High-Efficiency Circularly Polarized Light-Emitting Diodes Based on Chiral Metal Nanoclusters

Circularly polarized light-emitting diodes (CP-LEDs) are critical for next-generation optical technologies, ranging from holography to quantum information processing. Currently deployed chiral luminescent materials, with their intricate synthesis and processing and limited efficiency, are the main bottleneck for CP-LEDs. Chiral metal nanoclusters (MNCs) are potential CP-LED materials, given their ease of synthesis and processability as well as diverse structures and excited states. However, their films are usually plagued by inferior electronic quality and aggregation-caused photoluminescence quenching, necessitating their incorporation into host materials; without such a scheme, MNC-based LEDs exhibit external quantum efficiencies (EQEs) < 10%. Herein, we achieve an efficiency leap for both CP-LEDs and cluster-based LEDs by using novel chiral MNCs with aggregation-induced emission enhancement. CP-LEDs using enantiopure MNC films attain EQEs of up to 23.5%. Furthermore, by incorporating host materials, the devices yield record EQEs of up to 36.5% for both CP-LEDs and cluster-based LEDs, along with electroluminescence dissymmetry factors (|gEL|) of around 1.0 × 10–3. These findings open a new avenue for advancing chiral light sources for next-generation optoelectronics.


Synthesis of Cu2Au2(R/S-BTT)4.
[Cu(CH3CN)4]BF4 (1.3 g, 4.13 mmol) was dissolved in 20 mL CH3CN.Add R/S-BTT (1.0 g, 4.78 mmol), 20 mL MeOH, 10 mL CH2Cl2, and (CH3)2SAuCl (0.8 g, 2.72 mmol) into the solution successively under vigorous stirring for 5 min.After that, the solution was centrifuged. 1 mL Et3N was rapidly added into the supernatant, yielding a crude product of Cu2Au2(R/S-BTT)4 in yellowgreen precipitate.The precipitate was washed with 30 mL CH3CN twice.Dissolve the precipitate into CH2Cl2 to get a saturated solution and then add CH3CN into the solution, keeping the ratio of CH2Cl2:CH3CN as 5:1.The solution was kept in an uncapped vial at room temperature for five days, giving pale-green crystals (~ 1.5 g).

Precursors and films preparation.
The precursors were prepared by dissolving Cu2Au2(R/S-BTT)4 into CH2Cl2 at a concentration of 15 mg mL -1 , filtered with 0.45μm PTFE before use.As for those incorporating TCTA, different weight ratios of TCTA were mixed with Cu2Au2(R/S-BTT)4 before dissolving into CH2Cl2, while maintaining a total solute concentration of 15 mg mL -1 .To prepare the films, 80 μL precursor was spin-coated on the substrates at 5000 rpm.for 60 s with an acceleration of 5000 rpm s -1 and annealed at 40 °C for 2 min.

CP-LED devices fabrication.
Glass substrates coated with patterned ITO (15 Ω sq -1 ) were first cleaned with detergent, then ultrasonically cleaned with deionized water, acetone, isopropanol, and ethanol for 20 min in sequence.The ITO-glass substrates were dried with nitrogen flow and treated with ultraviolet ozone for 15 min before use.150 μL modified PEDOT:PSS (m-PEDOT:PSS) was coated on the ITO-glass at 9000 rpm.for 40 s and annealed at 150 °C for 15 min.(m-PEDOT:PSS was synthesized according to the literature 1 and filtered with 0.45 μm PTFE before use.)After annealing, the substrates were quickly transferred to a nitrogen-filled glovebox.80 μL precursors were spin-coated on the substrates at 5000 rpm.for 60 s with an acceleration of 5000 rpm s -1 and annealed at 40 °C for 2 min.Finally, the substrates were transferred into a high-vacuum thermal evaporator, where TPBi (35 nm), LiF (1 nm), and Al (110 nm) were deposited layer by layer at a pressure under 10 -6 Pa.The device active area was 2.5 × 4 mm 2 as defined by the overlapping area of the ITO and Al electrodes.

Single-crystal XRD measurements.
Single-crystal X-ray diffraction data were collected using a Bruker D8 Venture diffractometer with a SMART APEX2 area detector (Mo Kα, λ = 0.71073 Å) at 120 K. Data processing was performed using Bruker's APEX3 software.Data integration and reduction were performed using SaintPlus.
Absorption corrections were performed by the multi-scan method implemented in SADABS-2016/2.Space groups were determined using XPREP implemented in APEX32.The structures were solved with the ShelXT 2 structure solution program using intrinsic phasing and refined with the ShelXL-2015 3 refinement package by least squares minimization using Olex2 4 .All nonhydrogen atoms were in difference-Fourier maps and were then refined anisotropically.All hydrogen atoms were assigned isotropic displacement coefficients U(H) = 1.2U or 1.5U, and their coordinates were allowed to ride on their respective atoms.

Photoluminescence and absorption characterizations.
An Edinburgh Fluorescence Spectrometer (FLS 920) was used for PL, PLE, and 3D-EEM spectra measurements.The PLQYs and PL spectra were measured with an integrating sphere and a monochromatized xenon lamp as the excitation source.The 3D-EEM spectra were measured by a step-excitation wavelength of 5 nm.The PL intensity evolutions were recorded using a Horiba Fluromax-4 spectrofluorometer.Ultraviolet-visible absorption was measured using a spectrometer (JAZ, Ocean Optics) and an ultraviolet-visible-infrared light source (DH-2000-BAL, Ocean Optics).

Femtosecond Transient absorption (fs-TA) measurements.
The transient spectra and their kinetics were obtained through fs-TA spectroscopy in a Helios spectrometer (Ultrafast Systems).For this purpose, the samples were excited with 365-nm pulses obtained from an optical parametric amplifier pumped by an amplified Ti:Sapphire laser (800 nm, 150 fs, 1 kHz, Astrella-Coherent).The probe pulses (white light) were generated by passing another fraction of the 800-nm beam in mechanical delay stage and later through a 2-mm thick CaF2 crystal for the visible range.The white light was split into two beams (signal and reference).
The excitation pump pulses were spatially overlapped with the probe pulses on the samples after passing through a synchronized mechanical chopper (500 Hz) which blocked alternative pump pulses.The obtained signal was sent to the detector through an optical fiber.The absorption change (ΔA) was measured with respect to the time delay and wavelength (λ).The kinetic traces were fitted using the Lavenberg-Marquart algorithm as implemented in Ultrafast System software.

Time-resolved photoluminescence (TRPL) decay measurements.
The TRPL was measured through the time-Correlated Single-Photon Counting measurements (TCSPC) that were performed in a Halcyone setup (Ultrafast Systems), the excitation wavelength was selected to be 365 nm using a parametric optical amplifier (Newport, Spectra-Physics) that was pumped with an Astrella femtosecond pulsed laser (800 nm, 150 fs, 1 kHz, Coherent).
Photoluminescence at 540 nm was collected, and recollimated by a pair of parabolic mirrors passed through a longpass filter (490 nm, Newport) and finally focused on an optical fiber coupled to a monochromator and a PMT detector.The energy at each excitation wavelength was set constant with the help of a pair of variable neutral density filters (Thorlabs) to ensure that less than 1 % of excitation events resulted in a detected photon.TCSPC histograms were fitted using the Lavenberg-Marquart algorithm as implemented in Ultrafast System software.The overall time resolution for the system was better than 120 ps.

Transmission electron microscopy (TEM) measurements.
Cross-section TEM samples were prepared by the focused ion beam (Helios G4 UX).TEM images and the corresponding EDS mappings were acquired by using an aberration-corrected FEI (Themis Z) at 300 kV.
SEM images were acquired using a Helios G4 UX dual beam scanning electron microscopy.
AFM using a Digital Instrument Multimode AFM (Veeco Metrology Group) equipped with specific AFM tips (Model: OLTESPA, nominal spring constant 0.5-4.4N/m).The AFM tips were coated with 50 ± 10 nm Al on back sides.

Ultraviolet photoelectron spectroscopy (UPS) measurements.
UPS studies were carried out in a Kratos Axis Ultra DLD spectrometer equipped with a monochromatic Al Ka X-ray source (hν = 1486.6eV) operating at 150 W, a multi-channel plate and delay line detector under a vacuum of ~10 -9 mbar.All spectra were recorded using an aperture slot of 300 x 700 μm.Survey spectra were collected using a pass energy of 160 eV and a step size of 1 eV.A pass energy of 20 eV and a step size of 0.1 eV were used for the high-resolution spectra.
Samples were mounted in floating mode to avoid differential charging.

Density functional theory (DFT) calculation.
The geometry of Cu2Au2(R/S-BTT)4 were optimized using the hybrid functional B3LYP as implemented in Gaussian09 (version D.01).The LANL2DZ basis set was used for Cu and Au atoms, and the 631G(d) basis set was used to describe H, C, N, and S atoms.The same hybrid functional and basis sets were used to optimize the geometry of Cu2Au2(R/S-BTT)4 in the excited state using the time-dependent DFT (TDDFT) method.Based on the optimized geometries, the dipole moments and charge densities for the frontier orbitals of Cu2Au2(R/S-BTT)4 were obtained.
Performance evaluation of LED.
The LED devices were measured in a nitrogen-filled glovebox at room temperature.The current density-voltage curves were recorded using a Keithley 2400 source meter.The EL and absolute radiation flux for calculating luminance, EQE, CE, PE, and CIE were measured using a commercialized system (LQ-100X, Enlitech) that was equipped with an integrating sphere and a photomultiplier tube (PMT).A commercial inorganic LED (Enlitech, RR2110501) to calibrate and cross-check the LED measurements.
CD and CPL spectra were recorded on a Chirascan V100 spectropolarimeter and a JASCO CPL-300 spectrometer, respectively.The CPPL was excited at 365 nm and the CPEL was obtained under a constant driving voltage at 5.5 V.

Figure S1 .
Figure S1.Ball-stick model of Cu2Au2(R/S-BTT)4 structure, quasi-rhombic geometry of the tetranuclear Cu-Au structure, and the spatial orientation of the benzyl groups.

Figure S8|
Figure S8| Top surface morphology of cluster film.a, Scanning electron microscopy (SEM) image, b,

Figure S10|
Figure S10| Electronic charge densities and energy levels for the highest occupied molecular orbital

Figure S13|
Figure S13| Performances of CP-LEDs based on different precursor concentrations.a, Current density-

Figure S14|
Figure S14| Calibration of the LED testing system.a, Current density-voltage, b, luminance-

Figure S16|
Figure S16| Operational lifetime performances of the CP-LEDs based on Cu2Au2(R-BTT)4 and

Figure S17| a ,
Figure S17| a, Absorption and b, PL spectra of Cu2Au2(R-BTT)4 films with different weight ratios of TCTA

Figure S19|
Figure S19| Summary of the maximum EQEs of state-of-the-art solution-processed LED technologies.

Figure S20| a ,
Figure S20| a, The schematic diagram of the CPEL spectra detection.CPEL spectra of the CP-LEDs based