Trap‐induced persistent luminescence in organic light‐emitting diodes

Abstract

Luminescence in organics that lasts for seconds to a few hours after light excitation has been reported recently, showcasing significant application potentials in flexible electronics and bioimaging. In contrast, long‐lasting luminescence that can be electrically excited, whether in organics or inorganics, is much rarer and often less efficient. In this study, we report persistent luminescence (PersL) in organic light‐emitting diodes (OLEDs) that lasts over 100 s and an energy storage effect beyond 60 min after charging with a direct‐current electric field. Thermoluminescence studies reveal that the PersL in OLEDs is induced by traps formed in a host‐guest molecular system serving as an emission layer (EML) with a trap depth of approximately 0.24 eV, consistent with the results from the same EML materials under light irradiation. Integrating results from electronic spin resonance, and density functional theory calculations, we propose a model delineating the charge carrier migration responsible for the trap‐induced PersL in OLEDs. This study on trap‐induced PersL in OLEDs may deepen our understanding of the luminescence mechanism in organic semiconductors and pave the way for expanding their applications in optoelectronics, energy storage and biological detection technologies.

INTRODUCTION

Luminescence in organics has garnered extensive interest over recent decades due to its inherent structural flexibility, high emission efficiency, controllable luminescence dynamics, and widely applicable semiconductor properties.^1,2 It has catalyzed the development of advanced displays, wearable devices, and flexible electronics technologies.^3–5 Historically, room temperature phosphorescence (RTP) and thermally activated delayed fluorescence (TADF), which involve spin-forbidden electronic transitions of triplet and singlet states, have been the primary strategies for achieving long-lasting luminescence in organic systems.⁶ Through well-studied strategies such as enhancing the production of triplet excitons, stabilizing triplet excited states, or mitigating nonradiative pathways, the development and application of organic materials with RTP and TADF properties have seen remarkable success.^7–18 Only recently, a new form of long-lasting luminescence, known as persistent luminescence (PersL), with decay characteristics distinct from RTP and TADF, has been identified in organics.^19–26 While the exact PersL mechanism in organics requires further in-depth exploration, it is observed that the long-lasting decay characteristics, energy storage capacity, thermoluminescence (TL) effect, and the relevance to traps in certain PersL organic materials are consistent with those of typical inorganic PersL phosphors.²⁷ Consequently, a model describing charge carrier migration in host-guest luminescent molecular systems after turning off light excitation, including charge separation, trapping, detrapping and recombination, has been proposed (Figure 1A,B). This model establishes a bridge between the widely accepted PersL mechanism in inorganic phosphors and organic materials with wide-bandgap semiconductor properties.^28–33

[IMAGE OMITTED. SEE PDF]

Light irradiation, particularly at short wavelengths, is the primary energy source for producing long-lasting luminescence in both organic and inorganic materials.^34,35 On the other hand, electric power, which is the most commonly utilized excitation method for devices offering exceptional user-friendliness and controllability, is the energy source for the majority of artificial light, including various light-emitting diodes (LEDs).^36–40 Despite this, progress in realizing electrically excitable long-lasting luminescence is slow, which is incongruous with the substantial application value of electrically driven light-emitting devices. Recently, several types of long-lasting luminescence have been reported in organic LEDs (OLEDs).^41–44 These OLEDs leverage the efficient charge carrier injection of organic semiconductors, and utilize organic luminescent materials, such as deuterated RTP emitters, TADF dimers, or PersL exciplex films as the emission layer (EML) materials to achieve long-lasting luminescent OLEDs that last for seconds or minutes after the electric power supply is turned off. The long-lasting luminescence in OLEDs opens a new avenue for organic luminescent materials to attain novel optoelectronic properties and applications.

In this study, we developed electrically chargeable long-lasting luminescent OLEDs by combining the inherent charge carrier injection structure of OLEDs with the charge carrier trapping capacity of EML materials, and investigated the intermediate products during long-lasting luminescence to unravel the underlying mechanism (Figure 1C). TL measurements on the OLEDs after electric charging were conducted to assess the generated traps and trap depths, which were compared with those of the same EML materials after light irradiation, yielding similar results. In addition, electronic spin resonance (ESR), and density functional theory (DFT) calculations were employed to further confirm the existence of traps as intermediate products after charging. The findings clearly indicate that the long-lasting luminescence in the OLEDs should be primarily attributed to the trapping-detrapping effect of traps, with the charge carrier migration in the OLEDs after electric charging similar to the trap-induced PersL model in organics after light irradiation (Figure 1D). The fabricated OLEDs exhibit long-lasting PersL over 100 s, temperature-dependent decay characteristics, and energy storage effects within the EML, thus enabling them to find new applications, such as time–temperature indicators (TTIs).⁴⁵

RESULTS AND DISCUSSION

Photophysical properties

To validate the proposed scheme, we adopted the typical device configuration of OLEDs and selected a host-guest molecule system as the EML material (see Figure 1C). As a representative example, 1,3,5-tris(1-phenyl-1H-benzimidazol-2-yl)benzene (TPBi) was used as the host molecule due to its fast electron mobility, and 4-(6-(4-(9H-carbazol-9-yl)phenyl)-1,3-dioxo-1H-benzo[de]isoquinolin-2(3H)-yl)benzonitrile (CPN) was chosen as the guest molecule due to its nearly 100% exciton harvesting capability as a potential TADF emitter (Figure 2A).⁴⁶ CPN was synthesized in two steps including dehydration reaction and palladium catalyzed Buchwald–Hartwig cross-coupling reaction, followed by comprehensive characterization via ¹H/¹³C nuclear magnetic resonance (NMR) spectroscopy, liquid chromatography-mass spectrometry (LC–MS), and elemental analysis (Materials and Chemicals Section of Supporting Information and Figures S1 and S2). The synthesized CPN shows excellent thermal stability with an elevated decomposition temperature (T_d, corresponding to a 5% weight loss) at 419°C and a high glass transition temperature (T_g) at 331°C (Figure S3). The doping ratio of CPN with respect to the TPBi host varied from 1 to 100 wt%. According to the photoluminescence quantum yield (Φ_PL) measurements, the optimal doping ratio of CPN was 10 wt%, and the maximum Φ_PL reached 96.4% (Figure S4).

[IMAGE OMITTED. SEE PDF]

The multilayered configuration and energy levels of the fabricated OLEDs are depicted in Figure 2B. The OLEDs consist of ITO/1,4,5,8,9,11-hexaazatriphenylene hexacarbonitrile (HAT-CN, 10 nm, hole injecting layer)/N,N′-bis(1-naphthalenyl)-N,N′-bis-phenyl-(1,1′-biphenyl)-4,4′-diamine (NPB, 40 nm, hole transporting layer)/1,3-bis(N-carbazolyl)benzene (mCP, 10 nm, EBL)/10 wt% CPN@TPBi (30 or 150 nm, EML)/TPBi (60 nm, ETL)/LiF (1 nm, electron injecting layer)/Al (120 nm). The EL characteristics of the OLEDs with 10 wt% CPN doping ratios and different EML thickness (30 or 150 nm) are illustrated in Figures 2C–E and Table 1. The OLEDs with 30 and 150 nm EML thickness both showed sky-blue emission with peaks at 478 nm (Figure 2C). The OLEDs with a 30 nm emissive layer (EML) exhibited a lower turn-on voltage of approximately 3.5 V with an EL brightness of 1 cd m⁻², whereas the device with a 150 nm EML showed a relatively higher turn-on voltage of 6.6 V (Figure 2D). The two OLEDs exhibited high luminescence output of 9764.4 and 9905.5 cd m⁻², respectively. As presented in Figure 2E, the OLEDs with 30 nm EML thickness showed a maximum current efficiency (CE) of 6.02 cd A⁻¹, a maximum power efficiency (PE) of 1.08 lm W⁻¹, and a maximum external quantum efficiency (EQE) of 2.59%. In contrast, the LEDs with 30 nm EML exhibited higher maximum efficiencies of 13.81 cd A⁻¹ (CE), 11.05 lm W⁻¹ (PE), and 12.69% (EQE). Notably, the maximum EQE of device with 30 nm EML surpassed the theoretical limit of 5% for fluorescent-type OLEDs, which verified the efficient exciton recombination probability of the host-guest molecule system used.⁴⁷ Additionally, compared to the OLEDs with 1 or 5 wt% CPN doping ratios, the devices with 10 wt% CPN@TPBi as EML exhibited the highest EQE_max (Figure 2E, Figure S5 and Table S1). Although the PersL intensity decreased with increasing guest concentration (Figure S6), a doping ratio of 10 wt% CPN was selected for subsequent experiments to optimize device performance.

TABLE 1 EL performance of the OLEDs fabricated using 10 wt% CPN@TPBi with different 30 and 150 nm EML thickness.

EML thickness (nm)	Vona (V)	Lmaxb (cd m−2)	CEmaxc (cd A−1)	PEmaxd (lm W−1)	EQEmaxe (%)	λELf (nm)
30	3.8	9764.4	13.81	11.05	12.64	478
150	6.6	9905.5	6.02	1.08	2.59	478

Subsequently, the PersL behavior of the OLEDs after turning off the driving electric field was studied. Four OLEDs with different thicknesses of the EML materials (10 wt% CPN@TPBi) were fabricated. As depicted in Figure 3A, all the OLEDs show PersL lasting for more than 10 s after ceasing the electric power. Notably, all the recorded PersL decay curves deviate significantly from the single-exponential decay functions (Figure S7), and they approximate straight lines in log–log coordinates (Figure 3A). This indicates that the PersL of the OLEDs after electric charging is clearly different from the typical RTP process. In addition, when we fix the EL intensity of the OLEDs at 11300 mcd m⁻² (under electric charging), the PersL intensity increases with increasing EML thickness. The variation in the PersL-to-EL intensity ratio suggests that the PersL and EL might be attributed to two different charge carrier migration processes (i.e., direct recombination to give EL and transport via traps to give PersL), and increasing the EML thickness (i.e., longer transport distances) is conducive to improving the probability of PersL. Impressively, the PersL decay time (defined as the time when the PersL intensity decays to the noise ratio of the photodetector) of the OLEDs recorded at RT exceeds 60 s when the EML thickness is greater than 150 nm. Considering that an excessively thick EML (e.g., 250 nm) significantly increases the turn-on voltage and reduces the EL efficiency of the OLEDs (Figure S8 and Table S2), we used 150 nm as the EML thickness in the following studies.

[IMAGE OMITTED. SEE PDF]

The PersL performance of the OLEDs after charging at different voltages was investigated. The PersL intensity is improved with increasing driving voltage, and it becomes saturated above 12.5 V (Figure 3B). In fact, the OLEDs show different PersL decay curves at various working temperatures. The initial PersL intensity (1 s after ceasing the electric field) increases as the working temperature decreases from 350 to 100 K, and the longest decay time (>100 s) is found at 150 K (Figure 3C and Video S1). This indicates that the PersL of the OLEDs is a temperature-dependent process and again it is different from the previously reported afterglow phenomenon.^41,43,44 After the PersL intensity decays to a low value, the intensity could be recovered to the initial value by electric recharging (Figure 3D and Video S2). From this point of view, one can consider that electric charging and PersL decay are two reversible processes. Finally, we recorded the PersL decay curves as the temperature increased by half. After the OLEDs were charged at 150 K for 1 min, they were kept power-off at 150 K for 30 s and subsequently heated to a higher temperature of 250, 300, or 350 K at a heating rate of 150 K min^–1. Interestingly, the PersL intensity increases during heating, while the intensity decays even faster after reaching the target temperature (Figure 3E and Video S3). To summarize the above results, the PersL of OLEDs after electric charging is an energy storage and thermally activated process, which is the same as that of trap-induced PersL in organics caused by light irradiation.²⁷ Therefore, the minute-level PersL of the electrically charged OLEDs should be attributed to the traps. It should be noted that the EQE of the delayed component (including phosphorescence and PersL) is too low to be quantified with our current testing technique. The PersL decay duration was calculated based on the time it takes for the PersL intensity to decrease to the detection limit of the photomultiplier tube (PMT). To obtain PersL intensity in an absolute unit, the PersL decay signals have been corrected for brightness using a luminance meter.

The frontier molecular orbitals (FMOs) and electronic states of CPN were obtained by density functional theory (DFT) calculations based on the optimized ground-state geometries. As shown in Figure 4A, the HOMO and LUMO are predominantly localized on the 9-phenyl-9H-carbazole segment and 4-(1,3-dioxo-1H-benzo[de]isoquinolin-2(3H)-yl)benzonitrile (NABN) cores, respectively. The calculated energy levels of the HOMO and LUMO are approximately −5.86 and −2.94 eV, respectively. These values are in agreement with the HOMO and LUMO levels (−5.68 and −3.05 eV) obtained from cyclic voltammetry (CV) measurements (Figure S9). The energy difference between the lowest triplet and singlet states of CPN (ΔE_ST) was calculated to be 0.20 eV. The small value of ΔE_ST is attributed to the significant separation of the HOMO and LUMO.

[IMAGE OMITTED. SEE PDF]

The fluorescence spectra of the pure CPN film at 77 K exhibit a broad band centered at 545 nm, and the delayed fluorescence spectra (delay time, 20 ms) show two bands at 545 and 610 nm (top panel of Figure 4B). The energy levels of S₁ and T₁ are 2.28 and 2.03 eV, respectively, with respect to the S₀ level, and the ΔE_ST is 0.25 eV. Moreover, the intensity of the delayed fluorescence is enhanced as the temperature increased from 77 to 300 K, which also verifies the TADF feature of CPN (Figure S10). The steady-state fluorescence spectra and delayed fluorescence spectra of the CPN@TPBi film (middle and bottom panels of Figure 4B) are similar to those of pure CPN molecules in neat film and CPN in toluene (top panel of Figure 4B), suggesting that CPN is the luminescent center in the CPN@TPBi guest-host system. Notably, the emission at 610 nm is identified as phosphorescence rather than TADF, distinguished by its nearly single-exponential decay profile, in contrast to the multi-exponential decay signature of TADF (Figure S11). Notably, both the steady-state fluorescence and delayed fluorescence in CPN@TPBi show a slight blueshift in emission. This could be ascribed to the effects of CPN molecule aggregation and TPBi polarization. Nevertheless, the ΔE_ST value is almost unchanged for pure CPN (0.25 eV) and CPN@TPBi (0.26 eV). Due to the small value of ΔE_ST, the probability of reverse intersystem crossing (RISC) is high at RT.³⁸ It is thus understandable that the delayed fluorescence spectra of the CPN@TPBi film (delay time of 20 ms, bottom of Figure 4B) and the PersL spectra at RT (1 min or 2 h after ceasing the light excitation, Figure S12) are dominated by the S₁–S₀ emission band at 510 nm.

To explore the origin of the PersL, thermoluminescence (TL) glow curves of the OLEDs after electric charging and those of the CPN@TPBi melt-casting film after UV light irradiation were measured (Figure 4C). We record similar TL glow curves in the two measurements, which show peaks in the temperature range of 220–250 K, and the peak temperature turns higher with a larger heating rate. The trap depths (ε) in the two measurements were estimated using the Randall–Wilkins model:⁴⁸ 1 $\frac{\mathit{\beta \epsilon }}{{\mathrm{k}}_{\mathrm{B}}·T_{\mathrm{m}}^2}=s·\exp \left(\frac{-\epsilon }{{\mathrm{k}}_{\mathrm{B}}·T_{\mathrm{m}}}\right)$ where k_B is the Boltzmann constant, T_m (K) is the peak temperature obtained from the TL glow curves, and s (s⁻¹) is the frequency factor. The results show that the estimated trap depths are close, with 0.24 ± 0.01 and 0.28 ± 0.01 eV for the electrically charged OLEDs and light-charged CPB@TPBi film, respectively (Figure 4D). The temperature-wavelength-intensity mapping (3D-plot) of TL in CPB@TPBi further indicates that the TL in the whole range of 100–400 K is primarily attributed to the S₁–S₀ transition of CPN (Figure 4E). It is reasonable to believe that in both cases of electric charging or light irradiation, a part of the input energy is stored by traps in CPN@TPBi, and the trapped charge carriers could be thermally released to give TL (PersL) at the luminescence centers of CPN.

Furthermore, ESR spectra of the CPN@TPBi melt-casting film were recorded (Figure 4F). After UV excitation, ESR signals with a g-factor of ~2.0241 are obtained. These ESR signals correspond to the formation of radical species with unpaired electrons.⁴⁹ The ESR intensity is lowered as the excitation temperature increases from 100 to 300 K, indicating that more radical species return to neutral states at higher temperatures. The decrease in the ESR intensity is consistent with the decrease in the PersL intensity with increasing temperature, as shown in Figure 3C. Furthermore, the FMOs of both closed-shell (TPBi and CPN) and open-shell (TPBi^•+, TPBi^•−, CPN^•+, and CPN^•−) species were analyzed via DFT calculations at the B3LYP/def2-SVP level of theory (Figure 4G, Figure S13 and Table S3). Considering two different electron spin densities in an open-shell system (spin α and spin β), the energy gaps between the LUMO levels of TPBi^•− and CPN^•− were calculated to be 0.08, 0.21, 0.29, and 0.42 eV, respectively (Figure 4G). The average energy gap (~0.25 eV) closely aligns with the experimental trap depth (0.24–0.28 eV), indicating that the traps in CPN@TPBi are probably due to the energy difference between the LUMO levels of CPN^•− and TPBi^•−.

Based on the above results, a possible model for the PersL mechanism in the CPN@TPBi OLEDs is proposed (see Figure 1D). Under electric charging, electrons and holes are injected into the EML from two sides, and they may combine with the CPN guest molecules to form radical species of ${\mathrm{CPN}}_T^{•-*}$ and ${\mathrm{CPN}}_L^{•+}$ (steps i and ii). The radical species are metastable, the electrons may escape from the guest to host molecules, and they require activation energy to cross the barrier between the energy levels of ${\mathrm{CPN}}_T^{•-*}$ and ${\text{TPBi}}^{•-*}$ (i.e., trap depth). Consequently, the trapped electrons could be released under thermal stimulation (step iii) and then recombine with the luminescent centers to give PersL (step iv). The four steps are briefly described as follows:

• Electron injection and radical anion formation:

2 $\mathrm{CPN}+e^{-}\stackrel{V_{\mathrm{ON}}}{\to }{\mathrm{CPN}}_T^{•-*}$

• Hole injection and radical cation formation:

3 $\mathrm{CPN}+h^{+}\stackrel{V_{\mathrm{ON}}}{\to }{\mathrm{CPN}}_L^{•+}$

• Detrapping:

4 ${\mathrm{CPN}}_T^{•-*}+\text{TPBi}\stackrel{\mathrm{\epsilon }}{\to }{\mathrm{CPN}}_T+{\text{TPBi}}^{•-*}$

• Charge recombination and photon emission:

5 ${\text{TPBi}}^{•-*}+{\mathrm{CPN}}_L^{•+}\to \text{TPBi}+{\mathrm{CPN}}_L^{*}\to \text{TPBi}+{\text{pTAP}}_L+\mathit{hv}$ where the asterisk denotes that the active site contains an electron in the LUMO levels, and the subscripts L and T denote that the CPN molecules work as luminescent centers and trap centers, respectively. The above PersL mechanism is similar to that of trap-induced PersL in organics after light irradiation. The proposed mechanism reveals that, in addition to light irradiation, an electricity supply is available for energy storage and long-lasting emission in organics, which may provide new possibilities for the application of organic luminescent materials.

Finally, we demonstrated the potential applications of the electrically chargeable PersL in the field of time–temperature indicators (TTIs).^50,51 In general, TTIs can transform the time–temperature history into accessible information for users, which facilitates the monitoring of perishable foods, pharmaceuticals, or specialty chemicals throughout the entire transport-storage process. Since the liberation of charge carriers from traps is a kinetic process dependent on thermal activation, which is similar to the temperature–time-related loss rate of active ingredients in perishable products, the trap-induced PersL has been considered a valuable principle for the development of novel TTI technologies.⁴⁵ As schematically shown in Figure 5A,B, and Video S1, charge carriers can be stored in the OLEDs by prior electric charging, and more carriers will be released in photon emission as the working temperature increases or the time is extended. Based on this principle, the time–temperature history of the OLEDs after charging can be reflected through the number of residual charge carriers in the OLEDs.

[IMAGE OMITTED. SEE PDF]

By utilizing the energy storage ability and time–temperature-related charge carrier release of the OLEDs, a new TTI method to monitor the quality of products for cold-chain transport is proposed, as shown in Figure 5C. In brief, charge carriers are prestored into OLEDs (TTIs) at 150 K under electric charging. The TTIs are subsequently kept at a certain temperature and transported along with the monitored products. Finally, the residual charge carriers in the TTIs are read out in the form of optical information (by means of TL spectra) and compared with the standard curves to determine the time–temperature history experienced by the products. Based on the proposed scheme, we charged the OLEDs at 12.5 V at 150 K for 5 min. We set the standard storage temperature at 150 K and recorded the integral TL intensity of the OLEDs after different storage times (Figure 5D,E). The integral TL intensity enables us to establish the standard curve of the TTIs working at 150 K. In the cases when the storage temperature is unexpectedly higher or lower than the standard temperature or in the cases when the storage time is longer or shorter than the preset time, the recorded TL intensity will be lower or higher than the values in the standard curves, respectively. The results will fall into the unsafe and safe regions determined by the standard curves, thus achieving the TTI function (Figure 5F). It should be noted that here, we simply demonstrate the conceptual applications of new TTIs based on the principle of the trap-induced PersL, which shows interesting electricity chargeability and reusability. However, more efforts, including increasing the trap density, enhancing the emission efficiency, and extending the storage time, should be made before these new-concept TTIs can be further considered for practical applications.

CONCLUSION

In this work, we fabricated OLEDs with host-guest CPN@TPBi molecules as EML materials, which show PersL lasting for 100 s and energy storage ability over 1 h after ceasing electric charging. We revealed that the electrically chargeable PersL was attributed to trapping-detrapping processes of charge carriers. The trap depth of the trap-induced PersL in the fabricated OLEDs after electric charging was estimated to be 0.28 ± 0.01 eV, which is consistent with the results of the CPN@TPBi film after light irradiation. The traps in the studied host-guest system were assigned to the energy difference between the LUMO levels of CPN^•− and TPBi^•− in radical anions, which was supported by DFT calculations, and ESR measurements. Finally, due to the energy storage ability and time–temperature-related charge carrier release of the fabricated OLEDs, a new TTI method based on the principle of trap-induced PersL possible for cold-chain transport applications was proposed. We believe that this work may deepen the understanding of PersL in organics chargeable with different excitation sources and extend the applications of organic semiconductors in multifunctional photoelectric devices.

METHODS

Synthesis of CPN@TPBi melt-casting films

A mixture of CPN and TPBi (~1 mmol in total) was heated to 350°C on a quartz substrate in a glovebox. Upon melting, the mixture was stirred homogeneously and rapidly cooled to room temperature. The sample was encapsulated in glass cases (~20 × 20 × 1 mm³) by using UV-light-cured epoxy resin as an encapsulant.

Fabrication of OLEDs

The OLEDs were fabricated via thermal evaporation onto ITO-coated glass substrates (Suzhou Fangsheng Optoelectronics) in a high vacuum environment (ca. 5.0 × 10⁻⁵ Pa). The substrates were carefully cleaned with detergent, distilled water, acetone, and isopropanol, dried in a drying oven at 80°C, and then treated with a plasma cleaner (Diener Femto) for 30 min before vacuum thermal deposition. When the vacuum degree of the evaporating cave reached 5 × 10⁻⁵ Pa, the organic materials were evaporated at a rate of 1–1.5 Å s^–1, the LiF was evaporated at a rate of 0.1 Å s^–1, and the aluminum was evaporated at 5 Å s^–1. All the OLEDs were encapsulated and measured in a standard dry nitrogen glovebox. The voltage, current, luminance, and radiant flux characteristics were measured using a source meter (2400, Keithley) and an absolute EQE measurement system (C9920-12, Hamamatsu).

Photophysical characterization

Emission spectra and luminescence decay curves were collected with a fluorescence spectrometer (FLS980, Edinburgh). ESR spectra were obtained using an X-band EPR spectrometer (EMXplus-9.5/12, Bruker). Photographs and videos of the samples were taken with a digital camera (EOS 5D Mark II, Canon and α7SIII, Sony). The PersL decay curves and TL glow curves were recorded using a homemade measurement system. Briefly, the samples were placed on a cooling–heating stage (THMS600E, Linkam) with a controllable temperature range of 100–800 K. The sample chamber was filled with dry nitrogen gas, and quartz glass was installed at the top of the chamber. The samples were excited by electric power (for the OLEDs, different voltages) or ultraviolet light (for the CPN@TPBi films, excitation power density ~5 mW cm⁻²) for 5 min. The intensity of the PersL or TL after ceasing the excitation sources was monitored with a filter-attached PMT (R928P, Hamamatsu). The intensity measured by the PMT can be calibrated using a luminance meter (LM-5, Evenfine) to obtain comparable luminance values. By operating the device at different voltages and simultaneously recording its PersL intensity with a PMT and brightness with a luminance meter, we established a calibration curve. The data from the two instruments exhibit a strong linear correlation: Luminance (mcd m⁻²) = 13.89 × PMT intensity (a.u.). Using this parameter, the PersL intensity decay curve obtained by the PMT was calibrated to provide luminance values. The spectra of PersL or TL were recorded simultaneously using a multichannel spectrometer (QE-Pro, Ocean Optics). In a typical TL measurement, the OLEDs or samples were cooled to 150 K and charged with the excitation sources for 5 min. After a preset waiting time of 20 s, the OLEDs or samples were heated to 350 K at a certain heating rate (100, 75, 50, 35, or 20 K min^–1), and the emission intensity was recorded by photodetectors. The above measurement system was driven by a computer program based on the LabVIEW system.

Cyclic voltammetry tests

The energies of the HOMO and LUMO levels were also measured via cyclic voltammetry with a PalmSens four electrochemical workstation using Pt as the working electrode, platinum wire as the auxiliary electrode, and Ag wire as the reference electrode standardized against ferrocene/ferrocenium. The solutions were prepared inside a glovebox to ensure an oxygen-free atmosphere, and the reduction and oxidation potentials were measured in anhydrous CH₂Cl₂ solutions containing 0.1 M tetrabutylammonium hexafluorophosphate (n-Bu₄NPF₆) as the supporting electrolyte at a scan rate of 0.1 V s⁻¹.

Energy level calculations

The time-dependent density functional theory (TD-DFT) method was used to evaluate the excitation energy and oscillator strength (f) of the excited states (S₁ and T₁) based on the equilibrium configuration of the ground state (S₀) at the B3LYP/def2-SVP level. We calculated the HOMO and LUMO levels using the Gaussian 16 package. All neutral molecules were optimized at the B3LYP/def2-SVP level for stable equilibrium geometries in the ground state (S₀) using the Gaussian 16 program. The geometries of the anions and cations were obtained on the basis of the neutral S₀ structures. All the molecular orbitals were depicted for the closed-shell and open-shell systems.

ACKNOWLEDGMENTS

This work was financially supported by the National Natural Science Foundation of China (Nos. 52172156, 12474412), the Natural Science Foundation of Fujian Province (No. 2023J06005), the Natural Science Foundation of Guangdong Province (No. 2024A1515011197), and the Fundamental Research Funds for the Central Universities (No. 20720240057).

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding authors upon reasonable request.

References

Zhou B, Yan D. Long persistent luminescence from metal–organic compounds: state of the art. Adv Funct Mater. 2023;33(19):2300735.

Li Q, Tang Y, Hu W, Li Z. Fluorescence of nonaromatic organic systems and room temperature phosphorescence of organic luminogens: the intrinsic principle and recent progress. Small. 2018;14(38):1801560.

Huang Y, Hsiang E‐L, Deng M‐Y, Wu S‐T. Mini‐LED, micro‐LED and OLED displays: present status and future perspectives. Light Sci Appl. 2020;9(1):105.

Chow PC, Someya T. Organic photodetectors for next‐generation wearable electronics. Adv Mater. 2020;32(15):1902045.

Ling H, Liu S, Zheng Z, Yan F. Organic flexible electronics. Small Methods. 2018;2(10):1800070.

Zhao W, He Z, Tang BZ. Room‐temperature phosphorescence from organic aggregates. Nat Rev Mater. 2020;5(12):869‐885.

Ye W, Ma H, Shi H, et al. Confining isolated chromophores for highly efficient blue phosphorescence. Nat Mater. 2021;20(11):1539‐1544.

Zhang N, Qu L, Dai S, et al. Intramolecular charge transfer enables highly‐efficient x‐ray luminescence in cluster scintillators. Nat Commun. 2023;14(1):2901.

An Z, Zheng C, Tao Y, et al. Stabilizing triplet excited states for ultralong organic phosphorescence. Nat Mater. 2015;14(7):685‐690.

Liang Y, Hu P, Zhang H, et al. Enabling highly robust full‐color ultralong room‐temperature phosphorescence and stable white organic afterglow from polycyclic aromatic hydrocarbons. Angew Chem Int Ed. 2024;63(10):e202318516.

Zhou B, Yan D. Simultaneous long‐persistent blue luminescence and high quantum yield within 2D organic–metal halide perovskite micro/nanosheets. Angew Chem Int Ed. 2019;131(42):15272‐15279.

Xie Y, Ge Y, Peng Q, Li C, Li Q, Li Z. How the molecular packing affects the room temperature phosphorescence in pure organic compounds: ingenious molecular design, detailed crystal analysis, and rational theoretical calculations. Adv Mater. 2017;29(17):1606829.

Chen C, Chi Z, Chong KC, et al. Carbazole isomers induce ultralong organic phosphorescence. Nat Mater. 2021;20(2):175‐180.

Ju C‐W, Wang X‐C, Li B, et al. Evolution of organic phosphor through precision regulation of nonradiative decay. Proc Natl Acad Sci U S A. 2023;120(46):e2310883120.

Miao Y, Lin F, Guo D, et al. Stable and ultralong room‐temperature phosphorescent copolymers with excellent adhesion, resistance, and toughness. Sci Adv. 2024;10(10):eadk3354.

Xie W, Huang W, Li J, et al. Anti‐kasha triplet energy transfer and excitation wavelength dependent persistent luminescence from host‐guest doping systems. Nat Commun. 2023;14(1):8098.

Zhou L, Song J, He Z, et al. Achieving efficient dark blue room‐temperature phosphorescence with ultra‐wide range tunable‐lifetime. Angew Chem Int Ed. 2024;63(22):e202403773.

Yu J, Sun Z, Ma H, et al. Efficient visible light activated ultralong room temperature phosphorescence triggered by multi‐esterification. Angew Chem Int Ed. 2023;135(52):e202316647.

Kabe R, Adachi C. Organic long persistent luminescence. Nature. 2017;550(7676):384‐387.

Jinnai K, Kabe R, Lin Z, Adachi C. Organic long‐persistent luminescence stimulated by visible light in p‐type systems based on organic photoredox catalyst dopants. Nat Mater. 2022;21(3):338‐344.

Li W, Li Z, Si C, et al. Organic long‐persistent luminescence from a thermally activated delayed fluorescence compound. Adv Mater. 2020;32(45):2003911.

Alam P, Leung NLC, Liu J, et al. Two are better than one: a design principle for ultralong‐persistent luminescence of pure organics. Adv Mater. 2020;32(22):2001026.

Liang X, Zheng YX, Zuo JL. Two‐photon ionization induced stable white organic long persistent luminescence. Angew Chem Int Ed. 2021;60(31):16984‐16988.

Alam P, Cheung TS, Leung NLC, et al. Organic long‐persistent luminescence from a single‐component aggregate. J Am Chem Soc. 2022;144(7):3050‐3062.

Jiang K, Wang Y, Lin C, et al. Enabling robust and hour‐level organic long persistent luminescence from carbon dots by covalent fixation. Light Sci Appl. 2022;11(1):80.

Zhou Y, Zhang P, Liu Z, et al. Sunlight‐activated hour‐long afterglow from transparent and flexible polymers. Adv Mater. 2024;16:2312439.

Lin C, Wu Z, Ma H, et al. Charge trapping for controllable persistent luminescence in organics. Nat Photonics. 2024;18(4):350‐356.

Liang L, Chen J, Shao K, Qin X, Pan Z, Liu X. Controlling persistent luminescence in nanocrystalline phosphors. Nat Mater. 2023;22(3):289‐304.

Ou X, Qin X, Huang B, et al. High‐resolution x‐ray luminescence extension imaging. Nature. 2021;590(7846):410‐415.

Pan Z, Lu Y‐Y, Liu F. Sunlight‐activated long‐persistent luminescence in the near‐infrared from Cr3+‐doped zinc gallogermanates. Nat Mater. 2012;11(1):58‐63.

Li Y, Gecevicius M, Qiu J. Long persistent phosphors—from fundamentals to applications. Chem Soc Rev. 2016;45(8):2090‐2136.

Wei Y, Gong C, Zhao M, et al. Recent progress in synthesis of lanthanide‐based persistent luminescence nanoparticles. J Rare Earths. 2022;40(9):1333‐1342.

Ueda J. How to design and analyze persistent phosphors? Bull Chem Soc Jpn. 2021;94(12):2807‐2821.

Xu J, Tanabe S. Persistent luminescence instead of phosphorescence: history, mechanism, and perspective. Journal of Luminescence. 2019;205:581‐620.

Walfort B, Gartmann N, Afshani J, Rosspeintner A, Hagemann H. Effect of excitation wavelength (blue vs near UV) and dopant concentrations on afterglow and fast decay of persistent phosphor SrAl2O4: Eu2+, Dy3+. J Rare Earths. 2022;40(7):1022‐1028.

Tang CW, VanSlyke SA. Organic electroluminescent diodes. Appl Phys Lett. 1987;51(12):913‐915.

Baldo MA, O'Brien DF, You Y, et al. Highly efficient phosphorescent emission from organic electroluminescent devices. Nature. 1998;395(6698):151‐154.

Uoyama H, Goushi K, Shizu K, Nomura H, Adachi C. Highly efficient organic light‐emitting diodes from delayed fluorescence. Nature. 2012;492(7428):234‐238.

Wang L, Xiao L, Gu H, Sun H. Advances in alternating current electroluminescent devices. Adv Opt Mater. 2019;7(7):1801154.

Lv H, Tang H, Cai Y, et al. Highly stable metal‐free long‐persistent luminescent copolymer for low flicker AC‐LEDs. Angew Chem Int Ed. 2022;134(30):e202204209.

Kabe R, Notsuka N, Yoshida K, Adachi C. Afterglow organic light‐emitting diode. Adv Mater. 2015;28(4):655‐660.

Feng H‐T, Zeng J, Yin P‐A, et al. Tuning molecular emission of organic emitters from fluorescence to phosphorescence through push‐pull electronic effects. Nat Commun. 2020;11(1):2617.

Tan S, Jinnai K, Kabe R, Adachi C. Long‐persistent luminescence from an exciplex‐based organic light‐emitting diode. Adv Mater. 2021;33(23):2008844.

Qiu W, Liu D, Chen Z, et al. Afterglow OLEDs incorporating bright closely stacked molecular dimers with ultra‐long thermally activated delayed fluorescence. Matter. 2023;6(4):1231‐1248.

Song Y, Du J, Yang R, et al. Recyclable time–temperature indicator enabled by light storage in particles. Adv Opt Mater. 2023;11(8):2202654.

Im Y, Kim M, Cho YJ, Seo JA, Yook KS, Lee JY. Molecular design strategy of organic thermally activated delayed fluorescence emitters. Chem Mater. 2017;29(5):1946‐1963.

Liu W, Zhang C, Alessandri R, et al. High‐efficiency stretchable light‐emitting polymers from thermally activated delayed fluorescence. Nat Mater. 2023;22(6):737‐745.

Bos AJ. Thermoluminescence as a research tool to investigate luminescence mechanisms. Materials. 2017;10(12):1357.

Xue J, Xu J, Liang Q, et al. Deepening insights into aggregation effect of intermolecular charge‐transfer aggregates for highly efficient near‐infrared non‐doped organic light‐emitting diodes over 780 nm. Adv Funct Mater. 2023;33(30):2301312.

McMinn W, Magee T. Kinetics of ascorbic acid degradation and non‐enzymic browning in potatoes. Food Bioprod Process. 1997;75(4):223‐231.

Singh R, Lund D, Buelow F. Storage stability of intermediate moisture apples: kinetics of quality change. J Food Sci. 1983;48(3):939‐944.