Copper antimony sulfide skinnerite thin-film solar absorber

Cu3SbS3 skinnerite represents a structurally distinct entry in the copper antimony chalcogenide space — an orthorhombic (space group P2₁2₁2₁) phase with a direct bandgap in the 1.0–1.5 eV range, placing it squarely in the single-junction solar absorption window. The compelling element of the IP position is not merely that skinnerite is an attractive absorber candidate in isolation, but that the patent family explicitly carves away its most obvious imitators. The two closest competitor phases — CuSbS2 chalcostibite and Cu3SbSe3 — have been demonstrated computationally to be dynamically unstable and are explicitly excluded from the claims, leaving a defensible whitespace around the single surviving Cu–Sb–S phase geometry. This architecture of positive claims combined with negative limitations on killed siblings is a deliberate strategy to prevent design-arounds through simple cation or chalcogen substitution. The timing of this asset is shaped by two converging pressures in the thin-film photovoltaics industry. First, cadmium telluride (CdTe) faces increasing regulatory pressure on cadmium handling and disposal, and the CIGS (copper indium gallium selenide) supply chain depends on indium and gallium, both of which are designated critical materials in the US, EU, and Japan. Second, the leading earth-abundant alternative — kesterite Cu2ZnSnS4 (CZTS) — has been trapped below 13% certified efficiency for over a decade due to intrinsic disorder on the Cu/Zn sublattice. Skinnerite occupies a compositionally simpler structural landscape (no mixed-site disorder analogous to CZTS) and, if the bandgap proves suitable post-HSE06 validation, could attract development effort from the same thin-film manufacturing community that has already invested in sulfide processing infrastructure. The patent family claims Cu3SbS3 as the primary composition and extends protection to structural analogs Cu3BiS3, Cu3AsS3, and Cu8Sb3S8, giving a licensor significant room to negotiate across the near-neighbor chemical space while the core composition remains the most computationally validated.

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.

Cu3SbS3 skinnerite crystallizes in the orthorhombic P2₁2₁2₁ space group, a chiral, non-centrosymmetric arrangement that distinguishes it from both chalcostibite (CuSbS2, monoclinic) and the tetrahedrite family. In the skinnerite structure, copper occupies two crystallographically inequivalent sites — a linear two-coordinate site and a trigonal three-coordinate site — while antimony adopts a pyramidal coordination characteristic of the Sb(III) lone-pair stereochemistry. This structural asymmetry creates an intrinsically polar environment that can favor carrier separation and suppresses some of the recombination pathways that undermine simpler binary sulfide absorbers. The (010) facet exposure in this geometry is computationally relevant both to the solar absorber application and to the dependent hydrogen evolution reaction (HER) use case, where adsorption-energy screening across (010), (001), and (110) surface terminations placed the (010) facet near the Sabatier optimum at approximately +0.05 eV hydrogen adsorption free energy — a value that is competitive with the best non-noble HER catalysts and forms the basis for the dependent, secondary claim in this family. The bandgap target of 1.0–1.5 eV is highly desirable for single-junction photovoltaics; the Shockley–Queisser limit peaks near 1.3–1.4 eV, and a direct gap ensures strong absorption without requiring the thick absorber layers that impede minority-carrier collection in indirect semiconductors. The current DFT estimates place the gap near the lower end (approximately 1.0 eV at the GGA level), and the HSE06 hybrid-functional recalculation — re-submitted on 2026-06-10 and still pending — is expected to revise this upward toward the more favorable part of the 1.0–1.5 eV window. GGA-level bandgap underestimation is a well-understood systematic error, and the HSE06 correction is the standard validation gate before optical-absorption coefficients can be reported with confidence. This is the single open proof gate for the asset. The dynamic stability picture for Cu3SbS3 skinnerite is the most technically nuanced aspect of this asset. Four independent machine-learning interatomic potentials — MACE, MatterSim, ORB, and CHGNet — were deployed to compute phonon dispersions across the Brillouin zone without any shared functional form or training trajectory. Three of the four potentials (MACE at +0.137 THz, MatterSim at +0.157 THz, and ORB at +0.152 THz) report positive minimum phonon frequencies across all high-symmetry paths, indicating the structure sits in a true local energy minimum with no imaginary-mode instabilities. CHGNet alone returns a small negative frequency (−0.201 THz), which is a soft-mode signal rather than a catastrophic instability. A 3-of-4 majority consensus among independent potentials is a substantive validation in computational materials science, where disagreement across potentials often flags genuinely problematic structures. The CHGNet dissent is documented candidly and does not overturn the majority finding, but it does motivate the ongoing HSE06 gate and would ideally be followed by a full DFT phonon calculation as part of the advancement roadmap. Two independent DFT source calculations already support the structural model, reinforcing confidence in the majority result. The killed-sibling phonon analysis is equally important technically. CuSbS2 chalcostibite, the most commonly cited member of this chemical family in the photovoltaics literature, was found to exhibit approximately 381 imaginary phonon modes — a result that is not a borderline soft-mode but a wholesale structural collapse under the phonon calculation, indicating the bulk phase is dynamically unstable under the conditions modeled. Cu3SbSe3 similarly failed dynamic stability screening. These results are the technical basis for the explicit exclusion of both phases from the claims. The exclusion is strategically valuable: by demonstrating that the closest alternatives are computationally dead ends, the IP position argues that skinnerite is not merely one of many viable Cu–Sb–chalcogenide absorbers but may be the principal surviving phase in this corner of composition space. The adsorption-energy screen across three low-index skinnerite surface terminations using DFT-calculated H* binding energies rounds out the computational evidence relevant to the dependent HER claim.

The addressable market for this asset straddles two segments. The primary application — thin-film solar absorbers — sits within a global thin-film PV market that exceeded $15 billion in annual revenue as of the mid-2020s and is projected to grow with continued utility-scale solar deployment. Within that market, the relevant sub-segment for an emerging earth-abundant absorber is the portion currently served by CdTe and CIGS that could be displaced by a sulfide-based alternative with comparable performance and lower materials cost. This is a contested and speculative displacement scenario; estimates of a serviceable market for a new absorber technology in the $0.5–2 billion range reflect the realistic portion reachable via licensing or materials supply agreements within a 10–15 year horizon, not immediate capture. Royalty logic for a composition-of-matter absorber patent would typically take the form of a per-watt or per-square-meter license fee to thin-film module manufacturers, or a materials licensing agreement with sulfide precursor suppliers. The secondary application — hydrogen evolution catalysis using the (010) facet — is a dependent use case that strengthens the family's overall commercial surface without requiring a standalone commercial pathway for the HER application alone. The green hydrogen electrolysis market is growing rapidly, and non-noble HER catalysts are a genuine unmet need, but this application depends entirely on the primary solar absorber validation succeeding first. Buyers who are primarily interested in the thin-film PV rights would acquire the HER dependent claim as part of the family package; buyers focused on electrocatalysis would have secondary interest. The customer base for the primary application is concentrated among a small number of global thin-film PV manufacturers — including established CdTe producers, CIGS module manufacturers, and emerging players in sulfide thin-film technologies — plus the upstream precursor and target-material suppliers who supply sputtering or evaporation feedstocks to those manufacturers.

earth-abundant copper-antimony absorber distinct from killed CuSbS2/Cu3SbSe3 + kesterite family

skinnerite phase + killed-sibling exclusion

The most natural acquirers or licensees for this asset are thin-film photovoltaic manufacturers who are actively diversifying their absorber material portfolios away from cadmium and indium dependence. This includes established CdTe module producers with manufacturing infrastructure already optimized for sulfide thin-film deposition, CIGS producers with selenization and sulfurization process capability who could adapt existing reactors to Cu–Sb–S targets, and the smaller cohort of companies and research consortia specifically pursuing CZTS and related earth-abundant sulfide absorbers. Precursor and sputtering-target suppliers who serve these manufacturers are also plausible licensees for the composition-of-matter claims, as they would need freedom to operate to supply Cu3SbS3 targets to PV device manufacturers. National laboratories and university spin-outs with active thin-film PV programs in the earth-abundant space represent a secondary audience who might seek non-exclusive licenses for academic development rights as a path to establishing skinnerite as a recognized absorber class. The dependent HER claim adds a small but real additional buyer segment in the electrocatalysis space — specifically, developers of non-noble hydrogen evolution catalysts for alkaline electrolysis systems. These buyers are unlikely to be the primary acquirers of the family but could be relevant in a sub-licensing or field-of-use split where the PV rights and the HER rights are commercialized separately. Given the current state of the asset (pending HSE06 gate, majority-stable rather than consensus-stable), the most likely near-term transaction structure is a license with milestone payments tied to HSE06 confirmation and subsequent experimental deposition results, rather than a full acquisition at current validation state.

The primary technical risk is that the CHGNet minority dissent on phonon stability represents a real physical effect rather than a model artifact. If CHGNet is correct and the other three potentials are overcounting stability in a region of chemical space where their training data is sparse, the manufactured skinnerite phase could exhibit a structural instability at growth temperatures or under epitaxial strain, manifesting as phase segregation, poor film morphology, or preferential formation of a competing phase. This risk is not resolvable at the machine-learning potential level; it requires a full DFT phonon dispersion calculation and, ultimately, experimental synthesis and structural characterization. The roadmap to de-risk this is well-defined: complete the HSE06 bandgap calculation currently in queue, follow with DFT-level phonon calculation on the relaxed HSE06 structure, commission a synthesis attempt (sulfurization of co-evaporated Cu and Sb precursors is the most tractable route given existing sulfide thin-film infrastructure), and characterize the resulting films by X-ray diffraction and Raman spectroscopy to confirm phase purity. If the skinnerite phase forms cleanly and retains the predicted direct bandgap, the asset moves from a computationally validated candidate to a demonstrated absorber material. The commercial risk is that even a fully validated skinnerite absorber faces a long development path to device-level efficiency, and the thin-film PV market tends to consolidate around technologies that have already demonstrated competitive efficiencies rather than rewarding early-stage composition-of-matter positions. A licensee will need to invest substantially in device engineering — contact selection, buffer layer compatibility, defect passivation — before any module-level performance is achievable. This means the asset's value is most likely realized as a foundational composition patent that grows in value as the field develops, rather than as an immediately productizable technology. The tetrahedrite exclusion — Cu12Sb4S13 excluded absent proof — leaves a gap in the family coverage that a competitor could exploit if tetrahedrite proves stable and active, making a follow-on tetrahedrite stability study a prudent defensive investment.