Intermediate-band photovoltaics (PV) rely on electronic states that are bound and shielded from thermal effects, yet still allow enough coupling between dots and proper energy separation. We analyse 495 tensile-strained Si/relaxed Si 1-x Ge x quantum-dot superlattices using a 3D BenDaniel–Duke Hamiltonian, incorporating a Δ 2 -valley electron mass tensor of diag(0.19, 0.19, 0.916) m 0 and applying Bloch-periodic boundary conditions. The design variables are the cubic-dot side a = 3.0–8.0 nm, barrier thickness b = 1.0–5.0 nm, and Ge fraction x = 0.20–0.38. The screen retains 171 core geometries and 170 strict geometries. At b ≤ 1.5 nm, all 110 geometries fail the 5 meV isolation threshold; 22 also lose full binding. The retained region shifts towards x = 0.35–0.38. Heavy-hole levels are combined with the electron minibands to define the VB → IB, IB → CB, and direct VB → CB thresholds. The ASTM G173-03 (2020) AM1.5G reference spectrum is partitioned into non-overlapping photon-energy windows at the bottom, centre, and top of each finite-width miniband while enforcing E 1 + E 2 = E g eff . For the 170 strict geometries, the median matched two-step photon fluxes are 2.910, 2.786, and 2.681 × 10 2 ⁰ m −2 s −1 at the respective placements, with 31 geometries showing variations exceeding 20% between them. Given that E 2 < ΔEc ≤ 0.2432 eV, the valence-band-to-intermediate-band transition window is inherently narrow, and the ASTM spectral integration confirms this pathway as the limiting factor for two-step photon absorption across all strict configurations. These values represent upper limits based on incident light, not predictions of device efficiency or carrier escape rates.
No takes yet. Share an insight, caveat, or question.
Lassina et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: