Empowering Computing Capacity – Electronics & TechnologyElectronics & Technology

When Matthew Marinella left his research job at Sandia National Laboratories to become an associate professor of electrical engineering at Arizona State University, he did not abandon his Sandia ties.

Since joining ASU’s Ira A. Fulton Schools of Engineering faculty, Marinella has continued to collaborate with Sandia on his research. As part of Sandia’s Grand Challenge series of scientific research, for example, he is the lead investigator on a project to increase computing power by making radiation-tolerant or “radiation-tolerant” electronic devices more power-efficient. The work at ASU builds on previous electronics research done by Marinella at Sandia.

Radiation hardening is a process that increases the durability of electronic products used in high radiation environments such as outer space. This can keep critical computer components, such as those on spacecraft, working where normal electronics fail due to radiation exposure.

Sandia’s Grand Challenge project is considered high-risk, high-reward and is funded for three years.

Said Marinella, “I started working at ASU in January 2022. She has been working on large projects like this at Sandia for many years.” “It’s finally funded, so of course I hope this will lead to bigger projects.”

Also participating in this Grand Challenge study are collaborators from the University of California-Berkeley, the University of Texas at Austin, and the University of Michigan. Rick McCormick, a senior managing sponsor of the project and a strategic customer of Sandia’s research, says the agency’s researchers have joined for their expertise in a new analog computing device that is a key part of the project.

Use in radiation-rich environments

To increase the computing efficiency of radiation-tolerant electronic devices beyond their existing capabilities, the research team is developing new analog devices using resistive random access memory (ReRAM) and electrochemical random access memory (ECRAM). These analog devices are combined in an array arranged on a complementary metal oxide semiconductor (CMOS) computer chip manufactured by Taiwan Semiconductor Manufacturing Co.

Once developed, these devices will be used in radiation-rich environments both in space and on Earth.

“If you’ve ever seen Chernobyl or anything like that, there are robots that try to go to places you don’t want to put people in,” Marinella says of the use of Earth-tethered devices.

Satellite cameras are another example of what increased processing efficiency can do for you. A satellite’s capabilities are limited by parameters such as size, weight and battery power. Making onboard computing more efficient, such as the technology the Marinella team is developing, frees up power for other tasks, such as increasing the resolution of satellite imagery.

Efficiency memory chips

Marinella said the Department of Defense is interested in radiation-enhanced computing technology, including uses such as image processing using edge computing. In edge computing, data is processed on computer systems immediately after it is collected, reducing the number of large raw data files that need to be transferred and speeding up file exchange.

The chips are currently intended for use in radiation-tolerant applications, but Marinella sees arrays of efficiency-boosting memory chips combined with conventional semiconductors to eventually become ubiquitous in consumer electronics.

“These will be chips in cell phones, self-driving vehicles and cloud computing systems,” he says.

Data transfer between separate elements

According to Sapan Agarwal, Sandia’s lead investigator on the grand challenge project, integrating memory devices with existing chips will push the boundaries beyond what is possible with existing chips alone. The processing efficiency of conventional CMOS chips is limited by the size and voltage of the transistors.

Agarwal says one of the main factors slowing down computing with conventional chips is the need to transfer data between separate elements for memory storage and processing. The integration of the team’s new ReRAM and ECRAM devices with CMOS chips allows processing and memory storage to all be done in one place. According to Agarwal, this would lead to 100 times higher computing power per watt than is currently possible.

pin field effect transistor

McCormick said it will help the team experiment with different analog computing devices to figure out what works best for their application. Marinella is also Hugh Barnaby, an ASU professor of electrical engineering, understands the radiation effect on transistors known as fin field effect transistors, or FinFETs, which are more efficient than older standard node transistors. The project also collaborates with and is funded by Sandia, but differs from Marinella’s work in the Grand Challenge program, which focuses on integrating CMOS chips with an array of analog devices.

The Grand Challenge project combining CMOS chips and analog devices does not use FinFETs in its research, but Marinella’s future goal is to use FinFETs as transistors for CMOS chips and integrated analog arrays, should FinFETs prove to be sufficiently radiation resistant in extreme environments. is to use it. Environment. A more efficient combination would yield far more radiation-enhanced computing power, making the system far more efficient than what the Grand Challenge CMOS and analog memory consolidation projects are currently investigating.

Support for the development of key supporting technologies

McCormick says that after hiring Marinella at ASU, followed by her stellar innovation record and postdoctoral training at Sandia, both institutions are well positioned for research collaboration opportunities.

“Matt has already connected us with other influential professors at ASU and has maintained a strong working relationship with Sandia,” he says. “We look forward to continuing to work with him and help him train the next generation of Sandia researchers.”

Agarwal agrees that the partnership with ASU is a great opportunity to expand Sandia’s research capabilities.

“The team at ASU is an integral part of our broader research efforts, helping to develop key enabling technologies and collaborating with us on our broader emerging microelectronics strategy,” he says. “At ASU, Matt continues to be an important collaborator.”