America’s only fatal reactor accident was the culmination of a costly battle between the army and navy at the dawn of the Atomic Age.
John Byrnes knelt atop the U.S. Army’s SL-1 nuclear reactor, poised to pull the 84-pound central control rod straight up. The procedure for reassembling the control rod drive mechanism called for lifting the rod “not more than four inches.” Byrnes was no nuclear engineer, but he was a trained army specialist—he knew that the central rod in SL-1 was enormously powerful, capable of starting up the reactor all by itself. But the boron strips inside the core were crumbling, occasionally jamming the control rods in their channels and making them almost impossible to move, a problem that had gotten worse in recent months.
Byrnes’ supervisor, Richard Legg, was nearby. A third crewman, Richard McKinley, was pacing around the vessel head.
Byrnes pulled the control rod up. Within 380 milliseconds, a chain reaction started in the uranium fuel. Within one-half second, the massive stored energy inside the reactor was released, most of it in the form of heat. The temperature of the fuel spiked to 2,000 degrees Celsius, far more heat than the water inside the core could absorb. The water closest to the fuel flashed into steam, pushing the water on top of it upward into the lid of the pressure vessel—on which Byrnes and Legg were standing—with a force of 10,000 pounds per square inch, ejecting the control rods and loose shield plugs with a velocity of 85 feet per second. A shield plug penetrated Legg’s groin and exited through his shoulder, propelling him straight up and pinning him to the ceiling for six days.
The explosion at the National Reactor Testing Station in Idaho Falls, Idaho, on Jan. 3, 1961, remains the only fatal reactor accident in American history. Within days of the incident, rumors of infidelity and a love triangle sprang to life, stories of an aggrieved husband who used a control rod as a peculiarly modern murder weapon. There was just enough truth to the story, and it explained the tragedy so neatly, that it rapidly obscured the underlying questions about why a reactor would be designed so perilously close to criticality and what really caused it to fail. But the saga ultimately turned out to be more than a murder mystery or a case of engineering gone awry.
The saga of the SL-1 nuclear accident is a war story, a tale of a bloody, costly struggle between the three branches of the United States military. It happened at a time when nuclear annihilation was a frighteningly real possibility—eight months after U-2 pilot Francis Gary Powers was shot down over the Soviet Union and three months before the Bay of Pigs invasion. Nuclear power was still almost exclusively a military enterprise, and each branch of the service made the case that it needed that power. The army wanted portable power plants for Arctic radar bases, the first line of defense against a Soviet air attack. The air force wanted a supersonic bomber with unlimited endurance, and the navy wanted a “true submarine” that would live beneath the waves. Each service was convinced that without perfecting a mission for the Atomic Age, it would become obsolete. More than that, the generals and admirals all believed passionately that the survival of the nation itself was at stake.
The explosion ended the army’s program and the dream of portable nuclear power. The air force program died almost simultaneously with the explosion, having never solved the considerable technical and safety challenges of keeping a massive nuclear-fueled bomber in the air. That left the navy, under the direction of Admiral Hyman Rickover, holding the torch for nuclear power and affected the DNA of the civilian industry in utero. The navy’s emphasis on pressurized water reactors eventually led to the development of similar civilian power plants. Moreover, the vast majority of navy nuclear engineers are submariners—a group that prides itself on being a “silent service”—and their influence has engendered a similar culture of secrecy in the civilian nuclear industry.
The army and Atomic Energy Commission were surprisingly candid in the reports they released to the press immediately after the accident. The Idaho Falls while the three operators had died from the Post Register informed readers that effects of the explosion itself, “radiation…was at such a high level in the reactor building that emergency crews could only enter the building for a minute at a time without exposing themselves to excessive radiation limits.”
Nonetheless, officials were perhaps overly quick to reassure residents that “there is no radiation danger to populated areas of Idaho and Utah.” Based on its portability requirements and proposed remote locations, SL-1 was not designed to keep the results of an explosion contained. The thin metal-walled building actually did a remarkably good job of staying intact, but it was not anywhere near airtight and its vent fans exhausted directly to the environment. For days after the explosion, a stream of highly radioactive contamination from the shattered reactor spewed over southeastern Idaho. Unable to contain it, the scientists did what they could to track it.
Sagebrush in the surrounding area was sampled daily until the contamination levels rose too high to be read by the standard method. Between January 4 and 19, 28 milk samples were taken from five different farms near the southern boundary of the site. Six of the 28 samples showed radioactivity in the milk greater than three standard deviations above background levels. It does not appear any warnings were issued, to area residents or to those farms with the contaminated milk.
Meanwhile, stories of marital infidelity—and bad blood between reactor crewman John Byrnes and his supervisor Richard Legg—surfaced as investigators searched for causes of the explosion. Byrnes’ wife, Arlene, had called John at the reactor building two hours before the explosion, and told him she wanted a divorce. She’d had enough of fighting and loneliness in Idaho’s Lost River Desert. Their last conversation ended with a discussion of how to split his paltry army paycheck.
Investigators also learned that Byrnes might have been uneasy to have Legg hovering so closely behind him. The two had clashed ever since they’d both arrived in Idaho in October 1959 and had even come to drunken blows at a sleazy bachelor party the year before. But Legg had since surpassed Byrnes professionally and qualified as both chief operator and shift supervisor, while Byrnes’ steady record of disciplinary problems all but guaranteed that his progress in the army was over.
Legg had instigated his own share of trouble along the way. Self-conscious about his height at 5 feet 6 inches, he was constantly physically asserting himself, challenging any and all to wrestling matches and goosing his comrades at inappropriate times. When Byrnes was hunched over the control rod, straining with effort, he might have made a tempting target for one of Legg’s pranks. Byrnes was certainly capable of pulling the central control rod too far in a fit of anger or to besmirch Legg’s reputation. But he might not have been aware that his actions would cause an explosion.
Clarence Lushbaugh, a University of Chicago– trained pathologist at the Los Alamos National Laboratory, led the team of medical investigators charged with the dangerous and grisly task of examining the highly radioactive bodies of the three victims. Lushbaugh had the exceedingly rare experience of having performed an autopsy on a radioactive corpse— that of Cecil Kelley, a Los Alamos worker who had died from a radiation dose in 1959. But Kelley’s body had been relatively intact, and the level of radioactivity low. By contrast, Byrnes, Legg and McKinley were ripped apart in the reactor explosion, and their bodies were so radioactive that it wasn’t safe to be in the same room with them, much less operate on them with anything resembling normal methods.
Lushbaugh and his team arrived in Idaho on January 8. The bodies were waiting for them inside the decontamination room of the Idaho Chemical Processing Plant. Byrnes and McKinley had been placed in stainless-steel tanks filled with alcohol and ice, while Legg, by far the most radioactive, was still in the lead cask that had carried him away from the explosion site.
The Chem Plant turned out to be an ideal radioactive mortuary. It was relatively close to the failed reactor. Its decontamination room was lined with stainless steel and contained drains and large tanks. An overhead crane traversed the room, which proved extremely useful in moving the bodies from tank to autopsy table and back while maintaining a safe distance.
Lushbaugh improvised an autopsy table by placing a 6-footlong stainless-steel tray on sawhorses. Even the least radioactive body was far too hazardous to approach from the normal position of a medical examiner. Lushbaugh crafted some crude autopsy tools by welding disposable knives and hooks onto 4-foot lengths of galvanized steel pipe. Members of Lushbaugh’s team wore protective gear during the procedure, including a 45-pound lead apron and a portable lead shield held between them and the body. Every aspect of the victims, from the texture of their bone marrow to the size of their adrenal glands, was scrutinized in a controlled rush.
Traditional means of decontamination, simply washing and rinsing the bodies in a variety of liquids, detergents and even citric acid, proved almost completely ineffective. The bodies were raised from the tanks, lowered and washed again and again in the improvised autopsy room, but the radiation remained dangerously, stubbornly high. This was especially true for Legg’s body, which was between 100 and 1,000 times more radioactive than Byrnes’ or McKinley’s, depending on which part of the body was measured. In the end, Lushbaugh found the only way to reduce the radioactivity of the bodies was to remove the most radioactive parts. They were placed in drums and buried in the Idaho desert as radioactive waste.
Efforts by physicists and engineers to reconstruct what happened during the accident were even more problematic. The reactor was destroyed. The scene of the accident was lethally radioactive. The 60-second limit for stays inside the building was hardly conducive to methodical investigation. Worst of all, the only three eyewitnesses were dead. Despite the challenges, the investigators soon identified two important facts about the explosion that approached scientific certainty: First, the mechanism of the explosion was nuclear. Second, the nuclear surge that caused the explosion was itself caused by raising the central control rod.
That the explosion was nuclear was not a foregone conclusion. While radiation alarms sounded as far as a mile away, a non-nuclear explosion could have spread radioactive material from inside the reactor—the same kind of havoc that would be wreaked by a so-called dirty bomb. There was speculation early on that the explosion might have been caused by a chemical reaction between aluminum and boiling water. SL-1 was a boiling water reactor, which the army preferred over the navy’s pressurized water reactor. Boiling water reactors created radioactive steam inside the core and were inherently less stable than pressurized water reactors, which Admiral Rickover had designed to produce nonradioactive steam in a more controlled environment. Boiling water reactors like SL-1 were compact and used fewer rods—vital qualities for a portable power plant—and the army believed the safety tradeoff was worth the risk.
Finally, some wondered if the accident had been an act of sabotage. In those paranoid Cold War days, many believed Soviet spies lurked around every corner. Could someone have planted an explosive charge under the central control rod? The Atomic Energy Commission enlisted the services of the prestigious Poulter Laboratories of the Stanford Research Institute, the country’s largest private explosives-research facility, but the lab found no evidence of conventional explosives.
It took some nifty nuclear detective work to prove definitively that an episode of supercriticality had occurred in the reactor. Investigators needed to identify isotopes with half-lives short enough to make them nonexistent in nature, but at the same time long enough for measurable quantities to be present days after the explosion. Interestingly, each of the three victims made a personal contribution to this part of the investigation.
From Richard McKinley came a Zippo lighter with a tiny brass screw that held the flint in place. Brass is an alloy made from copper and zinc. Copper as it occurs in nature is more than two-thirds the stable isotope Cu-63: a nucleus of 29 protons and 34 neutrons. In the presence of a neutron flux, however, some of the copper absorbs one neutron, becoming the rare Cu-64. A portion of the screw had become Cu-64, an event that could occur only in the neutron field of a critical reactor. Analysis of the brass buckle from John Byrnes’ watchstrap yielded the same results. Richard Legg made perhaps the most poignant contribution: His gold wedding ring was blisteringly radioactive, and analysis proved conclusively that Legg had died in the presence of a supercritical reactor.
Armed with this knowledge, the investigators needed to figure out the cause. The Atomic Energy Commission determined that slowly withdrawing the central rod to a height of 16.7 inches would have made the reactor critical, perhaps emitting lethal amounts of radiation but not causing an explosion. The height necessary to cause the kind of destruction seen at SL-1 was 20 inches. In fact, the distance may have been even less because of the deteriorating boron strips inside the core. In addition to fouling the control rods, each atom of boron that dropped to the bottom of the core moved the reactor that much closer to criticality. Everyone who knew anything about the SL-1 reactor recognized it as a serious safety issue.
Atomic Energy Commission special investigator Leo Miazga had been detailed to look into the rumors swirling around Byrnes and Legg. The rumors usually vaguely implied that one of the men was having an affair with the other’s wife, without specifying who was cheating on whom. Miazga discovered not one iota of evidence to support the stories, but he uncovered something even more troubling: No one that Miazga interviewed—not even the most experienced men in the SL-1 program—thought that yanking the central control rod straight up would create the kind of havoc that it did. The men all affirmed that they had been trained never to withdraw any rod more than 4 inches. Just as universally, they said that no one ever told them why that requirement existed, or what the consequences of violating it might be. Miazga asked to see the examination questions given to trainees about the handling of control rods; he was told that all examinations had been destroyed. It’s plausible that John Byrnes, like the rest of his colleagues at SL-1, was unaware of what could happen.
The Atomic Energy Commission declined to fix specific blame for the accident, but its final report, published in June 1961, made it clear that warning signs had existed: “The reactor core and the reactor control system had deteriorated to such an extent that a prudent operator would not have allowed operation of the reactor to continue without a thorough analysis and review, and subsequent appropriate corrective action.”
The fact remains that the army and Argonne National Laboratory, which designed and built the reactor, had constructed a core that could go critical with the motion of a single control rod. Most reactor designs adhered to the “one stuck rod criteria,” which held that no single control rod, even if fully withdrawn from the core, could push the reactor to criticality. Maintenance at SL-1 required a man to stand atop the core and pull that rod out manually, with something less than 16 inches being the difference between standard operating procedure and catastrophe. The crew charged with this dangerous maintenance on January 3 consisted of three young enlisted men whose sum total of experience with the reactor could be measured in months.
The majority of the commercial nuclear power plants in America today use pressurized water reactors evolved from designs pioneered by the navy. That is a direct consequence of the explosion that destroyed the experimental SL-1 reactor and with it the army’s dream of portable nuclear power. The demise of both the army and air force nuclear programs ensured that navy standards would become industry standards. Utilities with multiple nuclear plants refer to them as “fleets,” and many top executives in the industry have a ship’s plaque and officer’s ribbons displayed somewhere in their offices.
Many people mistakenly assume that the partial meltdown of a reactor at Pennsylvania’s Three Mile Island plant in 1979 ended nuclear energy development in America. In fact, exactly half the nation’s 104 licensed plants began operation after Three Mile Island, and the amount of the nation’s electricity supplied by nuclear power has increased from 12.5 to 19.4 percent in that same period. And the industry is on the verge of a renaissance, with preliminary plans for 32 new reactors in the works.
The philosophy that permeates the industry was epitomized by Admiral Hyman Rickover, who helped build a Cold War nuclear submarine fleet renowned for its safety, and often spoke publicly about nuclear power as a necessary evil. “The whole reactor game,” he said in 1958, “hangs on a much more slender thread than most people are aware. There are a lot of things that can go wrong and it requires eternal vigilance.”
Todd Tucker served as an officer with the U.S. Navy’s nuclear submarine force and is the author of Notre Dame vs. the Klan (2004) and The Great Starvation Experiment (2006).
Originally published in the April 2009 issue of American History. To subscribe, click here.
