Metal halide perovskites for blue light emitting materials

Metal halide perovskites are considered the most promising next generation optoelectronic materials with a general chemical formula of ABX₃,^1–3 where A represents a monovalent cation, such as methylammonium (MA⁺: CH₃NH₃⁺), formamidinium [FA⁺: CH(NH₂)²⁺], or cesium (Cs⁺), B is a bivalent metal cation, for instance, lead (Pb²⁺) or tin (Sn²⁺), and X stands for halides. Perovskite materials have attracted extensive interests in a light-emitting area since 2014 due to their high photoluminescence quantum yields (PLQYs), tunable spectral wavelength, outstanding carrier transport properties, etc.^4–8 Furthermore, facile solution-processable feature facilitates perovskite light-emitting diodes applied in the display and lighting field due to their high brightness and low cost.^9,10

Red,^11–13 green,^14–16 and blue^17–21 are the primary hues of color, which means all visible colors can be generated by combining different proportions of these three primary hues. Accordingly, developing efficient green-, red-, and blue-PeLEDs (perovskite light-emitting diodes) is highly critical for display and lighting applications. Tremendous progress has been made in green- and red-PeLEDs, and the external quantum efficiencies (EQEs) have already exceeded 20%,^11,15,16 which approach theoretical limits of conversion efficiency.²² The encouraging technologies will accelerate the development of PeLEDs to commercial applications. However, blue PeLEDs still lag far behind with poor efficiencies, which significantly impedes the commercialization process of PeLEDs. In order to overcome the problem, the significant improvement of EQEs (external quantum efficiencies) for blue PeLEDs is highly demanded.

The research of blue light emitters started with the invention of inorganic phosphors (Ba, Eu)²³ and Mg₂Al₁₆O₂₇ (BAM)²⁴ in 1974. Subsequently, blue light-emitting diodes built on GaN,²⁵ organic fluorescent and phosphorescent molecules, and core–shell quantum dots^26,27 were developed. Unfortunately, those traditional blue light-emitting diodes are typically complex. For example, most organic materials are subjected to aging problems, while quantum dot materials require precise synthesis control under stringent conditions. Thus, new materials should be discovered to replace them.

Typically, blue-emissive perovskites with a tunable spectral wavelength can be obtained via the halogen regulation strategy and the quantum confinement strategy (Fig. 1).^28–32 The spectra will exhibit blueshift by increasing the Cl⁻ content of mixing Br⁻ and Cl⁻ or forming a quantum-well structure. However, multiple obstacles still exist in these strategies, hindering the development of high-efficiency devices with blue emission. For instance, the spectrum is particularly unstable in the mixed halide PeLED system, where the phase separation induced by the halide migration under electric-field is observed. The phase separation thereby results in a large electroluminescence spectral shift from the blue region to the green region.^18,32 Furthermore, the misaligned energy level, which exists between transport layers and perovskite emission layers, leads to a poor charge injection efficiency and notably deteriorates the overall device performance.^16,21 In addition, the still undesirable optoelectronic property of the reported blue-emissive perovskites is another critical problem that restrains the development of high-efficiency blue PeLEDs.¹⁵ 

Herein, we address the recent progress toward fabricating highly efficient and stable blue PeLED technologies from aspects of the halogen regulation strategy and the quantum confinement strategy (Table I). Afterward, we discuss the challenges in blue PeLEDs, such as poor charge injection, spectral instability, and high defect-states density, all of which are the obstacles to the commercial applications. Finally, we suggest the promising future directions for highly efficient and stable blue PeLEDs.

Mixed halogen engineering is an effective strategy for spectral tunability. The tunable spectral wavelength ranges from 390 nm to 790 nm, which belongs to visible light wavelength.^33,34 Kumawat et al. reported the first mixed halide perovskite, MAPb(Br_xCl_1−x)₃ (0 < x < 1), for blue PeLEDs in 2015.³¹ They explored the relationship between the bandgap and the Cl⁻ content. Tunable emission from the sky-blue³² region to the deep-blue^16,30 region was realized via increasing the ratio of Cl⁻/Br⁻ [Fig. 2(a)]. Specifically, electroluminescence at 482 nm was obtained by employing MAPbBr_1.08Cl_1.92 as the active layer [Fig. 2(b)]. However, the performance of related devices was still poor, with a brightness of 2 cd cm⁻² and a EQE of 0.0001%. Furthermore, by increasing the Cl⁻ content, Shi et al. prepared a PeLED located at 450 nm. The deep-blue PeLED demonstrated an obvious EL (electroluminescence) at a low temperature and thermally activated EL quenching at room temperature. The EL quenching at room temperature stemmed from the nonradiative recombination process.²² 

Bulk perovskite films generally possess poor surface flatness and high-density dangling-bond defects, which increases the nonradiative recombination sites.^35,36 Perovskite nanocrystals with alkyl ligands are considered potential candidates for materials with effective light emission and spectral stability. The alkyl ligands on the crystal surface can effectively passivate dangling-bond defects. The first perovskite nanocrystal based blue PeLED was pioneered by Song and co-workers. A mixed halide perovskite nanocrystal CsPb(Br_1−xCl_x)₃ (0 < x < 1) was employed as the active layer, and the champion device delivered an EQE of 0.07% with a turn-on voltage of 4.2 V.²¹ The relatively higher turn-on voltage illustrated the existence of a huge charge injection barrier for operation perspective. Fortunately, the huge charge injection barrier can be partially eliminated through ligand engineering. Pan et al. employed a halide ion pair ligand, DDAB (docosyldimethylammonium bromide chloride), instead of conventional OAm (oleylamine) and oleic acid ligands. This ligand exchange process included an intermediate step to desorb protonated OAm, and then the ensuing DDAB effectively passivated the surface defects. As a result, the best-performing device turned on at 3.0 V with an improved EQE of 1.9%. The lower turn-on voltage indicated the improvement of charge injection after optimization.³⁷ However, the ligand exchange process deteriorated the perovskite nanocrystals’ crystallinity and induced rearrangement of crystals, leading to a poor film morphology with massive cracks and defects. Seeking to overcome the issue above, Li et al. replaced the long alkylchain ligands with the short alkylchain ligands named TMA (trimethylaluminum). After treating with TMA, the coated perovskite nanocrystals exhibited a strong bonding effect with ligands, as shown in Fourier transform infrared (FTIR) spectra [Fig. 2(c)]. TMA coordinated and reacted with the carboxylate and amino groups of the ligands, adjacent to the nanocrystals. The nanocrystals’ quality was enhanced as shown in TEM (Transmission Electron Microscope) images [Fig. 2(d)].³⁸ 

To date, most three-dimensional blue PeLEDs were created based on the halogen regulation strategy. The halogen regulation strategy offers the possibility to precisely tune the spectral wavelength. Nevertheless, phase segregation, originating from migration of mixed halogen, leads to large spectral shift. Therefore, other strategies are desirable for achieving efficient and stable blue PeLEDs.

Reduced-dimensional perovskites are widely utilized in various optoelectronic devices due to their better material stability than traditional 3D (three-dimensional) perovskites. The general chemical formula of reduced-dimensional perovskites is (RNH₃)₂A_n−1B_nX_3n+1, where R represents large organic cations³⁹ and ⟨n⟩ represents the number of stacking PbI_x octahedral. According to material properties, reduced-dimensional perovskites can be classified as 2D (two-dimensional) perovskites (⟨n⟩ = 1), expressed as (RNH₃)₂BX₄, and quasi-2D perovskites once ⟨n⟩ > 1. In the reduced-dimensional perovskite structure, the perovskite octahedral layer with high-dielectric constant is sandwiched between organic large cations with low-dielectric constant, which is called quantum-wells.

Liang et al. explored the first 2D blue PeLED at room temperature in 2016.⁴⁰ PEA (phenylethylammonium) was introduced as the spacer of (PEA)₂PbBr₄ perovskites. The 2D material naturally formed a quantum-wells structure where the excitons were tightly confined. Device architecture of ITO/PEDOT:PSS/(PEA)₂PbBr₄/TPBi [1,3,5-Tris(1-phenyl-1H-benzimidazol-2-yl)benzene]/Ca/Al was adopted here, and the obtained PeLED delivered a purple-blue electroluminescence located at 410 nm. However, due to the strong phonon scattering and the strong exciton–phonon coupling, the device exhibited an EQE as low as 0.04%.

Quasi-2D perovskites are considered excellent light emitting materials owing to their outstanding optical properties.⁴¹ Their exciton binding energy can be tuned by regulating the thickness of “quantum-wells” [Fig. 3(a)]. Efficient energy transfers between different ⟨n⟩-values species were observed in quasi-2D perovskite films [Fig. 3(b)]. The perovskite films consisted of a series of grains with different bandgaps, where the photo-carriers spontaneously concentrated into the lowest-bandgap species. Excitons in low ⟨n⟩-value QWs (quantum-wells) were efficiently energy-transferred to high ⟨n⟩-value QWs, which resulted in a high PLQY.⁸ Quasi-2D perovskites have achieved great success in green- and red-PeLED applications.^12,16 The structure and material engineering, thus, provided attractive approaches in realizing quasi-2D blue PeLEDs. Wang et al. added EABr (ethyl ammonium bromide) as large organic amine cations into MAPbBr₃, and then the layered perovskite structure could be formed. An energy-funneling process among the layered perovskite structure was observed. The highest EQE of 2.6% was obtained at 485 nm when the molar ratio of MAPbBr₃ and EABr was 1:1.3.⁴² 

Introducing a large amount of organic amine cations can generate low ⟨n⟩-value perovskite phases, which delivers a strong quantum confinement effect and a blueshift of emission. However, as the formation energy of low ⟨n⟩-value perovskite phases is small, an excessive amount of organic amine cations also causes inhomogeneous phase distribution. The increased low ⟨n⟩-value phases (⟨n⟩ = 1, ⟨n⟩ = 2) led to incomplete energy transfer, resulting in multiple emission peaks in the spectrum. Therefore, it is essential to modulate the ⟨n⟩-value distribution in quasi-2D perovskites. A careful selection of organic cations played an important role in modulating the ⟨n⟩-value distribution. Xing et al. reported a mixed cation, IPA (iso-propylammonium), and PEA, based quasi-2D perovskite (IPA/PEA)₂(MA/Cs)_n−1Pb_nBr_3n+1. The alloyed perovskite reduced the strong van der Waals interaction between stacked bulky organic cations, making the formation of ⟨n⟩ = 1 perovskites less favorable. Slower crystallization of ⟨n⟩ = 1 phases enhanced the purity and the monodispersity of ⟨n⟩ = 2, 3, and 4 phases. After a precise regulation of the ratio of PEA and IPA, undesired ⟨n⟩ = 1 2D perovskites were decreased while desired ⟨n⟩ = 2, 3, and 4 phases were increased. The desirable ⟨n⟩-value distribution inhibited phase separation in the sky-blue region. Ultimately, the corresponding device exhibited an EQE of 1.9% with a brightness of 2480 cd m⁻² at 490 nm.³² Similarly, Yuan et al. reported a quasi-2D perovskite mixing PEA and P-PDA [1,4-bis (aminomethyl) benzene] [Fig. 3(c)]. In this reduced-dimensional perovskite system, the formation of ⟨n⟩ = 1 phases was significantly suppressed and ⟨n⟩ = 3 domain was remarkably promoted, reducing the nonradiative recombination rates. The resulting LED exhibited an EL peak located at 465 nm with FWHM (full width at half maxima) as narrow as 25 nm. The maximum EQE of 2.6% represented one of the highest efficiencies for deep-blue PeLEDs.⁴³ Moreover, this group also obtained a stable quasi-2D blue PeLED by combining PEABr and NPABr [N-(2-bromoethyl)-1,3-propanediamine bromine].⁴⁴ 

Apart from the regulation of ⟨n⟩-value distribution, chloride-doping is a facile strategy. Vashishtha et al. tuned the spectral spanning sky-blue region (480–500 nm) to the pure-blue region (460–480 nm) via chloride-doping.²⁹ Li et al. also fabricated a high-efficiency blue PeLED using the mixed halide quasi-2D perovskite system. Besides, the authors modulated the recombination zone by adjusting the thickness of the hole transport layer, delivering a balanced hole–electron injection [Fig. 3(d)]. The corresponding device reached an EQE of 5.7% with a remarkable brightness of 3780 cd m⁻².⁴⁵ Furthermore, a new record EQE of 11.0% was obtained in the sky-blue region. Wang et al. reported a quasi-2D perovskite composed of PEACl:CsPbBr₃ (1:1) [Fig. 3(e)]. The alloyed films exhibited higher PLQY (photoluminescence quantum yield) of 19.8% than unoptimized films. The authors also added a small amount of YCl₃ to passivate defects on the grain boundaries, which further increased the PLQY up to 49.7% [Fig. 3(f)].⁴⁶ 

The quantum size effect also makes an important impact in reduced-dimensional perovskites. When the size of particles drops below the exciton Bohr diameter, the quantum size effect widens the bandgap due to the discrete energy levels.^47,48 The exciton Bohr diameter of CsPbBr₃ is about 7 nm, indicating that the small-sized quantum dot has the potential for blue light emission.⁴⁹ Yang et al. successfully manufactured the first blue perovskite nanoparticle LED by using ultra-thin CsPbBr₃ nanoparticles as the active layer [Fig. 4(a)]. Due to the strong quantum size effect, the LED showed the brightness of 25 cd m⁻² at 480 nm, with a maximum EQE of ∼0.1%.⁵⁰ Liu et al. reported a quantum-confined quasi-2D perovskite nanoparticle by embedding small-sized nanoparticles into quasi-2D perovskites. The energy-funneling process between the quasi-2D perovskites and the nanoparticles effectively suppressed nonradiative recombination. Moreover, using ethyl acetate as antisolvent facilitated the charge injection. As a result, the device based on ITO/NiOx/TFB/PVK/4-PBA (phenylbutylamine)-FACsPbBr₃/TPBi/LiF/Al delivered an EL peak located at 483 nm, accompanied by a high EQE of 9.5% [Figs. 4(b)–4(d)].⁵¹ 

The efficiency of blue PeLEDs is still very low, compared with green- and red-PeLEDs. One of the important factors limiting their efficiency is the serious charge injection issue. The theoretical minimum turn-on voltage of blue PeLEDs is larger than red and green analogs, which is accounted for the wider bandgap of the emission layer. Due to the wider bandgap, the conduction band and the valence band of blue emissive perovskite are likely to be higher and lower (Fig. 5). The mismatched energy level between the emission layer and the charge transport layer generally results in a large charge injection barrier (Fig. 5). It is tough to inject charges over the barrier in blue PeLEDs, and excessive charge accumulation reduces the device efficiency. Thus, it is pivotal to improve the device efficiency by minimizing the charge injection barrier.

The interface modification strategy was employed to achieve energy level matching. To improve the hole injection efficiency, Gangishetty et al. inserted a thin composite layer, TFB [poly (9,9-dioctyl-fluorene-co-N-4-butylphenyl) diphenyl]/PFI (perfluorinated ionomer), onto PEDOT:PSS film. TFB possessed a deeper highest occupied molecular orbital (−5.3 eV) than PEDOT:PSS (−5.2 eV), facilitating the hole injection. Moreover, PFI was considered to form a band-bending effect at the hole transport layer/perovskite interface due to the strong surface dipole effect. The energy level matching minimized emission quenching at the interface between the HTL (hole transport layer) and the perovskite layer. Benefiting from the improved hole injection, EQE of the corresponding device was enhanced from 0.03% to 0.5%.⁵² PVK [poly(N-vinylcarbazole)] was also used as an interface material between HTL and perovskite. Due to the deep HOMO (Highest Occupied Molecular Orbital) and shallow LUMO (Lowest Unoccupied Molecular Orbital), PVK not only facilitated hole transport but also blocked electron transport. As a result, holes and electrons could effectively recombine in the QD (quantum dot) emitting layers, significantly improving device’s EQE.²¹ Furthermore, Chen et al. employed TmPyPB (1,3,5-tri[(3-pyridinyl)-benzene-3-base] benzene) as ETL (electron transport layer) in their devices. TmPyPB (−2.7 eV) possessed a shallower LUMO than TPBi [1,3,5-tris(2-N-phenylbenzimidazolyl) benzene, ∼−2.8 eV], which effectively improved electron injection. Moreover, the deeper HOMO (−6.7 eV) could block the holes and suppress nonradiative recombination. Ultimately, an improved EQE of 3.6% was obtained in blue PeLED (∼505 nm), accompanied by a maximum brightness of 7320 cd m⁻².⁵³ Similar strategies were utilized in green PeLEDs, where PEI (polyetherimide) or PEIE (polyethylenimine ethoxylated) worked as the interfacial layer, to adjust the work function of transport layer and suppress interfacial quenching.^16,54–56

In addition to enhancement of hole and electron injection, the balance between hole and electron injection also plays an important role in achieving efficient blue PeLEDs. The carrier mobility of hole and electron transport layers, as well as the electrical performance, has a great influence on the balance. Excessive electrons or holes, accumulated at the interface between the transport layer and the perovskite layer, increase the decay pathways of nonradiative recombination via defect-assisted or Auger.^57,58 This phenomenon has been discovered in green PeLEDs. PVP (polyvinyl pyrrolidone) was inserted at the interface between ZnO and perovskite.⁵⁹ The corresponding device achieved an improved charge injection balance and an enhanced efficiency. A similar approach has been applied in blue PeLEDs. In order to obtain the balance between hole and electron injection, Li et al.⁴⁵ controlled the hole injection rate by changing the thickness of the hole transport layer. On the one hand, the injected holes would take a relatively long time to pass through the thick PEDOT:PSS layer. On the other hand, ultra-thin PEDOT:PSS film would contribute to a poor film morphology, which may allow the perovskite crystals to directly get contact with the bottom electrode, and therefore led to a severe leakage current in the PeLED. The precise modulation of PEDOT:PSS thickness (15 nm) ensured that holes and electrons have almost the same injection time. As a result, radiative recombination centers would be located at the perovskite layer rather than at the interface. The resulting device achieved a prominent EQE of 5.7%.

The spectral instability in mixed halide perovskites was observed by Li and co-workers. For CsPb(Cl/Br)₃ perovskite nanocrystal LEDs, red-shifted EL peak led to the movement of spectra from the blue region to the green region under a constant bias voltage.³⁸ In the reduced-dimensional perovskites system, it has been reported that the spectrum can be stabilized by the quantum confinement effect.⁶⁰ This method has been demonstrated in green and red regions. Xiao et al. added BA (butyl amine) into the precursor to naturally form reduced-dimensional perovskites. Owing to the suppressed ion migration by the layered structure, no obvious change of spectra was found after the devices were stored in a nitrogen-filled glovebox for exceeding eight months.¹² For blue PeLEDs, the spectrum could be stabilized by the mixed cation strategy. When IPA and PEA cations were simultaneously used, the growth of lowest ⟨n⟩-value phases and highest ⟨n⟩-value phases was inhibited. Thus, the concentrated ⟨n⟩-values distribution was obtained after a precise ratio modulation. Substantially, a pure-blue emission with narrow FWHM was achieved due to the suppressed phase segregation. The PL (photoluminescence) position had no shift under 325 nm irradiation (7 W cm⁻²) for 1 h.³² 

Organic molecules are highly sensitive to temperature, oxygen, and humidity;⁶¹ thus, inorganic cations are considered to improve the spectral stability. Cs-based perovskites were investigated due to their high thermal stability (stable up to 250 °C).⁶² To improve spectral stability, Kim et al. employed the triple cation, MA/Cs/FA, to stabilize the perovskite crystal structure.⁶³ Additionally, the Rb cation was also introduced into the Cs-based emission layer, and the alloyed PEA₂(Rb_xCs_1−x)_n−1Pb_nBr_3n+1 (0 < x < 1) perovskite avoided Ostwald ripening caused by Joule heating. As a result, no PL position shifting or broadening was observed in the device even after long-time thermal annealing, which demonstrated an excellent spectral stability.¹⁸ 

Defects on the surface of perovskites and point defects in perovskite crystal lattice are inevitable. Nonradiative recombination sites are formed by those defects, reducing device stability.^63–65 The surface defects caused the self-quenching effect in perovskite nanocrystals, leading to a low PLQY.^62,66–68 Moreover, emission from defect states of blue PeLEDs can be green or red, affecting the color purity. Considerable efforts have been made on reducing surface defects through the passivation strategy. Mn²⁺ was demonstrated to decrease nonradiative recombination by defect passivation.^69,70 Hou et al. prepared CsPb(Br_1−xCl_x)₃ shaped nanowires by incorporating a small amount of Mn²⁺. Reduced nonradiative recombination delivered an efficient energy transfer process. This passivation strategy significantly increased the PLQY of nanocrystal films from 9%⁵² to 28%.⁷¹ However, an excessive amount of Mn²⁺ resulted in a characteristic Mn phosphorescence peak,⁷² which influenced the color purity of the PL spectrum. Similarly, Mondal et al. reported a Cd²⁺-doping CsPbCl₃ perovskite. The nonradiative carrier trapping centers could be effectively suppressed by defect passivation, while introduction of shallow energy levels would facilitate the radiative recombination process. The resulting film achieved an increased quantum yield from 3% to 98% due to the elimination of defect states.⁷³ 

Lee et al. reported that the organic small molecule, EDA (ethylenediamine), could be used to passivate defects.⁷⁴ Due to the reduced nonradiative recombination, the resulting device achieved an outstanding performance. In particular, the encapsulated device exhibited a slight decrement of brightness even after 4 h of continuous illumination. Additionally, organic or inorganic ligands, such as TOPO (trioctylphosphine oxide),⁷⁵ PPA (3-phenyl-2-propen-1-amine),⁷⁶ and ZnX₂,⁷⁷ were applied to reduce surface defect-states density and suppress nonradiative recombination, leading to higher PLQYs. Moreover, additives such as PEO (polyethylene oxide) were demonstrated to passivate defect states on grain boundaries of the films.^78–80 

In conclusion, based on the halogen regulation strategy and the quantum confinement strategy, the performance of blue PeLEDs has been greatly enhanced. However, poor charge injection, spectral instability, and high defect-states density are still important challenges, impeding the commercialization of blue PeLEDs.

Recent reports about decreasing the charge injection barrier are certainly encouraging, but such materials need to be married with enhanced carrier injection efficiency to achieve improved device performance. (1) Functionalizing intermediate layers, such as PMMA (polymethyl methacrylate),¹⁵ as the tunneling layer is to control the electron injection. (2) Tunneling the band alignment and conductivity of HTL, PEDOT:PSS 8000 as an example, can be used to make a remedy for hole injection. (3) Discovering new HTL materials is also a method. Up to now, the first two methods were well developed in PeLEDs and successfully improved the devices’ EQE. These strategies partially alleviated the imbalance of electron and hole injection. Thus, new HTL materials are urgently desired to facilitate the hole injection.

Spectral instability is also an urgent challenge required for mixed halide blue PeLEDs. (1) The mechanism of ion migration under high build-in voltage is yet to be understood in terms of complex structure. (2) New additives should be designed to decrease the formation energy of ⟨n⟩-values, leading to a constrained distribution of ⟨n⟩-values. (3) A proper mixed cation strategy also shows great potential in stabilizing the mixed halide perovskites. (4) A suitable dimensional strategy needs to be explored to stabilize the material.

Therefore, as future directions, more efforts are still needed to fabricate perovskite nanocrystal films for obtaining great properties. (1) The perovskite nanocrystal films have an exceedingly fast crystallization process, leading to an uneven growth of components. This is not conducive to produce a high quality perovskite film with uniform crystal sizes. Thus, ion doping can be an effective means to slow down the crystallization process by modulation of crystallization dynamics, delivering a flatter film with less pin-hole. (2) Plenty of surface ligands get lost easily during fabrication in perovskite nanocrystal LEDs, resulting in agglomeration of crystals. Thus, an obvious leakage current appeared under operation. To decline the leakage current, it is particularly crucial to design ligands of strong bonding energy with the perovskite crystal structure.

Lead-free perovskites are proper candidates for traditional perovskite materials. Pb²⁺, as a cause of trap-induced nonradiative recombination, is detrimental to the environment. Thus, it is necessary to find non-toxic replacement for Pb²⁺. Accordingly, Sn²⁺ was utilized to replace Pb²⁺. Sn²⁺ species were susceptible to oxidation, so they were not good candidates for lead-free perovskites. It takes a long period to find compatible candidates for Pb²⁺ in the perovskite crystal structure without significant perturbation. Photonic engineering, such as patterning the substrate and the lens, could be applied to enhance out-coupling efficiency. The overall EQE with lens patterning could be double of the unapplied devices. A green PeLED was reported with an EQE of 28.2% through this method. In addition, as technology of perovskite solar cell matures, it is wise to expect that those novel strategies to improve perovskite solar cell PCE (power conversion efficiency) could be transferred to the application of LEDs. Furthermore, the operational stability is a main concern unsolved for the community of PeLEDs.

In short, through a variety of rational approaches, the EQE of blue perovskite LEDs illustrates a great potential in the future.

The authors acknowledge financial support from the National Natural Science Foundation of China (Grant Nos. 21771114, 51972175, and 91956130) and the Natural Science Foundation of Tianjin (Grant Nos. 17JCYBJC40900 and 18YFZCGX00580). M. Yuan acknowledges financial support from the Recruitment Program of Global Youth Experts of China.