Cerium-based ternary silicide group (CeCuSi / CeNiSi / CeRuSi) for cryogenic and superconducting device applications

The cerium-based 1:1:1 ternary silicides — CeCuSi, CeNiSi, and CeRuSi — occupy a distinctive corner of the heavy-fermion landscape where f-electron physics, structural tunability, and practical device integration intersect. Cerium's 4f electrons sit at the boundary between localized and itinerant behavior, which drives the correlated-electron ground states — including unconventional superconductivity, non-Fermi-liquid behavior, and quantum criticality — that make this compound class scientifically and technologically compelling. In a cryogenic electronics environment that is under sustained pressure from quantum computing buildout, photon-counting detector arrays, and next-generation defense sensing, the question is no longer whether these materials are interesting but whether they can be qualified, sourced, and integrated. This dossier addresses that gap. The strategic framing here is honest and precise. These three cerium silicide compositions are known to the literature; no composition-of-matter claim is asserted or assertable for the bare formulas. What Lattice Graph has constructed instead is a method-of-screening and device-use intellectual property position that is counsel-reviewed, freedom-to-operate-prescreened across more than 300,000 materials patents, and supported by computational phonon-stability validation across four independent machine-learning interatomic potential engines. The value to an acquirer or licensee is not a monopoly on the compound class — it is a ready-to-litigate, ready-to-license claim package covering how these materials are identified for device use and how they are integrated into cryogenic device architectures, packaged alongside a documented computational provenance that meets modern patent-office and due-diligence standards. This cerium arm is the first of three genera in the rare-earth silicide superconductor candidate portfolio. Its positioning as the lead arm reflects the depth of experimental literature on cerium heavy-fermion compounds, which gives any device-use claim a broader commercial read and makes the open validation gates — first-party DFPT calculations and direct calorimetric Tc measurement — achievable with well-understood experimental protocols.

Each candidate is validated by multiple independent machine-learning interatomic potentials. A material advances only when the engines agree on phonon (dynamic) stability — disagreement is surfaced, not hidden.

CeCuSi, CeNiSi, and CeRuSi each adopt variants of the layered PbFCl-type or CeFeSi-type structure (space group derivatives of the ThCr2Si2 parent), which is the same structural family that hosts some of the most well-studied unconventional superconductors in condensed matter physics. In these ternary silicides, alternating Ce-Si and transition-metal layers create a quasi-two-dimensional electronic environment where hybridization between Ce 4f states and conduction electrons from the Cu, Ni, or Ru sublattice can be tuned systematically. The transition-metal choice sets the effective hybridization strength (the Kondo coupling), which in turn controls whether the ground state is a Kondo insulator, a heavy Fermi liquid, or a magnetically ordered state with possible superconducting instability at its critical point. The CeRuSi member is of particular interest because Ru 4d states bring stronger spin-orbit coupling and a larger hybridization window, placing it closer to the quantum critical regime where unconventional Cooper pairing is most likely. All three compositions have passed a phonon consensus evaluation using four independent machine-learning interatomic potential engines: MACE, CHGNet, MatterSim, and ORB. Requiring a majority consensus across four distinct model architectures and training sets — rather than relying on any single potential — is a materially stricter standard than most screening campaigns apply, because it reduces the risk that stability artifacts from one potential's training distribution are mistaken for physical reality. A majority consensus across these four engines, with no imaginary phonon modes in the majority of models evaluated, indicates that the reported layered structures are dynamically stable at the harmonic level under the conditions simulated. This is the phonon (dynamic) stability criterion: the absence of soft modes confirms that the crystal will not spontaneously distort or decompose under small atomic displacements, which is a prerequisite for synthesizability and device reliability. The freedom-to-operate prescreen was run against more than 300,000 materials patent documents before claims were drafted, explicitly mapping the compositional and use-claim landscape to identify white space. The conclusion is that while the bare cerium silicide compositions appear in academic literature and potentially in broad compositional disclosures, the specific method-of-screening claims and device-integration use claims occupy defensible territory. This prescreen is a key part of the computational workflow: it happens in the knowledge graph layer before any claims language is generated, ensuring that the claim strategy is shaped by the actual patent landscape rather than retrofitted to it. The open validation gates are clearly defined and achievable. First-party DFPT (density functional perturbation theory) calculations on the cerium arm flagships would replace the machine-learning phonon consensus with direct quantum-mechanical phonon dispersion curves and, importantly, would yield dielectric tensors and Born effective charges — data needed to estimate electron-phonon coupling constants and assess whether phonon-mediated superconductivity is energetically plausible. Direct calorimetric measurement of the superconducting transition temperature (Tc) is the definitive experimental gate; this is straightforward for any group with a dilution refrigerator and high-purity synthesis capability. Neither gate requires novel experimental infrastructure, which is an important de-risking point for a prospective partner or acquirer evaluating the path from computational validation to device-ready qualification.

The addressable commercial opportunity for cryogenic and superconducting electronics materials sits in an estimated $1–5 billion range, spanning superconducting quantum computing components, cryogenic sensing and detector arrays (including single-photon detectors and bolometric arrays for radio astronomy and defense), and low-temperature signal processing hardware. These are estimates, not audited figures, and the actual realizable licensing revenue depends heavily on which device categories adopt ternary silicide components and on the specific performance benchmarks achieved in experimental qualification. The market is nevertheless real and growing: quantum computing hardware buildout is driving demand for any material that can operate at millikelvin temperatures with low noise and reproducible fabrication, and the heavy-fermion silicide class sits at the intersection of those requirements. The primary buyers in this market are companies and national laboratories building cryogenic electronic systems who need qualified, intellectually protected material choices at the component level. This includes quantum processor manufacturers, cryogenic detector array integrators for defense and scientific applications, and specialized foundries developing superconducting interconnect and sensor technology. Licensing logic in this space typically runs on a per-device or per-wafer royalty for incorporated materials or processes, with upfront milestone payments tied to experimental qualification events such as Tc confirmation and device yield demonstrations. The counsel-ready nature of this dossier — including documented FTO prescreen and computational provenance — shortens the time from acquisition to licensing execution. It is worth being direct about the current stage: this is a method-of-use and screening-method IP position supported by computational validation, not yet by device demonstrations. The market opportunity is contingent on completing the experimental validation gates. A partner who already works in the cryogenic device space and has synthesis infrastructure can close those gates rapidly; a financial acquirer without that capability would need to plan for a multi-year experimental campaign. The portfolio value here is partly in the claim package itself and partly in the negative-results atlas and computational workflow that Lattice Graph has built — the knowledge of which material variants fail and why is commercially valuable in its own right because it compresses the screening timeline for any follow-on development program.

Ce-anchored heavy-fermion arm with device-use whitespace

method-of-screening + device-use only; no composition-of-matter to literature-known Ce members

The most natural acquirers or licensees for this dossier are companies with active cryogenic electronics programs who need a documented, computationally supported qualification pathway for heavy-fermion silicide candidates. This includes quantum computing hardware manufacturers developing superconducting qubit platforms, defense-adjacent cryogenic sensor integrators, and specialized foundries working on superconducting interconnect technology. For these buyers, the value proposition is the counsel-ready claim package and documented computational provenance, which compress the internal IP development timeline and provide a defensible basis for incorporating cerium silicide materials into device qualification programs. National laboratories with long-range programs in quantum sensing and unconventional superconductor characterization are also plausible partners, particularly for co-development arrangements in which the laboratory completes the experimental validation gates in exchange for licensing rights. A strategic acquirer buying the broader rare-earth silicide portfolio — of which this cerium arm is the first of three genera — would gain a platform claim position across the RE-1:1:1 ternary silicide family, a reusable computational screening workflow, and the associated negative-results atlas that documents which compositional variants fail and why. That atlas has independent commercial value for any materials development program in cryogenic electronics because it eliminates redundant screening effort. Financial acquirers without device-development infrastructure should weigh the experimental validation gap carefully; the path to royalty revenue requires completing the DFPT and calorimetric Tc gates, which implies a multi-year timeline and meaningful experimental investment before the licensing position becomes fully monetizable.

The primary risk is the composition-of-matter ceiling: because these cerium silicide compositions are literature-known, the claim scope is permanently bounded to method-of-screening and device-use, which limits the breadth of the licensing position compared to a novel-composition asset. A competitor who independently identifies the same materials through their own computational workflow and implements a device without using the claimed screening method would not infringe. This is an inherent limitation of the asset class and is priced into the estimated addressable market; buyers should treat this as a supporting position in a broader IP strategy rather than as a standalone blocking patent. The experimental validation gap is the second material risk. Heavy-fermion cerium compounds are notoriously sensitive to sample quality, stoichiometry, and impurity phases; Tc values can shift substantially between synthesis batches, and non-Fermi-liquid behavior near the quantum critical point can mask or mimic superconducting signatures. Completing the DFPT calculations and calorimetric measurements with the rigor required for patent written-description support and commercial qualification is non-trivial for CeRuSi in particular, where relativistic effects on the Ru 4d and Ce 4f manifolds require careful treatment in any DFT approach. The roadmap to de-risking both issues is clear: engage a synthesis-and-characterization partner early, prioritize DFPT with spin-orbit coupling on CeNiSi and CeRuSi as the most computationally tractable leads, and structure any licensing deal with milestone payments tied to confirmed Tc demonstration rather than upfront lump-sum valuation.