Wendy Vogan McNeil stood in the control room of the Dual-Axis Radiographic Hydrodynamic Test (DARHT) facility as the machine operators began their countdown for experiment H3837. McNeil and fellow experimenters Matthew Briggs and Lawrence Hull had spent about 10 months preparing for this moment. They had chosen the target material—a newly developed alloy—and had it fashioned into a miniature hockey puck that McNeil later set atop 31 grams of high explosive. They calculated that once the explosive detonated, it would take about 8 millionths of a second (8 microseconds) for a shock wave to reach the puck, rush through it, and begin to blow the top surface off. A computer simulation predicted how much the alloy would be compressed by the shock wave's passing, and helped them determine when DARHT's two x-ray sources—it's dual axes—should fire to take the radiographs (x-ray images) that would capture the compression in action.
Now, before McNeil could take a breath or cross her fingers for luck, both sources fired, and the experiment was over.
McNeil waited for the radiographs to appear on the control-room monitors. She would process each image and measure the density of the alloy directly behind the shock-wave front. Briggs, meanwhile, would process velocimetry data to measure how fast the puck's top surface moved. (See figure on page 7.) Together, the two measurements would constrain theoretical models of how the new alloy responds to being shocked. Because the best models would be incorporated into the enormous computer codes that aim to predict the long-term behavior of the nation's nuclear weapons, the measurements were of interest to several of the Laboratory's materials and physics groups.
More people, however, were interested in whether DARHT's second axis had operated successfully in multipulse mode, that is, if it had fired off four high-intensity, well-defined x-ray pulses in a precisely timed sequence totaling less than 1.6 microseconds. Various beam diagnostics would tell that story to DARHT's machine diagnostics crew.
But the major story concerned the timing of DARHT's two axes. DARHT was designed to radiograph a target from two vantage points simultaneously. That's why the two axes are set perpendicular to each other, so that x-rays from the first axis can illuminate the target from the front, say, at the same time that one of the second-axis pulses hits it from the side. Nearly everyone wanted to know if the fifth radiograph—the one from the first axis—lined up with the third pulse from the second axis.
Subrata Nath, a group leader in the Weapons Experiments (WX) Division and DARHT's principal accelerator physicist, had a stake in how well the two sources performed. A few years back, when prospects for the second axis were grim, Nath (then deputy project director) was one of the principal architects of a plan to revive it. He knew that for the many people who had followed DARHT's progress over the years, experiment H3837 was a litmus test.
In the big picture of the nuclear weapons community, a sequential set of sharp, clear radiographs would help scientists understand the dynamic interactions between a shock wave and a target far better than would a single radiograph. One of the goals of H3837 was to demonstrate that the second axis could generate the intense x-ray pulses needed to produce four clean images. As for the pair of radiographs taken from different vantage points, if the pulses fired at the right time, scientists could take the two radiographs and do a three-dimensional reconstruction of the target to see if the shock wave compressed it as desired. Thus, another goal was to check the relative timing between the axes.
If something went wrong with either of DARHT's x-ray sources (and the sources are so complicated there's always the possibility of something going wrong), Nath would be one of the people who would decide if the problem was serious enough to cancel the next experiment, already scheduled for early December. That experiment was to be a so-called hydrotest, conducted on a full-scale mock nuclear warhead, and would be a live run of the events that trigger a nuclear explosion. McNeil's alloy-puck experiment notwithstanding, the primary motivation behind experiment H3837 was to demonstrate that DARHT was ready to conduct December's hydrotest.
A Good Day
Both the multipulse and 3-D capabilities needed to be working well and reliably if DARHT were to become the workhorse facility of the nuclear weapons complex, as intended. The word from the beam diagnosticians was that DARHT had performed beautifully.
Success for McNeil, however, was pinned to the radiographs, so when they finally popped up, she immediately focused on each image. And she couldn't for the life of her see the dark band that would be the shock-wave front.
The DARHT facility, home to two x-ray sources (axes) and a confinement vessel, took nearly 20 years to design, build, and get running. X-ray pulses are produced by first accelerating a pulse of electrons to nearly the speed of light with a linear accelerator. DARHT's first-axis accelerator (red) became operational in 1999 and produces a single, very intense electron pulse. The second-axis accelerator (blue) uses a magnetic "kicker" to chop its 1.6-microsecond-long electron pulse into four pulses of variable duration. The electron pulses enter a conversion region and are focused onto a tantalum foil. Whenever an electron scatters from an atom on its way through the foil, it radiates x-rays. The x-rays pass through a collimator then into the confinement chamber. Facing page: The large confinement chamber that holds the test device.
"I saw only the puck, and it looked the same in every image," recalled McNeil. "I thought that maybe I had misunderstood the pulse parameters and had done the wrong radiograph calculations. I started reworking the numbers."
Just then Steve Balzer, DARHT's radiograph-imaging expert, spoke to her beneath the din of the control room. "I think everything's all right," he said. "Give me a minute."
Just like a dental x-ray, a radiograph is both an image of a target's interior and a map of its density, with lower densities appearing lighter in this map than higher ones. Balzer realized that there wasn't much high explosive under the puck, so the shock wave was relatively weak, and increased the alloy's density by only a few percent. Uncompressed material in front of the shock wave looked about the same as the compressed material behind it. The shock wave could be made much more noticeable if the puck's density range coincided with the radiograph's gray scale. This could be achieved by removing from each pixel a "background" component corresponding to x-rays transmitted by the unshocked material. "We just needed to normalize," said Balzer.
The normalization was quickly done, and there it was—a dark band that progressed over four images from the bottom of the puck to its top. As had been planned, the third image from the second axis was aligned perfectly with the lone image from the first axis.
McNeil and Briggs both got excellent data from experiment H3837, and DARHT and the DARHT crew had performed flawlessly. The experiment's claim to fame, however, is that for the first time in the history of the weapons complex, a four-frame x-ray "movie" had been made of an exploding object while at the same time another radiograph had captured the action from a different vantage point.
An Inside Look
No matter what its other accomplishments turn out to be, DARHT must provide scientists with a view inside a mock nuclear warhead during a hydrotest. The heart of a nuclear warhead, its "primary," is essentially a hollow shell of plutonium surrounded by a thick rind of high explosives. In a mock warhead, the primary's plutonium shell is replaced with a surrogate—a shell of highly dense, "dead" material such as depleted uranium.
The test of a mock warhead is called a hydrotest because once the high explosives that surround the surrogate primary are detonated, a shock wave races inward at supersonic speeds, smacks into the hollow shell, and liquefies it on impact. It is then a problem in hydrodynamics to understand in detail what happens as the shell implodes, that is, as the liquid is driven inward at high speed and compressed intensely into a compact sphere.
In a real weapon, with a plutonium primary, the sphere would be bombarded with neutrons, which would initiate a fission chain reaction, ultimately leading to a nuclear explosion. With the surrogate primary, there is no fission and none of the accompanying radiation; the chemical explosion simply consumes the mock warhead and the hydrotest ends.
DARHT is currently the only facility in the world that produces x-ray pulses with enough energy and intensity to pass through the dense, depleted uranium shell in the mock warhead and take a radiograph. DARHT is also the only facility that can produce those pulses fast enough to capture the details of the implosion, to make a four-frame movie, and to allow for a 3-D reconstruction. That's why DARHT is considered to be such a remarkable facility and one of the crown jewels of the nuclear weapons community.
"It was not an easy task to build this facility," says Nath. "Consider coordinating the pulses from the first and second axes. We have to be able to customize the duration of each pulse coming from the second axis. The two accelerators have to run independently of one another, yet in the end their respective x-ray pulses have to hit the target within a few nanoseconds of each other."
Many scientists were skeptical that such exquisite timing could be achieved. Indeed, before McNeil's experiment, it had never been attempted on an exploding target. But the scientists, engineers, and technicians from the Los Alamos, Lawrence Livermore, and Lawrence Berkeley national laboratories who designed and built DARHT knew it could be done.
"The real challenge is that we have one chance to do everything right," says Nath. "Each accelerator has hundreds of triggers that have to fire in the right sequence to produce the x-ray pulses, and if any fail, the experiment is lost. But we've been able to overcome this and many more challenges because of the genuine partnership between the three laboratories. It's been the best collaboration I've seen."
Assurance without Testing
The United States stopped testing nuclear weapons in 1992, a decision that deprived weapons scientists of a means to obtain information about how warheads age, what their lifespan is, and whether refurbishing a device would introduce enough small changes to turn the world's most powerful weapons into duds.
The Deparatment of Energy supported the testing moratorium by sponsoring the massive Stockpile Stewardship program, which uses a combination of weapons surveillance (weapons are removed from the stockpile, disassembled, and inspected for mechanical problems), computer simulations of the moment-by-moment performance of an exploding nuclear weapon, and nonnuclear tests of warheads and their components to ensure that the stockpile remains reliable and safe.
Numerous physics models, including those describing nonlinear processes and a material's response to shock waves, go into the computer simulations, which are among the most complex in the world. It's not by accident, then, that Los Alamos is home to the world's fastest computers.
The simulations are benchmarked against results from past weapons tests, but their predictions about what happens to materials after years inside a radioactive warhead, or about how manufacturing changes affect the weapon's performance, need confirmation. Small-scale experiments, such as McNeil's alloy-puck experiment, provide data to validate the physics models that account for those changes.
But the ultimate test of a simulation is a comparison of its results with the results of a hydrotest, especially concerning how the primary performs. In a modern nuclear device, if the primary underperforms, the so-called "secondary," which generates the bulk of the nuclear explosion, won't ignite. The device fizzles.
Thus, when given the correct command sequence, primaries must detonate with 100 percent reliability, but they must not detonate or produce a nuclear explosion if accidentally jostled, dropped, hit with a hammer, set afire, or subjected to one of a thousand other unplanned actions. In one sense, for safety's sake, primaries are predisposed to fail, and even minute changes to their parts or materials—planned or unplanned—may cause them to do so. The only way to know for sure that a change is okay is to run a hydrotest, and DARHT is the only direct tool for conducting that test.
(A) As seen in these simulations of experiment H3837, the detonation of high explosives beneath the target (an alloy puck) launches an extreme high-pressure wave (shock wave) that traverses the quarter-inch-thick puck in about 1.4 microseconds. The pressure behind the wave front is sufficient to compress the alloy and increase its density. (B) X-ray pulses from DARHT's second axis are generated so fast that they "freeze frame" the shock wave. Each pulse generates a radiograph that records the intensity of x-rays transmitted through the target. Because denser regions attenuate x-rays more, the radiographs yield density information. (C) In H3837 the first-axis radiograph coincided perfectly with the second-axis' third radiograph. Scientists use the two perspectives to check the symmetry of an implosion. (D) Tomographic software can construct a 3-D model of the target.
Early December 2009 found everyone associated with DARHT preparing for the full-scale hydrotest. It was hoped that the test would find the cause of an anomaly discovered earlier, because even a hint of uncertainty would cast a shadow on the safety, security, and reliability of the stockpile. On the big day, both axes were producing pulses that surpassed their design specifications. The test went flawlessly and helped put the anomalous issue to rest.
Of late, DARHT seems to prove its worth with every test, hydro or otherwise. In a recent experiment, the science staff members were baffled by what they saw in the first of four radiographs but then saw in the image sequence that the test object evolved as expected. The only thing wrong was the scientists' initial expectations. If the first radiograph had been the only one available, it would have generated hundreds of hours of discussion, endless emails, and a potentially costly experimental program to reproduce and understand the event. DARHT's multiple images made the whole further-investigation moot.
Helping to maintain a nuclear stockpile—and saving the taxpayers money. Now that's something you don't hear about every day.
— Jay Schecker
Photon Doppler velocimetry (PDV) is used to resolve the speed and direction of the top surface of the puck. (A) Light from an optical fiber is focused onto the target's surface. Some of the light reflects right back into the fiber. The fiber is one arm of a Michelson interferometer, and as the surface moves, the interferometer reveals its speed. Several fibers placed at angles to the surface yield the direction. (B) The velocity as a function of time from two PDV fibers.
In this issue...
- Dynamic Vision
DARHT FULFILLS ITS DESTINY
- Solar Smart Grid in the Atomic City
TEST BED FOR LOCAL CONTROL OF RENEWABLE ENERGY
EXPOSING AND EXPLOITING THE SECRET LIFE OF SOIL
Clean Air and Abundant Fuel
Shooting Rocks on Mars
Better Fuel Cell Membrane Materials