Potassium palladium sulfide electrocatalyst for CO2 reduction to formate and CO

Potassium palladium disulfide (K2PdS2) is a polar-acentric alkali-palladium sulfide that sits at a compelling intersection: it is a visible-light-compatible semiconductor with a band gap of approximately 2.38 eV (resolved by HSE06 hybrid functional calculations) that can function as a cathode material for the electrochemical reduction of CO2 to formate and CO in a flow-cell architecture. The compound's low palladium loading relative to conventional palladium-metal CO2 reduction catalysts, combined with its sulfide coordination environment that modifies the d-band electronic structure of Pd, makes it a structurally distinct entry point into an electrocatalyst space currently dominated by either PGM metals or first-row transition-metal materials. The patent position — anchored specifically on the A2PdS2 sulfide composition with CO2 reduction as the method of use — has been screened against the materials-patent landscape and found to be clear, with the closest structural relatives (selenide and telluride analogs) explicitly distinguished in the claim boundary. The timing reflects a real commercial dynamic. Carbon utilization mandates and the economics of green hydrogen are together forcing serious investment into scalable CO2-to-fuel pathways. Flow cells using gas-diffusion electrodes are the dominant architecture for high-throughput CO2 electroreduction, and the cathode catalyst remains the principal techno-economic bottleneck: palladium metal is active toward formate and CO but is expensive at high loadings, while copper and silver suffer from selectivity and stability trade-offs. A Pd-sulfide phase that retains favorable CO2 activation energetics at lower PGM site density, and that could in principle harvest visible photons for a photo-electrochemical variant, addresses both the cost structure and the functionality envelope simultaneously. The asset sits within the catalysts and energy-conversion materials portfolio and is designated as a lead filing — the primary offensive position of the chalcogenide CO2-reduction family.

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.

K2PdS2 crystallizes in the polar, non-centrosymmetric space group Cmc2₁ (orthorhombic, point group mm2). In this structure, palladium adopts a square-planar or distorted coordination by sulfur, consistent with the d⁸ electronic configuration of Pd²⁺, while the potassium ions occupy interstitial positions that balance charge and define the layered, chain-like topology of the inorganic framework. The polar-acentric symmetry has implications beyond catalysis: it rules out centrosymmetric space groups and makes the material a candidate for second-harmonic generation and piezoelectric response, though those properties are not the primary claim target here. The coordination environment around palladium — all-sulfide rather than all-oxide or metal — significantly perturbs the local d-band center relative to bulk Pd, which is the principal lever for tuning adsorption free energies of CO2 reduction intermediates. Band structure calculations tell a critical story about the suitability of K2PdS2 for electrochemical and potential photoelectrochemical CO2 reduction. The standard PBE generalized-gradient approximation significantly underestimates the band gap, returning a value of approximately 1.05 eV — a number that would indicate a narrow-gap semiconductor with rapid recombination and limited visible-light photovoltage. A higher-level HSE06 hybrid functional calculation, completed in June 2026 with fixed occupations to handle potential convergence artifacts, resolves the gap to approximately 2.38 eV. This is a meaningful distinction: 2.38 eV places the absorption edge in the blue-green visible range, consistent with a material that could absorb a significant fraction of the solar spectrum if used in a photoelectrochemical cell, while still maintaining sufficient driving force for CO2 reduction thermodynamics. The HSE06 pipeline was validated against cubic boron nitride (c-BN, experimental gap ~6.4 eV; calculated 6.31 eV), confirming that the functional and basis set settings are behaving correctly before applying them to K2PdS2. Dynamic (phonon) stability has been evaluated by two independent machine-learning interatomic potentials: MACE and CHGNet. Both agree that the K2PdS2 structure in the Cmc2₁ setting is dynamically stable, with no imaginary (negative-frequency) phonon modes anywhere in the Brillouin zone. The minimum phonon frequency across the computed dispersion is +0.0851 THz (MACE), and CHGNet independently returns positive values throughout — consensus on the absence of soft modes confirms that the structure does not want to spontaneously distort to a lower-symmetry phase, which would otherwise call into question the relevance of the computed electronic structure. A second phonon recheck using a deep-GPU accelerated protocol was completed as an additional validation pass. The phonon calculations used the mp-1068941 reference entry as the structural starting point. A soft bond-valence-sum ionic proxy calculation (performed as an inexpensive consistency check on oxidation-state assignments) further supports the K⁺/Pd²⁺/S²⁻ charge balance underlying the electronic structure description. Surface and intermediate adsorption energetics were evaluated by DFT-level adsorption energy screening for key CO2 reduction intermediates: *CO (carbon monoxide bound to the surface Pd site), *OOCH (formate bound through oxygen in a bidentate geometry), and *COOH (carboxyl intermediate). These three adsorption calculations collectively map out the two dominant CO2 reduction pathways — the formate pathway proceeding through *OOCH and the CO/syngas pathway proceeding through *COOH — and allow a qualitative assessment of whether the Pd-sulfide surface sits at a favorable position on the activity volcano relative to known active CO2RR materials. The existence of these adsorption simulations (identified as Simulation Example F in the supporting computational record) provides the mechanistic grounding for the claimed operating voltage window of −0.7 to −1.1 V versus RHE. Two independent DFT sources underpin the computational record for this asset.

The addressable market for this asset sits within the broader CO2 utilization and e-fuels space. The immediate application — electrochemical CO2 reduction to formate and carbon monoxide in a flow cell — maps to two adjacent commercial segments. Formate and formic acid have near-term markets as chemical feedstocks, hydrogen carriers, and direct formic acid fuel cell feeds, with the global formic acid market estimated in the range of several hundred million dollars annually and growing as green chemistry substitution accelerates. Carbon monoxide and syngas (CO/H2) are foundational chemical intermediates for Fischer-Tropsch and methanol synthesis, anchoring a much larger industrial market; CO2-derived syngas is a direct drop-in for those synthesis routes. These estimates are approximate — the specific fraction addressable by a novel electrocatalyst material depends heavily on cost and selectivity performance at scale. The total addressable market for CO2 utilization and electrochemical CO2 reduction technologies is commonly cited in the range of one to three billion dollars annually within this decade, scaling substantially as carbon pricing and regulatory mandates expand. Lattice Graph's internal market sizing for this asset is consistent with that range. The royalty or licensing logic is straightforward for this class of material: a PGM-containing electrocatalyst with a clean FTO position and a composition-plus-use patent can be licensed per kilogram of material produced, or as part of a broader technology package to a flow-cell developer or CO2 utilization project developer. Customers are primarily CO2-utilization technology companies, e-fuel developers, and industrial electrochemistry companies building out CO2 capture-and-conversion infrastructure. Carbon capture and utilization project developers operating at industrial scale would represent the highest-value licensing targets because the cathode catalyst specification is a recurring procurement decision across the lifetime of a project.

reduced PGM content vs conventional Pd CO2RR catalysts; visible-light-compatible semiconductor

A2PdS2 sulfide + CO2RR method-of-use; selenide/telluride and non-CO2RR uses excluded

The most natural strategic buyers or licensees for this asset are companies operating at the intersection of carbon capture, CO2 utilization, and electrochemical technology. Flow-cell CO2 electroreduction developers — including startups and established industrial gas or specialty chemical companies building out CO2-to-formate or CO2-to-syngas infrastructure — represent the primary target because the asset is directly relevant to their cathode procurement decisions. Companies already working with palladium catalysts in electrochemical contexts (e.g., hydrogen evolution, fuel cell anode materials) have existing supply chains and characterization infrastructure that lower the barrier to adopting a Pd-sulfide variant. Industrial gas companies and chemical majors with stated CO2 utilization commitments (under regulatory or ESG pressure) are also plausible licensees, particularly if the visible-light-compatible band gap enables a photo-assisted mode that reduces electricity consumption in their CO2 conversion processes. On the licensing side, a per-kilogram royalty on catalyst material or a lump-sum license with milestones tied to demonstrated Faradaic efficiency in a flow cell would be the natural deal structures. The asset is well-suited for a development license at this stage — the FTO is clean, the computational record is solid, and the principal open gate (experimental Faradaic efficiency) is a defined, bounded experiment rather than an open-ended research program. A partner with synthesis and electrochemical testing capability could close that gap in a matter of months, at which point the asset's value would step up substantially.

The principal technical risk is the gap between computational projection and experimental Faradaic efficiency measurement. All the adsorption-energy calculations and band structure work are consistent with a promising CO2RR cathode, but the actual selectivity toward formate versus CO versus competing reactions (hydrogen evolution in particular) has not been measured on synthesized K2PdS2 under flow-cell conditions. Palladium is well known to be active for both CO2RR and hydrogen evolution, and the sulfide surface may shift the selectivity in either a favorable or unfavorable direction relative to Pd metal — this is the central empirical unknown. Synthesis of phase-pure K2PdS2 in the Cmc2₁ structure requires handling air-sensitive alkali-metal sulfide precursors and may present processing challenges that affect surface area and electrode fabrication. The stability of the Pd-S bonds under cathodic polarization in aqueous or mixed electrolytes is also an open question — sulfide ligands can be labile under reductive conditions. De-risking proceeds in a defined sequence: solid-state synthesis and phase characterization by powder X-ray diffraction to confirm the Cmc2₁ structure, followed by electrode fabrication on a gas-diffusion layer, chronoamperometry at the target potentials (−0.7 to −1.1 V vs. RHE), and product quantification by ion chromatography and gas chromatography. If Faradaic efficiencies toward formate or CO are demonstrated at competitive levels, the asset transitions from a computationally grounded claim with a prophetic example to one with experimental reduction-to-practice, substantially strengthening enforceability and commercial attractiveness. The PGM cost of Pd remains a structural risk relative to first-row alternatives, but the low-loading architecture is the mitigation strategy already embedded in the asset's commercial rationale.