Navigation Content
University of Wisconsin Madison College of Engineering
You are here:
  1. Home > 
  2. News > 
  3. News archive > 
  4. 2004 > 

Exploring the “shocking” physics of fusion fuel

Wisconsin Shock Tube

The Wisconsin Shock Tube (large image)

Riccardo  Bonazza

Riccardo Bonazza (large image)

Researchers who explore the inertial confinement method as a means to achieve nuclear fusion confine their fuel — deuterium and tritium — in pellets about the size of small peas. Their goal: to apply enough energy to the pellet that the atoms in each fuel fuse into helium atoms, giving off energy that can be collected and used for purposes such as electricity production.

However, what happens next is not well understood.

Huge laser beams, such as those used in experiments at the National Ignition Facility at the Lawrence Livermore National Laboratory and at other large laser laboratories elsewhere around the world, generate a spherical converging shock wave that implodes the fuel pellets, says Associate Professor Riccardo Bonazza. "In all these experiments, when this spherical shock wave hits the spherical shell, many, many things happen at once," he says.

Among the phenomena are phase change, radiation release, and plasma generation. And there are fluid mechanics issues. "Any ripples, perturbations, non-uniformities, roughness initially present on the shell will inevitably amplify when the shock hits it," says Bonazza. "As they amplify, they cause mixing of the shell material with the nuclear fuel contained inside the shell. And that ruins the process, because essentially fuel contamination leads to reduced energy yield or no yield at all."

To understand how the mixing occurs, what its controlling parameters are, and what can be done to mitigate it, Bonazza's group has devised experiments for the Wisconsin Shock Tube that decouple the fluid mechanics from the other phenomena. In addition, the group expanded the scale of its experiments, from the laser facilities' pea-sized target and time frame of femto- and picoseconds to a tennis-ball-sized target and time frame of milli- and microseconds.

The experiments' purpose is twofold: to provide an understanding for the physics being studied, and to develop a database that researchers can use to validate existing and new computer codes that eventually they will use to predict what will happen in a fusion-ignition experiment.

Although colleagues around the country conduct such experiments, UW-Madison's facility is unique because it has the largest cross-section and structural capability, says Bonazza. "So we can launch stronger shocks onto larger targets than anybody else can in this country," he says.

Previously, the group used a copper plate to separate heavy and light gasses in the shock tube. But, by observing the experiment from the side, the researchers found that when they retracted the plate, they were introducing motions they didn't want.

Now they've eliminated the copper plate entirely. Instead, they blow a soap bubble into the tube via a small retractable stainless-steel spout. Inside the bubble is one gas; a different gas fills the shock tube. As the bubble free-falls inside the tube, Bonazza's group sends a shock that strikes it.

Because they use spherical targets and spherical shock waves, the laser experiments produce results different from those of the UW-Madison shock tube, which uses linear shock waves. "The physics are not exactly identical," says Bonazza. "But things can be scaled appropriately, and our experiments can form a really good baseline to start collecting some data."

While most of his group conducted shock-tube experiments here, one of his students spent time working on a numerical simulation on a computer at the Lawrence Livermore Laboratory. When the researchers compared the experimental and numerical results, they were gratified to learn that the code reproduced most of the experiments' relevant large-scale features. It is the first step toward ensuring that scientists can trust the code to predict results of experiments they can't perform.