to the § 103 calculus either way.
5. § 112 issues (adjacent, not § 103): the claim-1 preamble contains a duplicated transition ("…comprising, comprising:") and the concurrency clause is grammatically garbled ("…occur concurrently determined using the fit quality determiner…"). The concurrency limitation is also arguably indefinite because no quantitative overlap is given. If construed narrowly to require simultaneous processing threads, the limitation becomes easier to attack as an obvious parallelism; if construed broadly, it excludes nothing.

7. Graham factor 4 — objective indicia

On the face of the record there is no evidence of unexpected results, criticality of the ≤1 thresholds, or commercial success attributable to the claimed combination:

• No data comparing class-pooled sub-region statistics against spectrum-wide FQR.
• No data showing that "1" is a critical value rather than the inherited normalization point from US 2006/0047444.
• No range-criticality data for the claim-3 wavenumber windows.
• The patent's stated advantage ("fit quality indication … so that the operator can rely on the model") is the same benefit already asserted by US 2006/0047444.

If a patentee were to attempt to save the claims, its best path would be secondary evidence developed during litigation — e.g., showing the claimed class-partitioned gate materially outperforms the spectrum-wide FQR of the assignee's own earlier art. That evidence does not exist in the record I have.

8. Bottom line

• Claim 1 is, in my assessment, vulnerable to a § 103 challenge, with the most probable combination being US 6,662,116 + US 8,682,597 + ASTM E1655 + US 2017/0023538 (Mertens/Tesoro), further in view of US 2006/0047444 A1 and, for the architecture/concurrency limitations, US 8,322,200, and with US 2018/0134978 and the applicant-admitted Ghosh 2006 papers supplying the naphtha-octane/class motivation.
• The pivotal reference is US 2006/0047444 A1, which supplies the Fit Quality Ratio with a cutoff at one and confidence-interval normalization — i.e., the numeric core of claim 1's "one or less" gate — and which is the same assignee's own published improvement to the patent's principal cited reference. It is not in the examiner's citation list, and its absence is the single most likely explanation for why claim 1 issued.
• The weakest link for a challenger is the class-partitioned, pooled PLS leverage/residual statistic itself; no located reference expressly keys fit statistics to compound-class sub-regions of the unknown spectrum. That limitation would have to be met by the combination rationale (ASTM E1655 band selection + 2017/0023538 band-wise PLS + 2018/0134978's class↔octane linkage), which is a KSR-type predictable-variation argument, not a reference-by-reference disclosure.
• Dependent claims 2–7 add essentially nothing: they recite conventional IR band ranges, a feedstock definition, an FT-IR selection, a library-spanning-site choice, and a library-size floor — each of which is disclosed or routine-optimized in the art of record (with US 2006/0047444's 545–562-crude cross-validation and ASTM E1655's minimum calibration-set sizes being particularly damaging to claims 6 and 7).

Confidence: High on the content and dates of 6,662,116, 8,682,597, 2017/0023538, 2018/0134978, and the specification's own Ghosh admission (all verified against the record/retrieved text). High on the existence and substance of the FQR/cutoff-of-one disclosure in US 2006/0047444 A1 (verified via retrieved text of that publication and its WO counterpart). Moderate-to-low, and explicitly flagged as unverified, on two points: (a) whether any additional art cited during EP 3861321 B1 prosecution or in any opposition would strengthen or weaken the case — I could not retrieve the EP file (my final searches were cut off); and (b) the precise scope the examiner attributed to US 2020/0103390 during prosecution, given that I conclude it is not available as prior art. No identifier has been auto-corrected: all patent/application numbers are reproduced exactly as found.