Skip to main content

Rotation Fascination: Keh-Fei Liu

By Guy Spriggs

From early childhood human beings have an understanding of rotational movement: we see tops spin and planets move and gain some comprehension of what rotation is. However, even the most gifted scientists don’t have a complete understanding of how rotation – or spin, a quantum analog – operates at a sub-nucleon level.

After being awarded a highly-competitive grant to perform Advanced Scientific Computer Research (ASCR) from the ASCR Leadership Computer Challenge (ALCC), UK physics professor Keh-Fei Liu and his collaborators (including colleague Terrence Draper, post-docs and students at UK, as well as 2 co-PIs at George Washington University) hope to resolve what has been dubbed the Proton Spin Crisis.

The ALCC awards grants to projects of interest for the Department of Energy, prioritizing high-risk, high-payoff simulations. The 69 million processor-hours Liu’s group has been allocated – which will be utilized at Oak Ride Leadership Computing Facility – will go a long way toward accomplishing the high payoff of understanding every component of proton spin.

As Liu explains, physics textbooks still suggest that protons are made of three quarks, with the total spin of the proton made up of the combined spin of the quarks. But groundbreaking experiments revealed that only about 25 percent of proton spin comes from quark spin. The constitution of the remaining 75 percent – the subject of the aforementioned Proton Spin Crisis – is still unsettled.

“Other studies were based on models, and these turn out to be very unreliable,” Liu explained. “The only tool which we have control over the systematic errors is called lattice quantum chromodynamics [QCD], dividing space and time into lattices. This turns out to be a huge numerical job.”

Liu’s calculations require tremendous amounts of computer resources, so the ALCC grant is instrumental in facilitating the simulations at the center of the group’s work. The processor-hours granted to Liu’s group roughly equate to 10,000 CPUs running every second for the entire year. The XK7 lab cluster located at Oak Ridge National Laboratory is the nation’s most powerful supercomputer for open science, theoretically capable of 27,000 trillion calculations per second.

“Studying and understanding the quark-gluon structure of the nucleons and their interactions are of fundamental importance to understanding the building blocks of our universe and life,” Liu said. “We were allocated precious time because people view this as an important theoretical effort.”

The simulations undertaken at Oak Ridge will yield new insights to proton spin and proton mass. As Liu explains, accounting for the mass of a proton is difficult since quarks only account for roughly 4 percent of the proton’s total mass.

“We used to think you break things up into building blocks, that everything is made of materials. But nucleons are made of things largely not of matter, but rather energy which gives rise to the nucleon mass – a concept dawned in Einstein’s famous formula,” he explained.

Liu believes the best candidate for answering these mysteries is the quark orbital angular momentum and gluon angular momentum. Gluons are the force particles which bind quarks together in the nucleon, contributing to the nucleon spin and mass. This led to the prediction of “glueballs” – particles with a mass made of pure energy – which are being actively searched for experimentally.

The grant gives Liu’s group access to the XK7 until July 2015, and he is hopeful that this time will allow them to obtain the bulk of their results and demonstrate enough success to continue the project.

“The outcome of this work will be a first-principle-based understanding of the quark-gluon structure of the nucleon which can be compared with experiments being conducted in high energy and nuclear physics laboratories around the world,” Liu said.