Acid‐etching induced metal cation competitive lattice occupancy of perovskite quantum dots for efficient pure‐blue QLEDs

Low efficiency and spectral instability caused by the surface defects have been considerable issues for the mixed‐halogen blue emitting perovskite quantum dots light‐emitting diodes (PeQLEDs). Here, an in situ surface passivation to perovskite quantum dots (PeQDs) is realized by introducing the metal cations competitive lattice occupancy assisted with acid‐etching, in which the long‐chain, insulating and weakly bond surface ligands are removed by addition of octanoic acid (OTAC). Meanwhile, the dissolved A‐site cations (Na+) compete with the protonated oleyl amine and are subsequently anchored to the surface vacancies. The preadded lead bromide, acting as inorganic ligands, demonstrates strong bonding to the uncoordinated surface ions. The as‐synthesized PeQDs show the boosted photoluminescence quantum yield (PLQY) and superior stability with longer lifetime. As a result, the PeQLEDs (470 nm) based on the OTAC‐Na PeQDs exhibit an external quantum efficiency of 8.42% in the mixed halogen PeQDs (CsPb(BrxCl1−x)3). Moreover, the device exhibits superior spectra stability with negligible shift. Our competition mechanism in combination with in situ passivation strategy paves a new way for improving the performance of blue PeQLEDs.

with the protonated oleyl amine and are subsequently anchored to the surface vacancies.The preadded lead bromide, acting as inorganic ligands, demonstrates strong bonding to the uncoordinated surface ions.The as-synthesized PeQDs show the boosted photoluminescence quantum yield (PLQY) and superior stability with longer lifetime.As a result, the PeQLEDs (470 nm) based on the OTAC-Na PeQDs exhibit an external quantum efficiency of 8.42% in the mixed halogen PeQDs (CsPb(Br x Cl 1−x ) 3 ).Moreover, the device exhibits superior spectra stability with negligible shift.Our competition mechanism in combination with in situ passivation strategy paves a new way for improving the performance of blue PeQLEDs.

K E Y W O R D S
acid etching, blue PeQLEDs, high-efficiency, in situ passivation, spectral stability

| INTRODUCTION
3][4][5][6][7][8] Remarkable achievements have been made in the fields of solar cells, [9][10][11][12][13] light-emitting diodes (LEDs), [14][15][16][17][18] photodetectors, [19] and lasers [20] in the past decade.23] However, the performance of PeQLEDs for blue emission is still far behind the achievement by state-of-the-art green/red emission devices. [22]Early efforts to design blue PeQLEDs in an emission range of 475-490 nm have achieved impressive progress with an EQE of ~14%, [15,18,[24][25] while the pure-blue PeQLEDs (460-470 nm) exhibit inferior performance with an EQE of ~6%.28][29] In general, the blue emission of PeQDs was generated by optimizing the ratio of the mixed halogen ions (Br/Cl), which suffers from the undesirable halide segregation, resulting in the poor spectral stability in devices. [16,30,31][37] To address these matters, the method that long-chain ligands (oleylamine [OAm] and oleic acid [OA]) removed by alternative ligands has been explored to acquire the intact surface with ultralow trap density by surface engineering.For instance, double-chain ligand, didodecyldimethylammonium bromide (DDAB), is proposed to passivate the surface defect of PeQDs because of the strong affinity between DDA + cation and surface sites. [38]Zwitterionic ligands is also employed as surface passivation materials due to the strong binding to PeQDs surface. [39]In addition, inorganic ligands (i.e., PbBr 2 , ZnBr 2 ) are introduced to replace the insulting organic ligands, demonstrating the enhanced photoluminescence quantum yield (PLQY) and carrier transport behavior. [15,40]However, the excessive removal of surface capping ligands will possibly affect the surface morphology and luminescence of PeQDs. [41]Therefore, it is more preferable to develop the surface chemistry of PeQDs to avoid the undesirable surface damage in the fabrication of high quality of PeQLEDs.
Herein, we introduced a competition mechanism to control the dynamic equilibrium of ligands by acid etching to construct the in situ surface passivation in pure-blue PeQDs.Octanoic acid (OTAC) is employed to remove excessive carboxylate ligands, which contributes to a neat surface in PeQDs.Simultaneously, sodium carbonate (Na 2 CO 3 ) dissolved in OTAC and preadded PbBr 2 provide the free metal cations (Na + and Pb 2+ ) and bromine ions, which are expected to spontaneously anchor to the surface vacancies that are left by ligand detachment, constructing the in situ passivation to the defect state.In combination with the density functional theory (DFT) calculation, we further reveal that the Na + cations show higher adsorption energy at the surface vacancies compared to the oleylamine cations (OAm + ), which effectively avoids the phase transition caused by the addition of acid.Besides, the anchored cations exhibit a passivating effect on the surface trap states, leading to the enhanced PLQY.The spectroscopy characterizations indicate that long chain ligands (OA and OAm) have been replaced by the inorganic ligands (PbBr 2 ), thus facilitating carrier extraction and transport.The resultant PeQLEDs (470 nm) exhibit a champion EQE of 8.42% with the Commission Internationale del'Eclairage (CIE) coordinate of (0.13, 0.08).Importantly, the device based on OTAC-Na PeQDs demonstrate the superior stability of electroluminescence spectra with negligible shift under working condition.

| RESULTS AND DISCUSSION
The blue-emitting PeQDs of CsPb(Br x Cl 1−x ) 3 were synthesized by hot-injection as illustrated in Figure S1.PbBr 2 stock solution was dropped into the PeQDs solution in advance.Subsequently, Na 2 CO 3 solution was completely dissolved in OTAC-Na followed by dropping it to induce the in situ passivation process by OTAC-Na etching.The ethyl acetate (EtOAc) was added to precipitate the PeQDs, and the as-synthesized PeQDs were dispersed in n-hexane X-ray photoelectron spectroscopy (XPS) and Fourier transform infrared spectroscopy (FTIR) were employed to investigate the evolution of surface ligands in PeQDs.The vibrations at 2921 and 2852 cm −1 represent for the sharp decrease of CH 2 group, accompany with the decreased functional group intensity of the OA and OAm (v N-H = 1642 cm −1 , v C=O = 1712 cm −1 , v COO − = 1542 cm −1 ), as shown in Figure 1A.The significant decrease of the relative intensity of organic group peaks are observed after the post treatment of OTAC-Na, which indicates the bulky long organic ligands have been mostly removed in the process of acid etching.Compared to the XPS spectra of Pb 4f, Cs 3d, and Br 3d in pristine PeQDs (Figure 1B and Figure S2), the peaks in OTAC-Na etched PeQDs slightly shift toward the higher binding energy due to the ligand detachment and the in situ formation of strong binding between the free ions and PeQDs surface.The appearance of Na 1s in  The structure evolution in the process of OTAC-Na etching was investigated.In comparison with the pristine PeQDs, the OTAC-Na PeQDs maintain the cubic shape with an increased size from 7.89 to 9.80 nm (Figure S4), which indicates the phase transition is refrained by the addition of organic acid (Figure 1D,E).It is worth noting that the shedding of surface ligands in PeQDs provides a pathway for preadded PbBr 2 to achieve more anion exchange in the process of etching, in which the larger ionic radius of Br ions replaces Cl ions, resulting in the increase in PeQDs' size.Moreover, the high-resolution transmission electron microscopy (TEM) images exhibit an interplanar spacing of 0.34 nm, which is in accordance with the X-ray diffraction (XRD) patterns (Figure 1F), suggesting the negligible impact on the phase structure of PeQDs in relation to in situ etching processes.This improvement in the phase stability against the acid etching is attributed to the in situ compensation of the cation and PbBr 2 groups.
It is noted that the dynamic equilibrium in the dispersed solution system of PeQDs are used to evaluate the ligand exchange, as shown in Figure 1G.Generally, the addition of acid is expected to break the equilibrium between the surface and free ligands by depleting the dissociated OAm, thereby accelerating the chemical reaction in the direction of desorption of the weakly bonded surface ligands. [42]Subsequently, the product of excess OAm + ions is expected to take the place of cesium ions. [43]The intercalation of long-chain molecules leads to a phase transition from cubic structure to nanoplates, accompanied by the disappearance of the luminesce property. [44]The characterizations of the sample treated only with OTAC also prove the damage to PeQDs.As seen in Figure S5, the cubic phase gradually transforms to the nanosheet structure in TEM image (Figure S5a), which indicates that organic acid can essentially deteriorate the phase stability of PeQDs.Moreover, the appearance of two absorption peaks in Figure S5b indicates the coexistence of cubic and nanosheet phases, which is consistent with the phase transformation in Figure S5a.In comparison with the pristine PeQDs, the solution feature of OTAC PeQDs changes to the turbid after the addition of OTAC accompanying with the luminance quenching (Figure S5c), which is attributed to the poor dispersion caused by the OTAC assisting ligand detachment.
In our approach, we introduced sodium ions dissolved in OTAC that competed with OAm + to interact with the undercoordinated surface atoms during the etching process, which subtly removed the long-chain surface ligands and maintained the cubic phase.The complete removal of ligands was achieved without destroying the structure of PeQDs.In addition, inorganic ligand (PbBr 2 ) was employed to anchor surface vacancies and further improve the luminesce and carrier transport properties, as shown in Figure 1H.
The DFT calculations were also introduced to further illustrate the competitive mechanism between metal cations and OAm + .The models of Cs vacancy (V Cs ) along the (001) surface for OAm and Na adsorption are displayed in Figure 2A,B.The calculated adsorption energy is 0.10 and 2.98 eV for OAm and Na adsorption model, respectively.The results show that sodium ions are more favorably filled into the surface vacancies, which can prevent the phase transformation arising from OAm + substitution.In addition, the surface vacancies originating from ligand detachment can deteriorate the optical properties of PeQDs and the performance of devices.Figure S6 displays the models of the introduced V Cs and Na filling into the vacant space with the corresponding density of state calculated.Generally, the trap states created by the surface point defects of V Cs lead to the nonradiative recombination based on the density of state (DOS) calculation in Figure 2C.In contrast, the unfavorable trap states within band gap are removed after the compensation with Na ions in Figure 2D, exhibiting a great passivation effect on the surface defects.
The density of trap states in PeQDs was characterized based on the space charge limit current (SCLC) method by constructing electron-only and hole-only devices, which can be evaluated by the following equation [45,46] : where L is the thickness of PeQD film.ε 0 and ε are the vacuum and relative permittivity, respectively.e is the elementary charge.V TFL is the trap-filled limit voltage.The reduced electron/hole trap densities of OTAC-Na etched samples (4.85 × 10 17 /2.36× 10 17 cm −3 ) compared with the pristine case (9.63 × 10 17 /5.45× 10 17 cm −3 ) in Figure 2E,F indicate that the strong binding of inorganic ions produced by the in situ etching and passivation approach can greatly avoid the ligand detachment during the film fabrication and effectively improve the optical and carrier transport properties of films.
To gain further insights into the in situ passivation, optical properties were investigated.Red-shift of the photoluminescence (PL) spectra is observed after OTAC-Na etching process in Figure 3A.This shifting behavior demonstrates that the free PbBr 2 subsequently adhere to PeQDs when the surface bulky ligands are etched, facilitating the ion exchange processes.The PLQY are enhanced from 14% to 72% for pristine PeQDs and OTAC-Na PeQDs (Figure 3B), respectively.Time-resolved photoluminescence (TRPL) measurements in Figure 3C also exhibit the enhanced lifetime (Table S1), indicating that the carriers injected into PeQDs are robust to surface traps, a benefit of in situ passivation.
In addition, we employed the temperature-dependent PL spectra to investigate the exciton dynamics from 60 K to 300 K (Figure 3D,G).As the temperature increases, the intensities of peaks gradually decrease, accompanied with a blue shift behavior in PL spectra.The exciton binding energy can be fitted by the following equation: [18]  excitons dissociation into free carriers without radiation, which is beneficial for the performance of LEDs.The enhancement of exciton binding energy means the increased probability of radiative recombination, which leads to the enhanced optical properties.In the high temperature range of 300-350 K, PL intensities of the OTAC-Na etched PeQDs decrease slowly, while the pristine PeQDs present negligible photoluminescence, which is attributed to the enhanced stability due to the strong bonding between inorganic ligands and PeQD surface.Moreover, the zeta potential analysis exhibits an improvement in ink stability after OTAC-Na etching process, in which the zeta potential increases from 0.92 to 5.41 mV (Figure S7).Atomic force microscopy (AFM) and scanning electron microscopy (SEM) were employed to reveal the quality of films processed by PeQDs ink.Compared to the pristine PeQDs films, OTAC-Na PeQD films show uniform, dense, crack-free morphology as well as smooth surfaces as seen in Figure S8.The root-meansquare (RMS) roughness of pristine and OTAC-Na PeQD films are 2.57 and 1.46 nm, respectively.Moreover, SEM images also show the enhanced film formation, which can be attributed to the removal of long, weak ligands and the strong binding of inorganic ions (Figure S9).
Interestingly, we also performed the OTAC-Cs etching process, and obtained the enhanced optical properties, in which the high concentrations of Cs + keep Cs + from being replaced by OAm + .Based on the in situ passivation of OTAC etching process, perovskite QLEDs were constructed with a multilayer structure of ITO/PEDOT:PSS/poly-TPD/Na + , Cs + -PeQDs/TPBi/LiF/ Al (Figure 4A).Typical cross-sectional SEM image and the corresponding energy diagram of device are shown in Figures S10-S11.A symmetric pure-blue electroluminescence (EL) peak at nm can be observed in Figure 4B with a full width at half-maximum (FWHM) of 20 and 21 nm for Na +and Cs + -PeQDs, respectively, which corresponds to CIE coordinates of (0.13, 0.08), satisfying the new Rec.2020 standard (Figure 4C).The high purity emission is ascribed to the homogeneous size distribution after OTAC-Na etching process.Note that the surface defects created by ligand detachment in the process of purification and film fabrication can migrate to interior of the material, which contributes to the charge accumulation, resulting in the ion migration and instability.Importantly, the EL of modified PeQ-LEDs exhibits excellent spectral stability with negligible changes in the shape of emission peaks and EL spectra during the device operation, indicating the surface defects are well passivated.Figure 4D displays the current density−voltage-luminance (J−V−L) curves of devices based on Na +and Cs + -PeQDs, in which the max luminesces are 102 and 137 cd m −2 for Cs + and Na + based PeQDs, respectively.The maximum EQE and current efficiency in the controlled device are 5.8% and 8.4% for Cs + and Na + based PeQDs, respectively, which is higher than state-of-the-art records in mixed halogen PeQLEDs at 460-470 nm pure-blue emission (Table S2).Furthermore, the operational stability test was performed at an initial luminance of 100 cd m −2 .The device exhibits an operational half-life T 50 of 6 min (Figure S12).A histogram of the EQE values for 20 LEDs is displayed in Figure 4F, which demonstrates the good reproducibility of device performance based on the Na +and Cs + -PeQDs.

| CONCLUSION
In summary, in situ passivation of PeQDs was presented by introducing the metal cations competitive lattice occupancy against protonated OAm in the process of OTAC-Na etching.The OTAC is expected to remove the long-chain ligands from surface, meanwhile the sodium ions show higher adsorption energy to the surface, which not only completely removes the ligands, but also subtly maintains the cubic phase.The introduced ions (Na + , Pb 2+ , Br − ) exhibit strong binding to the uncoordinated atoms and the surface vacancies, as well as the passivation to the defects state, resulting in the boosted PLQY from 14% to 72%.In addition, the longchain ligands replacing by inorganic ligands enables to further enhance carrier transport behavior of PeQDs.Consequently, the blue-emitting PeQLEDs (470 nm) exhibit the high EQE of 8.42% in the system of mixed halogen PeQDs.Moreover, the corresponding EL spectra demonstrates superior stability with negligible shift under the rising operating voltage.

| Synthesis of OTAC-Na solution
A quantity of 0.1 M Na 2 CO 3 and OTAC were mixed in the brown sample bottle, and continuously stirring at 50°C for 12 h until Na 2 CO 3 was completely dissolved in the glove box.

| Synthesis of Caesium oleate precursor
A volume of 1.4 mL OA, 20 mL ODE, and 0.4073 g Cs 2 CO 3 were loaded into a 50 mL triple neck bottle and dried at 120°C for 1 h under vacuum.The temperature was raised to 150°C until Cs 2 CO 3 was completely dissolved, and then the reaction was terminated and cooled to 90°C for further use.

| Synthesis and purification of CsPbBr x /Cl 3−x QDs
A quantity of 0.1632 g PbBr 2 , 0.1235 g PbCl 2 , 1.6 mL OA, 1.6 mL OAm, 10 mL ODE, and 2 mL TOP were added into a three-neck flask, and then dried at 120°C for 1 h under vacuum.Then the temperature of the precursor solution was increased to 180°C under N 2 atmosphere.Finally, the reaction was performed quickly by injecting the Cs-oleate precursor (1 mL) quickly.After 30 s, the mixture was terminated by ice bath.For further purification, EtOAc was added in a volume ratio of 2.2:1 and then centrifuged at 12 000 rpm for 5 min.A volume of 5 mL n-hexane was added to dissolve the precipitate and centrifuged at 4000 rpm for 5 min.The supernatant was collected for further use.

| Acid-etching process
The perovskite quantum dot (PeQD) solution was diluted with toluene and n-hexane in a ratio of 1:1.Subsequently, PbBr 2 stock solution was preadded and stirred for 5 min.Then 30-50 μL solution of OTAC-Na was added and stirred for 5 min.Finally, EtOAc was used to purify the treated PeQDs in a volume ratio of 2:1 and centrifuged at 12 000 rpm for 10 min.The precipitate was collected in octane and centrifuged at 12 000 rpm for 5 min to remove excess impurities.

| Device fabrication
Detergent, acetone, and ethanol were employed to clean the ITO glass under ultrasound, and then dried at 120°C for 30 min.After treatment with oxygen plasma for 10 min, the PEDOT:PSS were spin-coated at 4000 rpm for 40 s and baked at 140°C for 15 min under air condition, the sample was transferred to the glove box to spin-coat poly-TPD solutions (6 mg/mL) at 3000 rpm for 40 s, followed by baking at 130°C for 15 min.The PeQD layer was subsequently prepared by spin-coating an n-octane solution of PeQDs (~15 mg/mL) at 2000 rpm for 60 s.Electron transport layer (TPBi), LiF, and Al were prepared by vapor deposition, and the thickness of the layers were 60, 1.4, and 100 nm, respectively.

| Supplemental computational methods
All the calculations are performed in the framework of DFT with the projector augmented plane-wave method, as implemented in the Vienna ab initio simulation package.The generalized gradient approximation proposed by Perdew, Burke, and Ernzerhof is selected for the exchange-correlation potential.The long-range van der Waals interaction is described by the DFT-D2 approach.The cut-off energy for plane waves is set to 500 eV.The energy criterion is set to 10 −5 eV in iterative solution of the Kohn-Sham equation.A vacuum layer of more than 15 Å is added perpendicular to the sheet to avoid artificial interaction between periodic images.The Brillouin zone integration is performed using a 2 × 2 × 1 k-mesh.All the structures are relaxed until the residual forces on the atoms have declined to less than 0.01 eV/Å.

| Characterization
The absorption and fluorescence (PL) spectra were measured by Shimadzu UV-2550 and Hitachi F-4500, respectively.The photoluminescence quantum yield (PLQY) of the PeQDs was obtained by FL3, Horiba.The temperature dependent delayed PL spectra were measured by using Hamamatsu Photonics, C4334 with an excitation wavelength of 373 nm under vacuum with the QDs films spined-coated on the glasses.Steady-state PL spectra and time-resolved PL decays of the PeQDs were recorded with C11367-32, Hamamatsu Photonic.XRD patterns were measured by a Rigaku D/MaxrB diffractometer.The high-resolution transmission electron microscopy images were measured by JEM-2100F.Film morphologies of PeQD films were characterized by scanning electron microscope (Zeiss G500) and atomic force microscopy (AFM) (Veeco Multimode V).Current density-voltage-luminance (J-V-L) characteristics were measured simultaneously by Hamamatsu Photonics K.K. C9920-12.
Figure S3 increases from 1.87 to 2.35 after OTAC-Na etching process.The increased bromine content is due to the attachment of bromine ions from inorganic ligand (PbBr 2 ) to PeQDs surface.

F
I G U R E 1 (A) Fourier transform infrared spectroscopy spectra of the pristine and octanoic acid Na (OTAC-Na) PeQDs.X-ray photoelectron spectroscopy spectra of (B) Pb 4f and (C) Na 1s for the pristine and OTAC-Na PeQDs.TEM images of (D) pristine QDs and (E) OTAC-Na PeQDs, inset images show the high-resolution transmission electron microscopy.(F) X-ray diffraction patterns of pristine and OTAC-Na PeQDs.(G) Dynamic equilibrium of the capping ligands (oleic acid and oleylamine).(H) The process of OTAC-Na etching: ① ligands desorption, ② in situ passivation.
the PL intensity at 0 K, A is a pre-exponential coefficient, k is the Boltzmann constant, and E b is the exciton binding energy.The E b is fitted to be 64.71 and 70.46 meV for pristine and OTAC-Na PeQDs, respectively.It is worth noting that E b shows the increased behavior as the size of PeQDs increase.Thus, we believe that the E b is highly dependent on the size and surface state of QDs.Although the QDs show the increased size, the QDs also exhibit the stronger binding with inorganic ligands (PbBr 2 ) on the surface, which can further increase the quantum confinement of electrons.The higher binding energy indicates the reduction for F I G U R E 2 The adsorption model between (A) oleylamine (OAm) and (B) sodium (Na) on the (001) surface.Density of the state of (C) the model of V Cs and (D) model of Na-filled vacancy.Current density-voltage curves of (E) electron-only devices and (F) hole-only devices under dark conditions.

F
I G U R E 3 (A) Ultraviolet-visible (UV-vis) spectra and PL spectra of pristine and octanoic acid (OTAC)-Na PeQDs, inset images show the luminescence of solution under UV lamp.(B) PLQY measurement of pristine and OTAC-Na PeQDs.(C) Time-resolved PL decay spectra of pristine and OTAC-Na PeQDs.PL spectra from 60 to 300 K of (D) pristine and (G) OTAC-Na PeQDs.Integrated PL intensity of (E) pristine PeQDs and (H) OTAC-Na PeQDs.PL spectra of (F) pristine PeQDs, and (I) OTAC-Na PeQDs from 300 to 350 K.

F
I G U R E 4 (A) Device structure.(B) Electroluminescence spectra driven by voltages from 4 to V, inset images show the luminescence of devices.(C) Commission Internationale del'Eclairage coordinate of the device.(D) Curves of current density-voltageluminance of the device.(E) EQE curves of the devices.(F) EQE statistics based on 20 devices.QD, quantum dot.