Oxide Semiconductor Materials Genome Engineering for OLED Displays: from Design to Manufacturing

Zongkai Yan¹, XinYu Wang¹, Jun Zhu², Yong Xiang^1*

¹ School of Materials and Energy, University of Electronic Science and Technology of China, Chengdu, 611731, China；

² School of Electronic Science and Engineering, University of Electronic Science and Technology of China, Chengdu, 611731, China

ABSTRACT: The last decade has seen the rapid development of the OLED display technique which has been widely used in wearable electronics and flexible display device. OLED is a current-driven device controlled by each independent TFF pixel. Therefore, the display performance is sensitive to the TFT characteristics. The OLED display technique, especially the super Hi-Vision (8K × 4 K) and high framerate (>240 Hz) displays, requires the TFT with higher response frequency.  Generally, the operation frequency () of the TFT is decided by the channel length (L) and mobility (µ) according to the formula of . The improvement of µ may therefore increases the of the device. Among the channel layer materials, n-type amorphous oxide semiconductor (AOS) is considered as a promising candidate due to its low cost for mass production and relatively high carrier mobility. The high carrier mobility of the AOS could be attributed to the overlap of the spherically symmetric ns (n>4) electron orbitals, enabling relatively high carrier mobility even for its amorphous form. In addition, the preparation process also plays important role affecting the carrier mobility. The variation of the process parameters, such as deposition methods, heat treatment temperature, and oxygen partial pressure, may result in the nonuniform distribution of elements, crystal structures and electronic structures for the AOS thin film. These factors will lead to changes in the relative position of the Fermi level, the conduction band minimum, and distribution of the migration barrier, giving rise to influence in the carrier transport performance. To overcome the aforementioned limitations of conventional methods and to develop a new AOS with high carrier mobility, a combinatorial and high throughput methodology rather than the conventional trial-and-error approach would be more suitable. Firstly, it is necessary to perform calculation such as density functional to realize the design and screening of AOS compositions. Then, high-throughput experimental methods are conducted to prepare the material library and characterize its compositional, structure, and semiconductor properties of different AOS. Otherwise, the process conditions for volume production of AOS may vary largely from those in the laboratory. In such case, these process parameters require regular adjustment to maintain ideal values ensuring the yield. However, frequent adjustments will reduce the production efficiency. Therefore, the ideal material composition should have not only an excellent optoelectronic performance, but also a wide tolerance of process parameters to increase the production efficiency while maintaining a high yield. A collaborative network incorporating the universities, institutes, and companies is demanded to be established. Besides, a database consisting of the experimental, process, and production data will be essential to build the prediction model and the material development continuum. Finally, based on the established model, an AI brain for design of new AOS and TFT will be trained. After massive data accumulation, analysis, and learning, this brain will benefit the optimization of compositions and process in return for high-yield and high-productivity according to the requirement of commercial applications.

Keywords: Material Genome Engineering; Amorphous oxide semiconductor; Thin-film transistor; collaborative network; Artificial Intelligence Brain

* Corresponding author: xyg@uestc.edu.cn.