Dual-dopant copper alloy catalyst for selective CO2 electroreduction

CO2 electroreduction (CO2RR) has emerged as one of the cleaner pathways from captured carbon to industrial feedstocks — formate, CO, and eventually C2 products like ethylene — without the thermal energy penalty of thermocatalytic routes. Copper is the only elemental metal known to produce C2+ products at meaningful Faradaic efficiency, but its native selectivity is notoriously difficult to control. The prior art has converged on single-atom-alloy (SAA) strategies: doping a copper host with one foreign atom — bismuth, tin, indium — to shift the adsorption energetics along the CO2 activation pathway. What the prior art has not claimed, and what this invention occupies, is the co-doping regime where two post-transition metal atoms are introduced simultaneously into the copper matrix. That dual-dopant configuration is the subject of this filing. The strategic logic is tuning through pairing. A single dopant shifts binding energy in one direction; a second dopant, chosen from a different region of the periodic table, introduces a second degree of freedom. Seven preferred binary pairs have been identified and screened at the descriptor level — bismuth plus tin, bismuth plus indium, bismuth plus lead, indium plus tin, tin plus lead, antimony plus tin, germanium plus indium — each producing a distinct CO2-to-formate or CO2-to-CO adsorption fingerprint. This is not incremental variation on a known idea; the whitespace specifically lies above the single-dopant art, in a region that prior filers did not claim. The portfolio this asset belongs to — integrated packaging, storage, and PFAS-treatment systems — includes this asset as a lead composition filing precisely because the double-dopant approach opens a new compositional landscape without conflicting with existing granted claims. The timing argument is straightforward: CO2RR is entering the scale-up phase. Several commercial pilots targeting green formate and green CO are now operating or planned, with catalyst specification decisions due in the 2026-2028 timeframe. An issued patent covering seven distinct binary-dopant pairs, spanning the formate-selective and CO-selective regimes, positions the holder to participate in licensing conversations precisely as those pilots seek catalyst differentiation. The window to file and prosecute a blocking composition claim in this whitespace is narrow, as the field is active.

The base material is a copper single-atom alloy host — a face-centered cubic copper matrix into which post-transition metal substituents are incorporated at low concentration, each between 0.1 and 5 atomic percent. The key claim is the simultaneous presence of at least two dopant atoms drawn from the set {Bi, In, Sn, Pb, Sb, Ge}. Individually, each of these elements is known to modulate the adsorption energy of CO2 and formate intermediates on copper surfaces; together, they create a coupled electronic environment that shifts the d-band center and modifies the local coordination geometry around the dopant pair in ways that a single substituent cannot achieve alone. The formula Cu(Bi,Sn) is the prototype, but six additional binary combinations are computationally characterized and included as preferred embodiments. The computational characterization to date uses a cluster-expansion descriptor approach applied to all seven pairs. This framework maps the binding energy of CO2 and key formate intermediates as a function of dopant identity and local environment without requiring a full periodic-DFT slab calculation for every configuration. The resulting binding energy range runs from approximately -0.58 eV to -0.86 eV (all values are descriptor-level estimates, not converged plane-wave DFT results). The Bi+Sn pair sits at the strongest-binding end (-0.86 eV), favoring the formate pathway; the Bi+In pair (-0.73 eV) falls closer to the CO-selective window. These descriptor values are useful for rank-ordering candidates and establishing the claim breadth, but they are not equivalent to full periodic-DFT calculations with explicit solvent — a distinction the dossier makes plainly. On the stability side, the materials here are surface-doped alloys rather than extended crystalline compounds, which means the standard phonon-stability workflow applied to bulk periodic materials in the broader portfolio does not directly apply. The relevant stability question for a SAA is whether the dopant atoms are thermodynamically and kinetically stable against segregation to the surface or bulk under electrochemical conditions — a question that requires periodic DFT with explicit electrochemical boundary conditions, as noted in the open validation gates. No bulk space group is assigned, and no multi-engine consensus phonon screening has been performed for these alloy surfaces, because that workflow is designed for periodic bulk or 2D layered structures rather than dilute-alloy catalysts. This is an honest architectural difference, not a gap in the computational framework. The broader Lattice Graph computational validation stack — which elsewhere applies four independent machine-learning interatomic potentials (MACE, CHGNet, MatterSim, and ORB) in consensus for phonon stability — is not the primary validation tool for surface-alloy catalysis. Instead, the relevant simulation chain for this asset runs through slab-model periodic DFT with an explicit electrochemical interface, minimum-energy-path calculations for the CO2-to-formate and CO2-to-CO reaction coordinates, and ultimately membrane-electrode-assembly (MEA) device-level selectivity demonstration. Those are the two open proof gates identified in the current data: periodic DFT plus explicit solvent on each of the seven pairs, and an MEA selectivity experiment.

The addressable market for CO2 electroreduction catalysts sits inside the broader carbon-capture-and-utilization industry, which is at an early but accelerating commercial stage. Formate and formic acid represent the nearest-term commercial product: formate is already a commodity chemical used in leather processing, de-icing, and as a hydrogen carrier, with global demand well above one million metric tons per year. CO is a direct feedstock for Fischer-Tropsch synthesis and for syngas routes to methanol and ethanol. Ethylene and other C2+ products are longer-term targets. Market sizing for CO2RR catalysts specifically is estimated in the range of $1-5 billion addressable, dominated by catalyst supply to electrolyzer developers and by licensing royalties from technology integrators — these are rough-order-of-magnitude estimates consistent with the nascent stage of the industry. The buyer profile for licensing or acquisition of this asset is primarily the CO2-to-chemicals developer: companies building electrolyzer stacks for formate production, green CO generation, or carbon-negative syngas. These organizations need catalyst differentiation to justify their technology position with industrial partners and to defend their own intellectual property. A second buyer class is the catalyst manufacturer who supplies standardized electrode materials to stack builders; a composition patent covering seven binary-dopant pairs gives them a portfolio position without requiring them to develop the underlying science. Royalty logic is likely per-kilogram of electrode material or per-kilowatt of installed electrolyzer capacity, with rates typical for specialty catalyst materials (often in the low-single-digit percentage of product value). The race window is not formally quantified in the available data, but the competitive dynamics are visible: multiple academic groups and startups have published on single-dopant Cu-SAA for CO2RR in the 2020-2025 period, and at least some of that work is progressing toward patent filings. The double-dopant space is genuinely less crowded today, but that window narrows as the field matures. Getting a composition claim issued before the next wave of publications and filings is the primary timing driver.

tunable formate/CO/C2 selectivity via dopant-pair selection

double-dopant Cu-SAA whitespace over single-dopant art

The primary acquirers and licensees for this asset are companies actively developing CO2 electroreduction technology for commercial formate or CO production. This includes vertically integrated carbon-utilization startups that control both the electrolyzer stack and the catalyst formulation, as well as established chemical companies seeking to add CO2-negative feedstock options to their production networks. Catalyst manufacturers supplying standardized electrode materials to stack builders are a natural licensing target: a composition patent covering seven binary-dopant pairs, with a clean FTO position, gives them a defensible product line without requiring in-house materials discovery capability. Strategic fit is also strong for any company that has already licensed or acquired single-dopant Cu-SAA technology and wants to extend its IP position into the next generation of the catalyst class. A secondary buyer class includes electrolyzer companies and energy majors with CO2 utilization mandates, for whom this asset functions as both a technical option and a defensive IP position against competitors who might otherwise lock up the double-dopant space. The asset is at an early stage — descriptor-level proof, no MEA data — which means the most likely deal structure at this point is a sponsored research agreement or option-to-license arrangement, with the acquirer funding the completion of the two open proof gates in exchange for exclusivity rights. That structure aligns the buyer's R&D investment with the IP risk and is consistent with how early-stage composition claims in catalysis are typically commercialized.

The principal technical risk is that the descriptor-level binding energies do not survive translation to full periodic-DFT and explicit-electrochemical conditions. The cluster-expansion descriptor approach is a well-validated screening tool, but the CO2RR landscape is sensitive to solvent effects, pH, and electrode potential in ways that descriptor models can underestimate. A specific risk is that one or more of the seven pairs shows anomalous behavior — dopant segregation under applied bias, or competitive hydrogen evolution that erodes selectivity — when modeled or tested at the slab level. This is manageable: the seven-pair breadth means that even if two or three pairs underperform, the composition claim retains significant value through the remaining pairs. The FTO position is not at risk from this outcome, only the strength of the experimental evidence base for prosecution. The prosecution risk is that an examiner cites single-dopant prior art more broadly than the claim construction supports and requires narrowing to specific pairs. That risk is mitigated by the clear structural distinction between a single-dopant and a double-dopant SAA, and by the seven named preferred embodiments providing multiple fallback positions. The roadmap to de-risk both technical and prosecution uncertainty runs through the two identified open gates — periodic DFT with explicit solvent and one MEA selectivity demonstration — with the DFT calculations being the higher priority because they directly support claim prosecution. A 12-month sponsored-research program delivering DFT results on all seven pairs would substantially change the risk profile of this asset.