A Nuclear Family Vacation
Click here to see a slide show.
As one of the participants recalled, Teller, in declining health, attended the reception and even played piano, as he did during the early days of the Manhattan Project. Los Alamos employees pointedly snubbed him. Memories of Teller's 1954 testimony against revered Los Alamos founder Robert Oppenheimer ("I would like to see the vital interests of this country in hands which I understand better, and therefore trust more," Teller said), still lingered.
Some wounds never heal, and while Los Alamos and Livermore today often cooperate, rivalry and competition still divide the two labs.
Livermore traces its reputation to Teller's promise at a 1956 Navy conference in Woods Hole, Mass., to design a hydrogen bomb small enough to fit on a submarine-launched missile.
Bruce Goodwin, who heads Livermore's nuclear weapons program, described Teller's pronouncement as a "name that tune" moment in the history of the labs. A nuclear device in those days was the size of a truck; Teller was promising a quantum leap in weapons design.
Teller "made the whole room sort of stop," Goodwin said. "Because he had just said he could do something in between one-tenth and one-twentieth the size of what everybody else was saying could be done."
Goodwin understands the rivalry: He came to Livermore from Los Alamos in 1985, when he was offered the chance to work with Seymour Sack, a renowned designer recognized as a father of modern nuclear weapons design. It was an ironic professional evolution for Goodwin, who described himself as being a "mildly anti-nuclear person" back in graduate school when he attended lectures by activist Helen Caldicott, a vocal opponent of nuclear weapons.
While at Los Alamos in the early 1980s, Goodwin participated in two nuclear tests for the W-80 warhead, a major project at the time. But when he came to Livermore to work for Sack, Goodwin said, he "learned how to do everything all over again."
A stereotype persists to this day among weapons designers: Livermore designs are elegant but complex; Los Alamos designs are crude but effective. Both labs vehemently disagree over this characterization. Accurate or not, the two facilities each take a very different approach to training and mentoring weapons scientists.
When testing stopped in 1992, Goodwin said he was "beavering away" on questions that were considered interesting scientifically but were not very important during the era of testing. In those days, so long as the thing went off, most everyone was happy. But the Livermore approach to the craft of weapons design, which focused on understanding the way the devices worked in much more intimate detail, came in handy.
"Those little interesting questions turned out to be the big questions because now we're not going to test it and say, 'Well, we don't understand why it works, but it worked OK,' " he said. "Now we have to understand why this little fluke occurred over here, which we know, if it became really big, would cause the weapon to fail."
This question becomes even more challenging as the stockpile shrinks. In signing the Moscow Treaty, the Bush administration committed to reducing the country's deployed nuclear arsenal by 2012.
But while the number of nuclear weapons is declining, the administration has begun a new push to develop a simpler, cheaper weapon called the Reliable Replacement Warhead, or RRW.
Depending on which lab's version you believe, either RRW is needed because of problems with the Los Alamos-designed W-76 nuclear warhead or it was the brainchild of George "Pete" Nanos, the former Los Alamos director who rallied scientists around the idea at a 2003 conference to discuss the nuclear arsenal.
Whatever the truth, the current plan is to hold a competition between the two labs for the design of the new warhead. And critics of new nuclear weapons have again expressed concern that the Bush administration is trying to undermine what was once considered a grand bargain.
Stockpile stewardship—an effort to maintain the current arsenal without testing—formally began a decade ago, in what was a strategically crafted political bargain: Democrats got an ending to testing, Republicans got a commitment to a strong deterrent, and the labs got some expensive goodies.
Physicist Richard Garwin (who did calculations on the hydrogen bomb for Teller as a summer research project) described the bargain that laboratory directors struck with the Clinton administration when it was lining up support for the Comprehensive Test Ban Treaty.
"What could they get?" Garwin said. "Sandia got the microelectronics research center, which had minimal relevance to the CTBT. Los Alamos got the Dual-Axis Radiographic Hydrodynamic Test facility. Livermore got the National Ignition Facility—the white elephant eating us out of house and home. They all maintained these were essential to stockpile stewardship, which they are not."
The National Ignition Facility that Garwin was referring to is a massive 192-beam experimental laser facility under construction at Livermore. According to Livermore management, NIF is essential to stockpile stewardship, because once the facility is fully operational, they will be able to reproduce the fusion processes that take place inside a star or a thermonuclear weapon.
Congress, however, has been less than supportive of NIF. Senators recently moved to zero out funding for the facility, and Sen. Pete Domenici, R-N.M., the patron saint of Los Alamos, said NIF was draining resources away from other important research. On our tour of Livermore, one employee glumly observed that if opponents succeed in killing the project, Livermore would have the "world's biggest indoor soccer stadium."
The question Garwin likes to ask is this: "Will the labs stay bought?" In other words, will they continue to back the bargain and support the moratorium on testing? The U.S. Senate never ratified the CTBT, so the moratorium is a matter of policy, not treaty.
Livermore's Goodwin confessed that in 1995, when stockpile stewardship started, he was a skeptic. He did a back-of-the-envelope calculation on the amount of computing power they would need to combine experimental data, design, and engineering models to certify nuclear weapons without testing.
"I remember handing my answer in, thinking that they would kick me out of the room because it was insane at the time," Goodwin said. "It was 100 teraflops. That would be for a machine that could do one calculation only and take three months to complete it."
But computing power has well exceeded expectations. Livermore, for instance, has invested heavily in supercomputers like Blue Gene/L, which provide massive computing power to aid in research and simulations. Once fully operational, Blue Gene/L will provide 360 teraflops.
"So, stockpile stewardship worked," Goodwin said. "I'm glad to say that I was wrong."
Endnote: Lawrence Livermore National Laboratory conducts tours on Tuesdays and Thursdays. The tours last approximately two and a half hours (starting times vary), and tour participants must be at least 18 years old. U.S. citizens need to register two weeks in advance. Non-U.S. citizens must register 60 days in advance. Tour request forms are available online through the Livermore public affairs office.