If there's one thing scientists at Los Alamos National Laboratory know about it's nuclear fission. That's a reaction in which an overly large and heavy nucleus, such as plutonium, absorbs a single neutron and instantly breaks into two different nuclei (fission products) and a few neutrons, releasing nuclear energy. Los Alamos scientists expertly exploited the process, finding ways to unleash large amounts of energy slowly (in a nuclear reactor) or in a flash (in "the bomb").
In the 1940s, Los Alamos scientists developed a method for determining the fission yield—the amount of energy released by fission—of a nuclear device. Individual fission products were isolated from samples of the radioactive debris, and their beta-ray count rates were converted using conversion factors (so-called K's, Q's, and R's) into the number of fissions that occurred in the sample. The number of fissions fed into the fission-yield calculation.
Then 1952 saw the establishment of Lawrence Livermore National Laboratory. Fission gurus themselves, Livermore scientists designed their own bombs and initially, used the Los Alamos beta-counting method to determine fission yields. But Livermore had pioneered a gamma-ray counting method and toward the end of the 1960s, began using it exclusively to determine fission yields. Both methods produced the same results.
That is, until "the problem" showed up.
It began when Los Alamos remeasured some of their conversion factors in the 1970s and found some disagreements with the "historical" values. They determined a new, "modern" set of values and presented their results at a 1977 meeting sponsored by the Department of Energy (DOE), but they didn't exactly use the open literature to call attention to their findings, nor did they use the modern values in their fission-yield calculations.
"The differences between the modern and historical values were relatively small,"explains Don Barr, former head of the Los Alamos nuclear chemistry group (retired), "and when we calculated new fission yields, they always fell within the error bars of the historical yields. After discussions with some weapons designers, everyone agreed that it was better to maintain consistency and not use the modern values until there was a compelling reason to do so."
That reason began to surface in 1992 with the end of nuclear testing. Assurring the reliability and safety of the nuclear stockpile fell to the Stockpile Stewardship program, which among other things uses a combination of computer modeling, simulation, and non-nuclear tests to develop computer codes that could reproduce 45 years of nuclear test data. By 1995, the codes had become so sophisticated that they needed to be validated against the best data, and that, reasoned the Los Alamos analysts, meant providing the coders with fission yields calculated using the modern conversion factors. So Los Alamos began revising its previously determined yields and suggested that Livermore do the same.
Livermore politely declined, noting that its gamma-counting method didn't rely on K's or Q's or R's; therefore, its fission-yields didn't need revising. And Livermore was mystified as to why, after 40-some years, Los Alamos wanted to fix something that wasn't broken.
The dispute boiled down to fission-product yields (Y's), each of which is the fraction of fissions that lead to the production of a specified nucleus and is not to be confused with the fission-yield. The K factors are inversely proportional to the Y's, while the Q's and R's contain ratios of them. Whereas Livermore used the Y's directly in its fission calculations and Los Alamos used the rather clever but funky K's, Q's, and R's, it turns out that Livermore needed to multiply one of its Y's by a Los Alamos Q to put both methods on the same fission scale. Capiche?
Don Barr knew all this, but Livermore's post-testing-era analysts did not; that knowledge had not survived a round of retirements. Thus, when Los Alamos argued that at least one historical conversion factor was buried within the computer code that Livermore used to determine yields, Livermore replied it would have to look into it. And, by the way, just where was the documentation on the 1970s experiments, and would Los Alamos kindly share it?
Sifting through old data is about as appealing as doing one's taxes, and while both laboratories agreed to cooperate, their efforts lost traction, and the problem persisted into the next decade.
But at the urging of DOE and both laboratory directors, the effort to reconcile the differing fission yields gained momentum. Los Alamos, for its part, went through 60 years of in-house reports and laboratory notebooks and conducted an intense evaluation of its conversion factors. Livermore found a historical Los Alamos Q-value lurking among thousands of lines of its code and executed a thorough, unbiased review of Y's in the available literature.
In addition, Los Alamos staff from the Chemistry, Theoretical, and X-Computational Physics Divisions conducted a meta-analysis that included all of the Los Alamos and Livermore data, plus high-quality, independent data from other laboratories. Written up for the December 2010 issue of Nuclear Data Sheets, the meta-analysis showed that the results of the 1970s experiments are in excellent agreement with the best data known. Los Alamos and Livermore adjusted their conversion factors accordingly, and they once again agree on fission-yield estimates.
Though it took many years, the sister Laboratories turned what had become an entrenched scientific dispute into a fruitful and cooperative collaboration. And the final word from Barr? "It's about time."
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