Advanced · 30 min

Surface traps in a perovskite solar film

Explore a model perovskite crystallite layer by layer: count the under-coordinated surface sites that act as recombination traps and understand why passivation raises cell efficiency.

Goal

Estimate the fraction of atoms on the grain surface and reason about how it scales with grain size — and why small-grain films lose more carriers to traps.

Apparatus and reagents

Virtual microscope: a 320-atom model grain of a halide perovskite film with surface (trap) sites coloured gold; three view modes.

Procedure

View “all atoms” and rotate: a dense crystallite whose bulk looks well-ordered.
Switch to “surface sites only”: the hollow shell left behind is where Pb and I dangling bonds live.
Estimate the surface fraction by eye (~a shell one atom deep), then compare: for a 10 nm grain vs a 1 µm grain, which has more traps per unit volume?
Choose “crystalline interior”: photogenerated electrons and holes diffuse here safely — until they reach a grain boundary or the film surface.

What to observe

Only a thin shell of atoms is under-coordinated, yet that shell hosts the traps: defects, not the bulk, limit the cell.
The interior is uniformly coordinated — perfect perovskite $\ce{ABX3}$ stacking has no dangling bonds.
Smaller grains would make the golden shell a larger share of the volume — trap density scales with surface-to-volume ratio.

Explanation

Halide perovskites like $\ce{MAPbI3}$ absorb superbly and tolerate bulk defects, but the surface terminates the lattice: under-coordinated $\ce{Pb^{2+}}$ and halide vacancies create mid-gap states where photo-excited electrons and holes recombine non-radiatively. Passivating ligands (Lewis bases such as pyridine, ammonium salts forming 2D caps) heal these sites; together with larger grains and crystal orientation control they lifted certified efficiencies from ~3 % (2009) past 26 % — rivalling silicon.