Journal of the Society for Information Display最新文献

This study introduces recent innovative progress in white organic light emitting diode (WOLED) technology for premium OLED TV and IT displays. By fabricating a four-stack tandem device consisting of red, first blue, green, and second blue devices in sequence from the anode, we could achieve a current efficiency of the WOLED up to 134 cd/A at 10 mA/cm², resulting in enhanced efficiency for red, green, and blue subpixels after transmitting color filters. The new OLED TV implementing the four-stack WOLED technology reached luminance levels of 2,900, 1,000, and 400 nit at the average picture level (APL) of 3%, 25%, and 100%, respectively, and a flash peak luminance of 4,000 nit. Since the four-stack device uses red, green, and blue dopants of saturated color, eliminating yellow green (YG) dopant used for our previous WOLEDs, its color gamut in the Broadcasting service Television (BT)-2020 standard was widen up to 83%. We will explain how to enhance the efficiency and the lifetime in terms of device structure and materials.

The impact of pixel leakage current and liquid crystal flexoelectric effect on flicker were studied with simulation and measurement methods. The impact rule of pixel architecture, TFT voltage, electrode design, liquid crystal material, driving signal, and other factors is obtained, and the key factors under different conditions are identified. The pixel design and liquid crystal mixture development scheme under different conditions are proposed, and a strategy of data voltage modulation at the blanking period combined with frequency-wise ACC brightness compensation is established to reduce flicker. Based on the proposed total flicker optimization scheme, flicker lower than -50 dB is achieved on 55/65/75 in. products when swapping between 48/120 Hz every 200 ms, meeting the G-Sync flicker standard.

Large Language Models (LLMs) can be applied to many fields in the display industry. However, general LLMs lack domain-specific knowledge and specialized terminology understanding, which results in inaccurate responses when applied to industrial question-answering(Q&A) scenarios. To address this issue, this work introduces a framework of Large Language Model training to effectively import the Display Industry Knowledge. This framework is specifically designed to enhance the comprehension ability of LLMs on the knowledge from the display industry field by improving specialized data governance, knowledge distillation techniques, data augmentation strategies, and continual pre-training mechanisms. This approach not only significantly improves the model's performance in Q&A applications within the display industry but also prevents catastrophic forgetting of common knowledge. Experimental results demonstrate the effectiveness of these techniques. We hope that this work can be also helpful for the customization of LLMs in other specialized domains.

We fabricated light-emitting layers consisting of an inorganic phosphor and multicolored fluorescent dyes and successfully controlled the emission color of a powder electroluminescent device, ranging from reddish light-blue to purple, by inducing multistage energy transfer between them and controlling the transfer through the applied frequency.

This paper proposes a high-gain voltage-mode active pixel sensor (V-APS) for high-frame-rate dynamic X-ray imaging. Due to a source–follower structure, the proposed active pixel circuit exhibits superior linearity compared with current-mode APS. And a shared programmable-gain amplifier (PGA) is employed to provide secondary-stage amplification, mitigating the gain limitation of the source–follower structure. The charge-to-voltage gain of the proposed V-APS achieves 1.33 μV/e⁻ with a nonlinearity of 0.38%. Furthermore, threshold voltage (V_T) variation and shift of amplifying transistors in pixels and PGAs can be compensated by using dual-gate thin-film transistor technology. In comparison with conventional 3-T V-APS, the proposed one features a significant decrease of output voltage error rate from 43.67% to 0.79% at a V_T shift of ±0.5 V.

We present and validate an easy-to-setup approach to measure the reflection properties of displays. It is based on a wide field of view conoscopic lens in conjunction with an orientation detection algorithm. Using this approach, we can measure not only the specular, haze, and Lambertian components of display reflection but also the diffractive component. We then investigate the fundamental dependencies of this fourth reflection component through a series of measurements using a variable aperture source and an LC and OLED display. Through these experiments, we can show that the diffractive component scales linearly with the light source's luminance and depends on the angular subtense of the light source.

This study developed an organic light-emitting diode display featuring c-axis-aligned crystalline oxide semiconductor (CAAC-OS) field-effect transistors (FETs) monolithically stacked over silicon (Si) FETs. This display achieved a definition of 5,009 ppi and a luminance of 10,000 cd/m². Below the display pixels incorporating CAAC-OS FETs, both source and scan drivers and a central processing unit (CPU; Cortex-M0) were integrated using Si FETs without widening the bezel. The findings indicate that the built-in CPU enables amplifier offset compensation, which is necessary for image quality correction, and that functional circuits beyond the display driver can operate on a Si substrate.

Foldable displays have become integral to consumer electronics, yet they remain susceptible to mechanical failures, particularly cracks in the hinge region caused by repeated mechanical stress. This study introduces an artificial intelligence (AI)-based method to predict and prevent crack formation by analyzing hinge surface curvature measurements captured within 0.2 seconds of unfolding to 160°, at critical vulnerable angles, using Principal Component Analysis (PCA) and uncertainty quantification with k-Nearest Neighbors (k-NN). A mathematical model incorporating exponential distance-based weighting quantified classification uncertainties, distinguishing confidently classified samples from uncertain cases. Furthermore, an AI-driven scoring model was validated through leave-one-out cross-validation (LOOCV) on an expanded dataset of 200 samples. This model successfully translates complex curvature data into numerical scores, achieving an F1 score of 0.9692 under conditions ensuring zero false positives, thus preventing defective products from reaching customers. This approach significantly enhances quality control in foldable display manufacturing.

We develop both electrical and optical models to optimize the light efficiency and minimize the crosstalk of continuous multiple-quantum-well (CMQW) red AlGaInP μLED with a 2.4-μm pixel size for augmented reality (AR) applications. Our simulation results agree with the reported experimental data well. We also analyze the physical mechanisms and propose two methods to improve light extraction efficiency (LEE). By redirecting the reflected wave and implementing meta-atoms in the device structure, the LEE of CMQW red AlGaInP μLED is improved by ~30% while suppressing the crosstalk between adjacent pixels. In addition, by implementing a carbon black matrix, the crosstalk is reduced by ~5x while keeping a relatively high efficiency. As a result, the image blur is alleviated, and image quality is significantly improved. These high-efficiency red μLEDs will help reduce the power consumption of emerging full-color AR glasses.

The external quantum efficiency (EQE) of QLEDs reaches a theoretical limit of about 25%, primarily constrained by light outcoupling efficiency (ƞ_out). In contrast, quantum rods (QRs) provide directional light emissions that can exceed the ƞ_out limitations of spherical QDs, potentially double the EQE of LEDs. However, developing high-efficiency QRLEDs faces several challenges, such as low quantum yield of the QRs and charge imbalance in the device. Recent advancements in QRLEDs have achieved an EQE of 22%, and the study focuses on red QRs only. Expanding QRLEDs for full-color displays is essential for producing highly efficient green QRs. In this work, we synthesized highly efficient green CdSe/Cd_xZn_1-xS QRs at an emission wavelength of 518 nm, achieving a quantum yield of over 95%. We identified two pathways of electron leakage in QRLEDs at the organic/inorganic interface, ultimately limiting the device's efficiency. We improved the device efficiency by mitigating electron leakage and enhancing hole injection in QRLEDs using multi-stacked HTLs. As a result, QRLEDs achieved record-breaking performance metrics: an EQE of 19.4%, a current efficiency of 72 cd/A, and a luminance of 320,000 cd/m².