Hafnate, rare-earth sesquioxide, and borate companion dielectrics for packaging and photonic substrates

This asset covers a carefully assembled suite of companion dielectric materials — hafnate perovskites, rare-earth sesquioxides, borates, and nitrides — that extend the wide-bandgap dielectric platform within the catalysts and energy-conversion materials portfolio into two adjacent, high-value application domains: advanced semiconductor packaging and photonic/quantum substrates. The strategic logic is straightforward: the core platform already stakes a claim on a family of wide-bandgap dielectrics validated for radiation-hard and high-frequency environments; this companion arm widens the addressable target by populating an expanded material menu with compositions purpose-fit for high-k capacitor integration and photonic-substrate roles, covering the permittivity range from roughly 13 to 41 in a single claim family. Rather than being a speculative extension, these compositions are grounded in genuine first-principles computation — DFPT-derived total permittivities and HSE06 electronic structure calculations — which places them well ahead of the typical "hypothetical composition" filing. The timing argument for this asset is structural. Advanced packaging is undergoing a once-per-decade substrate technology refresh, driven by chiplet integration, heterogeneous packaging, and the need for low-loss, high-k dielectrics compatible with back-end-of-line thermal budgets. Simultaneously, photonic integrated circuits and quantum-memory modules are demanding substrate and cladding materials with precise refractive-index and optical-loss properties that conventional SiO2 and Al2O3 cannot simultaneously satisfy. Rare-earth sesquioxides and borates sit at a materials crossroads: they are optically active (relevant for laser-host and quantum coherence applications), they tolerate extreme radiation environments, and several members carry DFPT-computed dielectric constants that position them as drop-in or additive-stack high-k candidates. This portfolio captures that intersection with computed evidence, before experimental thin-film process development has commoditized any individual member. Honest framing matters: this is a companion arm, not a standalone flagship. Its role within the broader family is to provide strategic depth — additional claim coverage across a set of compositions that share the same dielectric and structural logic as the lead compositions but have distinct chemistries, supply profiles, and device-integration paths. That supporting role is itself commercially valuable: a licensing or acquisition buyer gains not just the lead compositions but a defensible envelope of alternatives that prevents design-around by competitors selecting closely related hafnates or borates not covered by narrower claims.

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.

The material universe covered here spans three chemical families with distinct dielectric physics. Hafnate and zirconate perovskites — BaHfO3, CaHfO3, Ba3Hf2O7, Ba2HfO4, Sr2HfO4, K2HfO3, Li2ZrO3, Li2HfO3 — are the high-k anchor. DFPT calculations (computational experiments CE13 and reference 0466) yield total dielectric constants of approximately 40.7 for BaHfO3, 31.85 for CaHfO3, 31.70 for Ba3Hf2O7, and 23.17 for HfO2. These figures exceed the dielectric constant of thermally grown SiO2 by roughly one to two orders of magnitude and compare favorably with HfO2 — the current semiconductor industry high-k standard — while offering the additional leverage of the perovskite structural framework, which allows A-site substitution to tune the permittivity-bandgap tradeoff. The hafnate perovskites are stable in their cubic or distorted-perovskite forms and are expected to remain amorphous-crystalline compatible with sub-500°C deposition, though this must be confirmed experimentally per member. The rare-earth sesquioxide sub-family — Dy2O3, Eu2O3, La2O3, Nd2O3, Sm2O3, Yb2O3 — was subjected to a four-engine machine-learning interatomic potential screen (MACE, CHGNet, MatterSim, and ORB), producing a consensus positive result on dynamic stability across all members (computational experiment CE18). This is the most rigorous computational stability check available short of full DFT phonon dispersion: four independently trained potentials, built on different architectures and training sets, must agree that no imaginary phonon modes are present before a material advances. The rare-earth sesquioxides present a dual opportunity — they carry moderate dielectric constants suited to interlayer dielectric applications in packaging, and their optically active 4f electronic manifolds make several members (Eu, Dy, Nd, Yb) candidates for photonic and quantum-memory substrates where the dielectric host must be both low-loss and optically functional. The borate and nitride sub-family (AlBO3, ZnB4O7, AlN) was computed at the HSE06 hybrid functional level (CE12 and reference 0463), which corrects the bandgap underestimation endemic to standard GGA-DFT and is the appropriate method for wide-gap insulators where accurate band alignment and dielectric response are linked. YBO3 received the most detailed stability treatment: three independent phonon calculations (three-engine screen under reference 0241h) yielded all-positive phonon frequencies, with the lowest acoustic branch bottom sitting between +0.38 and +0.60 THz — a narrow but unambiguous positive margin that confirms the compound is at least a local energy minimum on the potential energy surface. The dielectric constant of YBO3 falls in the 12.7–13.8 range, which positions it as a moderate-k photonic substrate candidate rather than a high-k capacitor material, but its wide bandgap and borate chemistry offer transparency windows suitable for ultraviolet-to-visible photonic integration. Ca2MgSi2O7 (akermanite) is included as a known-composition reference arm with published microwave dielectric data; it serves a negative-control and prior-art anchoring function rather than a novelty claim, and is treated accordingly within the claim structure. Two important limitations bound the computational picture. First, DFT sources for this asset come from two independent calculation sets, providing cross-validation on the DFPT permittivity numbers but not yet the breadth of experimental verification. Second, the four-engine consensus on dynamic stability does not by itself guarantee thermodynamic stability against competing phases (phase-diagram stability, or convex-hull position) or thin-film process compatibility — both of which require experimental thin-film coupons. The hafnate permittivity values are particularly sensitive to crystallographic phase: BaHfO3 in a cubic versus orthorhombic polymorph carries meaningfully different dielectric tensors, and deposition conditions that stabilize one phase over another are not yet characterized. These are the primary open validation gates before any member of this suite can be specified in a device flow.

The addressable market for this asset sits at the intersection of two segments: advanced packaging dielectrics and photonic/quantum substrate materials, with an estimated combined opportunity in the $1–3 billion range. This is an estimate, not a measured figure, and reflects the combined spend by OSAT (outsourced semiconductor assembly and test) providers on dielectric materials and photonics manufacturers on substrate consumables, not the full value of packaged devices. The dielectric-materials market is dominated by commodity SiO2 and Si3N4, with HfO2 and Al2O3 occupying a premium-priced niche in gate-dielectric and capacitor integration; the addressable fraction for novel high-k companions grows as chiplet packaging moves to finer pitch and higher capacitance density requirements that incumbent materials cannot meet at acceptable leakage levels. The customer base for this asset is bifurcated. OSATs and back-end-of-line integrators are the primary buyers for the hafnate and zirconate members: these firms source dielectric materials through chemical vapor deposition and atomic-layer deposition precursor supply chains, and a validated new high-k composition entering the supply chain requires both materials-science qualification and process engineering investment — which is precisely the kind of barrier-to-entry that IP coverage reinforces. Photonics and quantum-technology manufacturers are the primary buyers for the rare-earth sesquioxide and borate members: these customers are typically smaller-volume, higher-value-per-substrate buyers who are actively searching for rare-earth-compatible hosts that are patent-clear and have some computational pedigree. The licensing logic in both cases is a per-wafer or per-substrate royalty layered on top of precursor supply agreements, or a one-time technology transfer to a materials supplier building out an ALD precursor catalog. Supply-chain considerations add a secondary dimension to market sizing. Several rare-earth sesquioxides — Dy2O3 and Yb2O3 in particular — draw from heavy rare-earth supply chains where China controls a substantial majority of refined production. YBO3 is notable in this context because boron and yttrium together present a more geographically distributed supply profile than the heavy lanthanides, which is why it is flagged in this asset as a supply-safe alternative. This supply-security angle is increasingly a purchasing criterion for defense-adjacent and radiation-hard electronics customers, who face explicit domestic-content requirements under NDAA and related procurement rules and who represent a captive premium segment within the broader OSAT and photonics customer base.

broad DFPT/HSE06-validated companion menu spanning high-k hafnates to wide-gap photonic substrates

companion stack members; reserved for forthcoming Family G companion claims

The most natural acquirers or licensees for this asset are advanced-packaging materials suppliers and OSAT-adjacent IP holding companies seeking to build a defensible position in next-generation dielectric materials before the packaging industry's high-k integration wave fully matures. Specific buyer profiles include ALD precursor suppliers (companies that synthesize and qualify hafnium, yttrium, and rare-earth precursor chemistries for semiconductor deposition tools), who would benefit from patent coverage on the compositions their precursors are used to deposit; and OSAT integrators with internal R&D programs in dielectric stack optimization, who would license to block competitors from claiming the same hafnate or borate compositions in future filings. Photonics and quantum-technology platform companies — particularly those developing rare-earth-doped photonic integrated circuits or solid-state quantum-memory modules — represent a second buyer tier with a premium willingness to pay for FTO-clean, computationally pedigreed rare-earth sesquioxide and borate substrates. A defense or government-funded acquirer is also plausible given the radiation-hard electronics application context and the supply-safe framing of YBO3. Programs operating under domestic-content mandates for microelectronics and photonics are actively seeking alternatives to heavy-rare-earth-dependent materials, and an asset that provides IP coverage on a supply-safe borate alongside computational validation has direct relevance to those procurement contexts. The asset is best positioned as a companion licensing package to the lead wide-bandgap dielectric stack family rather than as a standalone transaction, since its full value is realized when a buyer controls both the core and companion claims and can enforce the full compositional envelope.

The primary technical risk is the gap between DFPT bulk permittivity and thin-film measured permittivity. Hafnate perovskites in particular are sensitive to crystallographic phase — the high-permittivity cubic phase of BaHfO3 may not be the thermodynamically stable phase under low-temperature ALD conditions, and the actual deposited film may be amorphous or adopt a lower-symmetry polymorph with a materially lower dielectric constant. This risk is not unique to this asset — it is endemic to the high-k dielectric field — but it means that the computed values of ~40.7 for BaHfO3 should be understood as upper-bound estimates pending thin-film validation, not guaranteed performance specifications. The YBO3 narrow phonon stability margin is a secondary technical risk: a positive-but-small margin means the computed stability is real but not robust to perturbation, and thin-film strain or compositional off-stoichiometry could push the structure toward instability. The rare-earth sesquioxides carry supply-chain risk for the heavy lanthanide members (Dy, Yb) that is structural and geopolitical rather than technical, but which affects commercial deployment timelines even if the materials perform as computed. The roadmap to de-risk this asset is relatively well-defined: thin-film coupon synthesis for the three to four highest-priority members (BaHfO3, CaHfO3, YBO3, and one rare-earth sesquioxide representative) by ALD, followed by ellipsometric and capacitance-voltage characterization. This experimental program would confirm or bound the DFPT permittivity predictions and provide the thin-film stability data needed to advance companion claims from reserved to asserted status. The akermanite negative-control designation is already appropriately scoped and requires no further experimental validation — its role is structural, not investigational. A parallel patent prosecution timeline that files the companion claims ahead of thin-film confirmation is viable given the strength of the multi-engine and DFPT computational record, which represents the kind of enabling disclosure that supports filing prior to experimental reduction to practice in jurisdictions that permit it.