Benjamin Gallant1,Elisabeth Duijnstee1,Joel Smith1,Dominik Kubicki2,Harry Sansom1,Silvia Collavini3,Bernd Sturdza1,Philippe Holzhey1,Alexandra Ramadan1,Adam Wright1,Chelsea Xia1,Murali Gedda4,Santanu Saha1,Marina Filip1,Thomas Anthopoulos4,Michael Johnston1,Robin Nicholas1,Henry Snaith1
University of Oxford1,University of Warwick2,POLYMAT: Molecular and Supramolecular Materials Laboratories3,King Abdullah University of Science and Technology4
Benjamin Gallant1,Elisabeth Duijnstee1,Joel Smith1,Dominik Kubicki2,Harry Sansom1,Silvia Collavini3,Bernd Sturdza1,Philippe Holzhey1,Alexandra Ramadan1,Adam Wright1,Chelsea Xia1,Murali Gedda4,Santanu Saha1,Marina Filip1,Thomas Anthopoulos4,Michael Johnston1,Robin Nicholas1,Henry Snaith1
University of Oxford1,University of Warwick2,POLYMAT: Molecular and Supramolecular Materials Laboratories3,King Abdullah University of Science and Technology4
Despite its attractive low band gap, a key challenge for formamidinium lead triiodide (FAPbI<sub>3</sub>) is to overcome its inherent ambient phase instability. A range of chemical additives have been employed in FAPbI<sub>3</sub>-based solar cells to produce record efficiencies and enhance the stability of the photoactive FAPbI<sub>3</sub> α-phase. Here we grow FAPbI<sub>3</sub> single crystals from a solution containing one such common additive (FAPbI<sub>3</sub>-A), which has been widely employed in high-performance solution-processed perovskite solar cells. We demonstrate that FAPbI<sub>3</sub>-A crystals are stable against transformation to the photoinactive δ-phase for more than one year under ambient conditions, while FAPbI<sub>3</sub> crystals grown from a neat solution degrade within days. Critically, we reveal via nuclear magnetic resonance (NMR) spectroscopy that this common additive is unstable in solution and does not play an active stabilising role itself, instead degrading fully within minutes to a range of stable products via a complex chemical pathway, which we present. We show that FAPbI<sub>3</sub> crystals grown from a solution containing a certain one of these degradation products (FAPbI<sub>3</sub>-D) replicate the enhanced α-phase stability of FAPbI<sub>3</sub>-A. By performing a combination of liquid- and solid-state NMR measurements, along with X-ray diffraction (XRD) techniques, we find that previously unreported organic species are present in the FAPbI<sub>3</sub>-A and FAPbI<sub>3</sub>-D crystals. Remarkably, however, these correspond to neither the common additive, any of its stable degradation products nor any of the intermediates in its degradation pathway. Instead we are able to show that interruption of the degradation pathway by reaction of these intermediates with excess FA<sup>+</sup> generates alternative organic species in-situ. It is these alternative products that are incorporated into the FAPbI<sub>3</sub> bulk material and lead to α-phase stabilisation. We anticipate that presentation of our new understanding of the complex solution chemistry associated with this common high-performance additive will provide crucial insight for other researchers, immediately enabling more effective and controllable use of it, and direct use of its degradation products, in the fabrication of highly phase-stable α-FAPbI<sub>3</sub> perovskite materials.