Lanthanum orthoborate (LaBO3) mid-permittivity halogen-free redistribution-layer dielectric

Lanthanum orthoborate (LaBO3) occupies a deliberate and strategically important position in the PFAS-free dielectric and process fluids portfolio as a halogen-free inorganic mid-permittivity dielectric candidate for advanced-packaging redistribution-layer (RDL) stacks. The driving force behind this work is the accelerating regulatory and supply-chain pressure on halogenated dielectrics — materials that have historically dominated interlayer dielectric and RDL applications but face mounting restrictions under REACH, RoHS successor frameworks, and voluntary PFAS phase-out commitments by major semiconductor OEMs. Advanced packaging now demands dielectric integration at the RDL level that can coexist with chiplet architectures, fan-out wafer-level packaging, and 2.5D interposer stacks — all of which require materials with dielectric constants meaningfully above those of polymer alternatives (typically 2.8–3.8) to support finer feature routing and reduced via dimensions without unacceptable signal-propagation penalties. LaBO3 addresses this need with a dielectric permittivity of approximately 16.5 and a wide bandgap of 4.50 eV, placing it in the practically useful intermediate-permittivity zone that bridges the gap between conventional polymer dielectrics and the high-k oxides (hafnia, alumina, titania) that carry their own integration challenges. The 4.50 eV bandgap is critical: it implies low intrinsic leakage at operating voltages relevant to RDL structures and provides adequate isolation margins. Crucially, the composition is completely halogen-free, satisfying the substitution requirement that motivates this portfolio as a category. The role of LaBO3 within the portfolio is candidly that of a dependent species — it is a named member of a broader rare-earth orthoborate genus (RE-BO3, where RE = La, Y, Gd, Ce, Pr, Nd) and is claimed as a specific composition confirming genus coverage, not as a standalone blockbuster filing. Its value is twofold: it provides a computationally validated anchor species for the genus that demonstrates the orthoborate structure class can deliver mid-k dielectric performance in a thermodynamically stable, halogen-free form, and it establishes freedom-to-operate whitespace in a corner of the patent landscape that incumbent polymer-dielectric IP does not reach.

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.

LaBO3 belongs to the rare-earth orthoborate structural family, a class of boron-oxygen framework compounds in which the rare-earth cation (here, lanthanum) coordinates with isolated BO3 trigonal-planar units rather than the condensed polyborate networks seen in other borate glasses and ceramics. This structural motif confers a combination of moderate polarizability — arising from the La 5d and O 2p contributions to the dielectric response — and a relatively large electronic bandgap, because the BO3 unit's strong covalent B–O bonds push the conduction-band edge to high energies. The result is a dielectric tensor with components that can reach the high teens while maintaining an optical bandgap around 4.5 eV, a combination that is genuinely difficult to achieve in halogen-free inorganic candidates without resorting to high-k oxides that carry leakage and integration problems. The computational validation of LaBO3 was conducted with three independent machine-learning interatomic potentials — MACE, CHGNet, and MatterSim — all three of which independently agree that the structure is dynamically (phonon) stable. Concretely, the maximum imaginary-frequency mode found by each potential is below 0.55 THz in absolute value, with MACE returning 0.49 THz, CHGNet 0.42 THz, and MatterSim 0.54 THz — all consistent with a structure that is at or within numerical noise of full phonon stability, carrying zero soft modes that would indicate a lattice instability. This three-engine consensus is the portfolio's minimum bar for advancing a material to filing consideration; no composition advances on the basis of a single potential alone, preventing false positives from known artifacts of individual models. The phonon calculations were performed using Phonopy interfacing with each potential on the Materials Project entry mp-8216, which places LaBO3 on or near the thermodynamic convex hull — meaning no known competing phase lies lower in energy, consistent with a synthesizable compound. Dielectric characterization is sourced from a Materials Project DFPT (density-functional perturbation theory) calculation, which yields a total dielectric permittivity of approximately 16.5 with a z-axis tensor component of 19.1, indicating mild dielectric anisotropy. DFPT at the DFT level provides the electronic plus ionic contributions to the static dielectric constant and is the standard computational benchmark for screening dielectric candidates before experimental synthesis. One honest limitation here is that this is a single-source dielectric value — the DFPT result comes from one calculation on one structure, and independent DFT confirmation or experimental ellipsometry has not yet been performed. This is the primary open validation gate for this asset. The phonon stability result — consensus across three independent potentials with no imaginary modes — is the strongest computational evidence in hand. The bandgap of 4.50 eV is consistent with the optically transparent, electronically insulating character expected of the orthoborate class and is broadly corroborated by the wide-gap behavior reported for related La-containing oxides and borates in the literature. The remaining open simulation work includes a loss-tangent measurement or calculation (tan δ is not yet computed, which is a key process-qualification figure for any RDL dielectric) and experimental thin-film deposition to confirm the computed bulk properties translate to integrated layers — a standard gap between bulk DFT and device-relevant thin-film characterization.

The immediate commercial opportunity for LaBO3 and the broader rare-earth orthoborate genus sits within the advanced-packaging RDL dielectric market. Redistribution layers are the patterned dielectric-and-metal stacks that reroute die I/O pads to package-level interconnects in fan-out and chiplet architectures. As heterogeneous integration scales and pitch tightens, the dielectric constant of the interlayer material increasingly determines achievable line capacitance, signal velocity, and cross-talk margins. The current dominant approach uses photosensitive polymer dielectrics (polyimide, polybenzoxazole, and related materials) with dielectric constants in the 2.8–3.8 range. These are mature, manufacturable, and OSAT-qualified, but they are not halogen-free by default in all formulations, and their dielectric constants leave significant performance headroom uncaptured for applications where capacitive loading is a binding constraint. The serviceable addressable market for alternative RDL dielectric materials — including inorganic, hybrid, and low-halogen polymer options — is estimated at roughly $200–500 million annually, reflecting the dielectric material spend within the global advanced-packaging equipment and materials market, which itself is growing at high-single to low-double-digit rates driven by AI chip packaging demand. These are estimates, not audited figures, and the specific inorganic mid-k segment is nascent. The licensing and royalty logic for this asset is structural rather than transactional: the primary mechanism is not selling LaBO3 powder but licensing the composition claim and process integration know-how to OSATs (outsourced semiconductor assembly and test houses), IDMs integrating advanced packaging in-house, or dielectric material suppliers formulating next-generation RDL stacks. Royalty rates on specialty dielectric compositions in the semiconductor packaging space typically range from 1–4% of material revenue or are structured as cross-license value in IP portfolios assembled ahead of design-in decisions. The dependent nature of this asset within the broader RE-BO3 genus claim means its value is additive to, not independent of, the parent genus filing — a buyer licensing the genus claims the LaBO3 species validation as proof-of-concept evidence, strengthening enforceability of the broader claim set.

intermediate-permittivity (~16.5) halogen-free RDL orthoborate option

halogen-free mid-k RDL orthoborate species; dependent posture

The natural acquirers and licensees for this asset divide into two groups. The first is advanced-packaging OSATs and IDMs with active RDL development programs — companies such as ASE Group, Amkor Technology, JCET, and Intel Foundry Services (among others), all of which are actively scaling fan-out and chiplet-based packaging where dielectric material choice is a qualified engineering decision. These buyers would value the composition-and-device-use claim as a design freedom asset: licensing it ensures they can develop inorganic mid-k RDL processes based on the orthoborate class without encountering blocking IP, and the computational validation record reduces their own materials discovery cost. The second group is specialty materials suppliers and dielectric formulation companies — including those supplying advanced polymer dielectric precursors today who are building next-generation halogen-free portfolios — who would acquire the orthoborate genus IP as a hedge against polymer regulatory risk and a competitive differentiator in the PFAS-free materials transition. Strategic fit is strongest for buyers already invested in halogen-free advanced-packaging materials roadmaps, particularly those responding to customer ESG mandates or anticipating regulatory tightening on halogenated dielectrics. The asset's value is most naturally realized as part of a broader portfolio license covering the full rare-earth orthoborate genus and related PFAS-free dielectric families, rather than as an isolated single-species transaction.

The primary technical risk is that the dielectric permittivity value rests on a single computational source, and thin-film properties — which govern real device behavior — routinely diverge from bulk DFT values due to deposition-induced disorder, interfacial effects, and non-stoichiometry. Until independent DFPT confirmation and, ultimately, thin-film ellipsometry are completed, the permittivity value should be treated as a well-founded estimate rather than a confirmed specification. The loss-tangent gap is equally important: if tan δ proves high at GHz frequencies characteristic of advanced-packaging signal environments, the material's RDL utility is substantially reduced regardless of its static permittivity. These gaps are bridgeable through targeted simulation and modest experimental effort, but they represent real milestones that must be cleared before commercial engagement with OSATs. The secondary risk is the dependent claim posture: as a species claim within a genus filing, LaBO3's enforceability is coupled to the health and scope of the parent genus claim. If the genus claim is narrowed during prosecution — for instance, if prior art forces the claim to a specific structural sub-class that excludes certain RE-BO3 members — the scope and value of the LaBO3 species claim could be affected. The roadmap to de-risk this is completing per-species validation for the remaining RE members (Y, Gd, Ce, Pr, Nd analogs) to build a richer enablement record, and pursuing experimental synthesis validation for at least one or two species to ground the claim set in reduced-to-practice evidence. The competitive risk from incumbent polymer dielectrics is real but slow-moving: entrenched OSAT qualifications create inertia, and the window for inorganic mid-k dielectric adoption is tied to a generational shift in packaging architecture rather than a near-term substitution event.