Over the past few decades, the upgrading of electronics has become so common that consumers expect this year’s technology to have significant increases in speed and capacity than the previous year. A large part of the constant increase in functionality is dependent on memory, which is crucial to the processing and storage of information.
Modern electronics use semiconductors—crystalline or amorphous solids with distinct electrical characteristics—for memory. The favorite semiconductor is silicon, but there’s a limit to its usefulness. “Right now semiconductor memory is very cheap and very dense,” says Robert A. Buhrman, Applied and Engineering Physics. “But we’re running into a fundamental problem: we can’t make it much smaller. So for the past two decades, many of us have been working on new types of magnetic memory, which have the potential to be smaller, faster, and denser than semiconductor memory.”
What’s Magnetic Memory?
Semiconductors and other electronics depend on harnessing the electrical charges of electrons passing through circuits, but Buhrman’s area of research—known as spintronics—seeks to harness the spin of an electron. Spin is the action of an electron spinning on its axis while rotating around the nucleus of an atom, similar to a planet spinning as it orbits a sun. Electrons may spin in two possible directions, up or down, and this spin gives magnetic properties to the electron. One aspect of this is the magnetic moment, the tendency of the electron to align with a magnetic field.
Some of Buhrman’s earliest work in spintronics was a collaboration with Daniel C. Ralph, Physics, to investigate spin transfer torque, the process that allows for the manipulation of magnetism and the magnetic moment. To study spin transfer torque, researchers run an electrical current through a ferromagnet, a substance with a high susceptibility to magnetism, and that causes the electrons to align their magnetic moments in the same direction. “Each electron now becomes a little magnet,” Buhrman explains. “When it flows out and hits another ferromagnetic material, it exerts a force on that material and causes the material to flip its orientation. That’s the basis for magnetic memory.”
Magnetic Memory versus Electronic Memory
Magnetic memory is denser than electronic memory, allowing it to retain more information. It is also nonvolatile, meaning it can retain data even if there’s a break in the power supply. “Right now we have a form of electronic memory called flash, which is used in things like thumb drives,” Buhrman says. “Flash memory is nonvolatile, but it is also relatively slow, and you can only write and erase from that memory a certain number of times before the material wears out. The magnetic memory we are working on would not only be faster, it would be infinite in endurance.”
Magnetic memory would also be more energy efficient, which is another of its advantages. ”There’s a great desire to have chips that consume less energy,” Buhrman explains. “Partly this is because we want to have green technology and partly it’s because we don’t want our cellphone or laptop battery to run out of charge, especially if we’re going to have an internet of things where everyday objects have network connectivity.”
“Each electron…becomes a little magnet,” Buhrman explains. “When it flows out and hits another ferromagnetic material, it exerts a force on that material and causes the material to flip its orientation. That’s the basis for magnetic memory.”
More recently, Buhrman and his group have investigated the spin Hall effect, which is a way to split electrons into two groups, those that spin up and those that spin down. When these groups of electrons are run along a ferromagnet, the magnetic poles can be flipped back and forth. “That can reverse the magnetic moment at even higher speed and lower energy cost,” Buhrman says.
Manipulating Magnetic Moments
Buhrman and his coresearchers discovered that the spin Hall effect in tantalum, a corrosion-resistant metal used in many electronics, is twice as strong as it is in any other previously tested metal. This was a major finding that suggests ways to manipulate magnetics moments—switching poles back and forth—with greater efficiency. The researchers wrote a paper on their findings and submitted it to the prestigious journal Science. The three initial pre-publication reviews surprised Buhrman. “I have never gotten such three uniformly and very positive reviews,” he says.
The paper was published in 2012 and has since been cited 500 times. “It was the culmination of work that I, along with the students and postdocs in my group and our long-term collaborator Dan Ralph, had been doing for years. It was a great pleasure to see it,” Buhrman says. “As applied physicists and engineers, we do things for scientific discovery, but it’s also nice when there’s a chance our work will have a technological impact that will matter to the world at large.”
Scientist and Administrator for Collaborative Research
Along with his research, Buhrman has also made an impact through his administrative efforts. In 2001, he was the leader of the team that won multimillion-dollar funding from the National Science Foundation for what would become the Cornell Center for Nanoscale Systems (CNS). Cornell’s proposal for the center was an answer to then President Bill Clinton’s call in 2000 for a National Nanotechnology Initiative. “The government wanted new research centers where people would collaborate on nanoscience and nanoengineering to look at things on the near-atomic scale,” Buhrman says. “We were one of six centers that were funded, out of 90 that applied.”
Buhrman became the director of the CNS from 2001 to 2007. After that, he took on the position he still holds today as senior vice provost for research. His office oversees and supports all aspects of research at Cornell—from assisting individual faculty members who are pursuing research funding, to the oversight of campus research facilities.
These administrative positions underscore Buhrman’s interest in collaborative research. “Every university I visit says they’re collegial and collaborative, but Cornell really is,” he says. “In the areas where we are the strongest, we have faculty who are real collaborators. That’s why we were so successful with the CNS. We had a great bunch of people that enjoyed working together. We each did our own research to some degree, but we were all on the same page. It was a pleasure for me to work with them to put together the proposal for the center, get it funded, set it up, and operate it.”
Serving Cornell Faculty Research
In his role as senior vice provost for research, Buhrman has been working to streamline and optimize the process researchers at Cornell must go through to apply for grants, receive funding, and conduct their research. He’s also been working to raise awareness among funders and others about the important findings coming out of Cornell. “We’ve been doing all we can to convince funding decision-makers that the money the nation spends on research is well spent and that Cornell is a particularly effective and efficient place to spend it,” Buhrman says. “My hope is that 50 years from now Cornell will still be an indispensible research institution.”
Despite the extensive time and commitment demands of his position as senior vice provost, Buhrman continues to carry out his own research. “It’s important that I stay active in research,” he says. “My experience as a faculty member, researching in applied physics may not be the same as the experiences of someone in the veterinary college, for example, but I still have to go through the process of writing proposals, applying for grants, and such. That means I’ve got more in common with other faculty doing research than I would if I hadn’t written a proposal in the past 20 years. I know what they’re up against when they’re trying to get funded, and I want to help them be successful.”