Lattice Graph × Vulcan Elements

Lattice Graph operates a computational materials-discovery platform built around a knowledge graph spanning millions of compositions, connecting crystal structure to thermodynamic property to synthesis route to patent claim in a single queryable fabric. When Vulcan's team evaluates a candidate rare-earth separation chemistry, they are not reading a vendor datasheet — they are navigating a governed graph where each composition traces directly to its proof type, its multi-engine validation record, and the patent landscape that surrounds it. That provenance is what separates a credible engineering claim from an optimistic prediction. The validation layer runs candidate materials through multiple independent physics engines in consensus — machine-learning interatomic potentials including MACE and CHGNet, cross-checked against density functional theory for phonon stability and thermodynamic properties. For separation materials like ion-imprinted resins or selective sorbents, this means predicted binding geometries, coordination environments, and selectivity drivers are evaluated against physically grounded stability criteria before any bench time is committed. Cross-engine disagreement is flagged explicitly, so Vulcan's technical team can distinguish a confident computational prediction from a contested one that needs experimental confirmation before scale-up. Freedom-to-operate and patent-whitespace screening runs continuously across more than 300,000 materials patents at composition and claim level. For Vulcan's situation — government-loan-backed plant construction with mandatory lender and offtake diligence — this is not a legal nicety. An auditable IP posture on every separation chemistry in the feedstock plan is part of the risk file that survives scrutiny. The platform also carries a large atlas of labeled negative results from failed experiments, the kind of institutional memory that prevents a domestic magnet manufacturer from re-investing engineering time in separation chemistries that have already been exhausted by others without public record of the failure. Collectively, the platform translates a hard rare-earth separation and supply-chain problem into a set of computationally validated, IP-cleared candidates with traceable evidence — exactly the artifact a government lender and an OEM offtaker need to see before a 10,000-ton plant is committed.

Vulcan Elements is not primarily a magnet-technology company — it is a feedstock-and-supply-chain bet wearing a manufacturing hat. A 10,000-ton-per-year sintered NdFeB plant backed by a government loan has to answer two recurring, auditable questions: where do the light rare earths — neodymium and praseodymium — come from, and how is heavy-rare-earth exposure managed when dysprosium and terbium are simultaneously the most critical inputs for high-temperature coercivity and the most supply-concentrated elements in the chain? Both answers have to survive lender diligence and OEM offtake scrutiny on a continuing basis, not just pass a one-time internal review. The strategic pressure on Dy and Tb is structural, not cyclical. Supply is dominated by a small number of non-domestic sources, and the elements that define magnet performance at elevated temperature are precisely the ones over which a US manufacturer has least procurement leverage. Domestic and recycled feed is the policy-favored path, but recycling only matters if its separation economics close — pulling Dy and Tb cleanly out of mixed NdFeB leachate, and splitting adjacent rare-earth pairs, without reproducing a capital-heavy solvent-extraction cascade that defeats the economics of domestic production. That is a chemistry problem before it is a manufacturing problem, and it is a problem that Lattice Graph has mapped at the computational level. The fit with Lattice Graph is direct and concrete. The critical-mineral recovery and recycling separations portfolio contains computationally screened, freedom-to-operate-cleared chemistries that address Dy/Tb separation from magnet leachate as a named target. The supply-chain intelligence layer — covering more than 300,000 mineral deposits with concentration and criticality data, waste-to-product conversion routes, and element-level supply risk — gives Vulcan the quantitative backbone for a domestic and recycled feed thesis that lenders can actually audit. Together, these turn an aspirational supply narrative into a defensible, traceable, IP-covered position.

Vulcan Elements is not primarily a magnet-technology company — it is a feedstock-and-supply-chain bet wearing a manufacturing hat. A 10,000-ton-per-year sintered NdFeB plant backed by a government loan has to answer two recurring, auditable questions: where do the light rare earths — neodymium and praseodymium — come from, and how is heavy-rare-earth exposure managed when dysprosium and terbium are simultaneously the most critical inputs for high-temperature coercivity and the most supply-concentrated elements in the chain? Both answers have to survive lender diligence and OEM offtake scrutiny on a continuing basis, not just pass a one-time internal review. The strategic pressure on Dy and Tb is structural, not cyclical. Supply is dominated by a small number of non-domestic sources, and the elements that define magnet performance at elevated temperature are precisely the ones over which a US manufacturer has least procurement leverage. Domestic and recycled feed is the policy-favored path, but recycling only matters if its separation economics close — pulling Dy and Tb cleanly out of mixed NdFeB leachate, and splitting adjacent rare-earth pairs, without reproducing a capital-heavy solvent-extraction cascade that defeats the economics of domestic production. That is a chemistry problem before it is a manufacturing problem, and it is a problem that Lattice Graph has mapped at the computational level. The fit with Lattice Graph is direct and concrete. The critical-mineral recovery and recycling separations portfolio contains computationally screened, freedom-to-operate-cleared chemistries that address Dy/Tb separation from magnet leachate as a named target. The supply-chain intelligence layer — covering more than 300,000 mineral deposits with concentration and criticality data, waste-to-product conversion routes, and element-level supply risk — gives Vulcan the quantitative backbone for a domestic and recycled feed thesis that lenders can actually audit. Together, these turn an aspirational supply narrative into a defensible, traceable, IP-covered position.

The center of gravity for a NdFeB magnet manufacturer sits squarely in the critical-mineral recovery and recycling separations portfolio. Within that portfolio, the highest-leverage chemistry for Vulcan's situation is Dy/Tb separation from magnet leachate — an explicit design target for the ion-imprinted phosphonate-bis-picolinamide resin that achieves predicted single-pass Dy/Tb separation factors of four to eight and Dy/Nd factors of thirty to one hundred, reducing solvent inventory compared to conventional solvent-extraction cascades. For a manufacturer whose high-temperature coercivity specification depends on the two least-controllable elements in its supply chain, a resin route with a clear freedom-to-operate path is directly relevant to both procurement strategy and the IP file a government lender expects. Broadening the feedstock window is the other major theme from the same portfolio. The universal chelating-resin platform recovers a dozen critical-mineral targets from named industrial streams — zinc, copper, and Bayer refinery — under a single, freedom-to-operate-clean IP genus. For a 10,000-ton plant that cannot be hostage to one feed type or one geography, a single recovery platform that widens the input window and supports co-product capture alongside rare earths turns a recovery cost center into a feedstock-flexibility and netback story for lenders and offtakers. The integrated flowsheet platform, which bundles the magnet-recycling separation train with a broader critical-mineral recovery cascade as an ordered system, provides the system-level IP architecture that makes isolated point chemistries defensible as a combined process rather than a collection of individual licences. The solid-state battery electrolytes and interfaces portfolio is a secondary adjacency at this stage, with relevance growing as Vulcan's recovered feedstock strategy diversifies beyond NdFeB scrap. As battery recycling volumes grow and co-location of magnet and battery material recovery becomes economically attractive, the adjacent recovery chemistries in the portfolio provide a path to spread plant economics across a wider critical-mineral base — an option worth preserving in any long-term licensing structure.

Vulcan's team would work primarily across two platform surfaces: the supply and IP workflow and the knowledge-graph explorer. In the supply workflow, analysts run element-level supply-risk and deposit searches for neodymium, praseodymium, dysprosium, and terbium, layer in conversion-route and captivity-pair intelligence for recycled and byproduct feed, and export a feed-risk picture that maps directly onto the government-loan diligence narrative. The composition-intelligence and confidence views let Vulcan inspect any candidate separation chemistry with its provenance, cross-engine validation record, and proof gates visible, so technical, commercial, and finance stakeholders share the same evidence trail rather than translating between an R-and-D summary and a lender's risk matrix. On the IP side, the freedom-to-operate and patent-whitespace screening interface and the knowledge-graph explorer let Vulcan validate the clean status on the portfolio's separation assets, trace each patent claim back through the composition-structure-property-recipe graph, and identify whitespace for proprietary filings around Vulcan's specific magnet-leachate recovery process. Batch screening and knowledge-graph neighborhood queries support systematic comparison of the ion-imprinted resin, the universal chelating-resin platform, and the integrated flowsheet architecture against Vulcan's actual feed compositions before any pilot commitment is made — compressing the pre-licensing technical diligence cycle from months to weeks.

The natural structure is a phased assets engagement anchored on the critical-mineral recovery and recycling separations portfolio. Because several lead assets carry explicit bench proof gates — most importantly the Dy/Tb-selective ion-imprinted resin, which needs validation on real NdFeB leachate before a full license commitment — a sensible first step is a scoped co-development and option phase that takes the lead chemistry to confirmation on Vulcan's actual feed before finalizing license terms. This de-risks both sides: Vulcan gets evidence-based confidence before a capital commitment, and Lattice Graph completes a validation milestone that strengthens the asset's commercial record. Alongside the lead resin, the universal chelating-resin platform and the integrated flowsheet platform are available as concurrent or sequential scopes as the feedstock strategy widens and the plant design matures. The data and intelligence layer — supply and conversion-routes intelligence, mineral-deposit and critical-minerals data, and freedom-to-operate and patent-whitespace screening — would run as an annual subscription alongside the IP engagement, scoped to the feed-traceability and IP-clarity work that the government loan and offtake conversations require on a continuing basis. The typical engagement structure pairs a co-development option fee on the lead chemistry with milestone and royalty terms on license, plus the annual data subscription for the supply and freedom-to-operate layer. All commercial figures are subject to scoping and diligence; nothing here should be read as a quote or commitment, but the structure is designed to match Vulcan's capital deployment timeline with the validation and plant-build sequence.