Abstract
Many of the best-performing perovskite photovoltaic devices make use of 2D/3D interfaces, which improve efficiency and stability – but it remains unclear how the conversion of 3D-to-2D perovskite occurs and how these interfaces are assembled. Here, we use in situ Grazing-Incidence Wide-Angle X-Ray Scattering to resolve 2D/3D interface formation during spin-coating. We observe progressive dimensional reduction from 3D to n = 3 → 2 → 1 when we expose (MAPbBr3)0.05(FAPbI3)0.95 perovskites to vinylbenzylammonium ligand cations. Density functional theory simulations suggest ligands incorporate sequentially into the 3D lattice, driven by phenyl ring stacking, progressively bisecting the 3D perovskite into lower-dimensional fragments to form stable interfaces. Slowing the 2D/3D transformation with higher concentrations of antisolvent yields thinner 2D layers formed conformally onto 3D grains, improving carrier extraction and device efficiency (20% 3D-only, 22% 2D/3D). Controlling this progressive dimensional reduction has potential to further improve the performance of 2D/3D perovskite photovoltaics.
Highlights
Many of the best-performing perovskite photovoltaic devices make use of 2D/3D interfaces, which improve efficiency and stability – but it remains unclear how the conversion of 3D-to2D perovskite occurs and how these interfaces are assembled
The crystallinity and quality of the interface is greatly affected by solvent: mixtures of isopropanol and nonpolar solvents like chlorobenzene or chloroform are often used, the isopropanol component can cause some redissolution of the 3D perovskite interface that is deleterious to performance[13]
The diffraction peak positions we observe for n = 1 and 2 are very similar to those we have previously measured and simulated for reduced-dimensional perovskites (RDPs) using the ligand phenethylammonium, PEA21, where we found n = 3 PEA RDPs to diffract at q = 0.22 Å
Summary
Many of the best-performing perovskite photovoltaic devices make use of 2D/3D interfaces, which improve efficiency and stability – but it remains unclear how the conversion of 3D-to2D perovskite occurs and how these interfaces are assembled. We instead use this technique to probe the transformation of the already fully-formed 3D perovskite surface into RDPs, by collecting diffraction patterns before, during, and after exposure of the 3D perovskite to a solution containing 2D ligand cations.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
