Recently, surface passivation has been proved to be an essential approach for obtaining efficient and stable perovskite light-emitting diodes (Pero-LEDs). Phosphine oxides performed well as passivators in many reports. However, the most commonly used phosphine oxides are insulators, which may inhibit carrier transport between the perovskite emitter and charge-transporter layers, limiting the corresponding device performance. Here, 2,7-bis­(diphenylphosphoryl)-9,9′-spirobifluorene (SPPO13), a conductive molecule with two phosphine oxide functional groups, is introduced to modify the perovskite emitting layer. The bifunctional SPPO13 can passivate the nonradiative defects of perovskite and promote electron injection at the interface of perovskite emitter and electron-transporter layers. As a result, the corresponding Pero-LEDs obtain a maximum external quantum efficiency (EQE) of 22.3%. In addition, the Pero-LEDs achieve extremely high brightness with a maximum of around 190 000 cd/m2.

All‐inorganic and lead‐free CsSnI3 is emerging as one of the most promising candidates for near‐infrared perovskite light‐emitting diodes (NIR Pero‐LEDs), which find practical applications including facial recognition, biomedical apparatus, night vision camera, and Light Fidelity. However, in the CsSnI3‐based Pero‐LEDs, the holes injection is significantly higher than that of electrons, resulting in unbalanced charge injection, undesired exciton dissipation, and poor device performance. Herein, it is proposed to manage charge injection and recombination behavior by tuning the interface area of perovskite and charge‐transporter. A dendritic CsSnI3 structure is prepared on the hole‐transporter, only making a bottom contact with the hole‐transporter and exposing all other available crystal surfaces to the electron‐transporter. In other words, the interface area of perovskite/electron‐transporter is significantly higher than that of perovskite/hole‐transporter. Moreover, the embedding interface of perovskite/electron‐transporter can spatially confine holes and electrons, increasing the radiation recombination. By taking advantage of the dendritic structure, efficient lead‐free NIR Pero‐LEDs are achieved with a record external quantum efficiency (EQE) of 5.4%. More importantly, the dendritic structure shows great superiorities in flexible devices, for there is almost no morphology change after 2000‐cycles of bends, and the fabricated Pero‐LEDs can keep 93.4% of initial EQEs after 50‐cycles of bends.

Metal halide perovskite films are prepared mainly by solution-based methods. However, the preparation process is prone to produce massive defects at the interface between the perovskite emitting layer and the charge transport layers, limiting the perovskite light-emitting diode device performance. Aiming at this problem, researchers have proposed many effective strategies to passivate these interface defects. However, most previous research studies only focus on modifying the perovskite top interface, and very few reports deal with the buried interface. Here, we deposited triphenylphosphine oxide (TPPO) molecules between the perovskite and the hole transport layer (HTL) and realized the buried interface modification. Adding TPPO avoids the contact recombination of the perovskite and HTL and improves the film quality by increasing the substrate wettability. Moreover, the lone pair electrons of PO can interact with the uncoordinated lead (Pb2+) of the perovskite and passivate halogen vacancy defects, and the insulation property of TPPO helps to balance the injection of holes and electrons. As a result, a maximum external quantum efficiency (EQEmax ) of 21.01% was obtained with an average of 18.4 ± 0.9% over 30 devices, and the device reproducibility was greatly enhanced.

Highlights The modification of perovskite precursor by a series of phosphoryl chloride molecules can indeed improve the performance of perovskite LEDs (Pero-LEDs). The bis(2-oxo-3-oxazolidinyl) phosphinic chloride can not only regulate the phase distribution by controlling the crystallization rate but also passivate the defects of the quasi-2D perovskite. Highly efficient and reproducible Pero-LEDs are achieved with an maximum external quantum efficiency (EQEmax) of 20.82% and an average EQE (EQEave) of around 20% on 50 devices. Abstract Quasi-2D perovskites have attracted tremendous interest for application as light-emission layers in light-emitting diodes (LEDs). However, the heterogeneous n phase and non-uniform distribution still severely limit the further development of quasi-2D perovskite LEDs (Pero-LEDs). Meanwhile, the increased defect density caused by the reduced dimension and grain size induces non-radiative recombination and further deteriorates the device performance. Here, we found that a series of molecules containing phosphoryl chloride functional groups have noticeable enhancement effects on the device performance of quasi-2D Pero-LEDs. Then, we studied the modification mechanism by focusing on the bis(2-oxo-3-oxazolidinyl) phosphinic chloride (BOPCl). It is concluded that the BOPCl can not only regulate the phase distribution by decreasing the crystallization rate but also remain in the grain boundaries and passivate the defects. As a result, the corresponding quasi-2D Pero-LEDs obtained a maximum external quantum efficiency (EQEmax) of 20.82% and an average EQE (EQEave) of around 20% on the optimal 50 devices, proving excellent reproducibility. Our work provides a new selection of molecular types for regulating the crystallization and passivating the defects of quasi-2D perovskite films.