University receives $20 million grant for cloud-based semiconductor lab

A team of Virginia Tech researchers received a $20 million grant from the U.S. National Science Foundation to establish an artificial intelligence (AI)-enabled cloud-based semiconductor laboratory.

The Virginia Tech Programmable Cloud Laboratory (PCL) Node will allow the team to make the university’s semiconductor facilities remotely accessible to a range of users as part of a nationwide network of AI-enabled automated laboratories. As one of 20 teams to receive funding in what the National Science Foundation calls "a core ... contribution to the U.S. government's Genesis Mission, a national effort to harness AI for scientific discovery," Virginia Tech will use its PCL Node to strengthen U.S. semiconductor innovation, manufacturing resilience, and workforce development.

“The chance of this serving the larger population is great,” said Professor Rose Hu, head of the Bradley Department of Electrical and Computer Engineering (ECE) and the project’s principal investigator. “Ultimately, the goal is for all 20 nodes to be made accessible across the entire United States. Broadened access is the whole purpose of building this powerful network.”

Hu compared the project to the computer networking experimentation done throughout the late 20th century: What began as four computer nodes initially restricted to academic research later became the internet as we know it today. 

In the same way, the team anticipates that the new network of remote laboratories, including Virginia Tech's PCL Node, will expand job opportunities and access to scientific research, especially for underserved rural communities.

Students from a variety of disciplines and universities will be able to access the node to conduct research and build skills with advanced semiconductor equipment unavailable at their home institutions. The node also will be accessible to startups and small businesses without their own equipment through the Small Business Innovation Research and Small Business Technology Transfer programs, two federal initiatives that support research development and technology commercialization. Rather than unsupervised, remote operation, this access would be secure, structured, and supported by trained facility personnel.

The research team

Hu is collaborating closely with faculty members and co-principal investigators across Virginia Tech, including:

• Wei Zhou, associate professor in electrical and computer engineering, director of the Micro-/Nanofabrication Cleanroom and Laboratory, and special advisor to the senior vice president for research and innovation for AI-guided semiconductor manufacturing and nanotechnology 
• Zhenyu “James” Kong, the Ralph H. Bogle Jr. Professor in the Grado Department of Industrial and Systems Engineering and an executive member of Virginia Tech Made
• Matthew Hull, research professor and director of the Nanoscale Characterization and Fabrication Laboratory within the Institute for Critical Technology and Applied Science
• Alberto Cano, associate professor in computer science and associate vice president for Advanced Research Computing

Connecting semiconductor design, fabrication, and testing

Semiconductors are tiny devices that form the foundation of modern electronics, from smartphones and data centers to vehicles and medical systems. Developing new semiconductor devices often requires iterative, expert-guided work in design, fabrication, characterization, and data analysis. These cycles can be time-consuming and difficult to reproduce across tools and facilities.

The primary goal of Virginia Tech’s PCL Node is to transform these labor-intensive work cycles into AI-guided workflows that make micro/nanodevice production faster, repeatable, and easier to predict. The node will use resources from the university's Micro-/Nanofabrication Cleanroom and Laboratory, Nanoscale Characterization and Fabrication Laboratory (NCFL), and Advanced Research Computing to accomplish this. 

“For nearly 20 years, the NCFL has provided critical infrastructure and personnel supporting some of Virginia Tech’s most prestigious research programs,” Hull said. “For the PCL node, our NCFL team will support users’ analytical needs and leverage our unique experience scaling networked user access programs through the National Nanotechnology Coordinated Infrastructure.”

The node's shared infrastructure includes advanced fabrication tools, digital twin models, high-performance computing, a sharable data system, and more. Zhou said the facilities derive value not only from specialized instruments, but also from validated process recipes, experienced engineers and staff, calibration and maintenance programs, contamination controls, environmental health and safety procedures, and coordinated fabrication-to-characterization expertise.

“The machines in ECE’s clean room, together with the NCFL and Advanced Research Computing, provide a nanofabrication, characterization, and computing infrastructure useful for semiconductor device innovation,” Kong said. “But they’re currently in a fragmented setting and challenging to use in a connected and holistic way. This program is meant to bring siloed tools and devices online and apply AI machine learning to make processes more automated, easy to use, and remotely accessible.”

Accelerating scientific innovation and discovery

Virginia Tech’s investments in research computing and AI can directly impact scientific innovation, Cano said. The team’s responsibility is to provide infrastructure that enables AI-driven science and supports a national research community working together to facilitate discovery. 

This work to improve semiconductor manufacturing across the country has the potential to translate into increasingly efficient domestic supply chains and more reliable, less expensive technology, including laptops, cellphones, and medical imaging tools.

“We’re excited to bring Virginia Tech’s nanofabrication clean room facility and advanced materials-device characterization capabilities into a new era of research and deepen collaboration among universities, national laboratories, and industry,” Zhou said. “The improved infrastructure is expected to accelerate research, broaden access to advanced tools, prepare an AI-ready workforce, and translate lab discoveries into reproducible and scalable technologies.”

The PCL Node is led by Virginia Tech in collaboration with various partners.