Photovoltaics converts light directly into electricity in a semiconductor absorber, and the research frontier visible in current work is not silicon but the hunt for lead-free, defect-tolerant absorbers plus light-management schemes that beat the standard detailed-balance assumptions.
Summary
A photovoltaic (PV) cell is a semiconductor absorber in which photons above the band gap create electron-hole pairs that are separated by a built-in asymmetry (usually a junction) and extracted as current. The ceiling on efficiency comes from detailed balance, the Shockley-Queisser argument that relates the non-equilibrium operating cell to an equilibrium radiative reference state, with the angular and spectral radiative channels conventionally set by the cell’s equilibrium absorptivity and emissivity. The parameters that decide a given material are therefore: band gap matched to the illuminating spectrum, absorption strength, carrier density and mobility, and the rate at which photogenerated carriers recombine, radiatively or via defects.
Most of the current literature on this concept is upstream absorber search. Lead toxicity and ambient instability in lead halide perovskites are driving work on alternatives: solution-deposited 350-450 nm films of the bismuth oxide double perovskites Bi2FeCrO6 and Bi2MnCrO6 show monoclinic P21/c structure, strong visible absorption (alpha around 10^4-10^5 cm^-1) and carrier densities of 10^17-10^20 cm^-3, but also mixed cation valences and deep-level defects. Trigonal selenium has re-emerged because its wide gap suits tandem top cells and indoor use; first-principles work across a large set of point defects finds it intrinsically defect tolerant, because non-radiative multi-phonon capture is suppressed by large lattice reorganisations and energy releases of at least 0.5 EG per event, so bulk Shockley-Read-Hall recombination cannot account for measured device losses. Adjacent work targets the charge-separation step itself: in epitaxial WS2-graphene heterostructures, a 0 degree twist gives efficient charge separation while at 30 degrees electron and hole transfer proceed at similar rates, which cancels the separation.
One result in this set cuts the other way. Yb3+-doped CsPb(Cl,Br)3 has been widely reported as a broadband quantum cutter with photoluminescence quantum yield above 100%, a candidate for improving the blue-green response of silicon cells; photon-correlation analysis shows no bunching, and instead anti-bunching under focused excitation, consistent with Auger quenching rather than quantum cutting. A reader should treat that as the calibration for how much of this frontier will survive independent measurement.
Evidence base
Open questions
- If bulk point-defect Shockley-Read-Hall recombination cannot explain the efficiency limits of trigonal selenium cells, what does: interfaces, contacts, grain boundaries or transport?
- Can the deep-level defects and mixed cation valences in solution-deposited Bi2FeCrO6 and Bi2MnCrO6 films be suppressed enough to yield a working cell, and what open-circuit voltage does that give?
- Does the luminescent angle restrictor’s predicted non-equilibrium emission profile survive an experimental test, and what measured efficiency gain does it produce over a flat-plate control?
- How many other reported photovoltaic enhancement mechanisms with high replication rates fail a photon-statistics or equivalent mechanism-level test, as Yb3+:CsPbCl3 did?
Assessment drafted 2026-08-31 from up to 18 KB sources using the technology-scorecard framework; scores are a draft read pending review.