Perovskite is a crystal-structure family (ABX3 and its double-perovskite and layered variants) whose chemical substitutability lets one lattice motif be tuned into solar absorbers, photodetectors, quantum light emitters, catalysts and thermoelectrics; the current research frontier in these sources is not efficiency but escaping lead and taming defects.
Summary
Perovskite names a structure, not a compound. The parent ABX3 arrangement puts a large A cation in a cage of corner-sharing BX6 octahedra, and it tolerates an unusual amount of chemical substitution: the halide branch (CsPbBr3, methylammonium lead iodide, layered (PEA)2PbI4) underpins solar cells, LEDs and photodetectors, while the oxide branch (SrTiO3, EuTiO3, La0.6Sr0.4Fe1-xCoxO3, Ba2NiTeO6) delivers ferroelectrics, magnets and electrocatalysts. Ordered double perovskites (A2BB’X6) double the unit cell to accommodate two different B-site cations, opening up compositions with no lead at all.
The parameters that decide device behaviour are consistently the same across these sources. First, A/B/X chemistry sets the band gap: machine-learning models trained across A2BX6, A2BB’X6, A3B2X9 and A4BX6 halides identify B-site and X-site elemental properties as the dominant descriptors. Second, octahedral tilting and distortion matter as much as composition: in Cs2KInI6 the ideal cubic phase is dynamically unstable, and the distorted phases that are stable widen the gap, shift it from direct to indirect and flatten the band edges; under pressure, Ba2NiTeO6 crosses from rhombohedral to monoclinic with an accompanying direct-to-indirect gap change. Third, defect chemistry sets whether a good band structure translates into a good device. Bismuth-based oxide double perovskites Bi2FeCrO6 and Bi2MnCrO6 absorb strongly in the visible (10^4-10^5 cm^-1) with high carrier density but carry mixed cation valences and deep-level defects; iron doping introduces midgap 3d states that quench photoluminescence completely in CsPbCl3 while CsPbBr3 retains finite emission.
The framing in these sources is telling. Lead-halide perovskites are treated as the performance benchmark that non-toxic alternatives are trying to reach, and the stated reasons to leave them are lead toxicity and ambient instability rather than any shortfall in performance. Pnictogen chalcohalides such as BiSBr are described as promising non-toxic, stable substitutes but severely limited by carrier localisation, which is now traced to vacancy-derived defect-bound hot polarons. Beyond photovoltaics, the same lattice family is being pushed into quantum optics (room-temperature chiral superfluorescence, single-quantum-dot multiexciton spectroscopy), thermal management (ultralow thermal conductivity), oxygen-evolution catalysis and magnetocaloric cooling.
Note for the reader: the supplied sources are entirely fundamental physics and materials-science preprints. They contain no solar-cell efficiency figures, no module lifetimes, no manufacturing cost or capacity data. Several dimensions below therefore have to be scored on lab evidence only, and some cannot be scored at all.
Evidence base
- First reported observation of chiral superfluorescence at room temperature from vertically aligned chiral perovskite superlattices over more than 100 by 100 micrometres, with circular polarisation up to about 14 per cent that reverses with handedness and responds to a weak magnetic field, 5 May 2026
- Solution-deposited 350-450 nm Bi2FeCrO6 and Bi2MnCrO6 double-perovskite films show visible absorption of 10^4-10^5 cm^-1 and carrier densities of 10^17-10^20 cm^-3 but contain mixed cation valences and deep-level defects, with lead toxicity and ambient instability given as the motivation for leaving lead halides, 6 May 2026
- Cs2KInI6 has a calculated 1.94 eV direct gap but a dynamically unstable cubic phase; machine-learning-accelerated search found 42 dynamically stable structures, the most stable lying 13 meV/atom above the convex hull and lacking octahedral cation coordination, with distortions widening the gap and turning it indirect, 4 June 2026
- In BiSBr, carrier localisation is extrinsic: vacancies introduced during synthesis or post-treatment produce defect-bound hot polarons and self-trapping, the mechanism limiting non-toxic pnictogen absorbers from approaching lead-halide performance, 29 May 2026
- A one-step-fabricated lead-free K2CuBr3 photoconductor with about 10^-10 A dark current and 38.82 / 61.94 microsecond response and recovery reconstructed near-UV single-pixel images with signal-to-noise improving from 16.4 to 31.7 dB, 2 June 2026
- Symbolic-regression screening selected GuaPbI3, which was then synthesised mechanochemically and measured at room-temperature lattice thermal conductivity of about 0.088 W m^-1 K^-1 while retaining bulk electrical conduction, 15 June 2026
Open questions
- Can any lead-free perovskite or perovskite-derived absorber be shown to reach lead-halide-comparable device performance, given that the closest candidates currently fail on either dynamic stability (Cs2KInI6) or deep-level defects (Bi2FeCrO6, Bi2MnCrO6)?
- What controls the divergent doping response of CsPbCl3 versus CsPbBr3, since neither midgap electronic structure nor phonon calculations account for it, and can that mechanism be turned into a design rule?
- Can vacancy formation during synthesis and post-treatment be suppressed enough to eliminate defect-bound hot polarons in pnictogen chalcohalides, which the sources identify as extrinsic rather than intrinsic?
- Does the roughly 14 per cent circular polarisation of room-temperature chiral superfluorescence scale toward the values a quantum light source would need, and does it survive at wafer rather than hundred-micrometre scale?
Assessment drafted 2026-08-31 from up to 18 KB sources using the technology-scorecard framework; scores are a draft read pending review.