Semiconductor Academy

This course introduces students to the foundational concepts and emerging trends in neuro-inspired computing. Students will explore how biological principles inform the design of neural networks, learning algorithms, and energy-efficient hardware architectures. Topics include spiking neural networks, continual learning, compute-in-memory systems, and cross-layer design for edge applications. Ideal for students interested in the intersection of neuroscience, machine learning, and semiconductor technologies. 

Learning Outcomes: 

• Describe the fundamental differences between biological and artificial neural systems, including neuron models, synaptic behavior, and energy efficiency. 
• Identify the roles of various neural network architectures (e.g., ANN, SNN, LSTM) and their suitability for different types of tasks such as object recognition, optical flow, and continual learning. 
• Explain the principles and benefits of cross-layer neuromorphic system design, including the integration of sensors, algorithms, and hardware. 
• Compare different hardware fabrics (e.g., GPUs, compute-in-memory, neuromimetic devices) in terms of their architectural features, energy efficiency, and suitability for neuromorphic computing. 
• Summarize biologically inspired learning strategies such as spike-timing dependent plasticity (STDP), gradient projection memory, and experience replay, and their relevance to continual learning. 

Faculty: Kaushik Roy, Purdue University 

The machine learning chip has been widely applied to various applications of edge computing in recent years. Machine learning chip is one of the branches of the multicore system design. This class will first introduce the fundamentals of the machine learning chip design method. Then, we will use Network-on-Chip (NoC)-based Multi-processor SoC (MPSoC) and introduce some popular on-chip interconnection as well as data reuse methods. Besides, we will introduce some protocols for on-chip data communication. 

Learning Outcomes: 

• Explain the fundamental principles of Network-on-Chip (NoC) interconnection.
• Analyze architectural design concepts of neural network accelerators.
• Examine data reuse methodologies and their architectural implications in neural networks.
• Implement the design flow of NoC-based neural network accelerators from algorithm to architecture

Faculty: Kun-Chih Chen, National Yang Ming Chiao Tung University (NYCU)

This course provides an overview of the failure concerns and their causes in advanced packaging. The common failures that occur in a package due to thermal, mechanical or chemical drivers are highlighted. Experimental methods to quantify reliability together with real world case studies are discussed. 

Learning Outcomes: 

• Develop simplified models for electronic packages and perform first-order accurate calculations for thermal stress. 
• Understand and identify key thermal, mechanical, and diffusional causes that influence package reliability. 
• Apply experimental methods and statistical approaches to quantify reliability in semiconductor packaging. 
• Develop insights in to package reliability assessment through real-world case studies. 

Faculty: Ganesh Subbarayan, Purdue University 

This course explores the intricate reliability challenges faced by modern 3D-HI packaged semiconductor integrated circuits. Students will gain insights into the physical models that explain chip-package degradation and understand the impact of environmental factors such as self-heating, moisture ingress, and corrosion. The course also covers the reliability considerations for packaging in harsh environments, including automotive systems, field-deployed sensors, and biomedical devices, providing a comprehensive understanding of application-specific reliability issues. 

Learning Outcomes: 

• Understand the reliability challenges of modern packaging, with focus on chip-package reliability issues. 
• Explain the physical models that capture the essence of the chip-package degradation modes. 
• Develop an appreciation for the application-specific package reliability in extreme, uncontrolled environments, such as automotives, field-deployed sensors, and implantable biomedical systems. 

Faculty: Muhammad Ashraful Alam, Purdue University 

This online course, “Characterization Techniques for Semiconductor Materials,” provides an in-depth exploration of various analytical methods used in the characterization of semiconductor materials. Through a combination of theoretical concepts and practical applications, students will gain a comprehensive understanding of key techniques such as Electron Microscopy (EM), Atomic Force Microscopy (AFM), X-ray Diffraction (XRD) Analysis, and X-ray Photoelectron Spectroscopy (XPS). The course emphasizes the importance of these techniques in elucidating the morphology, structure, composition, and properties of semiconductor materials. 

Learning Outcomes: 

• Understand the principles and operating mechanisms of Electron Microscopy (EM) techniques, including Transmission Electron Microscopy (TEM) and Scanning Electron Microscopy (SEM), for nanoscale imaging and elemental analysis of semiconductor materials. 
• Describe the working principles of Atomic Force Microscopy (AFM) and its applications in surface topography imaging and roughness analysis of semiconductor materials. 
• Explain the fundamentals of X-ray Diffraction (XRD) Analysis and its role in determining the crystallographic structure, crystal phases, and grain size of semiconductor materials. 
• Analyze the elemental composition and chemical state of semiconductor materials using X-ray Photoelectron Spectroscopy (XPS), and interpret XPS spectra to obtain valuable insights into surface chemistry and electronic properties. 
• Apply the acquired knowledge to interpret experimental data, critically evaluate characterization results, and make informed decisions in the context of semiconductor material analysis and research. 

Faculty: Yu-Sheng Su, National Yang Ming Chiao Tung University (NYCU)

Reliability is, after all, engineering in its most practical form said by James R. Schlesinger – Former US Secretary of State for Defense. Reliability is an essential part of a successful semiconductor product because it is related to trust. No reliability is equal to no business. Especially, the More-than-Moore technologies, such as wide bandgap technologies and Ferroelectric, cannot be commercialized without the proven reliability. Therefore, a good R&D engineer should recognize the statistical tools/characterization methods, identify the reliability issues, and evaluate failure physics, further designing for reliability. In this course, the statistical tools and characterization methods for reliability evaluation are covered. Furthermore, two most important main reliability challenges (Time dependent dielectric breakdown and Bias temperature instability & interface traps) will be addressed in this class. Last, reliability challenges in More-than-Moore devices, such as GaN/SiC wide bandgap technologies, will be discussed as well. 

Learning Outcomes: 

• Apply the basic statistics/statistical tools for reliability analysis 
• Evaluate the failure physics of reliability issues 
• Analyze the Time dependent dielectric breakdown and Bias temperature instability and interface traps 
• Analyze the reliability issues in GaN/SiC wide bandgap technologies 
• Design for reliability (DFR) 

Faculty: Tian-Li Wu, National Yang Ming Chiao Tung University (NYCU)

This course explores the design and analysis of high-resolution incremental delta-sigma analog-to-digital converters (IADCs) for sensor interfaces. You will learn how IADCs enable multiplexing across multiple sensor channels with minimal latency, and gain a deep understanding of their architecture, operation, and performance trade-offs. 

Learning Outcomes: 

• Differentiate incremental delta-sigma ADCs from conventional delta-sigma ADCs in terms of architecture and application. 
• Analyze integrator behavior in the time domain and derive corresponding difference equations. 
• Design incremental delta-sigma ADCs from first-order to higher-order configurations and implement digital filters to reconstruct oversampled bitstreams. 
• Evaluate the relationship between signal-to-quantization noise ratio (SQNR) and oversampling ratio for various loop filter orders. 

Faculty: Derek Chia-Hung Chen, National Yang Ming Chiao Tung University (NYCU)

Electronic Design Automation (EDA) techniques are essential for modern complex IC designs. In the EDA flow, automatic synthesis for generating the required designs is an important step. It can help designers to handle millions of transistors and realize your ideas in a short time. This course gives a brief overview about those synthesis techniques, which include: (1) logic synthesis for digital designs (2) circuit synthesis for analog designs. Our focus is the insight techniques and algorithms to develop the synthesis tools, not how to use those EDA tools, which can give important background knowledge to the related researches. For the students whose research directions are not EDA-related topics, better understanding of EDA tools also helps to use those tools better. 

Learning Outcomes: 

•  Describe the major steps in the front‑end Electronic Design Automation (EDA) design flow.
• Identify the core concepts and techniques used in digital logic synthesis and analog circuit synthesis tools.

Faculty: Chien-Nan Liu, National Yang Ming Chiao Tung University (NYCU)

This course provides a comprehensive exploration of power management integrated circuits, focusing on two critical designs: Low Drop-out Voltage Regulators (LDO) and Switching Buck Converters. The curriculum is designed to offer a deep dive into the fundamental principles, operational mechanics, and design strategies of these regulators. It emphasizes the importance of stable voltage regulation across varying power demands and input fluctuations, essential for enhancing the efficiency and performance of electronic devices. Through theoretical understanding and practical applications, the course prepares students to tackle challenges in designing and optimizing power management systems suitable for a wide range of electronic products. 

Learning Outcomes: 

• For Low Drop-out Voltage Regulators (LDO): 
  • Design simple and efficient LDO circuits suitable for various applications. 
  • Analyze and optimize key performance parameters, such as drop-out voltage and efficiency. 
• For Switching Buck Converters (SWR): 
  • Develop proficiency in designing and controlling high-efficiency buck converters. 
  • Understand and apply advanced feedback and control techniques to enhance converter performance. 

Faculty: Chung-Chih Hung, National Yang Ming Chiao Tung University (NYCU)

Two-dimensional (2D) semiconductors are seen as a future candidate to replace Si for further scaling of physical gate lengths or to work alone with Si in transistors, sensors, and memory. 2D transition metal chalcogenide (TMD) such as WSe2 and MoS2 is one of the 2D semiconductors widely used in optoelectronic research because a combination of its large bandgap and high effective mass suppresses a high off-state leakage usually seen on Si and other traditional bulk transistors. Additionally, their direct bandgap, tunable colors for emissions, and valley properties can be applied to innovative photonic research and applications. In this course, we will start with a brief introduction to 2D semiconductors in terms of basic properties and the usefulness of their clean surface and ultrathin body. Next, we will go over several topics centered on 2D TMD, including the synthesis by metalorganic chemical vapor deposition, van der Waals epitaxy, doping and implantation, dielectric integration, air stability, and devices. The end of this course will discuss the roadmap and challenges for the utilization of 2D materials in CMOS technology in the near future. 

Learning Outcomes: 

• Understand the basic properties and optoelectronic applications of TMD.  
• Have the  knowledge of synthesis, characterization, and integration of 2D materials  
• Learn the mechanisms of epitaxy, cation and anion doping, and MOCVD of 2D TMD.  
• Identify issues and challenges for device scaling of 2D semiconductors from material engineering perspectives 

Faculty: Yu-Chuan Lin, National Yang Ming Chiao Tung University (NYCU)

The course provides an introduction to optoelectronic device operation and design, which are important components in optical communication systems, consumer electronics and IT systems. Assuming an electrical/electronic or physics undergraduate level background in semiconductor physics, the course begins with a recap of this essential topic, followed by the study of pn junctions and LED fundamental operating principles. A variety of designs and configurations of these devices are considered with application-specific characteristics. The short course is ‘applied’ in nature. 

Learning Outcomes: 

• Posses core knowledge relating to pn junction and LEDs. 
• Apply these principles to new engineering projects. 
• Delve deeper into the topic of optoelectronics. 

Faculty: Niall Tumilty, National Yang Ming Chiao Tung University (NYCU)

This course aims to give an overview of magnetism and magnetic materials and the associated spintronics applications. It begins with the fundamentals of magnetic materials and the methods of characterization. Second, the principle of spintronics and the issues of developing spintronics devices will be included. Third, the evolution of magnetoresistive random access memory and the trilemma issues regarding storage, reading, and writing will be reviewed. Finally, the advanced spintronics applications, including neuromorphic computing, physically secured applications, and the digit-analog converter, will be demonstrated. 

Learning Outcomes: 

• Describe the properties of magnetic materials and the basis for using magnetic materials for spintronics. 
• Identify key challenges in applying spintronics technology to MRAM devices. 
• Summarize advanced applications of spintronics, including neuromorphic computing and secure hardware solutions. 

Faculty: Chao-Yao Yang, National Yang Ming Chiao Tung University (NYCU)

Ferroic materials, including ferroelectrics and multiferroics, exhibit intriguing properties that have led to their widespread application in various electronic devices. This short course provides a comprehensive introduction to the principles, theories, and applications of ferroelectric and multiferroics devices. Beginning with the fundamentals of ferroelectric materials and their phase transitions, participants will delve into the Landau-Devonshire theory, which serves as a cornerstone in understanding the behavior of ferroic materials. The course will explore the mechanisms behind switching currents and the factors influencing ferroelectric switching. Participants will also gain insight into advanced characterization techniques such as Piezoresponse Force Microscopy (PFM) for probing the electromechanical responses of ferroic domain configuration. Moving beyond theory, the course will spotlight key ferroic devices, including Ferroelectric Random Access Memory (FE-RAM), Ferroelectric Field-Effect Transistors (FE-FET), and Ferroelectric Tunnel Junctions (FTJ), elucidating their operating principles and technological implications. Throughout the course, attention will be given to the burgeoning field of multiferroics, which promise enhanced functionalities by coupling different ferroic orders. Emerging challenges in fabrication, integration, and scaling of ferroic devices will be discussed, offering participants a comprehensive understanding of the current state and future directions of ferroic technology. 

Learning Outcomes: 

• Understand the fundamental properties and phase transitions of ferroic materials. 
• Gain insights into the operation and applications of key ferroic devices such as FE-RAM, FE-FET, FTJ and MESO. 
• Recognize the challenges and opportunities in the development of ferroic devices, including those based on multiferroic materials. 

Faculty: Yen-Lin Huang, National Yang Ming Chiao Tung University (NYCU)

The course provides a comprehensive exploration of the semiconductor industry’s development from basic technologies such as diodes to advanced quantum computing. It highlights the significant role of Moore’s Law in driving technological advancement and establishing industry standards. This curriculum focuses on Research and Development (R&D) management, examining strategic resilience and the complexity of the global supply chain. It details the shift from straightforward value chains to intricate value networks, which is essential for success in the highly competitive semiconductor industry. The course aims to equip learners with a solid framework for strategic innovation, preparing them to take on leadership roles in the fast-paced semiconductor sector, thus promoting technological progress and strategic insights. 

Learning Outcomes: 

• Understand Semiconductor Technology Evolution. 
• Develop expertise in R&D management, emphasizing strategic resilience and innovation in the semiconductor industry. 
• Navigate complex value networks and gain skills for leading in a competitive global landscape. 

Faculty: Hsin-Ning Su, National Yang Ming Chiao Tung University (NYCU)

Course Description: This course is designed for students who are not familiar with, but have a keen interest in understanding, the semiconductor industry. We will begin by introducing the basics of semiconductors, followed by an overview of the current global semiconductor ecosystem. Subsequently, we will delve into the business models of fabless, foundry, and IDM, providing detailed explanations for each. Additionally, we will introduce other significant players in the global semiconductor ecosystem, such as semiconductor wafer manufacturers, semiconductor equipment manufacturers, and Outsourced Semiconductor Assembly and Test (OSAT) vendors. 

Learning Outcomes: 

• Identify fundamental concepts related to semiconductor technology.
• Describe the structure and characteristics of the global semiconductor ecosystem.
• Summarize the business models used by fabless, foundry, and IDM companies.
• Recognize the roles of key industry players, including wafer manufacturers, equipment manufacturers, and OSAT vendors.

Faculty: Anthony Kuo, National Yang Ming Chiao Tung University (NYCU)

Supply Chain Management (SCM) is the backbone of virtually every industry, ensuring the smooth flow of goods and services from raw material suppliers to end consumers. In the semiconductor industry, SCM is especially crucial due to the complexity of the production process, the rapid pace of technological advancements, and the global nature of the market. Throughout this course, we’ll explore the fundamental principles of SCM and how they apply within the semiconductor context. In particular, to truly grasp semiconductor SCM, it’s essential to understand the value addition in the broader ecosystem, which encompasses a multitude of interconnected processes, from design and fabrication to testing and distribution. Each stage in this chain plays a critical role in delivering high-quality semiconductor products to market. In addition, in recent years, the semiconductor industry has witnessed a paradigm shift towards platform-based approaches to product development and collaboration. Platforms serve as catalysts for innovation, enabling companies to leverage shared resources, technologies, and expertise to accelerate time-to-market and drive down costs. Finally, we’ll reflect on some of the overarching mega-trends that are reshaping the landscape of supply chain management. Therefore, learners can position themselves and their organizations for success in an increasingly dynamic and competitive marketplace. 

Learning Outcomes: 

• Recognize the critical role of Supply Chain Management (SCM) and grasp fundamental SCM principles. 
• Identify the unique features and challenges inherent in semiconductor SCM and be prepared to develop strategies to address them effectively. 
• Understand the components of the semiconductor value chain and analyze its ecosystem dynamics to enhance competitiveness and innovation. 
• Analyze the impact of semiconductor platforms on collaboration, innovation, and efficiency within the industry value chain. 

Faculty: Kuan-Cheng Huang, National Yang Ming Chiao Tung University (NYCU)