By Susan Dieterle
Two years ago, Ames Lab assistant scientist Dan Barnard uncovered a scientific riddle that has yet to be solved.
Barnard and Otto Buck, former director of the Metallurgy and Ceramics Program, were conducting experiments with acoustic harmonic generation, a technique in which ultrasound frequencies provide information about a material's strength and other characteristics. As Barnard plotted the data from his experiments, he noticed a pronounced downward curve as the amplitude of the frequency decreased. He showed the graph to Buck, who asked in his thick, German accent, "Vat's dat hook?"
Thus was born "the hook" -- and the controversy about whether it even exists.
As Barnard and other scientists have grappled with what the hook means, two schools of thought have emerged. One side thinks the sensitive testing equipment is picking up noise from the instrumentation itself, and that the hook doesn't really exist. The other side believes the hook may indicate something new about the microstructure of the materials being tested.
Which camp is Barnard in? "I don't have any preference as far as the real source of the hook," he says. "I'd just like to know what's causing it."
The debate is difficult to resolve because researchers are still trying to understand acoustic harmonic generation. The technique comes from the realm of nondestructive evaluation in which the reliability of materials is studied without the use of invasive procedures. Ultrasonic testing, which involves transmitting high-frequency sound waves into a material to detect imperfections and microstructural changes, is a standard evaluation method. It is often used in a linear fashion, which involves pumping sound energy into one end of a material and then measuring how much energy comes out the other end.
But there is a growing interest in nonlinear acoustic techniques like harmonic generation, Barnard says. While the approach is similar, scientists also measure the faint frequencies emanating from the microstructure itself at double, triple and even quadruple the rate of the original frequency.
"For example, we put in energy at a fundamental frequency of 10 megahertz and then we look at the energy in the harmonics, usually at 20 and 30 megahertz," Barnard says, noting that the harmonics are usually 1/100th or 1/1,000th the strength of the fundamental frequency.
Cracks, dislocations and fatigue in the material produce the second harmonics, but no one fully understands how or why. "We don't have a complete understanding of the sources of harmonic generation and we don't know how they relate to each other," Barnard says. "We want to find out how each source affects harmonic generation, and how they affect it when they're combined."
What they do know is that harmonic generation is more sensitive than linear methods and can detect cracks that aren't visible -- even through the most powerful optical microscope. "There are limitations on what kind of information you can get out of the materials with linear acoustic techniques," Barnard says. "Harmonic generation is a lot more sensitive to very fine changes that can occur at the microstructural level and even at the atomic level."
That kind of information could be crucial in settings such as power plants, where high temperatures and radiation can make materials brittle. "You don't ever want a brittle material when you have lots of power or pressure or temperature behind it," Barnard says. "Theoretically, they could use harmonic generation to determine the state of a material and know when to retire it and put in new material."
First, though, scientists need a better understanding of harmonic generation, and they need to adapt the technique for use in the field. Currently, the procedure requires painstaking, time-consuming measurements that wouldn't be possible in industrial settings.
The experiments that Barnard and Buck began two years ago were part of a collaboration with Oak Ridge National Laboratory to establish a deeper understanding of the phenomenon by measuring samples of copper-aluminum alloys.
Theory predicted that the data from the experiments would move in a straight line until reaching the "noise floor," or the level of background noise, which would then cause the data to shoot upward. That's why the emergence of the downward hook at low amplitudes was so puzzling.
"We'd never seen it go down and then go back up, and we couldn't think of any experimental reason why it would do that," Barnard says, adding that he has also seen the hook with samples of pure aluminum. In his tests with fused silica, however, there was no hook.
So far, Barnard and others have been unable to provide a definitive answer about the nature of the hook. Scientists at Oak Ridge produced an equation in support of their belief that the hook came from instrument noise, but Barnard says their equation doesn't fit all of the data.
Also, researchers at NASA's Langley Research Center have found curved deviations -- some upward, some downward -- at low amplitudes in aluminum alloys. John Cantrell, senior materials physicist in the Nondestructive Evaluation Sciences Branch at Langley, says that much of the curvature in his data appears to be linked to instrument noise. When corrected for that factor, the curves disappear -- almost.
"If we go down to sufficiently low amplitudes, we still see some variation," Cantrell says. "But we're not sure whether that's due to an instrument offset or whether that's real."
Cantrell, who uses different harmonic-generation techniques and equipment, says he plans to test the Ames Lab copper-aluminum samples to corroborate Barnard's findings.
"If it turns out that the hook, or at least part of the hook, is real and not due to instrumentation offset, that would give us some insight into the material physics going on at low acoustic amplitudes," Cantrell says. "But even if it isn't a physical process in the material, the very fact that there's an instrumentation situation has to be recognized. In some cases, you're going to be publishing data from low-amplitude tests, and you have to make sure you don't have instrument offset giving you false positive readings.
"Either way, Dan has made the nondestructive evaluation community aware of a very important situation," he adds.
Notes Jim Foley, who now serves as Barnard's Ames Lab group leader, "The hook is a result -- not that we understand that result, but it's a result. Now we have to see what we can do to critically evaluate whether it's a real phenomena or an instrumental phenomena."
Since Buck's death in November 1997, Barnard has continued working with Oak Ridge to better understand the nature of harmonic generation and its contributing sources. In addition, Ames Lab has received funding from the Department of Energy to research ways of adapting the technology for use in the field.
"I think this technique has a definite use in the real world," he says. "And there aren't a whole lot of other people doing this kind of research. It's always fun to be one of the first to find something new."
The Lab recently purchased a more sensitive amplifier system that Barnard hopes will give him new insights into harmonic generation. And it may also help him solve the perplexing riddle of the hook. If it turns out to be a true material phenomenon, Barnard would like to call it the Otto Buck Hook in memory of his mentor.
"If it's an instrumental effect, then I'd like to know what it is and develop a correction for it," Barnard says. "But if it's a material effect, then that's something new, and we want to know what's going on."
And although much of Barnard's time is spent on other research, the hook remains firmly embedded in his thoughts. "When I have time, I go back and do some measurements to see if the hook is still there," he says with a smile. "And it's always there."
For more information:
Dan Barnard, (515) 294-9998
dbarnard@ameslab.gov
Current research funded by:
DOE Office of Basic Energy Sciences
Last revision: 12/17/99 sd
Home | Comments
| Search | Disclaimer