Communications of the ACM
Quantum Materials: A New Paradigm For Computing?
Credit: iStockPhoto.com
Moore's Law enabled smaller, cheaper, faster electronic devices for five decades, but it will take a new paradigm like quantum materials to make the next technological leap, Materials Processing Center Director Carl V. Thompson told the annual Materials Day Symposium at MIT.
A new family of quantum materials, including graphene, hexagonal boron nitride, and molybdenum disulfide, and nitrogen vacancy centers in diamond, are at the forefront of recent scientific research. They are being explored for their unusual electronic, optical, and magnetic properties with special interest in their potential uses for sensing, information processing, and memory.
While such materials "may not necessarily be the basis for a replacement technology, [they] will certainly be the basis for a complementary technology to conventional integrated circuit technology," Thompson said. In addition to serving as MPC Director, Thompson is the Stavros Salapatas Professor of Materials Science and Engineering at MIT.
Diamond's Promise
Diamond spintronics holds promise for a wide range of applications that include nano-magnetic imaging, gyroscopes, and quantum information processing, Element Six Chief Technology Officer Daniel Twitchen told the symposium on Oct. 14, 2015. Twitchen reported progress in growth of synthetic diamond is helping to realize applications using diamond's desirable properties that range from its strength, wide optical transparency, chemical and biological stability, to having the highest bulk thermal conductivity of any material.
While pure diamond forms a lattice consisting only of carbon atoms, an impurity called the nitrogen vacancy defect can be created in the diamond lattice. Nitrogen vacancy center (NV center) diamond is so named because it pairs a nitrogen atom in place of a carbon atom next to an empty space where a carbon atom is missing in the lattice. Nitrogen vacancy pairs turn diamond a pink color. These nitrogen vacancy color centers can hold their quantum/spintronic properties for up to several seconds at room temperature. "If I apply a microwave sweep across this, while looking at the light as I'm exciting, coming off from these states, as I hit the zero state, it's bright," Twitchen said. "As I hit the one state, it's darker, so I have this optically detected magnetic resonance signal, and that's my readout."
Sensors Everywhere
Tomás Palacios, Associate Professor of Electrical Engineering and Computer Science, sees an opportunity for 100 times more electronics by embedding two-dimensional (2-D) materials into every object people use or wear. Among the wired objects Palacios envisions are: electronic wallpaper and desks to charge devices like cell phones wirelessly; ceilings that light up to replace traditional lighting; windows that double as transparent displays; large area distributed speakers; and sensors everywhere.
"I strongly believe that two-dimensional materials are a key element for the future of microsystems that will be able to bring electronics to every object and increase the performance density of electronics, not by 10 percent but by five or six orders of magnitude," Palacios said. His lab grew single-layer graphene on copper foil with chemical vapor deposition and developed a robotic system to transfer graphene to any substrate including paper and clothing. Performance of 2-D materials is now good enough for many applications, he said. Graphene, he noted, is 200 times stronger than structural steel and conducts electricity better than any metal. Graphene products include high-performance tennis racquets and conductive ink for flexible electronic systems. His latest work involves folding 2-D materials to demonstrate the next generation of micro and nano systems.
Material Revolution
Pablo Jarillo-Herrero, Mitsui Career Development Associate Professor of Physics at MIT, said layered atomically thin materials are a revolution in condensed matter physics. He discussed his work showing layers of graphene and hexagonal boron nitride create a moire pattern and having an electronic bandgap. This moire pattern, which is created by the overlapping honeycomb structures of these ultrathin materials, varies with the twist angle of the graphene/hexagonal boron nitride layers, and it turns out their bandgap varies with the twist angle of their crystal structures. "What we hypothesized is that there is actually a bandgap which is opening at the charge neutrality point and it is due to the coupling between this graphene and this hexagonal boron nitride," Jarillo-Herrero said.
Jarillo-Herrero's lab at MIT also created the world's thinnest diode with a related ultra-thin semiconductor material, single-layer tungsten diselenide, less than 1 nanometer in thickness. This single-layer tungsten diselenide diode can also act as a photo-detector and a light emitting diode (LED).
Exploring Insulators
Nuh Gedik, the Lawrence C. (1944) and Sarah W. Biedenharn Career Development Associate Professor of Physics at MIT, explained his groundbreaking work on materials such as bismuth selenide, which are known as topological insulators because they are electrically insulating in their bulk but freely conducting along their edge. "On the surface, if you apply an electric field, you get a magnetic response and vice versa," Gedik said. "You can think about switching memory bits using this effect, etc."
Gedik combines ultrafast laser optics with angle resolved photoemission spectroscopy (ARPES) to measure surface electrons in a topological insulator. Gedik showed a movie from his lab experiments that he described as the first demonstration of Floquet-Bloch bands in a solid. "You can see that as the light comes in, you see replicas of the electronic dispersion forming in energy and as the light goes out, they go out. When the light is there, you see formation of these extra coherent bands," Gedik said. "The photons that you are sending are hybridizing with the electrons, and they form these hybrid bands, photon-electron bands, which for all practical purposes actually behave like real bands." Subsequent experiments demonstrated that circularly polarized light breaks time-reversal symmetry and opens a bandgap in the material.
Nanoscale Magnetism
Harvard University Professor of Physics Amir Yacoby described his work to develop a new sensor, based on nitrogen-vacancy center diamond, to measure magnetic fields and capture the structure and charge of skyrmions, which are of potential interest for computing. He likened this to reinventing magnetic resonance imaging (MRI) for the nanoscale. Because of the tiny size of many new materials, such as a flake of graphene, there isn't a large enough sample to apply many traditional measurement tools.
"What we're hoping in developing these kind of new approaches to characterize materials is that it will give us new insights into the material properties but also aid in our ability to design materials with pre-determined properties," Yacoby said. He applied the lessons of quantum information science to make the sensor for detecting magnetic fields. "If these qubits are so sensitive to their environment, why not use them as really sensors of the environment?" Yacoby said. "The detection initialization is done optically and the control is done using microwaves," he said. "We have a very simple way of comparing what the signal is relative to the uncertainty and this gives us our sensitivity." Using a scanning tip, his group imaged a static electron spin in space, he said. The new tool can be used to investigate defects in materials, interfaces between materials, and nuclear spins inside solids.
Beyond Boolean Logic
George Bourianoff, a Special Consultant to Intel's Microprocessor Research Labs, said as more and more people and devices connect to the Internet of Things, computers that run the Internet will need to move beyond exact Boolean logic functions toward cognitive functions. "What you'd like in beyond-CMOS computing are patterns and image streams, not bits and bit streams. We want approximate and probabilistic; we want nonlinear operators; and then we want complex connection topologies which can be learned and suited to the task at hand," Bourianoff said.
One possibility for this new paradigm is reservoir computing, and skyrmion arrays are one possibility for implementing reservoir computing. "The basis of reservoir computing is that the dynamical system reduces multiple image streams to a single snapshot which is much simpler to recognize," Bourianoff said. "Reservoir computing has a mathematically rigorous foundation." One possibility to implement reservoir computing is magnetic skyrmion arrays. He defined magnetic skyrmions as chiral spin structures within a spin texture that has a whirling configuration. "They can encode as little as one quanta of information or they can encode thousands of quanta. They can be created and read and ideally provide a physical system to implement reservoir computing," he said.
Solving the Unsolvable
Nitrogen-vacancy center diamond pillars can provide unlimited memory storage, Center for Integrated Quantum Materials (CIQM) Co-Principal Investigator Gary L. Harris told the Symposium. Harris, Dean of the Graduate School at Howard University in Washington, D.C., developed a new technique to grow NV center diamond pillars for use in qubit (quantum bit) applications. These pillars can be formed in an array around a wafer. "The idea would be to make all of these little memory cells, and so we've solved the unsolvable problem because we have an infinite amount of memory storage and capability," Harris said.
The process uses self-assembled diblock copolymer masks to grow pillars of diamond with nitrogen-vacancy centers. Nitrogen-vacancy color centers create a single spin system that can be easily read. "You bring in light at one frequency, you get out light at another frequency. You bring green light in, you get red light out. The intensity of the red light gives you an idea of what the state of spin of the system is," Harris said. "If you are going to make this diamond technology viable, you're going to have to have large scale substrates. That was the secret weapon in the development of silicon technology. The size of the substrate grew and grew and grew," he said.
For the "Quantum Materials" symposium, MPC collaborated with the Harvard-based Center for Integrated Quantum Materials. The multi-site CIQM, a five-year, National Science Foundation-funded project, is led by Robert M. Westervelt, Mallinckrodt Professor of Applied Physics and of Physics, at Harvard University. MIT participation in CIQM, led by Co-Principal Investigator Ray Ashoori, who is a Professor of Physics, is managed through the Materials Processing Center. About 223 attended the Materials Day Symposium.
No entries found