Documented excluded compositions and phonon-instability controls for portfolio integrity

The PFAS-free dielectric and process fluids portfolio rests on a claim architecture that is only as defensible as its screening methodology is honest and symmetric. This asset — a formally documented set of excluded compositions and phonon-instability controls — is the backbone of that intellectual integrity. Rather than a product or a claimed invention in the conventional sense, it is a curated, adversarially generated record demonstrating that the computational screening pipeline applied the same rigorous pass/fail criteria to every candidate material evaluated, including those that failed. In patent prosecution and licensing contexts, a well-structured candor record of this kind is not a weakness; it is a prosecutorial asset that insulates the portfolio's positive claims against obviousness attacks and supports the enforceability of the negative limitations woven into the independent claims. The compositions documented here — spanning AlClO, broad LiAlO2, Na3PS4, black phosphorus, Li3OCl anti-perovskite, Yb-silicates (Yb2Si2O7, Yb2SiO5), AlLaO3, and LaTaO4, among others — were each surfaced as candidates by the materials knowledge graph, subjected to multi-potential phonon stability screening and prior-art review, and then deliberately excluded from broad independent claim scope for specific technical, stability, or freedom-to-operate reasons. This documentation was compiled under the discipline of Rule 56 (37 C.F.R. § 1.56) candor obligations: material information known to applicants that is adverse to patentability must be disclosed. Preserving these failures as a structured, searchable corpus transforms a legal obligation into a competitive differentiator. The timing argument here is subtle but real. As AI-driven materials discovery accelerates, patent examiners, inter partes review petitioners, and licensing counterparties are increasingly sophisticated about claim scope and prior art. A portfolio that can demonstrate — with computational receipts — not just what it claims but what it consciously did not claim, and why, commands a higher credibility premium. For a potential acquirer or licensee evaluating the PFAS-free dielectric and process fluids portfolio, the existence of this candor record is evidence that the positive claims survived a genuine gauntlet, not a cherry-picked selection.

The engines did not fully agree here — the asset carries that uncertainty openly rather than overstating confidence.

The excluded compositions fall into two mechanistically distinct failure categories: prior-art exclusions and dynamic-instability exclusions. Understanding both is necessary to appreciate the screening rigor. The prior-art exclusions comprise compositions that are scientifically plausible as dielectric or process fluid candidates but are anticipated or rendered obvious by existing literature and patent filings. Li3OCl anti-perovskite is excluded because it appears directly in WO 2017/139827 and in the foundational work of Zhao and Daemen (2012), as well as in a 2026 preprint by Xia — three independent prior-art sources that make any broad claim covering this material vulnerable to anticipation. Yb2Si2O7 and Yb2SiO5 are excluded on a combined prior-art and property basis: US 12,312,282 covers Yb-silicate compositions, and computational evaluation of the bandgap reveals a metallic artifact inconsistent with dielectric applications, making exclusion doubly justified. AlLaO3 is excluded as anticipated. LaTaO4 and broad LiAlO2 are excluded on compositional scope grounds — "broad LiAlO2" captures a class whose boundaries overlap with prior art in ways that would compromise the independent claim's novelty perimeter. The dynamic-instability exclusions are the more computationally intensive category, and they constitute the most technically transparent portion of the candor record. The paradigm case is Fe3O4 doped with La and Mn: this composition was subjected to full phonon dispersion calculation and returned 441 imaginary phonon modes — a catastrophic result indicating the structure is mechanically unstable in a fundamental sense. A material with 441 imaginary modes does not represent a local energy minimum on the potential-energy surface; it would spontaneously distort or decompose under any realistic thermodynamic condition. This is not a borderline case. Similarly, Bi2Te2S, InSbS3, and BiSeClO3 each failed phonon stability screening and are preserved as documented negative controls. For the multi-engine screening specifically, the machine-learning interatomic potential (MLIP) evaluation of the excluded set returned a result equivalent to instability consensus: the imaginary phonon frequency reaching -10.6 THz (a strongly negative imaginary mode, indicating the structure sits at a saddle point rather than a minimum). Li3OCl was included in the cross-validation table for completeness, reinforcing that even a literature-prominent composition receives the same computational treatment as novel candidates. The methodological significance of the instability controls extends beyond the individual failed compositions. In the positive claims of the PFAS-free dielectric and process fluids portfolio, each asserted material has passed a consensus stability screen across multiple independent machine-learning potentials — MACE, CHGNet, MatterSim, and ORB — with no imaginary phonon modes. The existence of documented failures like the Fe3O4+La/Mn case demonstrates that this consensus requirement is not a rubber stamp. The screening pipeline generates genuine failures at a meaningful rate, which validates the statistical credibility of the passes. From a patent-litigation standpoint, this corpus makes it substantially harder for a validity challenger to argue that the applicants' phonon stability criterion is an arbitrary post-hoc limitation; the failure record shows it has real discriminatory power. Na3PS4 and black phosphorus represent a third sub-category: compositions that are scientifically interesting but excluded on functional or compositional-scope grounds related to the specific claim architecture of the PFAS-free dielectric and process fluids portfolio. Black phosphorus, while a genuine two-dimensional dielectric candidate of current research interest, lies outside the compositional perimeter defined by the independent claims and carries its own significant prior-art exposure. AlClO is excluded at the compositional scope level as a negative limitation — a deliberate boundary drawn to prevent claim read-on to a class of compositions that would otherwise create prior-art or freedom-to-operate entanglement.

This asset does not itself address a commercial market in the sense of a product a company would manufacture or license for use. Its value is prosecutorial and portfolio-structural: it supports the enforceability and credibility of the positive claims in the PFAS-free dielectric and process fluids portfolio, which do address substantial markets. To understand the commercial context, one must understand what those positive claims protect and what this candor record defends. The PFAS-free dielectric and process fluids space is shaped by accelerating regulatory pressure. Perfluoroalkyl and polyfluoroalkyl substances (PFAS) face increasingly restrictive regimes in the European Union, the United States, and several Asian markets. Semiconductor process fluids, immersion cooling dielectrics, and specialty insulating fluids that currently rely on PFAS chemistry represent a multi-billion-dollar annual market in aggregate. Replacement chemistries that can be demonstrated — computationally and experimentally — to meet demanding dielectric constant, bandgap, thermal stability, and chemical inertness requirements are the target of significant industrial R&D investment. The positive claims in the broader portfolio are positioned to capture whitespace in this replacement market. The candor record in this asset supports that commercial position in two concrete ways. First, it provides prosecution history that narrowly scopes the claims to compositions with genuine computational validation, reducing the risk that broad, later-filed competitor patents can sweep in the same materials. Second, it documents that the portfolio applicants evaluated and explicitly rejected specific PFAS-adjacent or technically marginal compositions, providing a defensible claim perimeter that a licensing counterparty or acquirer can rely on. While no direct royalty rate or total addressable market figure is attributable to this candor record itself, its role in preserving the enforceability of the portfolio's positive claims makes it a material component of portfolio value for any buyer conducting due diligence on the PFAS-free dielectric and process fluids asset set.

evidences symmetric adversarial screening across asserted and excluded compositions

not asserted; documented as exclusions / negative controls under 1.56 candor

The primary buyers for the PFAS-free dielectric and process fluids portfolio — and the entity most likely to require and value this candor record — are large specialty chemical companies, semiconductor process materials suppliers, and advanced packaging materials companies navigating PFAS replacement mandates. For any acquirer or exclusive licensee conducting legal due diligence on the positive claims of the portfolio, this candor record is a material document: it demonstrates that the prosecution team's Rule 56 obligations were met, that negative limitations have a documented factual basis, and that the claim perimeter was drawn with specificity rather than aspiration. A buyer that intends to enforce the portfolio's claims commercially, or to license them defensively against competitors, benefits directly from the prosecutorial rigor this record provides. Secondary interest may come from companies evaluating the Lattice Graph computational platform itself — the methodology underlying the screening pipeline — rather than the specific PFAS-free dielectric claims. For a platform acquirer, the existence of a structured, machine-readable corpus of labeled negative results (failed experiments and unstable candidates with full computational provenance) is a training and validation asset for next-generation machine-learning interatomic potentials and materials-property predictors. The negative data is systematically underrepresented in published literature and public materials databases, making a curated, adversarially generated set of this kind a genuinely scarce resource for model development.

The principal risk associated with this asset is that its value is entirely derivative of the positive claims it supports. If the PFAS-free dielectric and process fluids portfolio's positive filings are not granted, are narrowed significantly in prosecution, or are invalidated post-grant, the candor record retains little standalone commercial value. It is a supporting document, not an independent revenue-generating asset. This dependency means that a buyer evaluating this asset in isolation should do so only in the context of evaluating the full portfolio; pricing it separately from the positive claims would be methodologically unsound. A secondary risk is completeness: the candor record documents the exclusions that were identified during the screening process, but it cannot guarantee that every potentially anticipatory prior-art reference or every unstable candidate was surfaced. The freedom-to-operate and prior-art landscape in PFAS-free materials is evolving rapidly, and new publications or patent filings could create anticipation risks for compositions not yet documented as exclusions. The de-risking path is continued monitoring of the prior-art landscape — particularly the Yb-silicate, anti-perovskite solid electrolyte, and layered oxyhalide spaces — and updating the candor record with additional exclusions as warranted during prosecution. The methodology for doing so is already in place: the same knowledge-graph and patent-whitespace screening infrastructure that generated the current exclusion set can be run on an ongoing basis against new prior art as it is published.