Understanding the Beginning of the Universe


The proton-spin crisis had perplexed nuclear and particle physicists for years.

Since the 1960s, physicists had generally agreed that the proton is made up of three quarks. Therefore, they had concluded that the spin of the proton is made up of the spins of the quarks inside it. Then in 1987, an experiment discovered that the quark spin contributes only 20 to 30 percent to the proton spin. This left a lot of unanswered questions for physicists such as Keh-Fei Liu, a professor of particle physics.

Finding the answer to this question is part of Liu’s research in quantum chromodynamics (QCD), widely accepted as the fundamental theory of the interaction governing the behavior of particles called quarks and gluons which form elementary particles known as hadrons. In this theory, quarks are the fundamental building blocks of the proton, and gluons are the force mediators which bind the quarks together. The theory is the conceptual and the calculational tool used by particle physicists to describe the structure of the hadrons and the beginning of the universe.

Through computer simulations, Liu was able to discover that quark and anti-quark pairs which are created from the vacuum have a big effect on the spin of the proton. The discovery was not only the answer to a baffling question but an example of how major leaps in knowledge are achieved today.

Liu and two postdoctoral students began work on the theory in 1991. It took more than 6,000 CPU hours on a supercomputer (CRAY-YMP) at the National Energy Research Supercomputer Center at Livermore to solve the problem, but it might never have been solved using models as researchers had done in the past. In fact, researchers had attempted to address the problem in more than 1,000 manuscripts, but were not able to do so from QCD - the fundamental theory directly. Using a supercomputer, Liu and his collaborators were able to employ the Lattice Gauge Monte-Carlo method of QCD to calculate the quark contribution to the proton spin which enabled the team to solve the problem quantitatively.

Today, Liu is addressing other issues with similar structures. In order to make the numerical calculation tractable, physicists are involved in developing algorithms themselves. Liu’s previous work relied on a Noise Estimator method which shaved computer time by as much as a factor of 6,000.

The new method Liu and a graduate student have just published enables physicists to calculate the quark determinant to a high accuracy. Known as the Pade-Z2 method, it allows a very accurate estimate of determinants needed for the computer simulation. This breakthrough could be faster than existing algorithms and will allow physicists to address quantum chromodynamics at a finite density. The algorithm will also have practical applications in other fields such as engineering as well as other branches of physics and essentially any field in which researchers need a reliable estimate of a large determinant.

Like the resolution of the proton spin crisis, the new algorithm is a leap in knowledge, addressing questions no less fundamental than how our universe is structured.


About Keh-Fei Liu: In 1997, Keh-Fei Liu was named a Fellow in the American Physical Society for his pioneering work in lattice gauge calculations which checked nuclear models quantitatively. In 1990, he received the Alexander Von Humboldt Award for Senior Scientists. Currently, he is serving on a committee for the Department of Energy (DoE) to recommend projects for DoE funding as well as a committee to recommend a dedicated computer to carry out DoE initiatives.

LIU.GIF (38436 bytes)

Keh-Fei Liu’s research team: Terrence Draper, UK associate professor, physics; John Sloan, UK assistant professor, physics; Murat Ates, UK graduate student, physics; Shao-Jing Dong, UK senior computing system scientist; Lee Lin, ChungHsin University, Taiwan visiting professor, physics; Walter Wilcox, Baylor University visiting professor, physics.


Next Page