High Brightness Electroluminescence of Non‐Carrier‐Injection QLEDs with Precise Layer Processing by Spontaneous Spreading Method

The luminescent properties and mechanisms of non‐carrier injection (NCI) mode quantum dot light‐emitting diodes (QLEDs) are explored in this work. The intermediate insulator electric layer, Poly(vinylidene fluoride‐trifluoroethylene‐chlorofluoroethylene) (P(VDF‐TrFE‐CFE)), effectively blocks carrier injection from the electrodes. Carriers for radiative recombination in the quantum dot (QD) layer are generated by the corresponding carrier generation complex layer under an AC electric field. In this investigation, the emission layer (EML), comprising distinct layers of Cd‐based quantum dots, is precisely regulated using the spontaneous spreading (SS) method. The work reveals that the thickness of the QDs in NCI‐QLEDs significantly influences the device's luminescent performance. In NCI‐QLEDs with a double QD layer as the EML, the device exhibits a maximum brightness of 1003.6 cd m⁻2 and a start‐up voltage of 7 root mean square voltage (VRMS). This brightness level represents the highest reported for vertical emission NCI‐QLEDs. All devices exhibit a broad range of driven voltages. Interestingly, luminescence is detected only during a half‐cycle of the driven signal, as indicated by transient time‐resolved spectrum test results. A system is established to analyze the luminescence mechanism comprehensively. Finally, a proposed carrier compounding mechanism sheds light on the behavior of NCI‐QLED devices.


Introduction
In the realm of advanced display technology, the pressing need for high display resolution has become evident, driving efforts to DOI: 10.1002/admi.202300801reduce pixel size.[3] Consequently, there is a demand to explore new driven systems or technologies for the next generation of high-quality displays, beyond simply adjusting the electric current as the driven signal.One viable approach involves utilizing the electric field to govern pixel luminescence, mitigating the risk of electric current crosstalk. [4]NCI-QLEDs, operating as devices driven by the electric field, distinguish themselves by avoiding external carrier injection into the device.Instead, they depend on the periodic oscillations of intrinsic carriers generated within the material by the electric field to achieve periodic light output. [5,6]Typically, the electric field in NCI-QLEDs is established by introducing an additional insulating layer between two electrodes to prevent the injection of external electrons and holes from the electrodes. [7,8]As carriers in the emission layer oscillate in tandem with the periodic electric field, accumulated carriers at the emission layer interface are effectively eliminated. [9]Furthermore, NCI-QLEDs, shielded by a double insulating layer, can be employed stably in ambient air.This saves time and costs associated with packaging the device and allows for a simplified device structure once again. [10,11]. Wood et al. achieved the inaugural illumination of red, green, and blue NCI-QLED devices in 2011, marking the first application of an injection-free structure to CdSe/ZnS and PbS/CdS core-shell structured quantum dots. [12]However, owing to the absence of external carrier injection, the luminescence efficiency of NCI-QLEDs tends to be modest, with devices exhibiting low brightness ranging from 1 to 1000 cd m −2 . [13,14]Understanding the root cause of this low efficiency has been a central focus in the research field, driving efforts to enhance the performance of NCI-QLEDs.To realize high-performance AC electric field-induced electroluminescent devices, recent research has concentrated on two primary areas.The first focuses on efficiently generating more excitons in the emission layer to increase light intensity and reduce driving voltage. [13,15]The second explores the identification of insulating layer materials with higher dielectric constants to elevate the breakdown voltage. [7,16]o create excitons efficiently, Tsutsui et al. proposed to embed metal nanoparticles between the luminescent layer and the dielectric to increase the charge generation. [17]By incorporating ITO nanoparticles, the device achieved a peak brightness of 261.6 cd m −2 at 300 kHz for a wavelength of 600 nm.Chen et al. proposed the application of the solution-processable high dielectric constant polymer P(VDF-TrFE-CFE) to single-insulator Organic Light Emitting Diode (OLED) devices in 2013, significantly boosting the performance of AC-emitting devices. [18]This device emitted 590 nm orange-red light at 60 kHz with a maximum brightness of 1600 cd m −2 .Despite these advancements, research on the influence of light-emitting layers in NCI-QLED devices remains limited.In DC devices, it is reported that a layer with double to triple quantum dots serves as an efficient emitting layer, offering optimal carrier balance and a low nonradiative relaxation probability. [19,20]However, in NCI-QLED devices, which lack direct injection of external carriers, the impact of the QD layer on the device is uncertain.Precise control of the thickness of the emission layer containing the QD layer is essential for comparative luminescence studies, a challenging task with the conventional spin-coating method.In our prior work, we proposed a novel mechanism for NCI electroluminescence (NCI-EL) in single-layer quantum dots and investigated the effects of drive voltage amplitude, frequency, and QD size on NCI-EL performance. [21]It is imperative to explore the influence of the light-emitting layer on NCI-QLED devices to realize highperformance NCI-QLEDs.
In this work, we employed the Spontaneous Spreading (SS) Method to fabricate red quantum dot films, aiming to develop high-performance Non-Carrier Injection Quantum Dot Light-Emitting Diode (NCI-QLED) devices with the following structure: ITO/P(VDF-TrFE-CFE)/PEDOT:PSS/ PF8Cz/RQD/ZnMgO/MXENE/P(VDF-TrFE-CFE)/Al.Our findings highlight the substantial impact of the number of quantum dot layers on the performance of NCI-QLED devices.The quantum dot layers were meticulously prepared and controlled using the SS process, layer by layer, to construct the emission layer with monolayer to quadruple layers of quantum dots.We systematically investigated the influence of driven voltage amplitude and frequency on the performance of NCI-QLEDs with varying light-emitting layer thicknesses.A time-resolved spectral test system was established to explore the compounding process in NCI-QLED devices, and the impact of the light-emitting layer thickness on device heat generation was discussed.Finally, we propose the carrier compounding mechanism of NCI-QLED devices.The NCI-QLED devices, exhibiting a maximum luminance of 1003.6 cd m −2 and a start-up voltage of 7 V RMS , represent the highest values reported for vertical emission NCI-QLEDs.Notably, these NCI-QLED devices were successfully illuminated on flexible substrates.We anticipate that our work can serve as a guiding reference for subsequent enhancements in the performance of NCI-QLED devices.

Results and Discussion
We previously disclosed the application of the spontaneous spreading method for crafting high-performance, solutionprocessed OLEDs without encountering issues of crossdissolving.Subsequent investigations revealed that incorporating a minute quantity of polymethyl methacrylate (PMMA) into the quantum dot solution enables the creation of monolayer quantum dot films (Figure 1a) on the water surface through the SS process. [22,23]As shown in Figure 1c, the SS film's thickness aligns with the size of individual QDs.Furthermore, the vertical view in Figure 1b, observed under scanning electron microscopy, displays a uniform and closely spaced monolayer of quantum dots.Additional TEM images of monolayer quantum dot films at varying scales are presented in Figure S1 (Supporting Information).This approach allowed for the preparation of an emission layer with diverse thicknesses by manipulating the number of SS films in solution-processed NCI-QLED, featuring the structure ITO/P(VDF-TrFE-CFE)/ PEDOT:PSS/ PF8Cz/RQD/ZnMgO/MXENE/P(VDF-TrFE-CFE)/Al. Figure 1d shows the energy level structure of the double QD layer NCI-QLED.As P(VDF-TrFE-CFE) functions as an insulating material, it induces a "charged effect" when observed under the SEM.Consequently, we conducted a cross-sectional SEM scan of the double-layer QD NCI-QLED without the insulating material, revealing a clearly defined hierarchy between the layers of the thin film.
In NCI-QLEDs, carrier injection from the two electrodes is prevented by the presence of insulator layers.[26] However, the intrinsic carrier density of QDs remains insufficient for producing intense luminescence even under a high electric field.To address this limitation, we introduced PEDOT:PSS/PF8Cz as the hole-generating layer (HGL) and ZnMgO/MXENE as the electron-generating layer (EGL).These layers were designed to generate carriers under an AC electric field inductively.The induced carriers subsequently inject into the quantum dot layer, contributing to the light emission.Therefore, the precise location of the electric field and the frequency at which it reverses direction are crucial factors influencing carrier density and ensuring that the induced carriers effectively recombine within a QD. [27]To delve into the carrier dynamics of NCI-QLEDs under an AC electric field, we utilized the SS process to create emitting layers of Red Quantum Dots (RQD) with varying thicknesses.
To confirm the quality of the QD emission layer by SS processed, time-resolved photoluminescence (TRPL) measurements under excitation of 373 nm were used to examine the emission decay of quantum dot layers with the same thickness prepared by spin-coated (SC) processed films.The decay curves and fitted curves (fitted using a double exponential model) are shown in Figure 2b, denoted as where A 1 and A 2 refer to the decay amplitudes (i.e., weighting factors) of the fit, and  1 and  2 refer to the decay time constants of the fast and slow components, respectively.Average lifetime  avg was calculated by  avg = . [28] The fast decay  1 denotes the exciton relaxation parameter of QDs, and the slow decay  2 indicates the interaction between ex-citons and defects of QD films.As shown in Table 1, the exciton lifetimes of QD films prepared using the SS process are consistently more prolonged than those prepared with the SC process.This indicates a lower defect density in the SS films.However, with the increasing thickness of SS films, the TRPL lifetime gradually shortens, with the  avg decreasing from 12.14 to 11.16 ns.This is attributed to the gradual increase in defect concentration within the film as its thickness grows.These results suggest that QD films prepared using the SS process exhibit high quality with fewer defects, resulting in enhanced luminescence efficiency and a lower non-radiative probability. [29]The PL spectra of SS and SC films, depicted in Figure 2a, reveal a stair-step increase in PL intensity with the thickness of SS films, further demonstrating the precise control achievable through the SS process for QD film modulation.The PL intensity of QD films produced via the SC  process is slightly lower than that of the triple-layer QD SS film, implying that the thickness of QD films prepared by the SC process is roughly equivalent to the triple-layer QDs.Moreover, the SC-prepared QD films exhibit more internal defects, corroborating the TRPL data.The PL spectra of QD films prepared using the SS process exhibit no redshift.
To validate the absence of external carrier injection in the NCI-QLED, we conducted I-V-L (current-voltage-luminance) curve measurements of the NCI-QLED under direct current voltages ranging from 7 to 50 V, as depicted in Figure S2 (Supporting Information).Observations indicate that the direct current levels of the NCI-QLED remain within the order of 10 −8 A (the minimum measurable current level of the source meter being 10 −8 A).The device did not exhibit any luminescence, conclusively demonstrating that the emission within the NCI-QLED solely originates from the driving alternating electric field.Figure 2c shows the normalized electroluminescence (EL) spectra of devices subjected to AC bias, with an amplitude corresponding to V RMS 40 V and a frequency of 70 kHz, are presented.The thickness variation in the emitting layer of the device does not influence the spectrum.The spectra remain essentially the same for NCI-QLED devices with single, double, triple, and quadruple quantum dot film layers (abbreviated as S-NLED, D-NLED, T-NLED, and Q-NLED).The inset in Figure 2c illustrates a physical diagram of a flexible D-NLED prepared using the SS process on a PFN substrate, and the CIE coordinates of the device are shown in Figure S3 (Supporting Information).In Figure 2e, we maintained a fixed AC voltage of 35 V RMS while altering the driven frequency from 10 to 200 kHz.The device's brightness exhibited a pattern of initially increasing and then decreasing, reaching its peak at ≈70 kHz.The luminance of AC electroluminescence, influenced by the electric field direction changes with the driving frequency, is associated with the insulator material and its thickness in devices. [30]As carriers are injected into the QD layer, their injection direction and velocity are influenced by the periodic electric field on the carrier generation layer and QD layer.Furthermore, we maintained a constant drive frequency of 70 kHz while adjusting the driven voltage of devices within the range of 7-60 V RMS .As shown in Figure 2f, the luminance of NCI-QLEDs initially experienced an upward trend, followed by stabilization.Notably, D-NLED outperformed all other devices, illuminating at 7 V RMS and reaching stability after 40 V RMS , with a peak brightness of 1003.6 cd m −2 .To the best of our knowledge, this achievement marks the highest reported value for a vertically emitting NCI-QLED (Figure 2d). [12,17,21,31]To substantiate the superiority of the SS process, we conducted a comparative analysis of the performance of NCI-QLEDs prepared using the SC process and T-NLEDs, as illustrated in Figure S4 (Supporting Information).The NCI-QLEDs fabricated through the SC process exhibited comparable turn-on voltage and peak brightness to those produced via the SS process.However, while increasing voltage, the device performance exhibited a slower improvement, a phenomenon closely correlated with the higher defect concentration within the SC thin film.The device's brightness tends to stabilize at elevated electric fields because the carrier generation layer reaches a saturation point.Additional strength does not result in more carriers at high electric fields, as indicated by the device's behavior. [15]This stands in stark contrast to DC driver devices, highlighting the superior voltage tolerance of NCI-QLEDs.Specifically, we observed a notable correlation between the brightness of the devices and the thickness of the quantum dot film layer.D-NLED exhibited the highest brightness across both the L-V and L-F curves while demonstrating the lowest turn-on voltage.In contrast, the performance of S-NLED lagged that of NCI-QLED devices with multiple QD layers.Thus, we posit that the thickness of the emitting layer in NCI-QLED devices could significantly impact the overall device performance.To delve into the mechanism underlying the influence of the number of QD layers on device performance, we established a transient spectrum test system, as illustrated in Figure 3a.In this setup, a signal generator drives the NCI-QLED device, sending a synchronous signal to the ICCD.The ICCD camera captures the signal, transmitting it to the computer.Subsequently, the computer employs a spectrometer for data analysis.We conducted transient spectrum tests on the NCI-QLED at 50 kHz and 35 V RMS .The time-resolved spectrum of the D-NLED is depicted in Figure 3c, revealing that all NCI-QLED devices exclusively emit light during the positive half-cycle of the sinusoidal signal.We generated the EL relative intensity-time curve using the transient spectrum test system, as depicted in Figure 3b.The device's brightness declines before the sinusoidal signal reaches its peak.The NCI-QLED with a single QD layer exhibits considerably higher responsiveness than devices with multiple QD layers, and the response speed gradually diminishes with an increase in QD layers.Combined with the device performance data, the S-NLED shows markedly lower performance than the multilayer device, yet it demonstrates greater responsiveness.We attribute this phenomenon to the presence of numerous voids in the monolayer quantum dot film, facilitating direct contact between the HGL and the EGL. [32]This results in leakage currents in NCI-QLED devices akin to those observed in DC devices, leading to a significant reduction in exciton concentration within the QD layer and subsequently causing a decrease in device brightness.Nevertheless, the device exhibits the fastest response time due to direct carrier injection into individual QDs.
Analyze the Joule heat generated by the device through the device temperature to characterize the effect of non-radiative compounding or leakage current.The temperature-time curve of NCI-QLED at 35 V RMS , 70 kHz is shown in Figure 4a, while the box plot depicting the maximum temperature is presented in Figure 4b.The temperature variation of the device follows a similar pattern as the transient time-resolved spectrum, as shown in Figures 4a and 3b.As the film thickness increases, the device temperature rises.The temperature of the S-NLED is significantly lower than that of the multilayer NCI-QLED.To investigate the correlation between device heat generation and leakage current, DC-QLEDs with varying QD layer thicknesses were fabricated.The DC-QLED device structure comprises glass/ITO/PEDOT:PSS/PF8Cz/RQD/ZnMgO/Al.The EL spectra of DC-QLED and NCI-QLED are shown in Figure S5 (Supporting Information), and their spectra overlap completely without broadening or redshirting.The relationship between leakage current and device heat generation is explored by altering the number of layers of RQD.Current density-voltage curves for the devices are depicted in Figure 4c.The leakage current in the device decreases with an increase in the number of QD layers.The box plot illustrating the maximum temperature of the DC-QLED at 4.4 V is presented in Figure 4d.As the leakage current diminishes, the heat generation in the device gradually decreases, which is contrary to our NCI-QLED case.Therefore, the heat generation in NCI-QLED is not correlated with the leakage current scenario.
In NCI-QLED, there is no injection of external carriers in the device due to an insulating layer, which can be equated to a series connection of capacitors.The current generated within the device is induced.According to the capacitor voltage division formula where V in is the input voltage, V 1 and V 2 are the partial voltages on the two capacitors,  is the dielectric constant of the capacitor, and d is the thickness of the capacitor.As shown in Figure 5a,b, we simplify the device as a series combination of two components: C1 (representing the insulating and carrier-generating layer) and C2 (representing the QD layer).We analyze the impact of varying the QD layer on voltage division within the device.The  of the QD film is 3.35, and the thickness of a single QD film is 11 nm, which is calculated as C 2 .The ɛ of the insulating material P (VDF-TRFE-CFE) is 35, and that of the carrier-generating layer is ≈3, which is uniformly calculated as C 1 .Using the voltage division formula, we can roughly calculate that as we transition from single-layer RQD to quadruple layers RQD, the percentage of voltage distributed across the QD layer (V2 / V in ) changes as follows: 3.23%-6.25%-9.09%-11.77%(as shown in Figure 5c).The voltage division on the QD layer gradually increases with an expanding QD layer count, with the most significant change occurring when moving from a single layer to a double layer, approaching a factor of two.This observation also results in a decrease in voltage division across the carrier generation layer (CGL), providing additional support for our transient time-resolved spectrum data from a voltage division perspective within the device.The S-NLED exhibits the highest partial voltage on the CGL, making it more prone to illumination in an AC field and allowing for a quicker response.In 2013, O'Yarema et al. proposed that high electric fields induce a separation of electron and hole wavefunctions in QDs, diminishing radiative compounding efficiency. [33,34]Consequently, as the voltage division on the QD layer increases, the radiative compounding efficiency of the QD decreases.This is manifested at a macroscopic level as heat generation and a deterioration in device performance.The observed trend in voltage change on the QD layer aligns with the device temperature change trend (refer to Figures 4b,5c).In summary, D-NLEDs outperform NCI-QLEDs with more layers, consistent with our experimental data.In subsequent experiments, we measured the T 50 lifetime of NCI-QLED, as illustrated in Figure S6 (Supporting Information).The lifetime of D-NLED was found to surpass that of other NCI-QLEDs.However, owing to the dielectric losses induced by organic materials under high electric fields, leading to material performance degradation, the current lifetime of NCI-QLEDs still falls short of that achieved by DC-QLEDs. [4]e believe this work will help clarify the operation mechanism of NCI-QLEDs and guide the development of high-performance NCI-QLEDs in the future.

Conclusion
In summary, this work prepared single-layer quantum dot films using the spontaneous spreading (SS) method.It demonstrated that the QD layer in NCI-QLEDs significantly affects the performance of the devices.Among them, D-NLED has the best performance.The maximum brightness of the D-NLED device was 1003.6 cd m −2 with a start-up voltage of 7 V RMS .This is the highest value reported for vertical outgoing NCI-QLEDs, which enables NCI-QLEDs to satisfy the requirements of flat panel displays.And NCI-QLED was also successfully lit on a PFN flexible substrate.The effects of driving voltage amplitude and frequency on the performance of NCI-QLEDs were analyzed.Using a transient time-resolved spectrum test system and the heat generation of the device, the carrier complexation mechanism of NCI-QLED devices is proposed.We expect this work will be a catalyst for the development of NCI-QLEDs.Device Fabrication: NCI-QLED was prepared by the solution method, in which the RQD light-emitting layer was prepared by the SS method.The device structure was ITO/ P(VDF-TrFE-CFE) (400 nm)/PEDOT:PSS (20 nm)/PF8Cz (45 nm)/RQD/ZnMgO (30 nm)/MXENE (80 nm)/P(VDF-TrFE-CFE) (400 nm)/ Al (100 nm).The P(VDF-TrFE-CFE) solution (100 mg mL −1 , Dissolved in DMF) was spin-coated as an insulating layer on the ITO glass substrate and annealed at 100 °C for 30 min.An aqueous solution of PEDOT:PSS (Clevios 4083) was spin-coated and annealed at 150 °C for 20 min.PF8Cz solution (8 mg mL −1 , Dissolved in DMF) was spin-coated on the PEDOT:PSS layer to form a hole-generating layer, and annealed at 80 °C for 20 min.The film of RQD (17.5 mg mL −1 , Dissolved in a solvent mixture of chlorobenzene: octane = 1:1, 2% PMMA doped) formed on the surface of the water was transferred to the top of the PF8Cz by stamping under atmospheric conditions and then annealed at 100 °C for 5 min to remove water and solvent residues.Subsequently, ZnMgO solution (20 mg mL −1 , Dissolved in ethanol) was spin-coated and annealed at 120 °C for 20 min.Further, the MXENE solution was spin-coated without annealing.P(VDF-TrFE-CFE) solution (100 mg mL −1 , Dissolved in DMF) was spin-coated onto the device as an insulating layer and annealed at 100 °C for 30 min.Finally, thermal vapor deposition deposited the AL electrodes under vacuum conditions.

Experimental Section
Device Measurements: The steady-state photoluminescence spectrum of QDs was obtained by Horiba Fluorolog-3 spectrofluorometer excited at 325 nm.FluoroCube-01-N1, Horiba Instrument carried out Time Resolution Photoluminescence (TRPL) at 373 nm excitation wavelength.The SEM image was measured by field emission scanning electronic microscope (GEMINISEM 500).The AC voltage is supplied by a waveform generator (RIGOL DG1022) connected to an amplifier (Pendulum F10A) and paired with an SRC-600 spectral radiance colorimeter to measure the device's Electroluminescence (EL) spectrum, Luminance-voltage curve (L-V), and Luminance-frequency curve (L-F).Device heat generation was measured by a thermal imaging camera (HIKMICRO K20).Transient time-resolved spectra were measured by an ICCD (ANDOR iStar) connected spectrometer (Zolix Omni-300i).Device leakage current measured by Keithley 2410 Source Meter.All measurements were conducted in the ambient atmosphere without encapsulation.

Figure 1 .
Figure 1.SS film preparation process and film quality.a) Preparation of SS film using the principle of spontaneous spreading.b) SEM image of a monolayer SS film.c) Cross-sectional SEM diagram of a monolayer quantum dot film.d) Energy level diagram of the NCI-QLED device.e) Cross-sectional SEM of a double-layer quantum dot film layer NCI-QLED without the insulating layer.

Figure 2 .
Figure 2. Comparison of film performance and NCI-QLED performance with different QD thicknesses.a) PL spectra of SS films and SC films.b) Transient PL spectra of SS film and SC film.c) EL spectrum at a drive voltage of 40 V RMS and frequency of 70 kHz.The inset shows a physical diagram of the NCI-QLED on a flexible PEN substrate.d) Comparison of the maximum brightness between our work and other vertically emitting NCI-QLEDs.e) Luminance-frequency curve (L-F) at a fixed drive voltage of 35 V RMS .f) Luminance-voltage curve (L-V) at a fixed drive frequency of 70 kHz.

Figure 3 .
Figure 3. Schematic diagram of the transient spectrum test system and transient time-resolved spectrum of NCI-QLED.a) Schematic diagram of the transient spectrum test system with ICCD, signal generator, and spectrometer.b) EL intensity-time curve of NCI-QLED device driven by a sine wave at 50 kHz and 35 V RMS .c) Transient time-resolved spectrum of D-NLED.

Figure 5 .
Figure 5. Device voltage division and working mechanism.a) Equivalent circuit of NCI-QLED.b) Schematic diagram of NCI-QLED equivalent circuit.c) Variability of voltage division on the QD layer in NCI-QLED.d) Schematic diagram of device operation mechanism.

Table 1 .
The TRPL lifetime of QD films prepared through the SC process and QD films with different thicknesses prepared via the SS process.