Atomic layer deposition (ALD) has long been an advanced technology in the fabrication of electronic devices, thanks to its atomic-level precision and conformality. However, the potential of ALD has extended far beyond the traditional applications in microelectronics. In this talk, I will explore unconventional and emerging directions in ALD research: what happens when we step off the beaten track. In particular, the talk will highlight three directions that may open new opportunities for materials design and processing. First, hot-wire atomic layer deposition (HW-ALD) will be discussed as a powerful platform for enabling novel ALD chemistries. By generating highly reactive species through catalytic filament activation, HW-ALD allows access to reaction pathways that are difficult or impossible to achieve using thermal or plasma-enhanced ALD, thereby expanding the precursor library and enabling low-temperature, selective and substrate-tolerant growth. Second, I will introduce the concept of temperature-variation ALD, in which a deliberate modulation of the substrate temperature during ALD cycles provides an additional degree of freedom to control the nucleation and growth of materials. This approach may offer new strategies to overcome the intrinsic limitations of narrow ALD temperature windows and enables precise tuning of material properties without changing precursors or reactants. Finally, the talk will highlight scalable ALD reactors and processes for powder engineering, with a focus on fluidized-bed and atmospheric-pressure ALD systems. These technologies translate the unique advantages of ALD to gram- and kilogram-scale powders, enabling applications in catalysis, energy conversion and storage, pharmaceuticals and functional composites

Hao Van Bui obtained his PhD in Electrical Engineering from the University of Twente, the Netherlands, in January 2013. He was a postdoctoral researcher at MESA+Institute for Nanotechnology, University of Twente (2013–2014) and the Department of Chemical Engineering, Delft University of Technology (2014–2017). In 2018, he joined Phenikaa University, Vietnam, where he currently leads the ALD Research Group. He was promoted to Associate Professor in December 2025 and has since served as vice director of Phenikaa Institute for Advanced Study. His research focuses on the application of ALD in semiconductor technology, energy and pharmaceutical systems.

오일권(Ph.D.) 교수는 아주대학교 전자공학과 및 지능형반도체공학과 부교수로 재직 중이다. 2016년 연세대학교 전기전자공학과에서 김형준 교수의 지도 아래 박사학위를 취득하였다. 이후 2년간 연세대학교에서 박사후연구원으로 연구를 수행했으며, 2018년부터 2021년까지 스탠퍼드대학교 화학공학과에서 Stacey Bent 교수의 지도 아래 박사후연구원으로 재직하였다. 2021년 아주대학교에 부임하였다. 오 교수의 연구는 나노구조 기반 차세대 전자소자의 제작 및 특성 분석에 중점을 두고 있다. 그의 연구는 다양한 반도체 기술 전반을 아우르며, 주요 연구 분야는 차세대 전자 및 메모리 소자의 설계, 제작 및 성능 평가와 AS-ALD을 활용한 표면 화학 반응의 정밀 제어를 통해 구현하는 나노구조 기반 신개념 전자소자 개발이다. 그의 반도체 연구 성과는 2017년 SK하이닉스 반도체 혁신 아이디어 공모전 대상과 2025년 김우중 학술상 등 다수의 수상으로 인정받았다. 특히 최근에는 위상 반금속(topological semimetal) 분야에서 박막 두께가 감소할수록 저항이 감소하는 초박막의 최초 실증 연구를 Science에 발표하였다.

MOSFET 소자의 문턱 전압 제어를 위한 ALD 금속 전극 연구를 소개하며, 차세대 배선 소재로 ALD 기반의 Co, Ru, Mo 계열 소재와 현재 사용되는 Barrier Layer의 대체 소재로 다양한 금속 질화물 기반 소재의 증착 및 평가 결과를 소개한다

생성형 인공지능과 핵생성 이론을 활용한 ALD 증착 레시피 설계

Sung Beom Cho is an Assistant Professor in the Department of Materials Science and Engineering at Ajou University, where he leads the Computational Materials Science and Informatics Laboratory. He earned his Ph.D. from Hanyang University in 2017 and completed postdoctoral research at Washington University in St. Louis (2017–2018). Before joining Ajou University, he served as a senior researcher at the Virtual Engineering Center in the Korea Institute of Ceramic Engineering and Technology (KICET).
Dr. Cho specializes in multiscale simulations and Large Language Models (LLMs). His research integrates diverse simulation methods—including DFT, MD, and FEM—to advance nucleation theory for thin-film crystallization. Currently, he focuses on leveraging LLMs and machine learning to predict practical, human-readable deposition recipes, bridging the gap between computational theory and experimental synthesis.

In the last years there have been a lot of interest in exploring the potential of AI agents that leverage large language models for autonomous materials synthesis. Given the long history of automation in semiconductor fabs, semiconductor processing techniques are ideal candidates to explore the potential of this tool. In this presentation I will introduce the design and performance of AI agents integrated with atomic layer deposition tools. From a hardware side, I will summarize how we are augmenting existing state of the art ALD tools to be interfaced with AI agents based on large language models, including the analysis of the performance of a Python interface that is compatible with model context protocol (MCP) interfaces used in agentic framework. I will also introduce our approach to evaluate the performance of AI agents based on leading models. Our research has focused on three types of tasks: basic instruction tasks, where the agent is asked to execute specific ALD processes, process discovery tasks, where the agent is asked to grow one or more materials and needs to identify the correct ALD processes compatible with the reactor configuration, and process optimization tasks, where the agent is asked to optimize the dose times of a new process.

Despite the simplicity of our agent design, we observed that most of the advanced models excelled at the instruction tasks. However, only recent models such as o1, o3, GPT-5, and Claude Opus 4, with reasoning capabilities, performed well in process discovery tasks. The performance of process optimization tasks, which were carried out using the reasoning models o3 and GPT-5, were on par with those of conventional machine learning algorithms. While the results obtained are promising, I will identify areas where AI research could help improve the performance of autonomous process discovery and optimization tasks involving atomic layer deposition.

Angel Yanguas-Gil is Principal Materials Scientist at Argonne National Laboratory and the Director of the Argonne Microelectronics Institute. He obtained a PhD in Physics by the University of Sevilla, Spain in 2006. After working as postdoctoral researcher at the Ruhr University in Germany and the University of Illinois at Urbana Champaign in USA, he joined Argonne National Laboratory in 2009. Angel’s research has been recognized by the US National Academies of Engineering, the Royal Society of Chemistry, and the American Vacuum Society.

우리는 현재 디지털 전환의 시대에 살고 있으며, 디지털 데이터의 총량은 기하급수적으로 증가하고 있습니다. 이를 위해 메모리 반도체의 저장 용량은 지속적으로 증가하고 있으며, 3 차원 구조에서의 메모리 반도체를 제조하기 위해서는 Atomic Layer Deposition (ALD) 기술이 더욱 중요해지고 있습니다.

하지만, 산업에서의 메모리 반도체에 적용하는 ALD 기술에서는 소자의 표면적이 과도하게 넓어지고, 다양한 표면이 혼재하여, 실험실에서의 이상적인 표면 포화 현상이 불가능하여, 300mm wafer 내 균일한 증착과 수직 구조에서의 정밀한 두께 조절이 점차로 어려워지고 있습니다. 이를 해결하기 위한 산업계에서의 적용되는 있는 기술들을 소개할 예정입니다. 또한, 현재 상용화되고 있는 수직 적층 NAND 구조와 새롭게 개발되고 있는 수직 적층 DRAM 소자 개발에서의 ALD 공정의 핵심 난제들을 소개하고자 합니다.

Advanced Process Development Team, Semiconductor R&D Center, Samsung, Korea

Kwangmin Park received the B.S. and Ph.D. degrees in Materials Science and Engineering from Seoul National University, Korea, in 2000 and 2005, respectively.

In 2005, he joined Samsung electronics in Korea. From 2005 to 2014, he was involved in research on NAND process technologies such as Charge Trap Flash and dielectric reliability at the Semiconductor Research and Development Center.

From 2015 to 2018, he participated in the development of a new memory for Storage Class Memory. From 2018 to 2022, he developed the NAND device process related to memory cell performance, reliability and quality as a project leader in Process Development Team.

Since 2023, he has led the development of the future materials and processes for next generation of memory and logic transistors as a Master (VP of Technology)

Atomic layer deposition (ALD) is recognized as a key technology to meet the semiconductor industry’s increasing demand for continuous device scaling and sustainable innovation. This is owing to its inherent ability to achieve atomic-level precise film deposition with controllable atomic composition. One important application of ALD is an area selective deposition (ASD), which enables film growth only on desired regions, essential for state-of-the-art miniaturized devices with complex three-dimensional nanoscale structures. Combining plasma with ALD is also crucial to control desired surface reactions at low temperature by utilizing reactive species generated by plasma. This presentation will introduce the latest developments in ALD technology, with a focus on the deposition of dielectric films for advanced semiconductor devices, taking the above aspects into consideration.

Tetsuya Goto received the B.S., M.S., and Ph.D. degrees in physics from University of Tsukuba, Japan, in 1995, 1997, and 2000, respectively, for his research on high-temperature fusion plasmas. In 2000, he joined Tohoku University, Japan, where he became an Associate Professor in 2007, and a Professor in 2020. During his tenure there, he was involved in the development of semiconductor manufacturing technologies particularly related to plasma process equipment, as well as advanced semiconductor devices and processes. In 2023, he moved to Tokyo Electron Technology Solutions Ltd., Japan, where he currently serves as Vice President of the Fundamental Technology Development Dpt.1. He is engaged in fundamental research activities focused on thin film deposition. He published over 100 full papers and conference proceedings, and holds over 30 patents.