Graded Li5AlO4/LiAlO2 bilayer anode-side interlayer for garnet solid-state batteries

The solid-state battery industry is converging on garnet-type electrolytes (primarily LLZO) as the long-cycle, safe alternative to liquid electrolytes in lithium-metal cells. The single most persistent failure mode is the anode-side interface: bare garnet surfaces are poorly wetted by molten lithium, accumulating interfacial resistance that quickly exceeds 100 Ω·cm² and limits practical critical current densities (CCD) to well below what automotive and grid applications demand. This invention addresses that choke point directly. A graded lithium aluminate bilayer — Li₅AlO₄ on the garnet-facing side, LiAlO₂ on the lithium-metal-facing side, with a monotonic composition gradient of at least 2× between the two regions — creates an asymmetric interlayer architecture that simultaneously satisfies the ionic and chemical demands of two completely different interfaces. The garnet-facing primer wets oxide surfaces and densifies grain contacts; the lithium-metal-facing cap remains electrochemically stable against metallic lithium. No single-phase interlayer can optimize both faces simultaneously, and that structural constraint is exactly what this bilayer architecture is designed to exploit. The timing is favorable for several interconnected reasons. Automotive OEMs have publicly committed to solid-state battery introduction windows in the 2027–2030 range, creating enormous pressure on electrolyte cell stack suppliers to solve interface stability now rather than at pilot scale. Simultaneously, ALD and sol-gel deposition tools capable of graded oxide coatings are becoming standard equipment in battery R&D lines, lowering the manufacturing barrier to a two-region aluminate layer. The patent landscape for garnet interlayers is competitive but still navigable: most filed art covers single-phase coatings or sintering-aid residues (lithium-oxide or alumina distributed during co-sintering), leaving the explicit bilayer-with-gradient construct in a defensible whitespace. The performance targets written into the protected claims — CCD ≥ 0.8 mA/cm² and interfacial resistance ≤ 30 Ω·cm² — are not aspirational guesses; they represent the threshold that practical lithium-metal cells need to pass fast-charge protocols without triggering lithium filament (dendrite) penetration through the electrolyte. Within the solid-state battery electrolytes and interfaces portfolio, this asset is characterized as a lead filing. That classification reflects both the structural originality of the bilayer construct and the performance-gate claims that tie the protection to measurable, commercially meaningful metrics. The portfolio as a whole spans electrolyte bulk materials, sintering processes, and interface engineering; this invention sits at the interface-engineering apex, where the electrochemical performance of the full cell is most acutely determined.

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 two phases in this bilayer — Li₅AlO₄ and LiAlO₂ — are not interchangeable members of a generic "lithium aluminate" family; they have meaningfully different crystal chemistry, lithium coordination environments, and electrochemical windows. Li₅AlO₄ adopts an orthorhombic structure (space group Pmmn, No. 59, with a closely competing Pbca polymorph) and is lithium-rich, with a 5:1 Li:Al ratio that provides a reservoir of available lithium ions and strong wettability against garnet oxide surfaces. Alpha-LiAlO₂ is the stoichiometrically leaner phase with a tetragonal layered structure; it sits at a more oxidized point on the phase diagram and is kinetically and thermodynamically more stable at the lithium-metal contact potential. The engineering insight behind stacking them — Li₅AlO₄ garnet-side, LiAlO₂ lithium-side, with a monotonic gradient connecting them — is that each phase is deployed where its chemistry is best suited, and the gradient suppresses the sharp composition discontinuity that would otherwise create a stress concentration or an ionic conductivity bottleneck at the bilayer midpoint. The wide bandgap of the aluminate system, approximately 5.0 eV, is a structural feature ensuring the interlayer is electronically insulating and therefore cannot short-circuit the cell — a critical property for any anode-side coating that will sit in direct contact with metallic lithium. Computational validation for this asset was carried out using two independent machine-learning interatomic potentials, specifically MACE and CHGNet, which were applied to both the Pmmn and Pbca polymorphs of Li₅AlO₄ to evaluate phonon (dynamical) stability. The result is unambiguous: both potentials agree that the Li₅AlO₄ structure is dynamically stable, with no imaginary phonon modes detected — the key indicator that a structure will not spontaneously distort or decompose under ambient conditions. The MACE potential yields a representative phonon frequency of 1.41 THz at the zone boundary; CHGNet independently returns 1.69 THz. These values are not identical, which is expected from two different potential energy surfaces trained on different datasets, but the directional agreement — both positive, both in the physically reasonable range — is the meaningful result. Two independent DFT sources provide additional electronic structure grounding. This multi-potential consensus protocol, requiring agreement across independent models before a material advances, is substantially more stringent than single-model validation and eliminates false positives that arise from artifacts in any individual potential. Beyond phase stability, the simulation suite for this asset includes nudged elastic band (NEB) calculations of lithium migration barriers. NEB is the standard computational method for mapping the minimum-energy pathway for an ion hopping between adjacent vacancy sites in a crystal lattice, and the barrier height directly predicts ionic conductivity. Low barriers correspond to fast lithium transport; high barriers correspond to resistive interfaces. The NEB simulation here provides a direct mechanistic rationale for why the selected phases, in the selected orientation (garnet-side vs. lithium-side), support the claimed interfacial resistance specification of ≤ 30 Ω·cm². These simulations also support the gradient requirement: a sharp compositional step would present a mismatched migration landscape at the phase boundary, whereas a monotonic gradient allows the migration barrier profile to evolve continuously, suppressing local ion-blocking effects. Two remaining experimental validation gates have been identified and are openly acknowledged. First, time-of-flight secondary ion mass spectrometry (ToF-SIMS) on fabricated coupons is needed to confirm that the intended composition gradient — with the ≥ 2× asymmetry between the garnet-facing and lithium-facing regions — is actually present in deposited films and survives sintering. This is a process validation step, not a materials discovery step; the question is whether the deposition and thermal processing sequence preserves the bilayer architecture. Second, explicit-layer dielectric-function and phonon calculations via DFPT (density functional perturbation theory) on the full bilayer stack, rather than on each phase in isolation, are planned. DFPT on the combined interface model would confirm that the electronic and ionic transport properties predicted for the individual phases are not disrupted at the bilayer junction. Both gates are well-defined, achievable with standard characterization and simulation tools, and do not present conceptual obstacles — they are execution items.

The addressable market for anode-side interface solutions in garnet solid-state batteries is embedded within a broader solid-state battery ecosystem that multiple independent industry analyses project to exceed $5 billion in annual revenues by the early 2030s, driven primarily by electric vehicles, grid storage, and consumer electronics. Within that broader market, the anode-electrolyte interface is not a peripheral component — it is the primary failure locus and therefore the primary value driver. Cell makers who cannot solve the lithium-metal/garnet interface cannot build commercially viable cells at all, which means interface technology commands licensing leverage disproportionate to its apparent simplicity as a "thin coating." The direct customers for this technology are of two types. Garnet cell manufacturers — ranging from established automotive battery suppliers pursuing solid-state roadmaps to dedicated solid-state startups — need a manufacturable, scalable interlayer that can be incorporated into their stack without adding multiple process steps. ALD vendors and sol-gel coating specialists who supply deposition services to cell makers represent a second pathway: a graded aluminate bilayer is architecturally straightforward to deposit by ALD (where composition can be controlled cycle-by-cycle) or by sequential sol-gel steps, meaning coating equipment companies may seek to license the composition and architecture as part of a process package sold to cell manufacturers. Royalty structures in battery materials IP typically run in the range of 1–5% of cell value or are structured as per-layer-area fees in manufacturing agreements, though the exact rate depends on negotiation leverage and exclusivity terms. The performance specifications in the claims — CCD ≥ 0.8 mA/cm² and R_int ≤ 30 Ω·cm² — are strategically chosen as commercial thresholds, not purely academic targets. At 0.8 mA/cm², a cell can support meaningful fast-charging rates for automotive applications. At ≤ 30 Ω·cm², interfacial resistance is a small fraction of total cell impedance and does not dominate rate capability. Any competing interlayer technology that cannot simultaneously meet both metrics will produce cells that fail automotive qualification protocols, creating a hard selection pressure in favor of solutions that can. This performance-gate structure in the claims is also commercially self-sharpening: as the market tightens its qualification criteria, the claims become more valuable, not less.

dual-function graded interface; design-around moat over single-phase interlayers

anode-side asymmetry + bilayer/gradient distinguishes single-phase and sintering-aid residue

The most natural strategic acquirers and licensees for this asset fall into two categories. The first is vertically integrated solid-state battery manufacturers who are building garnet-based cell stacks and need proprietary interface solutions to differentiate their products from competitors using commodity single-phase coatings. This group includes both established Tier-1 automotive battery suppliers with solid-state programs and dedicated solid-state startups that compete on cell performance rather than scale. For these buyers, licensing or acquiring the bilayer architecture provides a defined, performance-specified interface solution that can be incorporated into manufacturing process specs and cited in customer qualification packages. The second category is ALD and advanced thin-film deposition equipment and process companies that supply coating services to battery manufacturers. These vendors benefit from IP that defines a high-value deposition target — the graded aluminate bilayer is architecturally natural for ALD, where alternating precursor cycles can be programmed to produce a compositional gradient — and can bundle the IP with their process expertise as a value-added service to cell makers. Beyond direct licensees, the asset also has defensive value for any company building a broad solid-state battery IP portfolio. The bilayer architecture is sufficiently distinct from prior art to serve as a blocking position against competitors who might otherwise converge on similar dual-phase aluminate interface designs. For a portfolio acquirer, this asset complements bulk electrolyte and sintering-process IP by covering the anode-side interface, the component most directly linked to cell-level failure modes and therefore most likely to be the subject of litigation or licensing disputes as the market matures.

The primary technical risk is process transfer: computational validation confirms that both Li₅AlO₄ and LiAlO₂ are individually stable phases with favorable ionic and electronic properties, and NEB calculations support low migration barriers, but the performance targets (CCD ≥ 0.8 mA/cm², R_int ≤ 30 Ω·cm²) have not yet been validated on fabricated cells. The gap between computational prediction and measured device performance in interface engineering is non-trivial, because grain boundary resistance, film morphology, and inter-diffusion during sintering all affect real-world interfacial resistance in ways that phase-level simulations do not fully capture. The ToF-SIMS gradient profiling gate is the most critical near-term de-risking step: if the deposition and sintering sequence fails to preserve the ≥ 2× gradient, the architectural advantage of the bilayer collapses toward a mixed-phase single-layer, and the claim differentiation weakens. The mitigation path is clear — coupon-level fabrication and characterization using ALD or sol-gel deposition followed by ToF-SIMS and electrochemical impedance spectroscopy — and this work is on the standard roadmap. The secondary risk is competitive convergence. The bilayer-with-gradient concept, once published or disclosed, may prompt competitors to file in adjacent composition spaces (different aluminate stoichiometries, indium-doped variants, mixed aluminate-zirconate bilayers) that could crowd the whitespace around this protected architecture. The explicit exclusions built into the claims (gamma-LiAlO₂ excluded, sintering-aid residue excluded) and the specific anode-side asymmetry requirement provide substantial differentiation, but the pace of filing in solid-state battery interfaces is high and ongoing monitoring is warranted. The freedom-to-operate status is currently clean, but as a dynamic field, that assessment requires periodic refresh against new publications and prosecution outcomes.