Crystalline Li2PNO2 phosphonitride for interface-engineered solid-state battery use

Li2PNO2 — also written Li2PO2N — is a crystalline lithium phosphonitride with an orthorhombic crystal structure (space group 36) that sits in the same chemical family as amorphous LiPON, the workhorse thin-film solid electrolyte that has underpinned thin-film battery technology for decades. Where amorphous LiPON derives its practical value from deposition process flexibility and acceptable ionic conductivity, the crystalline polymorph of Li2PNO2 presents a structurally ordered phase whose properties and stability have now been independently mapped by computational methods. Lattice Graph's solid-state battery electrolytes & interfaces portfolio holds this material as a written-description asset — meaning it is included to establish computational provenance and stake a defensible position on a specific non-anticipated architecture, not to assert ownership of the bulk composition or its straightforward use as a bulk solid electrolyte, both of which the peer-reviewed literature has already established. The strategic logic is straightforward: the crystalline phase of Li2PNO2 has been synthesized and characterized in academic publications, with measured room-temperature ionic conductivity around 1×10⁻⁷ S/cm and an activation energy near 0.57 eV. Those facts put the bulk composition and its plain-vanilla use as a bulk solid electrolyte firmly in the prior-art public domain. The value Lattice Graph preserves is narrower and more specific — the computational confirmation of dynamic stability across multiple independent models now provides a documented technical foundation for claiming only those non-anticipated embodiments, such as specific dopant variants, interface-engineered device architectures, or metastable processing routes, that fall outside what the existing literature has already disclosed. The asset is honestly categorized as defensive and written-description in nature; it is not a lead asset for licensing in its own right, but it contributes meaningful coverage to a portfolio that addresses the full design space around phosphonitride electrolytes. Timing matters here in a structural sense. The solid-state battery industry is undergoing a forced transition driven by energy-density targets and safety mandates that liquid electrolytes cannot meet. Thin-film electrolyte vendors, in particular, are under pressure to diversify their materials palette beyond amorphous LiPON as they push toward thinner, more conformal, and more interface-compatible layers. Crystalline phases of phosphonitrides occupy a design space that is technically adjacent to established practice but computationally underexplored. By documenting this phase computationally and anchoring it within the portfolio, Lattice Graph creates a credible prior-art-clearing and disclosure record that supports any future work on doped or architecturally differentiated variants.

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.

Li2PNO2 crystallizes in an orthorhombic structure (space group 36), a symmetry class that accommodates the tetrahedral phosphorus coordination typical of phosphonitride and phosphate families. The measured bandgap reported in the literature, and consistent with the computational data supporting this asset, is approximately 5.6 eV — a wide-gap insulator profile that is chemically appropriate for a solid electrolyte, where electronic blocking is as important as ionic transport. The room-temperature ionic conductivity of the crystalline bulk phase is approximately 1×10⁻⁷ S/cm with an activation energy near 0.57 eV, both of which are reported in prior published experimental work. These values are below what modern sulfide or halide solid electrolytes deliver, but they are consistent with the broader class of oxide-nitride materials and are not the basis for a claim of superiority; they are simply the known property landscape of this phase. The computational validation effort went beyond a single-potential energy minimization, which would be insufficient evidence of genuine stability. Three independent machine-learning interatomic potentials were applied sequentially: MACE, CHGNet, and ORB. All three returned consistent results confirming that the orthorhombic Li2PNO2 structure is dynamically stable — that is, the phonon density of states contains no imaginary modes at any point in the Brillouin zone, which would signal a lattice instability or tendency to reconstruct into a different phase spontaneously. MACE yielded a minimum phonon frequency of 0.71 THz and CHGNet 0.39 THz, both positive, with ORB independently confirming stability. This level of cross-potential agreement is meaningful: different ML potentials are trained on different reference datasets and carry different inductive biases, so agreement across all three substantially reduces the probability of a false-positive stability verdict arising from a single model's fitting artifacts. One DFT-level calculation was also used to anchor the structural reference, giving the phonon analysis a first-principles grounding. Two distinct simulation passes were completed: an initial stability screen (labeled internally as the first pass) and a subsequent exact-composition re-confirmation run on the precise Li2PNO2 stoichiometry. The re-confirmation step matters because exploratory materials discovery sometimes evaluates slightly off-stoichiometric or proxy compositions; the second pass removes ambiguity and confirms that the stability verdict applies to the exact formula being documented. Together, these runs establish a documented computational pedigree for the crystalline phase that is independent of the experimental literature, and that pedigree is what the written-description asset preserves. The open validation gate, going forward, is the computational and experimental characterization of any specific non-anticipated embodiment — a doped variant, a thin-film interface-optimized form, or a metastable processing product — that would be the subject of any narrowed device or use claim. The 5.6 eV bandgap also warrants comment in the context of interface engineering. Wide-gap electrolyte materials with well-characterized electronic structure are candidates for dielectric interlayer applications at electrode-electrolyte interfaces, where controlling both ionic transport and electronic leakage is critical. DFPT-based dielectric tensor calculations are a natural next computational step for any interface-engineered embodiment, as the dielectric response governs space-charge layer formation and capacitive behavior at the interface. Those calculations are not yet part of the documented proof set for this asset but represent a clear and tractable extension of the existing stability work.

The addressable market for this asset is bounded by the thin-film solid electrolyte segment of the broader solid-state battery industry. Thin-film batteries — primarily manufactured by physical vapor deposition or sputtering of electrolyte layers — represent a specialized market distinct from the bulk ceramic or polymer solid electrolyte segments being pursued by automotive players. The addressable market for thin-film electrolyte materials and processes, on a royalty-or-licensing basis tied to the specific non-anticipated architectural embodiments claimed here, is estimated at roughly $0.5 billion to $1 billion. This is an honest estimate reflecting the niche nature of thin-film battery applications (medical implants, IoT sensors, wearables, defense micro-batteries) rather than the multibillion-dollar automotive solid-state battery market, which runs on bulk electrolyte formats and is not relevant to the present asset. Who buys or licenses in this segment? Thin-film electrolyte vendors — companies that deposit LiPON or related materials as part of their battery stack fabrication — are the natural first-look customers. For them, a crystalline phosphonitride phase that is computationally validated and tied to a documented IP position on specific interface architectures has value as a materials option to qualify alongside their existing amorphous LiPON processes. Equipment suppliers and integrated device manufacturers working on conformal thin-film batteries are secondary targets. Academic or government lab partners seeking a documented computational dataset to anchor further experimental work on crystalline phosphonitrides represent a third channel, particularly given the negative-experimental and failed-result atlas that Lattice Graph maintains alongside its positive results. Royalty or licensing logic for this asset would not follow the standard per-unit electrolyte materials royalty that a flagship compound might command. Instead, it is more appropriately positioned as a written-description license bundled with broader portfolio access — a licensor gains the documented computational provenance, the freedom-to-operate analysis relative to the 300,000+ materials patents screened, and the right to develop the specific non-anticipated embodiments under the portfolio's IP umbrella. Standalone monetization of this single asset as a licensing product is not the recommended approach; its value is as a supporting element within a broader solid-state battery electrolytes transaction.

preserved for written-description + a possible interface-engineered niche

specific non-anticipated dopant/interface/metastable architecture only; bulk composition + bulk-SE use not available

The most natural acquirers or licensees for this asset are thin-film battery manufacturers and the electrolyte deposition equipment suppliers who serve them — companies with existing LiPON sputtering capability who are evaluating crystalline phosphonitride phases as alternatives or complements to their current amorphous processes. A second category is large integrated energy-storage companies with in-house thin-film battery programs (medical devices, IoT, defense micro-battery applications) who would value the portfolio's computed phosphonitride coverage as a freedom-to-operate clearing tool and a foundation for internal R&D on interface-engineered architectures. For these buyers, the asset's primary value is not as a standalone licensing target but as part of a broader acquisition of the solid-state battery electrolytes & interfaces portfolio, where it contributes documented phosphonitride coverage and a clean disclosure record for the specific architectural embodiments that remain non-anticipated. Academic and government research institutions seeking a computationally validated dataset on crystalline phosphonitrides — particularly given the cross-validated ML potential stability data and the documented failed-experiment record maintained in the broader Lattice Graph corpus — represent a secondary interest. For any buyer doing serious interface engineering work on thin-film batteries, the combination of multi-potential phonon stability data, explicit prior-art acknowledgment, and a clear written-description foundation for non-anticipated architectures is a more defensible starting point than building from scratch on a composition whose IP landscape is murky. The asset is best positioned as a portfolio component rather than a standalone transaction.

The primary risk for this asset is stated plainly in its own documentation: the bulk crystalline composition and its direct solid-electrolyte use are anticipated by prior art, limiting any claim to specific non-anticipated architectural embodiments that have not yet been fully defined or computationally characterized. Until the exact dopant variant or interface architecture targeted by the claim is specified and validated, the asset functions as written-description coverage rather than a granted or prosecutable claim with defined scope. A buyer or licensee who expects broad blocking rights over crystalline Li2PNO2 will be disappointed; the value is in the narrower and still-to-be-specified architectural carve-out. A second risk is that crystalline Li2PNO2 is inherently less conductive than both amorphous LiPON and the sulfide/halide electrolytes now competing for thin-film applications, which limits the addressable use cases to those where structural order at the interface matters more than bulk transport rate. The roadmap to de-risk is clear. First, specify the target non-anticipated embodiment — a concrete dopant identity, interface material, and processing route — and run the next-stage computational proof gates (DFPT dielectric tensor, NEB migration barriers, interface molecular dynamics) on that specific configuration. Second, initiate experimental synthesis of the targeted embodiment to generate enabling data for prosecution. If those steps return positive results, the asset advances from written-description to a prosecutable device-use or method claim with a defined and defensible scope. The computational infrastructure to execute those next gates already exists within the Lattice Graph platform, making the path forward an investment of computational and experimental resource rather than a fundamental technical unknown.