Multifunctional ligand engineering enables high-performance CsPb(Br/Cl)3 nanocrystals toward efficient and stable pure-blue perovskite LEDs
Hujiabudula Maimaitizi, Hans Ågren, Guanying Chen
TL;DR
Researchers developed a new method to improve the performance and stability of pure-blue perovskite LEDs using a special passivation molecule.
Contribution
A novel passivation strategy using a multifunctional fluorinated phosphonic acid molecule enhances the performance and stability of pure-blue perovskite LEDs.
Findings
HFPA passivation leads to a 14.8% external quantum efficiency in pure-blue PeLEDs.
Device half-life is improved 13-fold compared to unmodified nanocrystals.
Ion migration is effectively suppressed, enhancing device stability.
Abstract
The development of pure-blue perovskite light-emitting diodes (PeLEDs) still lags behind that of green and red-emitting PeLEDs. Mixed halide (Br/Cl) perovskite nanocrystals (PeNCs) are commonly employed for blue emission but suffer from halide vacancies and ion migration. Here, we present a passivation strategy using the multifunctional fluorinated phosphonic acid molecule (1H,1H,2H,2H-heptadecafluorodec-1-yl)phosphonic acid (HFPA), which possesses active functional groups that improve the stability and electroluminescence performance of CsPb(Br/Cl)3 NCs. The HFPA molecule is shown to interact with uncoordinated Pb2+ on the PeNC surface through the phosphonate groups, concurrently establishing hydrogen bonds with adjacent halide ions. Moreover, the presence of fluorine atoms promotes ionic bond formation with the halide octahedra, thereby stabilizing the octahedral structure. The…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6- —https://doi.org/10.13039/501100001809National Natural Science Foundation of China (National Science Foundation of China)
- —Key Technology Research and Industrialization Demonstration Project of Qingdao (Grant No. 25-1-1-gjgg-1-gx), the Outstanding Young Scholars Project supported by the Natural Science Foundation of Heilo
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsPerovskite Materials and Applications · Inorganic Chemistry and Materials · Luminescence Properties of Advanced Materials
Introduction
All-inorganic halide perovskite nanocrystals (PeNCs) have drawn considerable attention for next-generation LEDs owing to their exceptional optoelectronic characteristics, presenting an excellent photoluminescent quantum yield (PLQY), tunable emission spectra, exceptional color purity, and solution processability^1–3^. Despite the impressive external quantum efficiencies (EQEs) exceeding 25% in full-band perovskite light-emitting diodes (PeLEDs)^4^, the advancement of blue PeLEDs, especially within the pure-blue spectral range (460–470 nm), presents significant challenges for their commercial viability in next-generation display applications^5,6^.
Numerous approaches, including the use of ultra-small bromide-based PeNCs and quantum confinement-based dimension engineering to create quasi-2D perovskite structures, have been explored to achieve blue perovskite emitters^7–9^. However, ultra-small PeNCs tend to aggregate and exhibit relatively low emission purity. For quasi-2D perovskite, tuning pure-blue emission remains a challenge, and most fabricated PeLEDs based on this approach typically emit in the sky-blue (480–499 nm) region. Furthermore, the reduced dimension leading to a higher surface-to-volume ratio can result in numerous surface defects, thereby causing low PLQY. Comparatively, mixed halide CsPb(Br/Cl)3-based PeNCs are considered promising candidates for blue light emission due to their readily tunable emission spectra in the blue region by adjusting the Cl content^10,11^. Unfortunately, the incorporation of chlorine leads to abundant defect-intolerant vacancies with deep trap states in the CsPb(Br/Cl)3 nanocrystals, which dramatically suppresses the radiative recombination and degradation of PLQY^12^. Moreover, the migration of halide anions with low activation energy is widely considered a critical factor in the degradation of PeLEDs device stability^13–15^. These halide anions are susceptible to bias voltage and Joule heating during device operation^16–18^. Consequently, to achieve pure-blue PeLEDs with superior efficiency and spectral stability, it is imperative to simultaneously ensure the anchoring of halide ions and the passivation of surface vacancies within the mixed halide (Br/Cl) PeNCs.
In recent years, multifunctional ligands have emerged as an effective approach for enhancing both the performance and stability of PeNCs^19,20^. These ligands typically contain functional groups such as ammonium, sulfonic acid, or phosphonic acid, which enable strong interactions with the PeNC surface. Among them, phosphonic acid groups with hydroxyl (-OH) groups can effectively passivate surface lead ion defects and simultaneously stabilize surface halides through hydrogen bonding with halide ions on the PeNC surface^21–23^. Notably, the use of (1H,1H,2H,2H-heptadecafluorodec-1-yl)phosphonic acid (HFPA) as a passivation layer has been shown to improve the stability of SnO_2_ nanowire transistors by forming strong metal-oxygen-phosphorus (M-O-P) coordination bonds with the SnO_2_ nanowire surface, thereby reducing surface defects and improving device reliability^24^. The terminal -CF_3_ groups in HFPA impart hydrophobicity, effectively protecting the SnO_2_ nanowires from the influence of H_2_O molecules in the atmosphere^25–27^. In addition, HFPA possesses a shorter alkyl chain compared to commonly used ligands such as oleylamine (OAm) and oleic acid (OA), advantageous for charge transport in PeNC-LEDs. Based on these characteristics, we postulate that the HFPA can be a promising multifunctional passivating ligand capable of simultaneously anchoring halide ions and mitigating surface vacancies. Notably, this passivating ligand has not been previously utilized in PeNC-LEDs.
In this study, we report an efficient strategy for synthesizing high-quality CsPb(Br/Cl)3 PeNCs by incorporating (1H,1H,2H,2H-heptadecafluorodec-1-yl)phosphonic acid (HFPA) into the conventional hot-injection process. The phosphonyl (-P = O) group of HFPA anchors on the PeNC through coordination with Pb^2+^, which passivates surface defects and reduces nonradiative recombination. The hydroxyl (-OH) group in HFPA forms hydrogen bonds with adjacent halide ions, stabilizing halogen ions and inhibiting their migration. Furthermore, the terminal F atoms enable the formation of ionic bonds with the halide octahedra, thereby stabilizing the octahedral structure. Consequently, the resulting pure-blue PeLED achieves a peak external quantum efficiency (EQE) of 14.8% and a maximum luminance of 1052 cd·m^-2^ at 467 nm, along with excellent spectral stability and an extended operational lifetime. Additionally, the modified pure-blue PeNCs film demonstrates improved stability under ambient conditions, high temperatures and UV irradiation, thereby highlighting the potential of this method for developing high performance pure-blue PeNCs emitters suitable for display technologies.
Results
CsPb(Br/Cl)3 NCs, prepared using a modified hot-injection method^27–30^, act as the emissive layer in our fabricated Pe-QLEDs. The unmodified Pb(Br/Cl)3 NCs are referred to as pristine NCs, while the Pb(Br/Cl)3 NCs modified with HFPA are referred to as HFPA/NCs. The soft ionic nature of the CsPb(Br/Cl)3 perovskite lattice renders these PeNCs susceptible to surface defects during purification and storage (Fig. 1a), which can exacerbate nonradiative carrier recombination and promote ion migration in PeNC-based LEDs^17,31^. To address these issues, we introduced the multifunctional HFPA molecule, containing a phosphonic acid group and 17 fluorine atoms (Fig. S1a), during hot-injection synthesis to stabilize the PeNCs surface. As illustrated in Fig. 1b, the HFPA molecule coordinates with Pb^2+^ ions via the phosphonate groups, anchoring to the PeNCs and effectively passivating surface defects, thereby mitigating non-radiative recombination. Simultaneously, the -OH groups of the attached HFPA can establish hydrogen bonds with the adjacent halide ions, therefore inhibiting their migration and preventing the halide vacancy defect formation. Furthermore, the fluorine atoms in the chain readily form ionic bonds with the halide octahedra, reinforcing the stability of the octahedral framework.Fig. 1. The influence of HFPA on PeNCs.Schematic illustration of the surface of PeNC (a) without and (b) with HFPA passivation. c Comparison of the adsorption energies (E_ad_) of OA and HFPA on the PeNC surface. d XRD spectra of NCs and HFPA/NCs. TEM images of (e) NCs and (f) HFPA/NCs. Insets display the relevant HRTEM images and size distribution histograms. g Elemental distribution of HFPA/NCs
To elucidate the potential binding mechanisms between HFPA ligands and CsPb(Br/Cl)3 NCs, density functional theory (DFT) computations were employed to analyze the electrostatic potential (ESP) distribution of HFPA molecule. As illustrated in Fig. S1b, the ESP map reveals distinct regions of electron-rich (red) and electron-deficient (blue) areas, indicating the capability of HFPA to interact with both cationic and anionic species on the PeNC surface, thereby facilitating surface defect passivation. The electron-rich domain is primarily localized around the -P = O group, with a minimum ESP value of -6.01 × 10^-2^ a.u., suggesting that the oxygen atom can coordinate with undercoordinated Pb^2+^ ions on the PeNC surface, thus passivating cationic defects^23,32,33^. In contrast, the electron-deficient region near the -OH group displays a maximum ESP value of 8.77 × 10^-2^ a.u., enabling hydrogen bonding with surface halide anions. Furthermore, the adsorption energies (E_ad_) of HFPA and OA on the PeNC surface were also computed via DFT. As depicted in Fig. 1c, HFPA demonstrates a stronger interaction with Pb^2+^ compared to oleic acid, resulting in a higher adsorption energy. This observation corroborates the more robust binding of HFPA to the perovskite surface, enhancing charge transfer^10,32^. Consequently, such interactions stabilize the surface halide ions, inhibit their migration, and mitigate the incidence of halide vacancy defects.
To analyze the influence of the HFPA ligand on the crystal structure and morphology of CsPb(Br/Cl)3 NCs, X-ray diffraction (XRD) and transmission electron microscopy (TEM) characterizations were conducted before and after HFPA treatment. XRD patterns of the synthesized PeNCs (Fig. 1d) revealed a cubic crystal structure for both pristine NCs and HFPA/NCs, indicating that HFPA modification does not alter the intrinsic crystal phase of the NCs. Notably, according to the Scherrer formula (see Supporting Information for calculation details), the crystallite size slightly increased from 7.43 nm (pristine NCs) to 7.71 nm (HFPA/NCs), further supporting the conclusion that the HFPA ligand facilitates the formation of better-defined NCs^34^. To verify this, TEM was employed to examine the morphology and size distribution of both pristine NCs and HFPA/NCs. As depicted in Fig. 1e, f, both samples predominantly display monodispersed cubic morphology, with the HFPA/NCs showing better crystallinity. Additionally, the average size of the HFPA/NCs (8.02 nm) is larger than that of the pristine NCs (7.36 nm), in agreement with XRD estimates. Furthermore, compared to the pristine NCs, the HFPA/NCs exhibit fewer small black spots, which are typically attributed to electron-beam induced degradation associated with halide vacancies^35^. This result suggests that the introduction of HFPA ligands significantly suppresses the decomposition of PeNCs during electron irradiation. High-resolution TEM (HRTEM) images, presented in the insets of Fig. 1e, f, reveal well-defined lattice fringes in both types of PeNCs, with a slight lattice contraction observed in the HFPA/NCs. This lattice contraction can be ascribed to a fluorine-enriched surface layer, which may result from the passivation of surface halide vacancies by fluoride ions or from halide-fluoride ion exchange processes^29,36^. Given the higher electronegativity and smaller radius of F^-^ (133 pm) compared to Br^-^ (196 pm) and Cl^-^ (181 pm) promote stronger chemical bonding with Pb^2+^^37–39^. Elemental mapping (Fig. 1g) confirms the presence of F and P elements in HFPA/NCs. Notably, compared with other elements (e.g., Cs, Pb, Br), F and P exhibit a distinct distribution pattern and mainly localized in the peripheral region of the NC. This observation further verifies the successful introduction and specific distribution of F and P elements in HFPA/NCs, thereby providing direct evidence for the effective interaction between the HFPA ligand and CsPb(Br/Cl)3 NC.
To investigate the interaction between the HFPA ligand and the CsPb(Br/Cl)3 NCs, we performed Fourier-transform infrared (FTIR) spectroscopy and nuclear magnetic resonance (NMR) analysis. As shown in FTIR spectra (Fig. 2a), the HFPA/NCs exhibited characteristic signals corresponding to δ(-P = O), δ(-OH), and δ(-C-F) vibrations, confirming the presence of HFPA ligands on the CsPb(Br/Cl)3 surface. As shown in ^19^F NMR spectra (Fig. 2b), resonance signals of active F atoms in the C-F bonds of HFPA were observed at -81.03 and -81.12 ppm, likely reflecting distinct fluorine chemical environments. Following the addition of PbBr_2_, these two distinct peaks merged into a single peak, indicative of robust coordination interactions between the -P = O group of HFPA and Pb^2+^ ions. This coordination likely locks the conformation of the fluorinated alkyl chain, homogenizing the magnetic environments of the -CF_3_ and -CF_2_- groups and thereby inducing peak merging. Additionally, the interaction with Pb^2+^ may reduce the electron density of fluorine atoms, resulting in a chemical shift of the resonance signal. As depicted in the ^1^H NMR spectrum (Fig. 2c), the resonance signal of active hydrogen in the -OH group shifts from 6.3 ppm to 4.5 ppm. This chemical shift not only intuitively reflects a strong interaction between the -OH group and PbBr_2_ but also supports the hypothesis that -OH groups form hydrogen bonds with halide ions, providing direct evidence for the interaction mechanism between them. Furthermore, in the ^31^P NMR spectra (Figure S2a), a single phosphorus signal is observed at 22.26 ppm for neat HFPA. However, upon adding PbBr_2_ to the HFPA solution, the phosphorus peak shifts to 21.74 ppm. This chemical shift suggests the formation of a new chemical environment around the phosphorus atoms, which can be attributed to the interaction between HFPA and PbBr_2_, likely due to the coordination of the phosphonic acid group with Pb^2+^. We thus speculate that this interaction arises through bonding between -P = O groups and Pb^2+^ ions, a conclusion consistent with the results from FTIR and DFT analyses.Fig. 2. Interaction between HFPA and PeNCs.a FTIR spectra of pristine NCs and HFPA/NCs. b ^19^F NMR spectra and c ^1^H NMR spectra of neat HFPA and the HFPA/PbBr_2_ mixture in dimethyl sulfoxide (DMSO-d6) solution. High-resolution XPS spectra of d C 1 s, e P 2p, and f Pb 4 f of pristine NCs, and HFPA/NCs
Furthermore, X-ray photoelectron spectroscopy (XPS) was conducted to comprehensively characterize the surface chemical states of the PeNCs. XPS survey spectra (Fig. S2b) revealed Cs, Pb, Br, Cl, and C peaks for both types of PeNCs. After HFPA passivation, an additional peak at 689.1 eV appeared in the survey spectra, which can be attributed to the F 1s signal from the HFPA ligand. Moreover, a peak at 291.5 eV in the C 1s spectrum of the HFPA/NCs, corresponding to the -CF_2_- group (Fig. 2d), confirmed the interaction between HFPA and CsPb(Br/Cl)3 NCs. As illustrated in Figs. 2e, S2c, the high-resolution XPS spectra of P 2p and F 1s provide additional confirmation of these findings. Notably, the Pb 4f peaks in the pristine NCs (Fig. 2f), initially positioned at 138.08 eV and 142.94 eV, exhibit a discernible shift to higher binding energies of 138.38 eV and 143.25 eV after the HFPA modification, reflecting changes in the chemical environment of lead atoms. This shift indicates a strong interaction between HFPA and the CsPb(Br/Cl)3 NCs, likely arising from the high negative electrostatic potential of -P = O. Consistent with previous reports^23,33^, we hypothesize that the -P = O group interacts with undercoordinated Pb^2+^ ions. In addition, the characteristic peaks for Cl and Br shifted to higher binding energies (Fig. S2d, e), which may result from hydrogen bonds formation between halide ions and the hydroxyl group of the phosphonic acid. This observation, combined with the NMR and FTIR results, further supports enhanced interactions between HFPA and CsPb(Br/Cl)3 NCs. Such strengthened interactions may stabilize the interface between the HFPA ligand and the CsPb(Br/Cl)3 NCs, thereby stabilizing Pb^2+^ ions and reinforcing the overall material structure. Furthermore, the Cs 3d signal in the high-resolution XPS spectrum also shifted to higher energies after HFPA treatment (Fig. S2f). This shift could be attributed to the ability of HFPA molecules to provide F⁻ ions that occupy surface halide vacancies, thereby modulating the binding energy of Cs^+^ within the crystal structure. These observations align with related literature^27,36^, further confirming the stabilizing effect of HFPA on the PeNC surface.
To elucidate the impact of HFPA passivation on the optical behavior of CsPb(Br/Cl)3 NCs, a comprehensive set of spectroscopic measurements was conducted. Figure 3a presents the photoluminescence (PL) and ultraviolet-visible (UV-vis) absorption spectra comparing pristine NCs with HFPA/NCs. The HFPA/NCs exhibit negligible spectral shifts in the PL spectra, whereas the PL intensity is significantly enhanced, indicating that HFPA modification effectively suppresses non-radiative recombination within the CsPb(Br/Cl)3 NCs. Following HFPA treatment, the PLQY of the PeNC increases from 24.7% to 82.1%. Transient photoluminescence (TRPL) measurements further reveal a substantial improvement in the luminescent performance of the PeNCs (Fig. 3b). Fitting analysis of the TRPL curves confirmed that HFPA treatment significantly prolonged the exciton lifetime of the PeNC, with detailed fitting results summarized in Table S1. The radiative recombination rate of HFPA/NCs (9.79 × 10^7 ^s^-1^) is significantly higher than that of pristine NCs (3.60 × 10^7 ^s^-1^), which can be attributed to reduced surface defects and the consequent suppression of trap-assisted non-radiative recombination pathways. These results support the observed enhancement in PL intensity and quantum yield, and provide direct evidence for the efficient passivation of surface defects in CsPb(Br/Cl)3 NCs via HFPA. Additionally, we employed the space charge limited current (SCLC) method to assess the trap density in the as-prepared PeNCs (Fig. 3c). The current density-voltage (J-V) characteristics of hole-only devices, structured as indium tin oxide (ITO)/ Poly (ethylenedioxythi ophene):polystyrene (PEDOT:PSS) /PNCs/MoO_x_/Ag, were evaluated in the dark condition. A linear region at low bias indicates an Ohmic response, while a sharp increase in current injection at high bias reflects the trap-filling process^40^. The voltage at which the two regions intersect is defined as the trap-filling limit voltage (V_TFL_). For hole-only devices, V_TFL_ correlates with the trap density (N_t_) according to the following equation^41^:
\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\mathrm{V}}_{\mathrm{TFL}}=\frac{\mathrm{e}{\mathrm{N}}_{\mathrm{t}}{\mathrm{L}}^{2}}{2{{\rm{\varepsilon }}}_{\mathrm{r}}{{\rm{\varepsilon }}}_{0}}$$\end{document}Where, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\rm{\varepsilon }}$$\end{document} 0 denotes the vacuum permittivity, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\rm{\varepsilon }}$$\end{document} r represents the relative permittivity of the material, and V_TFL_ exhibits a direct proportionality to the N_t_. After HFPA passivation, V_TFL_ decreased from 0.85 V to 0.78 V, indicating a reduced trap density of the HFPA/NCs film. This observation is consistent with the enhancements in PLQY and PL lifetime discussed earlier.Fig. 3. Defect passivation of PeNCs by HFPA.a Normalized absorbance and PL spectra, PLQY, b PL decay curves, and c J-V curves in SCLC measurements of pristine NCs, and HFPA/NCs. Contour maps of the temperature-dependent PL spectra from 78 to 300 K for d pristine NCs and g HFPA/NCs films. Integration of the temperature-dependent PL intensity and corresponding fitting curves of the exciton binding energy for e pristine NCs and h HFPA/NCs films. FWHM as a function of temperature for f pristine NCs and i HFPA/NCs films. PL excitation was performed at 365 nm
Temperature-dependent PL measurements were performed over a range of 78 K to 300 K to further characterize the passivation effect of HFPA. As illustrated in Fig. 3d, g, the PL intensity decreased with increasing temperature, primarily attributed to thermal quenching, which arises from the activation of nonradiative trap states within the PeNCs films^42^. The exciton binding energy (E_b_) was extracted by fitting the temperature-dependent PL intensity with the Arrhenius equation^43,44^:
\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\rm{I}}(T)=\frac{{I}_{0}}{1+A\,exp\left(-\frac{{E}_{b}}{{KT}}\right)}$$\end{document}Where I(T) represents the emission intensity at temperature T, while I_0_ denotes the intensity measured at 78 K. A is the pre-exponential factor, and K is the Boltzmann constant. As presented in Fig. 3e, h, the E_b_ values for the pristine NCs and HFPA/NCs films are fitted to be 43.53 meV and 51.07 meV, respectively. The increased E_b_ of HFPA/NCs films indicates a reduced probability of exciton dissociation into free carriers, an effect that is beneficial for PL performance. Specifically, a higher E_b_ implies that excitons are more tightly bound and less likely to separate into free electrons and holes. This reduces the chance of free carriers being trapped by surface defects (a major cause of non-radiative recombination), thereby promoting radiative recombination. This enhancement in E_b_ can be attributed to the ability of HFPA to passivate both cation (e.g., undercoordinated Pb^2+^) and anion (e.g., halide vacancies) defects on the NC surface via its phosphonic acid group and F^-^ ions, which alleviates the defect-induced exciton dissociation.
The variation of full width at half maximum (FWHM) with temperature can be described using the independent Boson model^45^:
\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\Gamma }_{T}={\Gamma }_{0}\,+\frac{{\Gamma }_{{op}}}{A\,exp\left(\frac{{\hslash \omega }_{{op}}}{{KT}}-1\right)}$$\end{document}In this model, Γ_T_ refers to the FWHM at a given temperature T, while Γ_0_ reflects the intrinsic inhomogeneous broadening. The parameter Γ_op_ quantifies the strength of exciton-optical phonon coupling, ℏω_op_ represents the optical phonon energy, and K denotes the Boltzmann constant. As shown in Fig. 3f, i, the combined contributions of photons and phonons result in a nonlinear FWHM behavior of the PeNCs at relatively high temperatures, consistent with prior studies^40,42^. The fitted ℏω_op_ values for pristine NCs and HFPA/NCs are 43.53 meV and 51.07 meV, respectively. A lower theoretical ℏω_op_ value indicates an increase in optical phonon generation. These phonons act as scattering centers for nonradiative exciton relaxation, thereby increasing the probability of nonradiative transitions^40,45^. The significantly higher ℏω_op_ value for HFPA/NCs compared to pristine NCs indicates that the HFPA ligand effectively reduces surface vacancies and enhances dielectric screening within the PeNCs. This effect enables more efficient suppression of nonradiative pathways and improved exciton stability, which in turn contributes to the substantial improvements in PL lifetime and PLQY.
The stability of PeNCs is a prerequisite for their integration into optoelectronic devices. To quantify the stabilizing effect of HFPA, thermogravimetric analysis (TGA) was performed on CsPb(ClBr)3 PeNCs before and after surface functionalization with HFPA. Figure S3 showed that both pristine NCs and HFPA-NCs lose a small fraction of mass below 260 °C, ascribed to desorption of residual solvent and weakly bound ligands. A new, well-resolved decomposition step appears exclusively for the HFPA-NCs sample between 380 and 480 °C, corresponding to cleavage of the strongly bound C_8_F_17_- segment and confirming robust anchoring of the HFPA ligand. More significantly, the major lattice-decomposition event shifts from ~388 °C (pristine NCs) to >490 °C (HFPA-NCs) while the associated mass loss decreases by ~6%, evidencing the formation of a dense, carbon- and fluorine-rich barrier that suppresses halide volatilization and retards lattice collapse. Consequently, HFPA-NCs raise the onset of functional degradation by ~100 °C and increase the residual mass at 800 °C, demonstrating a pronounced enhancement in the overall thermal stability of CsPb(ClBr)3 PNCs, consistent with the improved PL stability of the corresponding films under continuous heating (Fig. S4). In line with this enhanced thermal robustness, HFPA-modified PeNC films also exhibit superior UV-light stability (Fig. S5) and storage stability (Fig. S6).
To evaluate the potential of the material for LED application, PeNC-based LED were fabricated utilizing HFPA/NCs as the emissive layer. The device structure, as shown in Fig. 4a, is ITO/ PEDOT:PSS / poly[(9,9-dioctylfluorenyl-2,7-diyl)-co-(4,4’-(N-(4-sec-butyl phenyl)diphenylamine)] (TFB) & tris(4-carbazoyl-9-ylphenyl)amine (TCTa) / PFI/PeNCs/4,6-bis(3,5-di(pyridin-3-yl)phenyl)-2-phenylpyrimidine (B_3_PyPPM) / 2,4,6-Tris[3-(diphenyl phosphinyl)phenyl]-1,3,5-triazine (PO-T2T)/lithium fluoride (LiF) / Al. The corresponding energy level alignment of each layer is depicted in Fig. 4b, where the Fermi level and valence band maximum of both pristine NCs and HFPA/NCs were determined based on analysis of Tauc plots and ultraviolet photoelectron spectroscopy (UPS) measurements (Fig. S7). In this configuration, B_3_PyPPM and PO-T2T act as the electron transport layer and hole blocking layer, respectively, while the TFB/TCTa blend serve dual functions as the electron blocking layer and hole transport layer. The fabricated devices exhibit pure blue emission with a peak at 467 nm, with Commission Internationale de IEclairage (CIE) color coordinates of (0.13, 0.08), closely matching Rec. 2020 standard for ultra-high-definition display applications (Fig. S8a, b). This EL spectrum aligns with the PL spectrum of the HFPA/NCs film, and the absence of parasitic emission confirms efficient, confined electron-hole recombination within the emissive layer. Notably, PeNC-LEDs based on HFPA/NCs demonstrate a significantly enhanced performance, achieving a maximum luminance of 1054 cd·m^-2^, outperforming pristine NCs-based PeNC-LEDs across the entire voltage range (Fig. S8c). Additionally, HFPA/NCs devices exhibit a reduced turn-on voltage of 3.97 V (vs. 5.02 V for pristine NCs), indicating improved charge injection efficiency into the emissive layer. This improvement is mainly attributed to the incorporation of short-chain fluorinated phosphonic acid ligands (HFPA), which coordinates with the undercoordinated Pb^2+^ on the NC surface and partially displaces the insulating long-chain OA due to its higher adsorption (Fig. 1c), thereby increasing the conductivity of the NCs film. As shown in Fig. 4c, HFPA/NCs-based PeNC-LEDs achieves a maximum EQE of 14.8%, which is markedly higher than the peak EQEs of 1.7% observed for pristine NCs-based devices. These substantial enhancements in both electroluminescence and EQE are most likely driven by mitigated surface defects and improved carrier injection, which suppress non-radiative recombination pathways within the emissive layer. These results underscore the critical role of surface engineering and defect passivation in optimizing the efficiency and color purity of PeNC-LEDs.Fig. 4. Performance of PeLEDs devices.a Device structure of PNC-LEDs. b Energy diagram of each layer in PNC-LEDs. c EQE-current density curves of as-prepared pristine NCs and HFPA/NCs based devices. d EL spectra of PNC-LEDs at different driving voltage of from 5.0 V to 7.0 V. e Operational lifetimes of pristine NCs and HFPA/NCs based PNC-LEDs. f Maximum luminance and maximum EQE of pure-blue LEDs based on mixed halide (Br/Cl) PeNCs reported in the literature and this work (see Table S2)
We then investigated the EL spectral stability of pristine NCs- and HFPA/NCs-based PeNC-LEDs under driving voltages ranging from 5 to 7 V (Fig. 4d). The pristine NCs-based device exhibited a significant 8 nm redshift in the EL spectrum with increasing bias. Conversely, the optimized HFPA/NCs device showed a stable EL spectrum with negligible shift, indicating effective suppression of ion migration. Moreover, the HFPA/NCs-based device exhibits significantly improved operational stability when subjected to a constant drive current of 1 mA·cm^-2^ (Fig. 4e), which can be ascribed to the effective inhibition of ion migration via HFPA passivation. Based on these observations and prior reports, we propose a mechanistic explanation for the stable spectral performance of the HFPA/NCs-based PeNC-LED (Fig. 5). In pristine NCs, surface vacancies act as pathways for ion migration, allowing halide ions to migrate toward adjacent vacancy sites when a voltage bias is applied. This facilitates the redistribution of Br and Cl ions^46,47^, resulting in the emergence of localized Br-rich and Cl-rich regions within the material. Consequently, charge carriers are preferentially transferred from the high-bandgap Cl-rich domains to CsPb(Br/Cl)3, and subsequently to the low-bandgap Br-rich domains, ultimately causing a large redshift in the EL spectrum^48–50^. In contrast, HFPA molecules effectively passivate surface cations and halide vacancies in the PeNCs, blocking the formation of ion migration channels. The HFPA molecules also form hydrogen bonds with the adjacent halide ions to stabilize the halide ions and provide F⁻ ions to passivate halide defects, thereby collectively contributing to significantly improving spectral stability. Figure 4f summarizes the device performance of recently published pure-blue perovskite LEDs based on mixed halides (Br/Cl) PeNCs (EL peak at 460–470 nm). It is evident that our device can achieve relatively high luminance and stability simultaneously in PeNC-based pure blue LEDs (Table S2), demonstrating that HFPA, as a multifunctional passivation ligand, is effective for achieving high-performance and stable pure blue PeNC-based LEDs.Fig. 5. Proposed mechanism for device stability enhancement via multifunctional HFPA treatment
Discussion
In this work, we present an efficient synthetic strategy for preparing high-quality CsPb(Br/Cl)3 PeNCs with superior optoelectronic properties, using a fluorinated phosphonic acid (HFPA) ligand with abundant active functional groups. HFPA anchors firmly onto the PeNC surface via coordination with Pb^2+^ and hydrogen bonding with adjacent halide ions, while simultaneously providing F⁻ ions to passivate surface defects and reduce the formation of ion migration pathways. This modification not only enhances PL efficiency and stability of the PeNCs but also suppresses ion migration within the LEDs. The resulting high-performance pure-blue PeNC-based LED exhibits a peak EQE of 14.8% at an EL of 467 nm, along with a peak luminance of 1052 cd·m^-2^. Furthermore, superior spectral stability and operational stability are achieved due to the effective halide vacancy passivation and the formation of hydrogen bonds that stabilize surface halide ions.
Materials and methods
Materials
Cesium carbonate (Cs_2_CO_3_, 99.99%, Macklin), lead (II) bromide (PbBr_2_, 99.99%, Macklin), lead (II) chloride (PbCl_2_, 99.99%, Macklin), 1-octadecene (ODE, 90%, Sigma-Aldrich), OA (90%, Sigma-Aldrich), OAm (70%, Sigma-Aldrich), methyl acetate (MeOAc, 98%, Aladdin) and HFPA (C_10_H_6_F_17_O_3_P, >97%, Aladdin) were used as received without further purification. PFI (Nafion perfluorinated sulfonic-acid resin solution) was purchased from Alfa Aesar. PEDOT:PSS, TFB, TCTa, PO-T2T, B_3_PyPPM, and LiF were supplied by Xi’an Polymer Light Technology Corp.
Synthesis and purification of PeNCs
Preparation of precursor solutions: 2 mmol of Cs_2_CO_3_, 5 mL of ODE, and 5 mL of OA were introduced into a 100 mL three-neck flask fitted with a reflux condenser and a thermocouple. The mixture was degassed under vacuum at 120 °C for 60 min, followed by heating to 150 °C under argon flow until complete dissolution of solids to obtain the Cs-OA precursor solution. For preparation of the HFPA-OA precursor, HFPA (2 mmol), OA (1 mL), OAm (1 mL), and ODE (8 mL) were combined in a separate 100 ml three-necked flask. This mixture was dried at 90 °C for 60 min and then further heated to 120 °C under argon atmosphere to ensure complete dissolution. The resulting precursor solutions were cooled to room temperature before being transferred to an inert-atmosphere glovebox for sealed storage.
Synthesis of pristine NCs and HFPA/NCs: 0.6 mmol PbBr_2_, 0.4 mmol PbCl_2_, 10 mL ODE, 2 mL OAm, and 2 mL OA were loaded into a 100 mL three-neck flask. The reaction mixture was subjected to three cycles of argon purging and vacuum degassing before being heated to 120 °C under a continuous argon flow. The mixture was maintained under vacuum at 120 °C for 60 min until a clear solution was formed. After re-inflating the flask with argon, the temperature was increased to 180 °C under an argon atmosphere. Subsequently, 0.5 mL of preheated Cs-OA precursor was swiftly injected. After 5-10 s, the three-neck flask was cooled to room temperature in an ice-water bath. The crude nanocrystal solution was collected and centrifuged at 10,000 rpm for 5 min with MeOAc (at a 2:1 volume ratio). The supernatant was discarded, and the sediment was resuspended in toluene to form a stock solution. This stock solution was further purified by an additional centrifugation step at 10000 rpm for 5 min, and the resulting supernatant was collected as CsPb(Br/Cl)3 NCs. For HFPA-modified PeNCs, varying amounts of HFPA/OA precursor solution were added to the reaction mixture. The optimal passivation conditions were determined based on the PLQY results. Subsequent characterizations and PeLED fabrication were conducted using NCs synthesized under these optimal conditions.
Device fabrication
The devices were fabricated with the following structure: ITO/ PEDOT:PSS/ TFB&TCTa/ PFI/ PeNCs/ B_3_PyPPM/ PO-T2T/ LiF/ Al. ITO substrates were sequentially cleaned in detergent (Decon90), deionized water, ethanol, chloroform, acetone and isopropanol in an ultrasonic bath for 15 min each, followed by plasma treatment at 80 W for 5 min. After cooling to room temperature, a filtered PEDOT:PSS (0.22 µm pore size) was spin-coated onto the ITO substrates and annealed at 150 °C for 15 min. The PEDOT:PSS-coated substrates were then transferred into a nitrogen-filled glovebox for subsequent layer depositions. The hole-transport layer was prepared by spin-coating a 1:1 (v/v) mixture of TFB (5 mg·mL^-1^ in chlorobenzene) and TCTa solution (5 mg·mL^-1^ in chlorobenzene) at 1500 rpm for 45 s, followed by annealing at 120 °C for 20 min. A thin layer of PFI (0.05 wt.% in isopropanol) was subsequently deposited by spin-coating at 3000 rpm for 45 s and annealed at 145 °C for 10 min. PeNCs were then spin-coated at 1000 rpm for 30 s and annealed at 60 °C for 10 min. B_3_PyPPM (5 nm), PO-T2T (30 nm), LiF (1 nm), and Al (100 nm) were deposited via thermal evaporation through a shadow mask under high vacuum (<5 × 10^-4 ^Pa). The active area of each device was 0.045 cm^2^, defined by the overlapping region of the ITO and Al electrodes.
Characterizations
TEM and HRTEM imaging were conducted on an FEI Talos F200X microscope operated at 200 kV. XRD patterns were recorded using a Bruker D8 diffractometer (Cu Kα radiation, λ = 0.15418 nm) operating at 40 kV and 40 mA, which was equipped with a Vantec-2000 area detector. Absorption spectra of different PeNCs were measured with a Cary 5000 spectrophotometer (Agilent). All PL spectra were acquired using an Edinburgh Instruments FLS1000 spectrometer, and PL lifetimes were determined with the same instrument using a nanosecond flashlamp. XPS measurements were performed on a Thermo ESCALAB 250Xi spectrometer (Thermo Fisher Scientific, USA). FTIR measurements were conducted with a Nicolet 6700 FT-IR spectrometer (Thermo Fisher Scientific Nicolet, USA). For the TDPL measurements of PeNCs, a liquid nitrogen cooling system (model: Oxford Instrument Optistat-DN) coupled with an FLS1000 fluorescence spectrometer was used. The current-voltage-luminance characteristics of the devices were measured at room temperature in a nitrogen-filled glovebox, using a Keithley 2602 source meter coupled with an Ocean Optics 2000 spectrometer. Further details of the characterization procedures are provided in the Supplementary Information.
Supplementary information
Supplementary information