Prolific IBM inventor takes on Penn State Research: Let's make this into Happy Silicon Valley

By Stephanie Kalina-Metzger

Photo: Provided.

All it took for Madhavan Swaminathan was a little encouragement to jumpstart his career in innovation. The new head of the Department of Electrical Engineering in Penn State’s School of Electrical Engineering and Computer Science comes to Penn State by way of Georgia Tech, where he spent 28 years leading the 3D Systems Packaging Research Center and the National Science Foundation Center for Advanced Electronics through Machine Learning.

Swaminathan, who was born in India, said that innovation wasn’t really encouraged when he was young, but that all changed when he came to the United States, earned a Ph.D. in electrical engineering from Syracuse University and joined IBM.

“At that time, IBM was investing in the next supercomputers, when I was asked to develop new packaging technologies and my manager told me to concentrate on developing innovative solutions in that area,” said Swaminathan.

For those who might conjure up the idea of boxes and bubble wrap when it comes to packaging, Swaminathan explains that the packaging he speaks of is an entirely different concept. “If you view any electronic system, it has two parts. The first part are the transistors within a chip and the second part is the packaging. The transistors within the chip act as the brain and then there’s everything around the chip that supports its function, similar to the human body,” he said.

Swaminathan goes on to explain how minute these transistors are. “There can be one billion transistors within a chip and those transistors have to communicate within the external world and with other chips as well. To reduce the size of the system, you have to work on new packaging technologies,” he said, adding that packaging is very important in 5G and 6G communications and computers, where efficiency and battery life are important to send the electrons back and forth between chips.

“Students these days are very different. They are very much interested in startups and commercializing ideas, but college faculty members don’t always have an eye towards business. You need to have both sides and I intend to meet with faculty to encourage them towards providing a holistic education focused not only on research, but also on marketing and business."

Being given the green light to begin innovating at IBM was a boon for Swaminathan and he took to the task with enthusiasm. From 1991 to 1994, while working there, he registered six patents related to advanced packaging.

Later, during a 28-year career at Georgia Tech as the John Pippin Chair in Microsystems Packaging and Electromagnetics at the School of Electrical and Computer Engineering, he developed 25 additional patents, mainly related to advanced packaging and heterogeneous integration of semiconductors.

“At Georgia Tech, we were studying new technologies for packaging and were licensed to create companies,” said Swaminathan, adding that Jacket Micro Devices was just one of those companies that began under his purview and was later acquired by a larger company.

Throughout all of these achievements, what excites Swaminathan most, he said, is living between the worlds of innovation and commercialization. “Being able to commercialize that which emerges from leading-edge research is a very important component, otherwise these great ideas can get lost,” he said.

A Penn State recruit

Swaminathan said that he wasn’t actively seeking a job when Penn State came calling. “I was actually very happy where I was at the time, but I was encouraged to apply via a search firm, so I did.”

“I can’t see any reason why Happy Valley can’t one day be known as Happy Silicon Valley."

Swaminathan explains that he made it through the first round and a final interview before the Dean contacted him and asked him what it would take to convince him to relocate to Happy Valley.

“Continuing my research was very important to me and what struck me about Penn State were the people. I connected with them and their dedication to cutting-edge research was truly outstanding, which is why I thought, why not?” he said.

Swaminathan brings with him to Penn State a large packaging research center, which will include 13 other universities and which he will continue to lead, paving a path for his students.

“Students these days are very different. They are very much interested in startups and commercializing ideas, but college faculty members don’t always have an eye towards business. You need to have both sides and I intend to meet with faculty to encourage them towards providing a holistic education focused not only on research, but also on marketing and business,” he said.

“Being able to commercialize that which emerges from leading-edge research is a very important component, otherwise these great ideas can get lost."

The reason Swaminathan cites commercialization as high on his list of priorities is because of the challenges faced within the United States.

“Whatever is done in the lab needs to translate into manufacturing. Right now, there are two challenges. [The first is] reshoring. Currently most manufacturing in semi-conductors happens outside of the United States. That needs to come back. Also, there’s the supply chain problem. The U.S. produces only 12% of chips here, while consuming more than 50%,” he said.

Swaminathan said that he’s looking forward to living and innovating in Happy Valley and is optimistic about the future.

“I can’t see any reason why Happy Valley can’t one day be known as Happy Silicon Valley,” he said with a smile.

FREE intrapreneurship conference focuses on employees as change makers

Mark your calendars for the Conference on Intrapreneurship Employees as ChangeMakers, taking place May 11 at the Eric J. Barron Innovation Hub. The conference, which is free, is presented by the Penn State College of Engineering’s School of Engineering Design and Innovation and will focus on how employees can create innovation from within a company.

Registration is limited to 80 attendees. RSVP NOW to save your spot! You can register here or at the QR code below. For questions, please contact Frank Koe, teaching professor, at ftk2@psu.edu.

"Innovation is a popular word these days,” organizer Frank Koe said. Koe is a teaching professor of engineering entrepreneurship. He said that, while all companies see themselves as 'innovative,' he is curious about how they define the term in their own firms.

“A recent Gallop poll discovered that nearly 54% of employees have been found to be unengaged at work,” he said. “Everyone matters these days and the companies that can bring out the best in everyone and see that employees have more to offer than fulfilling a job description to the letter will innovate at the speed of change and be true innovators."

The conference will bring together small and medium-sized businesses to discuss the inclusion of intrapreneurship within their company cultures.

Group discussions will include:

Registration is limited to 80 attendees, so RSVP now to save your spot! You can register here or at the QR code below. For questions, please contact Frank Koe, teaching professor, at ftk2@psu.edu.

Penn State’s AIMI ushers in Industry 5.0 with a bold vision for the future

By Caryn Anderson

AIMI connects you with Penn State. Photo: Provided.

Penn State’s Center for Applications of Artificial Intelligence and Machine Learning to Industry (AIMI), which launched on June 30, 2022, has a complex, multifaceted goal. In part, AIMI aims to help small and medium manufacturers (SMMs) and small and medium enterprises (SMEs) get the resources they need to leverage the power of artificial intelligence and machine learning. Additionally, the center strives to secure an indigenous supply chain, support robust workforce development and improve education and economies throughout the country.

To learn more about AIMI, we sat down with its director, Soundar Kumara, a pioneer in nonlinear real-time data analysis for the manufacturing industry, and assistant director Gretta Kellogg.

Meet the industry powerhouses behind AIMI

Kumara has been working in this space for decades, with research and teaching excellence that contributed to his recognition as one of Smart Magazine's "20 Most Influential Academics" in 2021. Widely considered the "father of smart manufacturing in the current era," Kumara also pioneered teaching manufacturing-related AI, databases and data analytics to undergraduates and graduates, while addressing the interdisciplinary approach needed for successful smart manufacturing.

Kellogg worked as a program manager for two Genomic Centers at Penn State. Her work was so successful that she was recruited to launch Cornell University’s Epigenomics Core facility. Then, she was asked to return to Penn State specifically to work with Kumara and support AIMI’s mission.

“The background that [Kumara] provides, his experience and his professional expertise over the decades is really important because I have never known a center or core facility that is able to launch without relying upon the expertise of the initial leader in that space,” said Kellogg.

"There is a passion and goal to reach out, in the state of Pennsylvania in particular, to fulfill our land grant mission [with] the small and medium manufacturing organizations. AIMI strives to help them get to the place where they can do advanced manufacturing and become leaders in the world.”

She added, “AIMI’s success really depends upon the breadth and depth of his knowledge in AI and ML and his ability to apply that specifically to manufacturing. There is a passion and goal to reach out, in the state of Pennsylvania in particular, to fulfill our land grant mission [with] the small and medium manufacturing organizations. AIMI strives to help them get to the place where they can do advanced manufacturing and become leaders in the world.”

But when ICDA initially approached Kumara to help AIMI get started, he was hesitant. After heading another center in the past, which he eventually left after spending too much time fundraising instead of performing research, he didn’t want to risk a repeat of history.

Kumara noted, “But then I realized this was an opportunity to parlay my experience and expertise into something that would make a difference. After talking to my friends around the country, I could see that SME, AI and ML are the future — and that excites me.”

Taking small and medium industries into the future

AIMI’s slogan, “We are the innovators for you,” reflects the center’s calling, which is to lead industry into the future. Part of that, Kumara acknowledges, is addressing small and medium industries, particularly in order for supply chains to remain local and within the United States, lessening dependence on other countries for materials and products.

“COVID has taught us that we need to be more self-reliant,” said Kumara. “We need to innovate small and medium industries, taking them to what used to be Industry 4.0, but today is Industry 5.0. Also, how do we bring SMEs together to be part of the manufacturing ecosystems? That’s where AIMI’s work begins.”

The center recently submitted a proposal to the National Science Foundation, outlining its project to revitalize underserved communities with a skilled workforce that can support a thriving United States-based supply chain for greater manufacturing resilience. This will be in conjunction with the Connecticut Center for Advanced Technology, a nonprofit organization helping Connecticut’s small and medium-sized industries. More locally, AIMI will also partner with PennTAP and Ben Franklin Technology Partners with the goal of bringing all these organizations together to fulfill Penn State’s land grant mission.

“COVID has taught us that we need to be more self-reliant,” said Kumara. “We need to innovate small and medium industries, taking them to what used to be Industry 4.0, but today is Industry 5.0. Also, how do we bring SMEs together to be part of the manufacturing ecosystems? That’s where AIMI’s work begins.”

But taking Industry 5.0 to SMEs is easier said than done. Kumara has experience working with a range of smaller organizations. What he discovered was that some smaller companies don’t even have access to the internet.

“You can’t go and just tell them they need AI and ML. They’ll show you where the door is,” he noted.

Understanding the unique problems and challenges these companies face and advancing the necessary research and upskilling to solve those challenges is mission critical, though. That’s why AIMI is bringing a consortium together to study the problems within SMEs and find a solution.

Kumara said, “It isn’t just our pipe dream; the country needs this to happen. My vision and dream is for AIMI to become the world’s leading AI and ML center, eventually becoming a global force within the next five to 10 years. We are bringing in the Commonwealth campuses and building a model with our partners to create a national force that SMEs can turn to for our expertise and resources.”

AIMI memberships benefit industry. Photo: Provided.

Immediate impact

AIMI is also considering how it can help build local economies by upskilling the local workforce, creating jobs and improving rural education.

“If we do a good job of building up small and medium enterprises across the state, then that helps the students and local communities immediately,” noted Kellogg. “Instead of having to recruit from people way outside, you can have workforce development pipelines that take advantage of the high school and college students in the community and the students that come out of our Penn State campus locations across the state.”

The benefit? Graduates can work immediately with small and medium manufacturing companies right there in their area. They don’t have to move, and the companies don’t have to recruit to go out and find them.

What about large industries?

Although AIMI will focus on SMEs and SMMs, it will also be a conduit for AI in larger industries.

“We realized that unless we build those small and medium industries, our large industries will not be successful in the next generation,” said Kumara.

He envisions in the next five years, small and medium industries will function as an extended arm of large industries. Today, a big industry might have a finance department, marketing department and assembly department. In the future, they will have an assembly SME as an extended arm of the company. Large hospital systems will tie into small hospitals.

He concluded, “If large industries really want to use AI and ML successfully, they need to advance small and medium industries. We are their in-between.”

AIMI plans to host a two-day workshop at Penn State in the spring, during which they’ll gather information from small and medium industries, to learn about their challenges and problems. As it’s still in its early stages, AIMI also partners with student groups already active in learning and applying AI and ML to various projects, including Penn State EnvironMentors, Penn State’s Readiness Institute and the Nittany AI Alliance.

To learn more about AIMI, visit https://aimi.psu.edu/ or contact AIMI to further discuss your interests.

Penn State engineers optimizing water purification system for the U.S. Marines

By Timothy Kelly

An experimental MCDI unit. Photo: Chris Arges.

A team of researchers, led by Chris Arges, Penn State associate professor of chemical engineering, is working to optimize a water desalination process called membrane capacitive deionization (MCDI) for mobile teams of U.S. Marines. We asked Arges to explain how the system works, why he chose Penn State for his research and why this work is so important. Here’s what we learned.

"The biggest draw to come to Penn State was the nanofabrication facility within the Materials Research Institute."

How MCDI works

Arges described the MCDI system like a sandwich. The two slices of bread are the electrodes. Two slices of cheese are the membranes stuck to each piece of bread. The bottom membrane only allows sodium to move through it and the top membrane only allows chloride to move through it. The water to be purified moves between the two membranes. When the electrodes are charged, the sodium and chloride move out of the water, into the membranes. The charge holds the ions to the electrodes, thus purifying the water. Purified water is collected in a container.

When the electrodes can’t hold anymore ions, the flow of water is slowed, the charge is reversed and the ions move back into the water, making it briny. The briny water is then routed to waste. As a bonus, this step generates electricity that can be stored for use in the next deionization cycle.

Making MCDI better

Arges said that his team wants to redesign the system so it can operate with more electric current flowing through the material. Pushing more current through the system purifies more water.

Increasing the surface area of the electrodes with microscopic wells allows them to hold more ions. It also increases the contact between the electrodes and the membranes, shortening the distance the ions need to travel.

“The MCDI … units could be found in homes or neighborhoods for purifying water or deployed in emergency areas, like after a hurricane, for producing potable water."

Arges further explained that the objective is to, “reduce all the resistances… [of] the bulk material [and] the interfacial resistances to a point that gets us the high current without too high of a voltage. That's… what’s really driving our research in this project.”

He continued, “By studying these interfaces and identifying the bottleneck resistance, we devise strategies to overcome the resistances. This leads to greater ion removal rates from the water during desalination.”

The Materials Research Institute housing the Nanofabrication Lab. Photo: T. Kelly.

Why Penn State?

“When starting my own independent career in 2016 at [Louisiana State University], I decided to apply my expertise in materials for fuel cells to water purification,” Arges said. But something was missing at LSU.

He said the biggest draw to come to Penn State was the nanofabrication facility within the Materials Research Institute. With tools in the Nanofabrication Lab, he said, “we use advanced metrology and fabrication techniques … to study the rate of ion transport and gas/liquid transport … at the nanoscale.”

MCDI solves smaller-scale problems, too

"It is important that we continue to improve the performance and durability of water purification units so everyone, whether rich or poor, has access to quality and potable water.”

Clean water and energy efficiency are in the DNA of Arges’s research, according to the Arges Research Lab website. As to why it’s so important to him, Arges said the water shortages around the world, which are being made worse by climate disruption, moved him to address the problem. He said, “Water is central to our way of life. It is needed for energy production, agriculture … and surviving. It is important that we continue to improve the performance and durability of water purification units so everyone, whether rich or poor, has access to quality and potable water.”

According to Arges, “By improving the materials and their interfaces in MCDI … we hope that there will be greater proliferation of these systems … since they are modular and only require electricity for water purification…” The end result could be a smaller unit with a lower capital cost for purifying water and reduced energy duty.

Even so, the MCDI system won’t solve the large-scale water problems we see around the world, including the American Southwest. Economy-of-scale makes reverse osmosis the most cost-effective solution for large, centralized desalination facilities. By contrast, “The MCDI … units could be found in homes or neighborhoods for purifying water or deployed in emergency areas, like after a hurricane, for producing potable water,” Arges said.

Finding inspiration in the tiny-but-mighty hummingbird

By Holly Riddle

You don’t need to be a mechanical engineering professor like Penn State’s Dr. Bo Cheng to recognize that hummingbirds contrast significantly with their avian relatives. The smallest of bird species, hummingbirds flit and dart their way from food source to food source in a way no other bird can. Flying backward, forward and upside down at speeds of up to 45 miles per hour, hummingbirds are the only vertebrae capable of hovering for a period of time during flight — and that’s exactly why hummingbirds have inspired modern aerial vehicles like drones.

Unfortunately, these modern aerial vehicles still have a long way to go to catch up with the biological superiority of the tiny-but-mighty hummingbird. However, Dr. Cheng’s recently published research provides new insights into how hummingbirds’ unique wing movements could be robotically replicated, making for more agile, stable and efficient aerial vehicles.

Dr. Cheng and his team reverse-engineered the hummingbird musculoskeletal system to create a model that simulates hummingbird flight motion, and found that hummingbirds’ primary muscles pull the birds’ wings in three directions, while the birds use multiple smaller muscles to tighten their shoulder joints.

In a Penn State release, Dr. Cheng detailed the findings: “It’s like when we do fitness training and a trainer says to tighten your core to be more agile. We found that hummingbirds are using [a] similar kind of mechanism. They tighten their wings in the pitch and up-down directions but keep the wing loose along the back-and-forth direction, so their wings appear to be flapping back and forth only while their power muscles, or their flight engines, are actually pulling the wings in all three directions. In this way, the wings have very good agility in the up and down motion as well as the twist motion.”

"State College is by far the best college town that I've ever lived in. It's a great place to focus on research, raise young children and explore the wilderness."

HappyValley Industry recently caught up with Dr. Cheng to learn more about his work, what it means for industry and why he chooses to live and work in Happy Valley.

What is your research topic and why is it important on a global scale? 

This research reveals how hummingbirds use their flight muscles to generate wing motions and the underlying fundamental principles, which can inform robotics on how to properly mimic hummingbird flight.

Both researchers and [the] general public are very impressed by the unique flight abilities of hummingbirds, in terms of both hovering capability and their agile maneuvers. Understanding and mimicking how hummingbird[s] fly and control their flight can potentially transform the design of aerial vehicles.

How do you envision your research impacting/changing your industry? 

This research can be potentially used to improve the drone technologies in the future, in terms of improving flight agility, stability and efficiency.

What inspired you to follow this line of research? 

This is a no-brainer. Hummingbirds are arguably the most agile fliers in nature, and state-of-the-art drone technology is far behind researching hummingbird-level flight performance.

Why did you choose to conduct this research at Penn State specifically? 

Penn State is a world-renowned research institution that attracts many talented faculty members and students. Here, I can recruit talented students [from] all around world and find world-class collaborators.

Would you encourage other researchers to make their home in Happy Valley? If so, why?

State College is by far the best college town that I've ever lived in. It's a great place to focus on research, raise young children and explore the wilderness. It is also not too far from the major cities on the East Coast.

Do you know a researcher who would be a great fit for our HappyValley Industry 5 Questions with a Happy Valley Researcher series? Let us know!