DEIXIS 2003 - 2004 THE DOE CSGF ANNUAL


A Calling Realized

By Jacob Berkowitz

In 1997, Mary Ann Leung sold her house, quit her successful 16-year-long career as a trainer and instructional designer in the computer industry and, at age 38, returned to college to become a physician. Or so she thought. On her way to the medical sciences she discovered a long unrealized passion for math and the computational sciences. Now a Department of Energy Computational Science Graduate Fellow (DOE CSGF), she is using high performance computing to understand the role of quantum mechanics in applications ranging from nanoscale engineering to the emerging field of quantum computing.

Leung found her groove during a National Science Foundation-sponsored undergraduate summer stint in the lab of Dr. William (Bill) Reinhardt, a University of Washington professor of chemistry and adjunct professor of physics.

“I fell in love with the work I did,” says Leung, now a third-year doctoral student in Dr. Reinhardt’s lab at the University of Washington. Her job that first summer was to develop a computer program to visualize the time propagation of solitons in the Bose-Einstein Condensate (BEC). While planning a slide presentation of her work, she came across an image of a BEC generated by one of Dr. Reinhardt’s computer programs. It was a moment of scientific and personal epiphany.

“The image had a strong likeness to a piece of artwork that I’d created 20 years earlier. I decided at that point that I’d been visualizing math much earlier in life, but I just hadn’t realized it. So, seeing this image made me feel like I’d found my calling,” recalls Leung.

Ironically, had she returned to school several years earlier she would never have seen her calling. In 1924, Albert Einstein and Satyendra Nath Bose hypothesized that noninteracting atoms chilled to supercold temperatures (about 190 nanoKelvin) would condense into a new form of gaseous matter, the BEC. What’s remarkable about the BEC, they explained, is that in this supercold state the atoms would all go into the same quantum state: They would behave as a single super atom, a macroscopic quantum object large enough to be visible to the naked eye.

However, it wasn’t until 1995 that three independent research teams were successful in creating the first BEC by supercooling rubidium-87, lithium-7, and sodium atoms. This breakthrough has set off a global wave of experimental work with BECs. “They’re a wonderful place for people to explore the fundamentals of quantum mechanics because you can see the quantum effects,” says Leung, noting that BEC research has applications to fields ranging from superconductivity to nonlinear optics.

In the process of her doctoral work, initially focused on developing parallelized codes to understand and visualize the dynamics of solitons and vortices in BEC, Leung found her own kind of quantum coherence – a fellowship program that was just right for her.

“When I found the DOE CSGF, I thought: This is perfect for me. I love all of these things,” recalls Leung. The Fellowship requires participants to take science, applied math, and computer science courses, and Leung was particularly inspired by an algorithms course.

But it was the Fellowship practicum that let her really test her mettle as an emerging computational scientist.

During a workshop she attended on the computational tools at DOE’s Lawrence Berkeley National Lab, Leung met Dr. Andrew Canning, a staff scientist in Berkeley Lab’s Computational Research Division. She was intrigued by his work: Although also exploring computational methods for quantum systems, Dr. Canning models solid state systems. It’s a realm that offered a new challenge for Leung, one that she readily took on when offered the chance of a three-month practicum with Dr. Canning.

Leung’s practicum research was part of an ongoing collaboration between, among others, Dr. Canning and Dr. Z.Q. Qiu, a solid state physicist who holds a joint appointment with the Lab’s Materials Sciences Division and the Physics Department of the University of California at Berkeley. Dr. Qiu’s group is a world leader in the creation of very pure, nanoscale metallic films. These thin metal films (sometimes only several atoms thick) are ideal for studying quantum mechanical effects and are at the frontier of a new realm of materials engineering.

“In the past, people have designed materials to have specific mechanical properties, such as strength,” says Dr. Canning. “Now, because we can engineer at the atomic level, the idea is to design materials in which we can control the properties of individual electrons.”

Using Berkeley Lab’s IBM SP supercomputer, Leung modeled Dr. Qiu’s experiments conducted at the Lab’s Advanced Light Source, to understand the effect of the addition of a nanoscale nickel monolayer on quantum well states in copper. Quantum well states are an energy state in which an electron is sandwiched between two layers of atoms so that its motion is confined to a single dimension. They are thought to be responsible for the giant magneto resistance effect, the basis for the creation of very high density disk drives.

Leung and Canning’s simulations have already provided a more detailed physical understanding of the nickel-copper quantum well state experimental results, as well as confirming some of the theoretical models used by Dr. Qiu’s group.

Leung says that it was great to work closely with Dr. Canning and his post-doctoral students in “a collaborative environment that was really about pursuing interesting science.” The experience provided Leung with a clearer sense of her potential.

“One of the important lessons that I learned from that experience was that you can apply yourself to a new field and within a relatively short amount of time, you can learn quite a lot and make progress,” she says.

It a lesson she’s taken back to her doctoral research at the University of Washington. Along with her supervisor Dr. Reinhardt, Leung is exploring the potential of the BEC and high performance computing for the study of quantum computing.

Dr. Reinhardt stresses that the work is currently at the very early stages of creating an original and challenging PhD research project. “It’s an exciting, frustrating and turbulent process,” he says.

But it’s one that Leung has eagerly grasped.

“My vision is to create a computational model that will simulate the behavior of a quantum computer,” says Leung. “So in my case, computational science not only means the application of high performance computing to a scientific problem, it also includes the investigation of new computational techniques using high performance computing.”

Amidst the enormous international scientific buzz about quantum computing, Dr. Reinhardt is cautiously optimistic that BEC research offers a unique approach.

“Because BECs are big and you can control them with lasers, we’re trying to figure out if we can use these very special macroscopic properties to make extra-stable pieces of quantum computers,” he says, noting that as a large quantum object the BEC might be less prone to the quantum decoherence that disturbs single atom quantum computing systems.

At the same time as she is working towards numerically modeling a quantum computer, Leung is also helping others push scientific boundaries. She volunteers with organizations dedicated to encouraging girls to study science and engineering.

“In hindsight, I realize that I was always really good in math when I was a kid,” says Leung. “But I never really received any encouragement to do anything with math. Quite the contrary, I was discouraged: ‘That’s not a girl thing to do’. Since I got such a late start in science, I really want to encourage young girls and women to go into science; it’s a fascinating field.”

The Krell Institute
http://www.krellinst.org