Additive engineering with organic molecules is of critical importance for achieving high-performance perovskite optoelectronic devices. However, experimentally finding suitable additives is costly and time consuming, while conventional machine learning (ML) is difficult to predict accurately due to the limited experimental data available in this relatively new field. Here, we demonstrate a deep learning method that can predict the effectiveness of additives in perovskite light-emitting diodes (PeLEDs) with a high accuracy up to 96 % by using a small dataset of 132 molecules. This model can maximize the information of the molecules and significantly mitigate the duplicated problem that usually happened with previous models in ML for molecular screening. Very high efficiency PeLEDs with a peak external quantum efficiency up to 22.7 % can be achieved by using the predicated additive. Our work opens a new avenue for further boosting the performance of perovskite optoelectronic devices.

A major efficiency limit for solution-processed perovskite optoelectronic devices, for example light-emitting diodes, is trap-mediated non-radiative losses. Defect passivation using organic molecules has been identified as an attractive approach to tackle this issue. However, implementation of this approach has been hindered by a lack of deep understanding of how the molecular structures influence the effectiveness of passivation. We show that the so far largely ignored hydrogen bonds play a critical role in affecting the passivation. By weakening the hydrogen bonding between the passivating functional moieties and the organic cation featuring in the perovskite, we significantly enhance the interaction with defect sites and minimize non-radiative recombination losses. Consequently, we achieve exceptionally high-performance near-infrared perovskite light-emitting diodes with a record external quantum efficiency of 21.6%. In addition, our passivated perovskite light-emitting diodes maintain a high external quantum efficiency of 20.1% and a wall-plug efficiency of 11.0% at a high current density of 200 mA cm−2, making them more attractive than the most efficient organic and quantum-dot light-emitting diodes at high excitations.

The rapid development of organic solar cells (OSCs) based on non-fullerene acceptors has attracted increasing attention during the past few years, with a record power conversion efficiency of over 13% in a binary bulk heterojunction architecture. This exciting development also enables new possibilities for ternary OSCs to further enhance their efficiency and stability. This review summarizes very recent developments of ternary OSCs, with a focus on blends involving non-fullerene acceptors. We also highlight the challenges and perspectives for further development of ternary blend organic solar cells.