many-body physics

Quantum physics is a notorious rule-breaker. For example, it makes the classical laws of thermodynamics, which describe how heat and energy move around, look more like guidelines than ironclad natural laws. In some experiments, a quantum object can keep its cool despite sitting next to something hot that is steadily releasing energy. A new experiment led by David Weld, an associate professor of physics at the University of California, Santa Barbra (UCSB), in collaboration with JQI Fellow Victor Galitski, shows that several interacting quantum particles can also keep their cool—at least for a time.

The quantum world blatantly defies intuitions that we’ve developed while living among relatively large things, like cars, pennies and dust motes. The quantum behavior of dynamical localization bucks the assumption that a cold object will always steal heat from a warmer object. Until now, dynamical localization has only been observed for single quantum objects, which has prevented it from contributing to attempts to pin down where the changeover occurs. JQI researchers and colleagues have investigated mathematical models to see if dynamical localization can still arise when many quantum particles interact. To reveal the physics, they had to craft models to account for various temperatures, interaction strengths and lengths of times. The team’s results, published in Physical Review Letters, suggest that dynamical localization can occur even when strong interactions are part of the picture.

Exotic physics can happen when quantum particles come together and talk to each other. Understanding such processes is challenging for scientists, because the particle interactions can be hard to glimpse and even harder to control. Moreover, modern computer simulations struggle to make sense of all the intricate dynamics going on in a large group of particles. Luckily, atoms cooled to near zero temperatures can provide insight into this problem.Lasers can make cold atoms mimic the physics seen in other systems—an approach that is familiar terrain for atomic physicists. They regularly use intersecting laser beams to capture atoms in a landscape of rolling hills and valleys called an optical lattice. Atoms, when cooled, don’t have enough energy to walk up the hills, and they get stuck in the valleys. In this environment, the atoms behave similarly to the electrons in the crystal structure of many solids, so this approach provides a straightforward way to learn about interactions inside real materials. But the conventional way to make optical lattices has some limitations. The wavelength of the laser light determines the location of the hills and valleys, and so the distance between neighboring valleys—and with that the spacing between atoms—can only be shrunk to half of the light’s wavelength. Bringing atoms closer than this limit could activate much stronger interactions between them and reveal effects that otherwise remain in the dark. Now, a team of scientists from the Joint Quantum Institute (JQI), in collaboration with researchers from the Institute for Quantum Optics and Quantum Information in Innsbruck, Austria, has circumvented the wavelength limit by leveraging the atoms’ inherent quantum features, which should allow atomic lattice neighbors to get closer than ever before. The new technique manages to squeeze the gentle lattice hills into steep walls separated by only one-fiftieth of the laser’s wavelength—25 times narrower than possible with conventional methods. The work, which is based on two prior theoretical proposals, was recently published in Physical Review Letters.

Machine learning, the field that’s driving a revolution in artificial intelligence, has cemented its role in modern technology. Its tools and techniques have led to rapid improvements in everything from self-driving cars and speech recognition to the digital mastery of an ancient board game.Now, physicists are beginning to use machine learning tools to tackle a different kind of problem, one at the heart of quantum physics. In a paper published recently in Physical Review X, researchers from JQI and the Condensed Matter Theory Center (CMTC) at the University of Maryland showed that certain neural networks—abstract webs that pass information from node to node like neurons in the brain—can succinctly describe wide swathes of quantum systems.