Strontium zinc phosphide Zintl photovoltaic absorber with earth-abundant composition

SrZn2P2 sits at a compelling intersection of materials physics and market timing. It is a Zintl-phase phosphide — an ordered intermetallic compound built from strontium, zinc, and phosphorus, all earth-abundant and none of them toxic — that computation now shows carries a direct bandgap of approximately 1.72 eV. That value lands squarely inside the Shockley-Queisser single-junction efficiency window, where the theoretical conversion limit peaks. The significance is straightforward: the thin-film photovoltaic industry has long been constrained to a narrow shortlist of high-performance absorbers — cadmium telluride, copper indium gallium selenide, and halide perovskites — each of which carries either critical toxicity burdens, supply-chain fragility rooted in rare or geopolitically concentrated elements, or, in the perovskite case, unresolved long-term stability questions. A well-matched, direct-gap absorber composed entirely of abundant and non-hazardous elements would address all three concerns simultaneously. The timing pressure is real but nuanced. Independent experimental thin-film growth of SrZn2P2 was reported on arXiv in late April 2026, just ahead of this filing's priority date — a development that compresses the claim perimeter for the bare composition but does not eliminate IP value. Lattice Graph's position is not on the composition itself or the basic growth technique; it is on the integrated device architecture: dopant schemes, passivation approaches, buffer-layer pairings, and halide-flux process variants that transform a proof-of-concept thin film into a manufacturable, efficiency-optimized photovoltaic cell stack. This is the kind of engineering-layer IP that typically captures the larger share of commercial value anyway, because device makers license stacks, not elements. The portfolio to which this asset belongs — integrated packaging, storage and PFAS-treatment systems — reflects a broader Lattice Graph strategy of holding claims that mature at different timescales, and SrZn2P2 serves as a near-term anchor in the renewable-energy materials segment of that portfolio. The forced-substitution dynamic underpinning this asset's commercial logic is regulatory and supply-driven. Cadmium telluride faces tightening end-of-life regulations in the European Union and increasingly in U.S. state jurisdictions. CIGS depends on indium, whose supply is concentrated and priced accordingly. Perovskites, despite extraordinary efficiency trajectories, have not cleared a decade-long outdoor stability threshold in any certified commercial product. Each of those constraints creates a procurement rationale for device manufacturers to qualify an alternative absorber. SrZn2P2, if device efficiency can be demonstrated, offers a drop-in argument: thin-film deposition tooling is largely substrate-and-process agnostic, and a 1.72 eV direct gap with absorption exceeding 10,000 per centimeter means the absorber layer can be thin, reducing materials cost per watt.

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.

SrZn2P2 crystallizes in the trigonal P-3m1 space group, adopting the classic CaAl2Si2-type Zintl structure in which an anionic [Zn2P2]^2- framework of edge-sharing tetrahedra is charge-balanced by the divalent strontium cation. This layered, ordered arrangement is responsible for several properties favorable to photovoltaics: the tetrahedral Zn-P bonding creates a covalent network with high carrier mobility potential, while the ionic Sr sublattice provides the charge-transfer character that opens a direct gap. The structure is isoelectronic with more-studied arsenide Zintl phases but benefits from the lower atomic number of phosphorus, meaning no arsenic toxicity concerns. Hybrid density-functional theory calculations place the direct bandgap at approximately 1.72 eV and confirm that SrZn2P2 lies on or very close to the convex hull of thermodynamic stability relative to competing binary and elemental phases — meaning the phase should be accessible in synthesis without exotic stabilization. The optical absorption coefficient exceeds 10,000 per centimeter in the relevant spectral range, which is characteristic of strong direct transitions and implies that absorber layers of only a few hundred nanometers to perhaps one micron would be sufficient for near-complete photon harvesting. This is thinner than typical CIGS or CdTe devices, directly reducing materials cost per unit area. Hybrid DFT (simulation reference U-SRZN2P2-001) was the core computational method; one DFT source underpins the electronic structure and convex-hull placement reported here. Dynamic stability — the question of whether the crystal lattice will sustain phonon oscillations without collapsing into a different structure — was assessed using two independent machine-learning interatomic potentials: MACE and CHGNet. Both agree that SrZn2P2 is dynamically stable, showing no imaginary phonon modes across the Brillouin zone. This two-potential consensus is meaningful: MACE and CHGNet are built on different architectures and trained on different subsets of the Materials Project and related ab initio databases, so their agreement constitutes a genuine independent cross-check rather than a redundant computation. ORB and MatterSim were not evaluated for this compound; the two-potential consensus is sufficient for a lead-stage assessment, and the open validation gate is experimental device efficiency, not structural stability. The excluded compositions in the AZn2P2 family sharpen the picture: BaZn2P2 was evaluated and found to be metallic at the DFT level, making it unsuitable as a photovoltaic absorber and thus explicitly excluded from device claims, while SrMg2As2 and CaMg2As2 were declined because they exhibit indirect bandgaps, which would substantially reduce absorption efficiency. These negative results are part of Lattice Graph's labeled dataset and strengthen the specificity of the SrZn2P2 selection. CaZn2P2, the calcium analogue, has been published in the academic literature and is treated as a fallback position rather than a primary claim target. The nitrogen analogue SrZn2N2 and mixed A-site compositions of the form (A1,A2)Zn2P2 round out the broader chemical family in the patent coverage, preserving optionality if manufacturers find processing advantages in alloyed or nitride variants. The Zintl framework's tolerance for A-site substitution is well established in other phosphide families, making these extensions chemically plausible, though they carry higher experimental uncertainty than the Sr-phosphide lead.

The addressable market for this asset sits within thin-film photovoltaics, specifically the absorber-layer materials segment. Global thin-film PV module shipments were on the order of 25-30 GW per year as of the mid-2020s, with CdTe (primarily First Solar) and CIGS together representing the non-silicon segment. At module prices in the range of $0.20-0.30 per watt and absorber-material licensing royalties that historically run in the low single-digit percentage of module value, the addressable royalty pool for a novel absorber with strong IP is plausibly in the range of tens to low hundreds of millions of dollars per year at meaningful market penetration — though at this stage in development, such estimates should be understood as illustrative upper bounds contingent on successful device demonstration. The total addressable market for the broader thin-film PV absorber opportunity is estimated at $1-5 billion, covering materials supply, process licensing, and device-architecture licensing across the thin-film manufacturing ecosystem. The customer base for licensing or sale of this IP is thin-film PV manufacturers, both established producers seeking alternatives to CdTe or CIGS and new entrants building around alternative absorber platforms. This group includes vertically integrated module makers, research institutes with commercial licensing programs, and equipment manufacturers who sell turnkey deposition systems and would benefit from absorber IP to bundle with process know-how. Government-backed energy laboratories are also plausible partners for co-development agreements that defray the experimental cell demonstration cost while preserving upside. Royalty logic tracks standard semiconductor licensing practice: a per-watt royalty on cells using a patented absorber stack, or a lump-sum license for a defined manufacturing capacity, with the specific device-stack and passivation claims giving licensors leverage at the cell-architecture level rather than merely the material level — which is precisely where SrZn2P2's IP has been positioned after the bare-composition publication narrowed the landscape.

earth-abundant non-toxic PV absorber with single-junction-matched gap

narrowed to device-stack/dopant/passivation/buffer + halide-flux process carve-outs over published bare-composition growth

The most likely acquirers or licensees for this asset are thin-film PV manufacturers with existing infrastructure for depositing chalcogenide or pnictide absorbers and an active interest in absorber diversification. First Solar, which has built its business on CdTe and is aware of the regulatory and supply-chain risks embedded in that choice, would have strategic motivation to hold optionality on alternative absorbers at a device-architecture level. CIGS manufacturers such as Solar Frontier (now reduced in scale) and emerging entrants in Europe and Asia similarly have deposition tool platforms that could be adapted to phosphide growth. Beyond module makers, thin-film equipment manufacturers — Applied Materials' Baccini division, Von Ardenne, and comparable systems integrators — have a business model built around licensing process know-how alongside capital equipment, and phosphide absorber IP would complement that offering if bundled with deposition process know-how. National laboratories or government energy programs (U.S. DOE NREL, Fraunhofer ISE, CSIRO) are plausible co-development partners rather than outright buyers, providing the experimental cell demonstration resources in exchange for co-inventor status or non-exclusive research licenses, while the commercial exclusivity is retained for sale to an industrial party.

The primary risk is the gap between computational prediction and experimental device performance. Hybrid DFT and MLIP-confirmed stability are strong indicators, but photovoltaic efficiency depends on properties that computations do not fully capture at this stage: the density and energy levels of point defects that act as recombination centers, the band alignment at the buffer-layer interface, minority carrier diffusion length, and resistance of the film to oxidation or moisture during processing and operation. The 1.72 eV bandgap is ideal for single-junction efficiency but means the absorber sees a relatively narrow slice of the solar spectrum — any significant sub-gap absorption from defect states would reduce open-circuit voltage and erode the theoretical advantage. The path to de-risking runs through experimental cell fabrication: a collaborating laboratory needs to grow SrZn2P2 by a controlled method, pair it with a candidate buffer layer (likely a II-VI or transparent conductive oxide with matched band offset), fabricate a rudimentary cell, and measure current-voltage characteristics under standard illumination. Even a preliminary power conversion efficiency above a few percent would substantially upgrade the asset's commercial status. A secondary risk is claim scope: the narrow FTO position means that competitors who publish cell results before IP prosecution completes may establish prior art that further reduces claim breadth. Continued monitoring of the arXiv and journal literature on Zintl phosphide photovoltaics, combined with rapid provisional filing of any new experimental results, is the appropriate mitigation.