Bandgap-tunable zinc germanium nitride alloy absorber for tandem solar cells and SWIR detectors

The global photovoltaics industry faces a structural materials problem at the top-cell layer of tandem stacks. The dominant thin-film absorbers — cadmium telluride and copper indium gallium selenide — either rely on toxic heavy metals or on indium, a critical mineral concentrated in a handful of supply chains. Perovskite-halide alternatives have attracted enormous research capital but face persistent stability and lead-toxicity concerns that have stalled regulatory approval for outdoor deployment. Meanwhile, the Shockley-Queisser optimum for a two-terminal tandem architecture sits near 1.7-1.9 eV for the top cell, a range that very few earth-abundant, non-toxic material families can continuously span while maintaining a direct gap. Zn(Ge,Sn)N2 and its close relatives within the II-IV-N2 ternary nitride family occupy precisely that vacancy. The Zn–Ge–Sn–N system forms in the orthorhombic Pna2_1 structure — a wurtzite superstructure with ordered cation sublattice — and alloy substitution of Sn for Ge, or Si for Ge, continuously shifts the direct bandgap across 1.1 to 3.5 eV. That range covers the tandem-PV top-cell region, but the short-wave infrared window (roughly 1.0–2.5 µm in wavelength) needs lower photon energies than the computed gaps, so detector relevance is limited to the near-infrared edge. All constituent elements — zinc, germanium, tin, silicon, nitrogen — are either genuinely abundant or available from geopolitically stable supply chains without the concentration risk that attaches to indium, tellurium, or cadmium. The strategic moment for this filing is the race to secure broad alloy composition claims before the HSE06 hybrid-functional optical-bandgap calculation closes the final proof gate. The portfolio that includes this asset — covering dielectric, ferroelectric, and wide-bandgap oxide and nitride materials — gives a buyer a drafted claim set on substituted II-IV-N2 alloys aimed at the tandem-PV and photodetector verticals, with the parent compound already in the prior art.

Candidates are screened for imaginary phonon modes with the machine-learning interatomic potentials named above. This is a computed harmonic check, not a measurement; where engines disagree, that is shown.

The parent compound ZnGeN2 crystallizes in the orthorhombic space group Pna2_1 (No. 33), which is a cation-ordered derivative of the wurtzite structure. In this arrangement, Zn and Ge occupy distinct tetrahedral sites on alternating cation sublattice rows, preserving the corner-shared nitrogen tetrahedra of wurtzite while introducing a doubling of the unit cell along two axes. The ordered superstructure is what makes bandgap engineering by isovalent substitution tractable: replacing Ge4+ with Sn4+ (larger, lower electronegativity) contracts the bandgap toward 1.1 eV in ZnSnN2, while replacing Ge4+ with Si4+ (smaller, higher electronegativity) widens it toward 3.5 eV in ZnSiN2. The alloy series Zn(Ge1-xSnx)N2 and Zn(Ge1-xSix)N2 therefore provides computed tuning across a 2.4 eV window in a single-phase orthorhombic host, with the parent mid-gap composition ZnGeN2 at a PBE gap of approximately 1.87 eV; since PBE underestimates nitride gaps, fit as a tandem top cell above a silicon or CIGS bottom cell awaits hybrid-functional gaps. The computational validation of the Pna2_1 structure proceeds through two independent machine-learning interatomic potentials, MACE and a second complementary potential, both of which confirm dynamic stability in converged relaxed phonon calculations: the lowest imaginary-mode-free phonon frequencies sit at +0.614 THz for ZnGeN2, +0.457 THz for ZnSnN2, and +0.851 THz for ZnSiN2. These positive values across all three anchor compositions indicate no imaginary phonon modes — the crystal does not spontaneously distort under small perturbations at the harmonic level. Additionally, ab initio molecular dynamics at 300 K was run on the structure, and no structural decomposition or cation disordering was observed over the simulation window, providing a finite-temperature check on the ground-state phonon stability. Two independent DFT source calculations corroborate the structural parameters and energetics. Taken together, the phonon and AIMD evidence places the parent and endpoint compositions of the alloy series on solid mechanistic ground. An important transparency note belongs here. A small-supercell phonon screen conducted across four compositional arms in May 2026 returned adverse results for all four. The team's assessment — and the available evidence supports it — is that these are supercell-size artifacts: small supercells in ternary nitride ordered superstructures are well-known to produce spurious imaginary modes from zone-folding and incompletely sampled force-constant matrices, and the adverse screen result contradicts the converged-supercell multi-potential stability consensus already in hand. The appropriate resolution is to complete DFPT phonon calculations on fully converged supercells, alongside HSE06 hybrid-functional calculations that will deliver the optical bandgap with the accuracy needed for direct comparison to experimental absorption spectra. Those two computations constitute the open proof gates before the full alloy claim can be considered computationally closed. The family also includes magnesium-analogue backups: MgGeN2 and MgSiN2, which carry independent two-potential phonon stability confirmation and widen the nitrogen-nitride chemistry palette toward more refractory compositions. ZrZnN2 is included as a majority-stable bandgap-widening arm. These backup members serve a dual strategic function — they deepen the claim set and provide fallback positions if any single alloy composition encounters synthesis challenges or prior-art complications. The critical distinction for device operation is that the claims are directed at PV absorber and SWIR detector applications on non-GaN-template substrates, which avoids the GaN-template LED space already occupied by III-nitride lighting manufacturers.

The addressable market for this asset spans two distinct but reinforcing verticals. In tandem photovoltaics, the top-cell absorber material market is at an inflection point driven by the impending commercial deployment of silicon-tandem modules. Perovskite-silicon tandem is the dominant research pathway, but regulatory, stability, and lead-toxicity concerns are opening space for alternative top-cell absorbers at the 1.7–1.9 eV range. The global PV market installed roughly 400 GW annually as of the mid-2020s and is on trajectories toward 1 TW per year within the decade; the materials input market for thin-film absorbers in tandem configurations is conservatively estimated in the low single-digit billions of dollars, with the specific top-cell absorber segment growing as tandem fraction of total shipments increases. The total addressable market across tandem PV and SWIR detector applications is estimated at $1–5 billion, a figure that reflects early-stage market formation rather than a mature incumbent base. In short-wave infrared detection, the incumbents are InGaAs (indium gallium arsenide) detector arrays, which require indium and are manufactured on expensive III-V substrates. A direct-gap absorber tunable across 1.0–2.5 eV on a lower-cost nitride substrate would address cost and supply-chain concerns in machine vision, LiDAR, spectroscopy, and defense applications. SWIR detector markets are smaller than PV in aggregate but carry higher unit margins and are less sensitive to cost-per-watt pressure. Both markets reward materials that are manufacturable by scalable vapor-phase deposition routes — sputtering, CVD, or MOCVD — for which the II-IV-N2 system is compatible in principle, though process optimization is not yet demonstrated. Licensing logic for this asset follows the upstream materials IP model: a foundational composition-plus-device-use claim set licenses into module makers and detector OEMs who need freedom to operate, or is acquired outright by a materials supplier seeking to control input specifications for next-generation tandem stacks. Royalty structures in thin-film PV typically range from 1–3% of cell revenue at volume, which at even modest tandem market penetration represents meaningful aggregate value. The window for securing these claims before third-party filings in the alloy space is time-sensitive, and the HSE-rescue gate closure is the near-term trigger.

earth-abundant indium/tellurium-free tandem top-cell absorber at Shockley-Queisser optimum

bandgap-tuned substituted alloy + PV/SWIR device on non-GaN-template substrate; parent composition not claimed

Preliminary screen only. This is not a freedom-to-operate opinion.

The most natural acquirers or licensees fall into two categories. The first is integrated tandem-PV manufacturers and their upstream materials suppliers — companies actively developing silicon-tandem or III-V-tandem modules who need a non-perovskite, non-toxic, indium-free top-cell absorber option. For these buyers, a composition-plus-device-use claim set covering bandgap-tunable II-IV-N2 alloys could, if granted, offer offensive value (exclusion within its granted scope) and some defensive value, though their own freedom to operate would still need counsel review. Strategic fits include thin-film PV equipment and materials suppliers, module OEMs building tandem roadmaps, and diversified semiconductor materials companies seeking to extend into the clean-energy space. The second category is SWIR detector manufacturers and the aerospace/defense prime contractors who procure focal-plane arrays. InGaAs detector supply chains have faced periodic indium-availability and export-control pressures, creating latent demand for alternative absorber materials with tunable response across the SWIR window. A company developing a direct-gap, earth-abundant SWIR absorber platform would find this IP a useful foundation for product differentiation and licensing leverage. Government-affiliated research programs (DOE, DARPA) working on next-generation photovoltaics or advanced sensing also represent non-traditional acquirers who fund IP portfolios as part of technology transition agreements.

The primary technical risk is that the HSE06 hybrid-functional bandgap calculations, when completed, return values materially different from the 1.1–3.5 eV window projected from lower-level methods and literature interpolation. If the actual HSE06 gaps are narrower — particularly if ZnSnN2 falls below 1.0 eV or ZnSiN2 falls short of 2.5 eV — the claim that the alloy series continuously spans the tandem-PV optimum weakens. The mitigation is straightforward: run HSE06 on the three anchor compositions as the immediate next computational step, assess the gap, and adjust claim language before prosecution if needed. The adverse small-supercell phonon screen is a known artifact risk in ordered superstructures and is de-risked by completing the converged-supercell DFPT calculation, which is also in the near-term work program. A second risk is synthesis demonstration: no experimental thin-film growth of the Sn-substituted alloy series has been reported in the device-relevant thickness and phase-pure form needed to support a working tandem cell, meaning the claims are composition-and-use filings ahead of experimental proof of concept. That is an appropriate position for a priority filing but does create vulnerability to a third party who demonstrates the device first with independent synthesis, potentially complicating enforceability in jurisdictions requiring utility. The roadmap to de-risk both gates — HSE06 bandgap confirmation and converged-supercell phonon closure — is well-defined and executable within a standard computational materials research program.