Applied Materials launches "new strategy" for process control based on big data and artificial intelligence

Latest update time：2021-09-06 08:30

Reads：

News

Ying·News| Push the latest news of Applied Materials in the world and China to you in a timely manner. During SEMICON China 2021, Applied Materials announced its major innovation in process control.

News Highlights

New Enlight ^® Optical Wafer Inspection System Combines Breakthrough Performance with New Optics to Capture More Yield Data on Every Wafer

ExtractAI ^™ technology uses artificial intelligence to quickly classify yield-killing defects and eliminate noise

Applied Materials' fastest-growing inspection system ever enables customers to accelerate process node advancements, accelerate time to volume production, and maintain higher yields

SANTA CLARA, Calif., March 16, 2021 – Applied Materials, Inc. today announced major innovations in process control that use big data and artificial intelligence to help semiconductor manufacturers accelerate node advancements, speed time to profitability and generate more profits throughout the lifecycle of a technology node.

Semiconductor technology is becoming increasingly complex and expensive, and reducing the time required for advanced technology node development and capacity growth is worth billions of dollars to global chip manufacturers. As line widths continue to shrink, once harmless tiny particles become defects that affect yield, making detection and defect correction increasingly difficult, and the ability to deal with this problem is the key to success. Similarly, the formation of 3D transistors and multiple process technologies have also brought subtle changes, resulting in an exponential increase in defects that reduce yield, and solving these difficult and time-consuming defects is the core of this technical research.

Applied Materials is addressing these challenges with a “new strategy” for process control that brings the power of big data and artificial intelligence to the heart of chipmaking technology . The solution consists of three components that work together in real time to find and classify defects more quickly, accurately and economically than traditional methods . The three components are:

New Enlight ^® Optical Wafer Inspection System

Five years in development, the Enlight system combines industry-leading inspection speed, high resolution and advanced optics to collect more yield-critical data with every scan.

The Enlight system architecture improves the economic benefits of optical inspection, making the cost of capturing critical defects per wafer three times lower than other similar inspection methods.

Through significant cost optimization, the Enlight system enables chipmakers to add more inspection points in the process flow. The resulting availability of big data enhances "inline monitoring", a statistical process control method that predicts yield excursions before they occur, detects excursions immediately, stops wafer processing to protect yield, and quickly traces the root cause of defects, quickly corrects and continues mass production.

The Enlight ^® optical wafer inspection system combines industry-leading inspection speeds with high-performance optics to deliver a breakthrough in capturing cost-effective yield-reducing defects.

New ExtractAI ^™ Technology

Developed by Applied Materials data scientists, ExtractAI technology solves the toughest wafer inspection problem: quickly and accurately identifying yield-reducing defects among the millions of unwanted signals, or “noise,” generated by high-end optical scanners.

ExtractAI technology is an industry-unique solution that connects big data generated by optical inspection systems in real time with e-beam inspection systems that can classify specific yield signals, inferring that the Enlight system resolves signals for all wafer maps, distinguishing yield-reducing defects from noise.

ExtractAI technology is so efficient that it can characterize all potential defects on a wafer defect map by only testing 0.001% of samples. This allows us to obtain an actionable, classified defect wafer map, effectively improving the speed, ramp, and yield of semiconductor node development. Artificial intelligence technology can adapt and quickly identify new defects during mass production, and its performance and efficiency are gradually improving as the number of scanned wafers increases.

SEMVision ^® Electron Beam Inspection System

The SEMVision system is the world’s most advanced and widely used e-beam inspection system. Based on industry-leading resolution, the SEMVision system trains the Enlight system with ExtractAI technology to classify yield-reducing defects and distinguish them from noise.

Working together in real time, the Enlight system, ExtractAI technology, and SEMVision system can help customers identify new defects in the manufacturing process, thereby improving yield and profits.

The widely installed SEMVision G7 system is now compatible with the new Enlight system and ExtractAI technology.

Applied Materials’ “New Strategy” for process control brings big data and artificial intelligence technologies to the forefront of chip manufacturing success. Key to this innovation is the new ExtractAI ^™ technology: an industry-unique solution that connects big data generated by Enlight® ^optical inspection systems and SEMVision® ^electron -beam inspection systems in real time to automatically and flexibly identify and classify yield-reducing defects faster, more efficiently and more cost-effectively than traditional methods.

For more than 30 years, fab engineers have struggled with how to quickly and accurately distinguish yield-reducing defects from noise. Applied Materials' Enlight system with ExtractAI technology is a breakthrough product that addresses this challenge. As the system is used more, the artificial intelligence is trained to become smarter, which can increase chipmakers' profits per wafer over time.

—— Chairman and CEO of VLSIresearch

Dan Hutchison

Applied Materials' 'New Strategy' for process control combines big data and artificial intelligence to deliver an intelligent and adaptive solution that helps customers save time and maximize yield. Combining Applied Materials' best-in-class optical inspection and e-beam inspection technologies, we have introduced an industry-unique intelligent solution that not only detects and classifies yield-destroying defects, but also learns and adapts to process changes in real time . This unique capability enables chipmakers to ramp up new technology nodes faster and efficiently capture yield-destroying defects throughout the process lifecycle.

——Group Vice President, Applied Materials

General Manager of Imaging and Process Control Division

Keith Wells

The new Enlight system with ExtractAI technology is the fastest growing inspection system in Applied Materials’ history and is already in production at customers’ logic nodes in the world’s leading foundries. For more than 20 years, SEMVision systems have been the industry’s leading e-beam inspection tools, with more than 1,500 systems deployed in customer fabs around the world.

For more information in English, please click "Read original text" below

About Applied Materials

Applied Materials, Inc. (NASDAQ: AMAT) is a leader in materials engineering solutions that power nearly every new chip and advanced display produced worldwide. With technologies that can transform materials at the atomic level at scale, we enable our customers to realize what’s possible. Applied Materials believes that our innovations will drive the advanced science of the future. For more information, please visit:

www.appliedmaterials.com .