February 23, 2021 | Written by: Chris Lutz
Categorized: Chemistry | Physics | Publications | Science
Share this post:
In physics, there is a line of thinking that, if a particular phenomenon is strange enough, surely there’s a way we can put it to good use. It happened with lasers, which seemed like just another exotic quantum phenomenon in the 1960s, until we started using them in the 1980s to play music in high fidelity on compact discs.
Not every quirky physics concept can be as successful as that, but then again you never know until you investigate. Our small team at IBM Research’s Almaden lab, in the Silicon Valley, for example, has spent the past several years studying the basic underlying physics of atoms and their interactions with each other, and how this phenomenon could lead to exciting new technologies.
Our work is exploratory science at the atomic scale, where the magnetism of neighboring atoms can interact in a subtle dance of patterns described by the rules of quantum mechanics. In our new publication “Probing resonating valence bond states in artificial quantum magnets,” recently published in the journal Nature Communications, our team demonstrated that an important class of quantum spin liquids can be built and probed with atomic precision. The team included lead author Kai Yang, who did his postdoctoral work at IBM Research but recently took an associate professor position at the Chinese Academy of Sciences in Beijing.
Diving into spin liquids
To understand the significance of this discovery, you first have to know what a “quantum spin liquid” is. For the uninitiated, it represents a new state of matter whose electron spins remain in perpetual, fluid-like fluctuation, unlike the conventional ferromagnets currently used in magnetic data storage. Ferromagnets are materials in which all the magnetic spins point in the same direction.
Magnetism arises from spin, the property that gives an atom its magnetic field, with a north and a south pole. Groups of magnetic atoms can join together to form traditional magnets, like a simple refrigerator magnet or the magnets that turn electric motors. In a similar way but on a much smaller scale, microscopic magnets store information in magnetic disk drives and magnetic RAM by holding their north pole stably in one direction or another to record a bit of information. Although these magnets are microscopic, they still contain about 100,000 atoms.
Four titanium atoms positioned into a square just 1 nm wide.
A spin liquid is “liquid” in the sense that it is a configuration of atoms whose spins don’t fit in a fixed state and are very responsive to what happens around them. The spin on each atom is either up or down, and two atoms will have opposite spins when they are neighbors. If you were to pin one of the atom’s spins in the up direction, for example, the surrounding atoms will flip their own spin in response. Each pair of neighbors competes to maintain this opposite pattern. A delicate balance results, forming a small quantum spin liquid.
The ability to create and probe quantum spin liquids creates some intriguing possibilities. A possible application to technology is that a spin liquid can have spin current without having any charge current—no electrons move, they just flip in place and carry spin. We plan to demonstrate this in future studies, which could help to reveal how information propagates in many-body systems and find applications in quantum spintronics (short for spin electronics).
I’ll use Moore’s Law to add some context: At a high level, Moore’s Law describes how researchers have been able to shrink transistors and memory bits over time using ever cleverer techniques. That’s a top-down approach to improving electronics. We’re more interested in a bottom-up approach — looking at atoms and considering what you could build when you connect them together.
When, or whether, this line of research could lead to a new breed of electronics is still unclear, but it could contribute to a better understanding of the qubits used in quantum computing, including what, exactly, causes them to lose their phase coherence and, thus, their ability to perform calculations.
In our new paper, we describe the type of spin liquid in our experiment using a pattern called a “resonance valence bond” (RVB) state, which is relevant to many interesting phenomena in the physics of interacting particles, such as high-temperature superconductivity. To study the RVB quantum liquid, we used a custom scanning tunneling microscope (STM). An STM — which IBM researchers in Switzerland invented, and later won the Nobel prize for, in the 1980s — can see and manipulate atoms at the nanoscale level.
We combined this capability with the technique of spin resonance, which is a high-sensitivity tool to measure magnetic properties. Together these methods allowed us to probe this quantum liquid on an insulating surface in precisely planned and constructed arrangements of atoms. We closely examined the quantum behavior of four interacting titanium atoms (their interaction is the liquid), a step towards answering basic questions in the field of quantum magnetism.
Precise arrangements of three or four titanium atoms arranged to show different magnetic properties. The grid of dots shows the lattice of atoms in the underlying surface, a thin film of magnesium oxide.
Usually atoms are studied by placing them on the ultra-smooth surface of a metal that allows the atoms to be moved to new locations by gently tugging on them with the STM’s tip. Moving atoms on the surface of an insulator, however, has been fiendishly difficult.
We spent several weeks figuring out how to place our magnetic titanium atoms with atomic precision by fine tuning the position of the STM tip and the sequence of voltages that we applied. Once we had the recipe to position the atoms, it took us about a week to assemble the square of four atoms that forms the quantum spin liquid. Our work offers a fresh path to understanding and manipulating entanglement in quantum materials having many interacting spins.
The next (not so) big thing
Physicists are excited about spin liquids because they could allow for spintronics with an insulator, a material that doesn’t have electric current. That could be an advantage when designing electronics because you wouldn’t need electric current or conduction that scatters electrons and disrupts an atom’s spin current.
One of our next projects will be to study magnetic molecules. That sort of work gets chemists involved, especially those looking to make electronics at the smallest possible scale. Often those researchers can make molecular magnets but struggle to add spin exactly where they want it. Our hope is that by sharing our fundamental research we can help advance their efforts, and the efforts of others, toward building atomic electronics and other innovations.
Yang, K., Phark, SH., Bae, Y. et al. Probing resonating valence bond states in artificial quantum magnets. Nat Commun 12, 993 (2021). https://doi.org/10.1038/s41467-021-21274-5
Inventing What’s Next.
Stay up to date with the latest announcements, research, and events from IBM Research through our newsletter.